BIOLOGICAL BULLETIN

OK THK

noannc Biological laboratory

WOODS II' '1 1 . MAS

Stntt

1 ' . ' , KI IN /'' . //r.

[xcqi/BS LOKB- /'//' A' / ' !/' .- A 1 -

T. I I . Mo| , A \ ' It '>:'' : i :(Y.

\V. M. U'll I ' .')'.

E. B. WILSON ' tmbia U .'r.

. !i^ltol
\\K K. I. ii. i. ii The University '<?o.

\". >LUM1 XX II.

\V. >DS I I'M 1 , MA-
JIM 1" N 'VI MI'.KU. 1912

PRESS OF

THE NEW En* PR: 1 . ::,:; cc"f*'.
LANCASTER. PA

CONTENTS OF VOLUME XXII

\". 1. 1 > l.i K \II4KK. I'll I.

PA<

SlIH.H'Kli. Yl I"K | 1

Cini D, C. M.. M Kn . I . V. M. 7 tral .\>

in I '...'"/>/;///<//'< mar in

I 1 ' ;<;

\\ ipoDki ||. l.i i|. \NM I ISS

I (I *

NO. 2 . I \ M \ K \ . [912.

1 1 \M .1 I I . ( II \-. \\ .

. !
HlGLEY, K' -l M . I ll \ I II. I I \KH| I'

l\ \\H \l I . |i ">l I HIM I ll \ I II. I I \Kiil I' . .1 A

RIDDLE, OSCAR. /I i

'"7 S~

\ i . ; . I RY, 1 91

McCLENDON, J. 1

i l >
Kl I'M K. \\ \l. \. / ':

''initory A pfxirdtn
I'M M K-I 'N.I 1 in .\i ^s. 1 Hi-

/) ( ;r ( ; .1 >'../" PI 17.1

NO. 4- M \K< II. I'M

\Virn M N . I > 1 '. ' // 11 \ilntiiiti

STEVENS, N. M. .TV c'// ///

219
STEVENS, N M . ''">-

r

|IM ERNEST! / Relation of the /'/ /'/.);' /.- ///

kntr<nii<: Pot " - "')

iii

iv CONTENTS Ol \ <>l.i \l I XXII.

\\'(ti>-Ki>A!.KK. J. I-".. Palmen's Organ and its I-' unction in Xymphs
tin- I-.plicmcridn. lleptagenia interpiinctata (Say) and f-'.cdy-
tirns maculipennis ( Walsh) ..............................

\n. 5. APRIL, nil 2.

KING, HELEN DEAN. 'I' lie . Some Amido-acids on the De-

tneniofth<- l-'.^"* <>f Arlwciu mid f ('In.'toptfnis ......... 273

Oi \\II:KI<^. KdiiKiM. JK. .1 JH.tcussion of Cyclops viridis Jurine 2<)i ^

I'l \KI.. K.\v\iiiM>. \otft OH the History of Burred Breeds of

I'l'nltry .................................... -''7

MON i n i \II-.KV. T. II., J k. Complete niselnir^e of Mi lot houdria from

the Sf)en)into:oini of I'eri ptititx ........................... .>")

\o. 6. M.\\ . i<)i2.

l\i INKI , l ; .n\\i\ I-'.. .1 rrelimituiry Account of the I'erelopmcnt
oftlieApyrcnc Spermatozoa in Strombus and of the Nurse-cells

in Litloriiin .......................................... ,U ' i

I. il. 1. 1 1-:. K.M.i-n S. Certain Mean.: hy ichieh Starfish /v^.v .\atn-

rally Resistant to l-'ertilization m.iy l>c Rendered \ormal and the
I'liysiological Conditions of this Action ................... 328 N

I'onrtcentli Annual Report of the Marine Biological Laboratory. . . ,U7

Vol. XXII. December, ryii. No.

BIOLOGICAL BULLETIN

KCOLOGICAL SUCCESSION.

III. A ki < \\ \i~- \\CE oi ii- CAUSES i\ POM*- \\ini
I'AK i ic n.AR Ri-.i i KI \< i ro FlSH.

Vic TOR E. SHK! I < <\<\>

I. IN imuji < 1 1
II I'k-i-i -.1 ' H \NU CONTENT OK !'

i . /' r . . i

.;ihy

tin- Bottom
I ' "lit . .

2.1 <lt ........

i(>arison

i _
ion u

..i ... i ^

u . . i i

KI

III I in ;MI if,

i. v ,-nt . ir,

- I ' >nn fill i s
!

I >. |.ili
''

1 -:\ ^. -'.ill' \,j

1-;

i

: i .

<lity
..unity

Relative Ii :

h

ties ol tlii- I- i-li

l\ < .1 M-.K \I. I M-. USSH 28

\ Si MMAKV
\l \i'KNi'\\ I Mii.MI M- \NH UlUMiM.K AI'IIY . 35

I

2 V.CI' >K I . -Ill I.F<KI>.

I. IXTRODl ' I K >N.

In tin- preceding paper we presented rrrt.iin facts concerning
ponds, together with a statement of suree^ion in tin- |>ond- ,u
the head of Lake Michigan, without entering into its c.uix -.
Succession in ponds is due to many causes. It is only under tin-
most favorable conditions that we can separate these causes one
from another without long and careful investigation. The first
attempts of ecologists in this line were considerations of tin-
obvious general facts, such as the accumulation of organic detritus
and the increased denseness of vegetation. We can give hen .1
hint at the more specific changes in the ponds and the relations
of these to fish. The subject is one for cooperative, research.
At present some of the workers and the necessary funds are not
available, and the ponds are being destroyed rapidly. It is
therefore improbable that the study can be carried further. Thi-
paper deals with the results of a preliminary investigation of the
ponds for the purpose of learning something of the causes of
distribution and succession of fish and other organisms in ponds.

II. THK PRESENT CHARACTER AND CONTENT or-~ THE PONDS.

The ponds with which we are concerned are shown on the maj>,
p. 1 31, of the preceding paper of this series. 1 This map i- een-
tial to the understanding of the data of the present paper. Tin
ponds here considered are an ecological age series, ecological
age being determined by a study of amount of sand bottom,
humus, etc., as shown in Table I. below. The physiographic
history of the region is in full accord with the facts used in decid-
ing age though in this case physiographic hi.-iorv is not essen-
tial to the decision. The pond designated as I is ecologically
youngest, 14 the oldest, and the others intermediate. The
measurements, analyses, and quantitative study were carried
out on Pond i. ^c (west section), 70, and 146 of the map. Some
qualitative records from the other parts of pond 5r, from 5/<, and
\4d, are included with those of the ponds in which the oiln-r
work has been done.

See "Ecological Succession," II. BIOL. Hei.i,., Aug., 1911. pp. 1.27-151.
Tliese are errors in the pond numbers of this paper which shouM I" corrected.
Page 1.52, line 13, for "56" read 58.
Pane 133, Table I., last line, last column, for " 15 " read

ECOLOGICAL -ITCESSION. 3

A- \ve have already -idled, the ponds which ha\'c been -tudied
e-pecially, are part- of the long sloughs which have been long
enough i-olated to ~ho\\ ihcir efficiency in -upporiing the fi-he-
which tlu-v now contain. The ti-hes found in tin- -eparated
pond- are-ho\\n in Table XXI. ' p. 171 of the pre-ent paper.

With the kno\\n habits of fre-h-water fi-he- as a ^uide. the
pond- h.i\e lu-en roughly mea-ured and area determined, depth
and angle of til- -lope of sides mea-ured. the character of bottom
determined, -ketr'n.-d, and the area- of the different kind- e-g-
niated on the ba-i- of the sketch--, and the di olved -olid- and
lie \\aier- have been determined by chemi-t-. The
plant and animal content of the pond- ha- been anab /ed <|tialila-

ti\el\ and e-timated quantitatively.

Tlie-e re-nlt- \\ill be |resenled under the main head- of. i
l'h\ -ii-al < har.t. Biological Content.

1. I'liy^iull ('h>: -(a) Topo-r.iphy. The chief topo-

phii features are -ho\\n in Table I.

TAHLK 1.

-IK .\\ i DEPTH AND ^

,^c
Depth.

I

0.3

0.5

Mix 1.

0-S

h

[0

\ cry lr

.ill

The figure- repre-enting depth ol water are the re-nlt- of
me.i-nn-meiit . \\ith e-timatioii in th- -\ averages. \ T

re-nlt- o| r..ii-li mea-iiring by ] lonntin- rail- in |iarallel

railroad-, etc. \\liile p^itioii- -tnditd dit'ler in -i/e, the\ pi. -i nt
Considerable nnifo|-mit\ of other feature-.

I llal.H 'lei' .-I Mottoill. The bottom i- ( onipn-rd of pure

sand, or sand more or less mixed with or deeply covered by humus.

The -and or tin- In mm- ha- a con -id era Me mi\i nre of marl at -ome
point- in the \oim^er pond-. Vegetation iuarl\ al\\a\- covers a

pure humus bottom. (.'lm><i -m<l bnlrn-1^ netimes gro^ on

-and and marl bottom-, but in -uch i ases the\ are -i altered and
in the table -uch area- are included \\ith bare sand, because -uch
sparse vegetation does nol intei tfre \\ ith the breeding of tish.

VICTOR E. SHELFORD.
TABLE 1 1 .

SHOWING K.INIIS ANI> AKKA* 01- H..IIMMV

Pond

Area Sand in Square
Meters.

\ > Uiunnv in Square
M. ;

1 >> pill Hum
in Cm.

i

1^
14*

I.OOO
50

Very little.
1 None.

-'.50O
3.450

Nearly all.
All.

--> .111.
2O. O i Hi
21.0 rm.

24.11 . in.

It will be noticed that the area of sand is much less in tin- older
ponds and the area of humus much greater, due to accumulation
of the latter from the decay of vegetation. The depth of humus
does not increase proportionately with age because it becomes
more compact with time. With the exception of the first pond,
the average depth of humu> \va> obtained by dividing the average
depth at the center by two. In the case of pond i, then- are
large areas with only two centimeters of humus and two deep
places which contain humus of considerable depth ; we give only
an estimate.

(c) The Dissolved Content of the Water. For a preliminary
study of the dissolved solids of the water we have had a single
analysis of the solids and four analyses of the ga^-^ made by
chemists, (i) Solids. The small value of single analyses of solids
is well known; sanitary analysts have pointed out the danger,^
arising from conclusions drawn from so little data. Ho\\ever, in
this particular case, the value of the roiilts is greater than in tin-
case of single analyses of drinking water, because of the following
conditions in and about the ponds.

(a) The ponds are without outlet and have no >t i vam> cmpi yin.^
into them.

(b) During rain they have little inwash because nearly all water
must filter into them through sand; in case rain falls in such tor-
rents as to actually run in from the >ides tin- area of drainage is
small, being a strip not more than fifteen meters wide on either
side of each pond. The ponds than are comparable to balanced
aquaria and any variation of dissolved solids muM be due in tin-
main to the effect of organisms, of evaporation, and of renewal
from rain.

It -hould be noted that these analyses were made at i he end <>\

ECOLOGICAL SUCCESSION.

ihc lr\- -t.-uson and just at the close of tin- probable plankton
maximum for the year.

TAHLI-: III.

I.TS IN Si|.i [ I' is' IN I'AKI- I'l.k M 11.1. ION.

Analysis b] Mariner and Ho<kii me.. III. i. Tlu- o.Iln timi- \\vrr nuuk-

mi i Wolf Lake contain^ all tin- -pi-. i-- ,>i ti-h ..t all the

ami i- a'Mi-'l tut ( i iinp.nison.

! ml.

i

'A

-iutn . .ir 1" iii.it>- ....

-

I I I.O

77-J

96.8

< ;ili iutn i ;irl" iii.it'-

CA.2

27.0

I 14-1

'47

.in - ulpliati- ....

1 74-3

. inn ( lili it i'l'- .

i i .4

^' ill nun -ulpl. '

4vO

*M .ilium . lili >i ji !< ....

10. 1

8 i.o

1 1.3

Jri MI

3.0

3-0

'

1 ,il -..lifl- in. Imlinn tin.-.- wi\-n in

', 1

420.0

Tin- tal.lt- -h<>\\- no utui-al <|iialitati\e features. Thi-rc i- a
iiotaMi- di in total solids in the older pond-. 1 In- ma\ U-

dm- to tin- lixiim ot tin- solids by organi-m-

j Gases Ph< gasanalysesweremade with two facts in view:

Water ma\ I H- al .normal in gas content . -i > a- to make it impos-
sible La ti-h t,. live Marsh, 'oH; Juda\ and Wagner, '08; Birge
and Iuda\.'ii Phe eggs of all food fishes known to be in

the pond- piol'.il.lv rest oil the bottom ..t \e-etation during
inrnl.ation. Nnnr ti-h remove the \egi-tatioii I mm tin- l.otii.m:
otln-r- di po-it eggs on bare bottom; a te\\ ma\ attach --L;- to
\ . ^rtation.

I- determine the general suitably of tin- pond- for tin 1
place- of li-h. t\\o determination- of the gas content of the
\\ater- \\ere made.

TAULI IV.

t >\yi;'-ii in cul.i. . 'iitiiuf -t.-rs per liter; collivti'in- i" i J .in. ln-li>\v tin- -in;
..! tin- fp.-n \\ a)

| 'll\ ' ' 1 '; 1 i >

7-44

April 26, i'ji i

.. 6.-

7 -

6.96

6

A\ erage

,

S

7.20

'

Thr tal.K- -ho\\ - an oxygen roiitrnt in all the pond-. Hiltirirnt
-ii|)port an> ot tin- tislu--.

6 VICTOR E. SHELFORD.

With reference to fish breeding plau->, the gas content of the
water was determined on tour occasions. To make collect ion^ in
ponds 5r over a sandy bottom required taking advantage of the
sand areas made by artificial filling. Since there is little ban-
sand in ponds ~ja and I4/; the collections were made over the
vegetation.

TABLK V.

Oxygen content in cubic centimeters per liter. Depth 35-40 centimeters.
Sample collected at the bottom or among the upper two inches of branches of
aquatic plants as indicated.

Over Sandy Bottom.

Over Vegetation. B""om Materials 1 listurbed,
| Vegetation K n >iinnner

Date.

6/27

7 '22

4/26

E/io

Aver-
age.

6.27

7.22

V V t 1

6/27

4/26

1

5/10

Aver-
age.

I

Sc

l<i
146

6-57
7.36

6.37
8.18

5-Qi
7-31

6.32
6.60

6.28
7-36

4.42
3-33

2.52
2.24

3.34
6.52

3.47 o.oo
2.78 i o.oo

5-91

6.74J
o.oo 6.26
o.oo 1.38

3.36

1-93

4.62
6-63
2.40
0.83

The table shows high oxygen content over sandy bottom,
oxygen probably sufficient for most fish over vegetation, and no
oxygen on the bottom where vegetation was removed in July and
August.

TABLE VI.

SHOWING CO; CONTENT OF WATER.

Depth 35-40 cm. (i) Over sandy bottom. (2) Over vegetation. (3) Over
bottom with vegetation removed.

>i

i

3

I

3

i

3

June 27,

1910.

o.o

i

I

o o

O.O

1 1

i.o

i. !

6.6

Julv 22,

IQIO

It, II

o

0.0

O.O

2-S

2 1

i

Average

0.0

2

o

0.0

o.o

3-4

i.S-.-i

J.7

.v.5

The water was alkaline at points showing no ("()._.. The tables
do not show the uniformity that might be expected. This ma\
be explainable on the basis of the place of collection. (.'/Kirn,
for example, which was the plant removed from the bottom ol
ponds i. and $c, grows on bottoms of mixed sand and humus or on
a bottom covered \\iili humus, and sufficient care \\a- not taken
in selecting places of collection, to make the dillcicncc^ heir ol
importance.

ECOLOGICAL SUCCESSION. . 7

There are apparently numerous factors which influence gas
content Birge and Juduy, '11, p. 54;. These are temperature,
light as affecting photosynthesis, distance between point of col-
lection and plants which are giving off o\\ m-n and u-in- CO*,
and dirt-ction and velocity of wind a- affecting circulation of
water Hir^e and Juday (p. 55) state thai Kun.pean Corker-
have noted marked diurnal changes in the amount of dissolved
o\\ vjen.

I- nun three io lour hours were required to make our collections.
( )n Jinn- 2~ . ihr collecting began at pond 14'' ai 9:30 A.M., and
ended ai pond 5* at 12:30 P.M., temperature: pond I, Jn (.'.;
v and ~n. 2~ : and I4/;, 25. Velocity of \\ind <> mile- per liour.
Thi- was a < I'.ndv <lay. The sun broke through the dm id- bet on
ihe!a-i i olle< i i"ii ua- made. July 22 was a similar day. Col-
lei tin n Li van at |ioild l^l> at 8:30 A.M.. and ended at pond ~n
at i-'"" M . temperature: pond I, 25 ('.; 5. and -<i. Ji> ('..
and 14'', -V} . Mining the forenoon, the Mm came out -e\vral
time^. I Hi i e\ai i n i ord was not kept of the time i>r len-ih of sinh

|ieri..i|- n| MIIIM'lille. All t llC Other ( ol li i I ji .11 - \\ el'e made ill full

Minlight . \\ ind and tem|KTature Were a- !olln\\-: April Jo, tem-
perature i, !;-,'_ ; 51, i4 l ->', 7a, 15' _ : -md 14''. 14"; \\ind:
3 miles per hour. May io, temperature, 23 . \\iml: )6 miles per
hour. Ju^t \\hat elTect distance from plain- \\hirh \\ere d.iiiu
photosynthetic \M>rk has on gas conieni i- not kno\\n. It i>
highlv jn-obable ih.it colli-ctions made near to Mich plant- \\ould
lillereni Iroiu those taken at a greater di-laiue 1-Jir-e and
|uda\ , ' I I , |ip. 54 and 60).

The MI miner col led ions from pond ;./ \\eie taken from beiieaih

the \\.iiet lil\ leaves at the extreme east end \\ here lilie> h i\
di-placed i In- ( 'liiirn.

Collections taken after scraping the \ <- elation fnun the bottom
sho\\ \ a rim i- reMilts depending upon the character oi the bottom

licile. illl the \ e-elation.

2. /}/(/(-/! : : '<i<l. (a) Qualitative Comparison.

'1 Specie- of Plants and their Abundance. The i|iialitali\e
difference- in ]>oiid- a- -ho\\ n in Table- VII. and \ 111.

(2) (imuih li'rm of the Plain-. Pmid I i- dominated b\-
plain-. There are no bn-ad-leax e<l -hade prodm

8

VICTOR E. SHELFOKD.

TABLE VII.

SHOWING THE PLANTS OF THE CENTERS OF THE PONDS.
Data by Mr. G. D. Fuller. D =dominant; A = abundant; C = common; F =few.

Name.

Scientific Name.

1't.ixl Numbers.

7"

Nitflld batrachosperma

C

Stoncwort

C-hcirn sp. i

D

D

(

Pot&mo^cton luccns L .

C

C

F

"Ml 'III ll 'I IKlUl^.

Naiasflexilis Rostk. & Schmidt.

C

D

F

F

1 *i <\\( Iwi'ed

Potamogeton peclinatus L

A

?

C

F

Filamentous green

alt?;p

F

F

F

A

Ceratophyllum demersum L. . . .

A

F

>

N\ l mph(Ea (idvcnd Ait

F

C

F

Potatnogeton americanus C. & S.

C

F

F

\fvriophvllum spicatiim L

F

Utricularia vulgaris L.

C

A

Castalia tuherosa Greene

A

A

Water shield

A

C

C

F

Duckweed

Lemna minor L

C

TABLE VIII.

SHOWING THE MARGINAL PLANTS.

Data by Mr. G. D. Fuller.
Roots Usually Submerged.

Common Name.

Scientific Name.

Poinl Nuiiil'crs.

i

5<~

F

70

, 4 /'

Bulrush

Scirpns validus Vahl

F

F
F

C
C
C

Cattail

Typha latifolia L

Mermaid weed ....

Proserpinaca palustris L

,,.

Roots Submerged at High Water.

Sedges

C

F

Pines

Pinus Banksiana I>amb

C

C

F

Shrubs (other than

F

F

Button bush

Ccphaldnthus occidcntdlis I.. . .

F

C

Sii{ i v spn

F

C

F

Pond 5c shows the beginning of shade producers such as the
water lily and of plants which reach above the surface of the
water. Pond "ja has a large number of emeri;iii^ plants. In one
end of this pond there are many more of these than in the oilier.
In pond 14^ emergents are dominant.

(3) Animals. The different >pccic> of animals and their ar-

ECOLOGICAL SUCCESSION.

TABLE IX.

LEECHES.

'.

Glos\iph<>nia fusca Castle

puniliit-i Leidy *

ft-r:-i'l'i Y'-rrill . . *

\I-:'rli,l,-llii -ay ...

//.- ' nill

I'lijiui'ili-llii para -i Say
'nlfllii rn^ ii \Vrrill

< ifill<ll : :ta

//.. n

irs and letters see p. 1 1 .

TABLI- X

SPH^ERID.C ASP

S

M

i NIONID/1

/ afn 1 am

i y

loiltil H:

. : :: / :tia Lea. . .
H'll 1 KIN V

.\f n m Lins

M .. 'i line

\l u :i. '.-.. '

TAHLE \ I

ULS.

ne.

i

' 1

*

* *

t

...

*

* *

*

if

*

i-h\

F

1 '

(

( '*)

1 y IMM.I i.l.r :

1

1-

.

* *

* *

* *

*

*

* *

*

1

i '

A

\

*

*

(

A

"it'ttthtti tjrw/A,vrti Say

*

1C)

\ h ink E. Sill i r< IRI).

TABLE XII.

Name.

1' .ml Nuinlii-t-.

i 5''

7"

.

.t

11 valellti KnickcTbockeri Bate

c

C
F

i

C

c

A

1-
A
*
*
C

?

ii H ^< >n \ v fffddlis Smith. ...

\fdncdsfllus ddnitlsi Rich . .

Asellus communis Sav

Cambarus itntnunis Hagcn

F

F

c

C

F

Cambarus blandiniti acutos Girard..

TABLE XIII.

AQUATIC INSECT LARVAE AND NYMPHS.

Name.

Pond Numbers.

i

S<

7"

I4 /-

3

May flies:
Ccenis sp

*
*

*
*

*

*

Siphlurus sp. .

*
*

*

Callibcetis sp

Neuroptera :
Chauliodes rasticornis Ram.

*

*
*

*

ifc

*

Damsel flies:
Lcstes sp

l".nallagma sp

*

*

*
*

Ischnura verticalis Say. . .

*

Dragon flies:
Tratnea laccrata Hagen . .

*
*

*

*
*

Celithctnis fponinn Drur\*

Libellula pullcella Drurv

*
*
*
*
*

Gomphus spicatus Selvs.

*
*

*

Leucorhiniu inlacta Hagen

*
*

.4 nax junius Drury

*

Sympcttuni rubii'unduliim Sav

SympelTum sp. ...

*

'

*

Caddice worms:
Gcora sp

*
C

LcplocerincB sp

F

Neuronio sp. ... ...

*
*

j

C

*
*
*
*

\

*
*

Diptcra larva?:
Chironomid larv;c

*
*

*

Stratiomyid larva?

Tanvpus sp

Tipulid larva? . . .

*

Cerutopogon sp.

Hemiptera:
Ranalra kirkuldvi Bucn

*

*
*

*
*
*
*

*

*

*

*

F

Corixa sp

Kanalra fusca P.B

*

i

*
*

*

Belostoma Jlntninca Say. . . ...

Notonefta undulnta Sav.

Jiuenoa platycnemis Firb

\\'ator striders:
Gcrris rufoscuffllalus Lat

*
*
*

*
*

Mesovelia l>i.\ienata I "hi..

ECOLOGICAL SUCCESSION. II

TABLE XIV.

HIGHER VERTEBRATES.
The fish are shown in Table XXI. page 17.

; Numbers.
Name.

if

*
*

1 / " i iita Lat *

Rini,i f't i

nnla Ag. ... * *

sogtaphica Les *

/';. u;/. J.V Raf *

Hill ' .' ' ' ?

I h> : 'lie muskrat is indicate*! !>y i; : h>li-. ii<'xi~. tt.uk-.

I iiit ii'iiu- li.t\ Ijcen seen except in tin- "M-t p"i;

with resjx'ct to tin- ages "I tin- pond- are -ho\\ n in
Table- IN. ! NI\'. Letters indicate ivl.tii\r abundance: l :
few; ( ommuii ; A - ahund.iin. The -i.tr i- n-ed t<> indi-
< .IN piv-i in \\ here relative abundance lia^ not been .iMvri.iitird.
l-i comparison, a fifth j)oiul (No. 30 U .idilrd ; tlii-- i- nldi-r ih.m
ill. others in rvery resfK'ct and coni.iin^ certain species <>' ini-
["'M.iiirr i<> fi-li which are not found in .ui\ o! tin- ntlu-r--.

I I >i-i ii imi of the Tables. The i.il-lr^ rcjur-cin nut <nl\
nun Ii ..iidul ulleftinjj, but long e\]H-riniv \\itli the "inmuii
forms of the ponds. An inspect inn <>i tin- t.il>K-- -lu>\\ - ih.it tlu-rr
are <lilt< n u. es in the species in tin- dilh-n-in |)umU .ind ili.it tin-
dilN-n-in -, .tre correlated with tin- ages <>| tin- pund-. l-'m- ex-
ample, in the case of the let lie-. T.iMe IN., page 9, nmie <>t
tlie ~|iet ie- ul the youngest pund i- !i>iin<l in .ill <>t the pmnU and
iiuiic u|' tin- -pccies uf the ul<le-t i-> I'unnd in (he youngest. \-

iillvK .1- \\e pass from the \ulll)^e-! to the oldest \\e lluli-

tll. II 5] di-.ippear and are replaced l.y uther species. The

same \\ill be seen to be true ul' the uther -ruiip-. A -imilar
relation i-> illustrated also wlu-re \\ e have been able to estimate
relatixi ibnndance. In -\\\^ cases the number i- greater in
the older ponds; in other.-, le in the older jiuiid- e. . 1 1 yullelti
Kni(kt-r!>txkfri, Table XII.. p ;.< [o).

The ( -a -i -of the caddice \\urm> an<l uther ai|iiaii< insects which
are plat ( d in the water by the laying tfmale, i- u!" e-pecial inte:
as the iv-nllin^ distribution is jirobabK either a matter of

12 VICTOR E. SHELFORD.

tion on the part of the female during tin- breeding reason or
striking elimination of all eggs laid in tin- ponds in which the
larvae are not found.

It is evident that ecological types (here represented by the
various species) succeed each other as the ponds change with age.
Succession is here as elsewhere, a succession of all, or at least a
majority of the animals present.

(b) Quantitative Comparison. (i) Vegetation. Vegetation
is evidently a good index of the content, or the relative numbers
of the different species of plants and animals. In Table I., page
3, we note that more than two thirds of the bottom of pond I
is covered with humus. Vegetation covers about 70 per cent,
of the area. In pond 5r vegetation covers about 95 per cent,
and in 7 a about 99 per cent, of the area and in 146 100 per cent.
If the plants of each unit area were equal in volume, these per-
centages would represent relative volume also. More of the
plants of the older ponds reach to the surface; plants are closer
together in the older ponds. It is obvious from inspection that
the volume per unit area is greater in the older ponds.

A single test was made with a large tow net. The net was
drawn a distance of 40 feet in three of the ponds and the volume
of vegetation torn off by the net was measured by displacement
and reduced to terms of 100. This would give relative volume
if all plants were torn with equal ease.

Finally Mr. G. D. Fuller and myself have made an estimate
based on several inspections.

TABLE XV.

SHOWING MEASUREMENTS AND ESTIMATES OF RELATIVE VM \n OF \ i DICTA-
TION PER CUBIC UNIT.

fond.

1

7"

'4''

On the basis of areas of vegetation

70

9S

99

IOO

Tow net collections. ... ....

14 c.c.

30 c.c.

IOO C.C.

Estimate

!<)

(0

60

I OO

(2) Plant and Animal Food. The plant and animal tood in
solution is expressed in a general way by the sanitary analysis.
The results of a single analysis with the total carbonaies added,
arc given in Table XVI.

ECOLOf.ITAL SUCCESSION. I .}

TABLT-: XV I.

SHOWING CONTAMINATION OF POND sc AND ELEMENTARY FOOD SUBSTANCES ANP

< \KIii>NAII-> IN Al.L.

Single analysis, Oct. 26. 1909.

'

\\

( liliirim-

JO 7

90

i j *

16 T.

siiiiiii iiiia

O I OO O 1 ~ ' '

Trace

o 005

All>iiiiiit)niil aiuinonia.

O I O I S"

O.I7C

O 2 S u

o oo

Nitrites.

1 ' 1 - lit It

I ' ICC

Nitrati--

o.i 60 o 030

o o^o

o 040

o 060

1 i.il . ,irl..i. '

138.800 139- s

160

if,,,

[11.500

'I In- chlorine content is regard I as a good index of the presence

or .ili-fin i- nl sewage contamination, f\cn-i.i lu-in- hijji in
clilnriiif i iiinpiiiiixls. 49.7 part- pi-r million in pond 5. \\uiild
such contamination. I "mil \i\ iv.vntlv a h<m-e was
mi ilu- margin of pond 5*: tin- pond i- -lill -ulifct i
< ..iii.iinin.iiion l.\ domestic fowls.

I ree .inniiniiia is tin- final sta^f in llx- l>ii.ikin- <l<>\\n \
pn.ii-iiU .uxl .i|i|M.-ars also in animal excreta. It is used by plants
.ind f \ iili-nlK plants consume it in |m i|>rti<m to tlu-ir \ ulun

\llniiix -imiil ainmontu prohabK n-pn -cut- mctalmluid- in -ulu-
li'Ui. because tin- \\atcr \\as filtered lHmf dfic nninatimi- were
m.xli . Sewage is rich in such compi.umU and -anii.ir\ analysts
ha\ i- I ui ml that the number of bacteria is closely correlated with
.mimim Hi albumenoid antmonia.

; I'.aiii-ria. 1 >i . I'. G. Hfiiifiiiann and Mr- Class, "I the
I >f ILII l nn nl nl Bacteriology of the l'ni\er^it\ o|' ("hie.i-o. \<i\
kindU made the counts of the bacteria. The re-nil- an- i;i\en

in fable X\ II.

XVII.

\l K, >l.|. H M IKRIA

I'KR cc.. CAI-AI - \ i

i

i

i >, tobei -'. 1909

April _>;. i<ji i

1. 2QO J.'iOO
1.1 SOO 4.

SO?

I In- minil'i-i In ir !- n'i "i i ---I ii mil tn tin- .ill'iiini-ii' '!! .umih mi.i. hut iua>
partially accounted for by the fact that tin- Imttli- \\ li-ntally n|ii-in-,l m-.u

llir -in i.i.-i- l in April J'i a oillrction at tin- -in I'.iri- nf tin- pii'l -lm\\i-i|
.it

14 VICTOR | . -Ill LI-UKD.

The table shows that the number of bacteria is greater in tin-
older ponds, except in 5* which is noncomparable because of
contamination.

(4) The Plankton. The study of the plankton has been prac-
tically limited to the Entomostraca the most important food
of young fishes. The presence of a larger number of rotifers
and protozoa, etc., is observable as we pass from the younger
to the older ponds.

The number of Entomostraca in approximately 90 liters of
surface water, to a depth of 10-12 decimeters, is given in the
table below. It was thought best to simply clip the desired
amount from the water while walking and strain the dippings
through a bolting cloth strainer. After the first collection this
was repeated in as uniform a manner as possible and Birge net
collections were made at the same time for comparison. There
was no great discrepancy in the results of the two methods of
collecting, except in the case of Ostracoda in pond i4/>. As com-
pared with dippings, some Birge net collections showed less
Ostracoda. Ostracoda w r ere probably started from the bottom
by the feet of the collector but were not by the drawing of the

Birge net.

TABLE XYIII.

THE NUMBER OF ENTOMOSTRACA IN 90 LITERS OF WATER.

i

5<--

7"

146

September 3, 1909

5S6

S ?9

2,77 !

November 13, 1009. ...

200

106

797

7 SO

March 26, 1910.

42

I 1 '

I 2

oo

Mav 31 1910

7,407

1,014

4,168

3,600

Julv 22, 1910

160

200

52O

''.480

April 26, 1911 . . . .

I.2SO

ISO

I4O

=> ^s

M av 101911.

IOO

800

I 2 :;

s. 1 25

Total of 6. ...

"5.249

2,310

S.v

16,080

Average of 6

874

385

927

180

The table shows that with the exception of pond 5^, which
is probably noncomparable because of contamination, the older
ponds contain most Entomostraca except in early spring when
conditions are somewhat reversed.

A large quantity of plankton in old ponds has been noted lor
sevcral years in connection with class work. For comparison
with the; ponds under consideration we have studied \Yoll Lake,

ECOLOGICAL SUCCESSION.

and two small ponds near it. The younger of the small ponds
will IK.- designated as I. and the older one, II. They differ with
the exception of the margin vegetation) in much tin- -aim- manner
as (In ponds I and ~a of tin- -fries of special -tudy. While Wolf
Lake i- in.i -trictly comparable to the other-, ii is ecologically the
youngest, I- -cause of its greater area of hare hoi torn. The collec-
tions made Sept. 3, 1909) were four j n numher in Wolt" Lake,
tour in pond I., two in pond II., one half irom the open water. , m d
one hall Irom among vegetation. S-\eral collection- \\ere made
Apr. The numbers given are the ,,\, of all collections

made on t ho-.- dates. They were net collections made in as uni-
form a manner as possible.

TAHI.E XIX

OWING DIFFERENCES IN NUMBERS OF ENTOMOSTRA< \>

\viin DIFFERENCES IN I

I.

II.

Septeml

.

>cera.

Total. . .

1 |KxJa.
1 'icera.

96

IS

I hi- table -hows the same feature- a- the pnvedii'
i) I'll* I irg< r Animals. Little ha- been d"iie in e
the relaii\e niimbfr or volume of the lai-er animal- in the dif-

i p.,nd-. A general idea is given below in Table XX. Phis

Nl'MBER OF TH! HI

COLLEl I l< '

ler.

i

( I.I' !

i
1

371

[Q

I. I

i. :

i.;
421

-

:- .lur nuiiiily tn -mall iiniiii pods.

16 VICTOR I. SHELFORD.

is based on the general impression which ha- been acquired in
taking classes to these and other ponds of similar character
several times per year during six years. Secondly, by taking
the time required to make a representative collection from the
different ponds. On the basis of this experience, the figures given
in the table are thought to be very conservative. That there is
a far greater number of animals and a greater volume of animal
substance in the old ponds is very easily demonstrated to any
one by inspection.

TABLE XX.

SHOWING AN ESTIMATE OF THE RELATIVE NUMBERS OF THE CHIEF ITEMS OF FISH

FOOD IN THE DIFFERENT PONDS.

i

$<'

7"

146

Entomostraca

T.2

I c

jc

IOO

Chironomid larva?

8o 7

80'

80^

IOO J

Sphaerida?.

o

^O

^O

I OO

Gilled snails

20

3Q

en

IOO

Pulmonate snails . . .

IO

-3Q

^O

IOO

Arnphipods

CQ

7O

i ,. i

IOO

Decapods

IO

7O

ZQ

IOO

Insects

40

60

oo

IOO

Fish. .

80

100

7O

^0

Previous to being drained pond 140 should be rated at 70 for fishes.

While the results here presented are not such as to justify
conclusions concerning details, we may state that the amount of
life per unit volume unquestionably increases as the ponds grow
older, at least up to stages like 146. Qualitative differences are
shown in the Tables VII. and XIV., and the total number of
species recorded in each pond is about the same, tin- actual
quantity is far greater in the older.

IM. THE CAUSES OF SUCCESSION 01 FISH.

A discussion of succession must be made 1 with reference to all
the organisms of the habitat, or at least a large number of them
considered in mass. Succession of one group of organisms t.ikin^
place without the succession of others in the ^une environment
seems improbable. A discussion with reference to tish UHIM
take other organisms into consideration.

i. Statement of the Problem. A rlrar understanding of the
problem at hand will perhaps be facilitated bv a careful Mate-

ECOLOGICAL SUCCESSK >N .

ment of the question before us, after which we shall di^eu tin-
available data with reference to the relations of fn-h to the dif-
ferent ponds, from the standpoint of their area, their depth,
mineral- and gases in solution and finally the available food for
youn.u and adults. Competition, living pla<v and breeding place
(A the li-h will be discussed as fully as data will permit.

TABLE XXI.

iM-lklhi IIOX OF THE FlSH AND THEIR R KI.A Tl> >N To Hull

I In- li-tt'-r- .iiul numbers at the heads of the column-; n-tVr tn the various isnlati-il
parts ol pon ; indicates the presence of the species; !' that vi-ry young

nii-n wi-ri- I'linul in numbers and the >; :<! in looc;, in <ir n. The

ii'iiiifii' l.itun- .tii'l bottom preference data a: : i.ud-on.

Name.

in.. ml.' ' ' TMS

'l.i . -ides. ... B

{ill fnlliJus B

Blue-spotted

-mi li-li .us H

Pumpkin iced '.-'tis

... H
\\.iiinouili I :$

B

V*

14*

.iiiil.

IH

iii'l

I

b

( In.

bullhead.

I ad

rel

Mini ininn...
i i. >l<l<-n -hincr.

\i-ll..\\ hllllh

HI.., k bullh.-.i-l

: escens .
i micella

us

Its

niiiitatiis

mi

:iCilS . . .

natalis
i . mrlas .

Mud.
r. i< . k an I sand. In ;

H B |...

In part, linn k

B B . . . Mini .in-1 -.in. I.
B B ... .
B B . . . Mud vAI>l...u I. Mu. k pi.

B B Mud.

* ...
B B "

Tin- pmbli-m of the causes of sun v ion max In- -tau-d in tuo
\\ a \ 3 :

liiM'Kin- interpretation: \\li\ are the piniieer ti-he- of
a (mud -ncrci .led as the pond ",r"\\- older, by ti>lle^ of dilteiciil

habits?

I in le pendent ol interpret a ti< m : \\ hy are tin- ti-he- ot pond
I. noi in the older ponds and the ti>he- ol the older |>oiid- not
in pond I .. u hen the channels between I hem ha\ e been o j ic' n until
the paM leu \

1 8 VICTOR E. SHKLFORD.

2. The Cause of Succession Environment. (a) Area of the
Ponds. A comparison of Table I., page 3, with Table XXI.,
page 17, and a comparison of Table I. of the preceding paper with
the map (p. 131 of the preceding paper) show that most of tin-
fishes are in ponds of all the available areas of the region, with
the exception of several species which are confined to pond I.,
and which, on account of their numbers, could find no advan-
tage in such close quarters. Evidently no part of the aiiM\er
lies in the matter of size.

(6) Depth of the Ponds. A comparison of the records of depths
given in Table II., page 4, with Table XXL, page 17, shows
a situation parallel to the one with reference to area. Species
are in ponds of various depths and are absent from ponds of
depths the same as and greater than the ones in which they are
found. These ponds are shallower than the waters which many
of the species commonly occupy. The matter of depth does not
seem to be of importance in the answer to the question.

(c) Minerals in Solution. The minerals in sqlution in t In-
different ponds on October 26, 1909, are given in Table 111.

(1) Qualitative Differences. The minerals represented in tin-
analysis are those normal to w r aters inhabited by fish and probably
important to fish. No zinc, lead, aluminum, silver, or coppi-r,
metals highly poisonous to fish (Marsh, '10), were found and there
is no reason to expect their presence at another time of the year. 1
From the qualitative standpoint there is no reason to assign
importance to minerals in solution.

(2) Quantitative Differences. The total solids given in Table
III., p. 5, lie between the two extremes given by Marsh, '10,
as probably not affecting fish and as "normal" for waters which
are known to support fish in numbers. He gives 484 parts per
million for the Potomac River and 242 for other fish waters.
Nor is a very great seasonal variation to be expected, because
most of the animals live through the winter and the vegetation
disintegrates very slowly, especially through the cold weather,

1 Because of the small amount of inwash, this sc-t of ponds afford an IIIIHMM]
opportunity for the study of the effect of a varying amount of vegetation n ilu-
chemical composition of the water. For a statement of the salts ti-.l up l>\ plant-
see Pfeffcr-Ewert, 'oo, page 410.

ECOLOGICAL SUCCESSION. 1|

in the spring it- place i- lakc-n by iu-\v vegetation a- rapidly
as the decomposition of the old takes place.

Prom our knowledge of the composition of river water in-
habited liy all the fish, before and after the Hood-. IK. -real
importance could be assigned to mineral-, even though the com-
plexion ot the analyses changed with the season. However, no
positive ' onclu-ion could be drawn without careful stndv of
the ln'hin'inr n-nctions of fish to minute quantitie- of -alt.

'/ Gas The results of gas determination are -i\en in
Table- IV.. \'., .md V|., pp. 5 and 6. Table- IV. and V. -ho\\ the
'"iitent ,f the open water, abo\ e the xe^etation and -andv
bottom, to be -ulticient for fish in all the pond-. Jndav and
liir^e. 'n. p. iv>. state: " Konig found that he could keep ti-h
'kind not -pei ilied i in water which contained j.o.S C.C. and i j8
ol di--o|\cd oxvgcn per liter without any apparent ill et!< . ts.
'I horner loimd that a fish epidemic \\a- i .m-ed by the ab-ence
of free oxygen. I loppe-Seyler and 1 >mx an -t.tte that trout \\ hich
\\ere kept from one and a half to t\\o and a quarter hour- in
\\aier ha\iiiL; <>nl\ Irom o.cjN to 1.71 c.< of o\\-eti per liter
shoued marke.l -uiis of dyspmra. r.itoii. in experiment- -MI
young i.iiubou in. i it, found that a fall in tin- amount of di >\\ e. 1
o\\-en be|o\\ one third of the normal amount, /'. e. t belo\\ j
per liter ol \\atei. is prejudicial and geiieiallx fatal. Some indi-
\idn.il- ho\\e\er, \\i-re able to sustain life l.-r lon^ period- in
\\atei \\hich (oiitained only minimal tr.ue- of <li--ol\ed oxygen.

" l\ nau i he I ou! id that carp kept for an hour and twenty minuie-
in \\ater \\hich i . .ntained 1.33 c.c. "i o\\-en per liter, did not
sh.\\ an\ -i^n- ol dyspild-a, \\hile other- became d\-pno-ic in
\\atei containinv; from 2 c.c. tt) ,V I C.C. of tin- gas."

I'.ii'^e and Juday state also that Mat kin, lu tn.ut ha\e been
taken tiom \\ater> with I C.C. per liter. Ti-h di-ea-e- .ti\ -aid
to be nion- pn-xaK-ni in low o\\-eii i oim-ni Knaiithe. '07 .
In thi- case I here is no reason for a--i^ninu importance to the
Oxygen content of the open \\ater- frei|iiented b\ ti-h. and this
factor i- nearly uniform in the different pond-. The oxyj
con t i'n t of the Lot torn i- of great import a IK e and u ill be di-cii-
laler in connection with breeding.

Temperature. A -in-le -el ..I reading- taken in the late

20 VICTOR E. SHELFORD.

afternoon of a warm sunny day showed less than i degree of
difference between the different ponds and the readings \\ere not
repeated.

(/) Excretory Materials in Solution. Dacknowski i'o6) (see
Cowles, 'li) found that certain unknown water soluble sub-
stances present in bog water are poisonous to plants. Colton
('08), and authors cited by him, found that the excretory prod-
ucts of animals are toxic to the producer, and sometimes to
other organisms. This is a physiological basis for succession.
Knauthe states that the effect of fish on their environments is
important, but little of definite character is known concerning it.

(g) Food. The food of the fishes from these ponds has not
been studied, but knowledge of the food habits of the same species
was acquired from the study of literature, especially the work of
Forbes and Hankinson. The species found in the ponds being
known, each pond was inspected with reference to the things
eaten by each fish species. Forbes gives the percentage which
each item constituted in the individuals which he studied.

(i) Qualitative. The method of obtaining the results con-
sisted in adding Forbes' percentages ['80, p. 38] for the different
items of food for each species found in each pond. For example,
take the food of lake specimens of the perch. These were found
to have eaten fish food existing in pond i as follows: decapod-
rated at 14 per cent.; unidentified fish, 50 per cent.; Acan-
thopteri, 8 per cent., giving a total of 72 per cent. Pond i
contains 72 per cent, of the food of lake perch; Cyprinidae rated
at 28 per cent, do not occur (see Table XXII). For the youngest
individuals (under one inch) of all the species, all the ponds are
qualitatively equal. Hankinson's data on Walnut Lake species
show that all our ponds arc about qualitatively equal tor the
fish which he considers.

An inspection of Table XXII, p. 21, shows that in no case an
the fish confined to the place where their food is qualitatively best,
in fact, as a rule, the fish are in the pond where the food is qualita-
tively poorest. The available data on the food of fishes shows that
the fish eat food available where they live, rather than that their
distribution is due to the presence or absence of certain foodspecies.
Excluding students of the food of animals, the idea that food
determines distribution is commonly, though erroneously, held.

TABLE XXII.

QUALITATIVE EXPRESSION VALUE IN FISH FOOD.

* indit Lin-s presence of the species being considered. The averages are not
avi-rag'-s <>f the figures given here, but of all Forbes' iu-ms taken <oparau-ly; their
number is given in the last column.

Pond Numhc

\

hems

-

ics.
IVr

M

ptt'ru. : : " 12 in. 98 IOO

IOO

IOO

2-4 in. 100 100

IOO

IOO

Adults.

Average.

1-3 in. 100 inn

ioq

Adults. 81

-

Adults. 60 60

80

lltS. QI

91

91

Average. *88

92

i in. 100

1-4 in. 96

96

too

ilts. 58 71

7i

Average. *88 <;i

14 in. 100

[ O.I

.Its. 8l

Average. *95 95

96

All. *ioo

I . ...

1-3 in. 100

[OO

3-4 in. 76

Adults. 72

Adults. 56 ']

Average. *83

All. *io.

ni'l mfliis. Various

young. 100

[OO

Adi

Average. *90 *./

Various

young. 100

[00

Adults.

Averag

-

i ' i in. IOO IOO

in.,

[00

A.I . 4"

A\ *;n *;o

*7i

71

/ limi. . . \ lult. *.*3 *33

*33

68

Ahr.i" Ailult. 86 *86

*86

natiilis Adult. '14

VICTOR E. SIIKLFORD.

(2) Quantity of Food. The quantity of food, like the quality,
is one of the reasons assigned for the distribution, migration, and
extinction of animals. Although my data on quantity of food
in the ponds is not as good as that on quality, a comparison
is presented in Table XX IV.

In the case of the young fishes, the table follows from a com-
parison of the tables of Forbes with our own on Entomostraca.
The quantity of food for the youngest individuals of all species is
practically that of the Entomostraca: Pond I, 32; pond 5<r, 75;
pond ja, jj; pond ib, wo. For the adults and young from
one inch to four inches in length, an estimate of the quantity of
food in each pond for each species has been made by averaging
the ratings of the principal articles of food given for each species
by Forbes.

TAULE XXIII.

METHOD RATING PONDS. Ameiurus natalis.

Diet According to Forbes.

Rating in Table XX.

S<

7"

14*

Insects, 30 pe
Fish, 34 per c
Decapods, 17
Average

r cent

40
80
IO

43

60
IOO

30
63

QO
70
50
70

IOO

30

IOO

ent.

per cent.

The ratings being only estimates, a more accurate method is
unnecessary.

An inspection of Table XXIV shows that the distribution of fish
is not correlated with quantity of the foods known to be eaten
by that species of fish in other localities. The fish are frequently
found only in the ponds where the food is least abundant and no
fish is found where its food is most abundant. Are the fish
the cause of the deficiency ol their own lood/ To answer this
question \\Oll Lake and the small ponds were studied. \\<>l!
Lake contains many more fish than any ol the other bodies ol
water thus far mentioned, but as it is a large body we cannot
compare it with the ponds. Pond I. (see p. 15), which has bci-u
artifically separated from \Volf Lake, contains few \\^\\ Ahniniis
crysoleucas, Umbra linii, and Ameiurus nebulosus arc the only
species and these appear not to be numerous. Pond 1 1. contain-

ECOLOGICAL SUCCESSION. 23

TABLE XXIV.

.THY up FOOD; THE RATING OF THE PONDS FOR THE DIFFERENT SPECIFS.

* shows distribution of ti<h.

i

pterus siihnoides

. . Young. . .

*37

45

72

IOO

Adult

65

60

65

Lepomis pallidtf,

. . Young. . .

*^7

-

62

IOO

Adult

*33

77

IOO

. Young. . .

4i

60

IOO

Adult

62

us. .

. . Young. . .

45

71

IOO

A.lult

50

73

IOO

us . .

. Young.

..'

71

Adult

73

-<

Young. . .

...

71

IOO

Adult

:....

. Young.

[,,.,

Adult

...

*7

:id mtlas. .

. Adult

...

*6 S

Si hill" ' < f

. . Adult

....

. . Young. . .

*6i

*6i

7"

Adult

...

*4-'

57

.. Adult

if. . . .

. Adult

. ..

. . .

. . Adult. . . .

J

I'tuhrn .in<l Esox vernncnlatus all fairly .ilumdani. h
.idem ili.it pond I. contains fewer fish per unit \olmne, still
it ha- less Kntomostraca. Evidently con-umpiiun l.\ \\-\\ <\
not -iv.it!> .illt-ct Entomostraca. 1 The i-oudiiimi \\iih rc~|H
to Kntonn.Mi.il .1 is paralleled by other elements oi ii-li loud.

(//' ( "omprtition of Species. On this poini \\c h.i\r ln-rn ,i1>l<
i" -< -i mi -.tlmo-t no data. The golden shiner is absent trum pund
I. Su I'.ir .1- the conditions are concerned, ii should \n-
in iiuinlu i-. It is an important article of dirt for ni.ni\ !"
ti-lir- fuiind there, which suggests that it h.i-> ln-< 11 cliniin.ued
I >\ i In- oilier ti-hes.

;,. /\(7.://.-v Importance of the Brccdi* '. 'iritics and General
The activities will be separated into ^i in-r.il and hn-ed-

( '.riu-ral Activities. This will be taken up with reference
to the depth of water, kind of bottom and surrounding \

1 My >t.iti-im-llt (Slu-llord. 'lO) to tin- rtt.vt th.it thf .ilimuni
xini! ..ut tin- same in all the i . w;i-

.- iK-en (liaiiu-il. Tal'N- XX

24 VICTOR E. SHELFORD.

with which the fish are commonly associated, according to the
various writers cited.

Microptents saJmoides.

Vegetation of the pond weed zone (Hankinson, '07, p. 2131;
3 to 25 feet plants: Potamogeton, Naias, Myriophyllnni,
Elodea (Davis in Hankinson's Report).

Generally prefers still and sluggish waters (Forbes and
Richardson, '08).
Lepomis pallidus.

5 to 15 feet of water, patches of Potamogeton and other
aquatic plants (Jordan and Everman, '02).

Pond weed zone, 3 to 25 feet of water (Hankinson, '07).
Lepomis cyaneUus.

Shoals where plants were abundant; bulrushes and aquatic
types (Hankinson, '07).

Small streams (Forbes and Richardson, '08).
Eupomotis gibbosus.

Plant covered shoals o to 3 feet (Hankinson, '07).
Ch&nobryttus gulosus.

Shallow mud bottomed ponds or lakes (Jordan and Ever-
man). Still water, muddy bottom, plenty of vegetation
(Meek, '08).

Deep pools and quiet water (Henshall, '03).
Perca flavescens.

Chiefly an inhabitant of the pond weed zone; seldom found
in less than two feet of water (Hankinson, '07).

Gregarious; moderate depths of streams and ponds (Hen-
shall, '03).
Erimyzon sucetta.

Limited to places where vegetation was abundant (Hankin-
son, '07).
Ameiiirus nebulosns.

Loves mud; lives in weedy ponds ami rivers without current
(Jordanand Everman, '02).

Fond of mud; weedy ponds and rivers without current
(Forbes and Richardson, '08, p. 206).

Pond weed zone, shallow water at night (Hankinson, '07).

ECOLOGICAL SUCCESSION. 25

Sch i U>eodes gyrin us .

( ' mimon in dense vegetation of the shallow, almost stagnant
\\ater of bays.

I lide-, under stones and logs (Ha> . '141.
Esox irnuifnlatus.

Situations with most aquatic vegetation (Jordan and Kver-
inaii. 'oj .

I'ret. nnce for quiet muddy \\ater: weedy streams Forbe-
and l\i< hard son, *o8).

Grassy -treamsand muddy bayou- Hen-hall. '03
I 'nilini linii.

Mevei seen swimming in the open water; onl\ \\here.iquatie
plain- lornit-d a dense growth in -hallou u.itt r 1 l.mkiu-on,
'07).

liiir\ them-elves in a hole in tin- mud -cooped out \\iili the
lail: rest there at an angle of 45 with tin- tail <lo\\n and the
lie-ad b.tivK |n)iruding (Abbott, ';

Mr. l>\\igln L. Gardner has shown by experimental -mdie-
in our laboratory that they avoid -iron- li-ln.

.'fiicas.

1 "111111011 iii all places where tln-n- an- mam \\ait-r plant-
I laiikiii-on, '07).

Muddii-i and apparently most nninxitin^ hole- ll.i\. '<ij .
.1 niciitrii* nuttilis.

'.Mici.ilK t"re(|iieiiting the pond \\n-d /on.- from \\lii.h it
\\t-ni into -h.illou w.iter at night. Yonn^ in -hallou \\ait-r
\\iili d-n-- \< Delation (Hankin-on, <;

Streams \\ith mudd\' bottom I itrbc- and Kirli.ird-oii. '.
Anicitf 'tis.

^mall |ioud- \\ith muck bottom Jordan and K\ crnian. 'O2 .
A. comparison of the data abo\ \\ ith that in Table I .. p. J, and
Table XXI . p. 17, shows that the lar-c mouthed black ba--. tin-
blue uill. tin- \\armouth, the perch and tin- \ello\\ .md -potted
bullhead- are not in water of the depth which tln-\ pn-l't-r in other
lo ( .ilitii--. Tin- other ti-he- are belter loe.ued a- to the depth of
the ualei.

The lar^e mouthed black ba , the blue gill, the perch, and the
-jiotted and yellow bullhead- are found chiclly in the pond \\eed

26 VICTOR E. SHELFORD.

zone of Walnut Lake. This is characterized by plants that do
not reach the surface. They are Chura, hormvort, bladderwort,
\vater millfoil, water weed, slender Xains, pond weeds, etc.
(Davis in Hankinson, '08). These same plants grow also in the
bays and coves in company with the water lily and othrr muT^nm
plants.

Ponds i and 5r are dominated by submerged plants. Here the
perch, bass and sunfish mentioned above are associated, with the
same species and the same growth form types as in Walnut Lakr.
The bullheads are found common in the ponds in which the
submerged and emerging vegetation are mixed, and which contain
the greatest number of species of the pond weed zone of Walnut
Lake. It seems impossible to draw any conclusion here as to the
relation of these species to either species or growth form in plants.
The whole subject is one for investigation. A comparison of
Tables II., p. 4, and XXI., p. 17, shows that black bass, the sun -
fishes and pumpkinseed are found only where a considerable
area of their preferred bottom is present.

Mud and muck are evidently not distinguished in the tables
of Forbes and Richardson ('08) and it is not possible to make
much use of their data here for this reason. We have noted in the
preceding paper thai the rliubsucker prefers coarse bottom
materials. It muck is included with mud (Forbes and Richard-
son, '08) with the exception of the warmouth and chubsucker.
all are well placed. The chubsucker, the mudminnow, ;md the
golden shiner, tadpole cats and the bullheads avoid strong li.uht,
and their association with dense vegetation which result >, brings
them into relations with bottoms of fine material, e. g., muck, because
they support dense vegetation (Pond, '05).

(b) Breeding Activities. We give below all that has been found
regarding the location of nest and eggs.

Micropterus salmoides: Sterile bottom of clay, sand or gravel,
fibrous roots of the parrot feather preferred to others (Titcomb,
07, p. 10 of separate, fide Slranahan); (b) blackened roots o|'
waterfoil i to 2^/2 feet of water, bulrush shoals in 12 to 15 inches of
water, among conspicuous growth of bulrushes, eggs on rom^
(Hankiiison, '07, p. 214); (c) leaves of trees, gravel; u.-rd \\ln-n
artificial fibrous nest was present (Reighard, '05, p. 4*1; ah sa

ECOLOGICAL SUCCESSION. 2~

^r.ivi-1 preferred, mud, clay, or surface of plants in absence of
these (Henshall, 'o,; : e) gravel, clay or mud from which all
foreign materials have 1 it-en remo\ed Smith, '07, p. 2^7

Lepomis pallidus: Barren -hoals; bottom pure marl or marl and
sand, bottom of marl or gravel; water 5 inches to 2 feet; marl
bottom with bulrushes (Hankinson, '07, p. 212 .

I.<-f>i>nris cyanellits: Swamp loosestrife, black bottom, I foot of
water; m.irl. marl and sand, also root- I lankiii-<>n. '07, p.2io ,

Eupoi ibbosiis: (a) Sand bottom ; i to 2 feet <>i water; -and

bottom; in. nl and sand bottom, -eant biilru-h growth; marl
bottom, bulrush covered (Hankinson, '07 sand and gravel

boiiom not infrequently on roots (Reighard in (till, '05, p. 51
< lt-.tr \\ater; sand and gravel bottom lien-hall, 'o,;

Perca //-.'.- wr/;s: (a) No nest; bare -ami ami ^ia\cl ri\ti .
.mum- a<|iiatic plants (Abbott, '75); (l Stones, \ (Delation, other

objects or loose in water no ne-t Smith. '07, p. 252 .
nebnlosits: (a) Stove pipe, etc.. 4 5 leet. -and, under
cover, in ^ 2 \ in. of water (rarely more than 24 in. I -.\ i ! -h\ nu-r,
'07 vel and ac|iiariiun bottom Kendall. '02; Smith and

I lain.!!, '<>2

S<i:i; zyrinns: In tin can, marl bottom. ^ feet \ \\aifi

1 lankiii-.tii. '07).
/ il>nt linii: Stuck to a(juatic plant- K\d i

TAIU.I-: XX\ .

IIII\\IM. mi RELATION OK KNOWN !..;: II M

IN THE SERIES OF I'HM>-.

v

I'r,

uitli I

ides. Sand.

Sand. o

-,.... Sand. i - o

<us. . . . Sand. i -' - i o

:s Sand and vi"_

tation.
In. .... Sand iiiiclt-i i ^

. . . .' t.ilinll.

The data on breeding habit- as -ummari/ed in Table XXV.
-Ixm clearlv that the distrihution of tin- species whoa
i Lpparei -li;illn\\

28 VICTOR E. SHELFORD.

habits are known is correlated icith the distribution of the conditions
necessary for breeding.

While our tables show that there is considerable bare bottom
in the pond $c, there is good evidence that this is largely due to
building of the road and of the Lake Shore and Mich. Southern
R. R. which separated this pond from the others and from the
lake and probably excluded fish since 1851. The exposures of
bare sandy bottom which are due to natural causes are usually
not covered with more than six inches ol water.

Turning to the perch which is abundant here we note that the
eggs are extruded in the open water or vegetation as well as over
terrigenous bottom. Terrigenous bottom is less necessary than
to the other food fishes.

Turning to the spotted bullhead we note that the nests are
probably usually made in water shallower than any of the other
fishes. Only one specimen has been taken from pond I.; they
are numerous in pond $c and ja. There are some old logs and
stumps and a very narrow zone of bare sand in o in. and less of
water in these ponds. This is commonly shaded by vegetation.

In connection with oxygen content we note that it is greatest
in 5c where the first four species of Table XXV. do not breed.
However, this pond must be regarded as in a measure non-
comparable because of contamination and small amount of plank-
ton.

The low oxygen content on the muck bottoms of the older
ponds, at depths used by the fishes present in pond I., and absent
from these older ones, certainly is a sufficient reason for their
absence, though it is not to be expected that this is the sole
cause. It is apparent also that A. nebulosits, which is present
in the older ponds, not only breeds in shallower water but also
has superior means of aerating the eggs (Smith and Harron, '02).

Succession of fish then becomes succession of breeding condi-
tions and breeding mores. While the major factors as indicated
here are related to deptli and bottom, there are doubtless others.

IV. GENERAL DISCUSSION.

There is great danger of error in dealing with such complex
problems when compilation is necessary and especially when the

ECOLOGICAL SUCCESSION. 29

point of view of the compiler differs from that of the original
investigator. To illustrate principles and methods we have relied
upon compilation far more than could otherwise be justified.
Still certain facts and relation- appear to be clearly indicated
by tin- reconnaissance. These will be roughly grouped under
tin- head- quantitative, economic and general.

i. (Jnantitntive. As has been pointed out in the body of the
paper, the quantity of living material in tin- form of plankton,
invertebrate-, and vegetation increases a- a pond '/rows eco-
lojcally older. In our data there an- t\\<> exception- to thU
which mu-i be noted: First the greater number of Kniomo-traca
in the younger ponds in early spring and the le er number in
pom I 5/ (in all occasions. The greater number in the early
.- prim: i- not ea-ily explained but may be due to the better con~
ditioii- on the bottom where the egg-, etc.. <>f the plankton
Kiiiomo-ir.K a art- found. Possibly the larger areas of clean
bottom pre\eiit their being buried and shut a\va\ from the
eitei ' oi tin sun's heat, oxygen, etc.

I'did 5. i-. as we have indicated, probablv not comparable mi
.K' cunt of the contamination; also plankton production i- mca-
Mired in < iii-tacea and Marsh ('03) ha- pointed cut pit ible
errors in thi- method. A study of all the plankton < -cii-t itueiii-
mi/In .-ho\\ a different relation of 5c. Here, however, low plank-
tcn content is associated with little COj (Birge and Juday, 'n .

The iccied /ross vegetation secures necessary --ib- Irom the
-oil and I 'end '05) jxiintcd out that it im Tea-e- plankton bei ause
the foods absorbed from the soil are added to thewaterwhen the
plain- decay. Our results are then in full accord with tho-ed
l'cn-1. Se< al-o Hirge \' Juday, 'i I . Knauihe, '07, p. 57-

The greater number of large invertebrate-, appears to I.,- gen-
erally clo.-elv related to the amount d" gross vegetation. .\Carl\-
all Mich animals cling in vegetation and main d the species
found in tlu- older ponds use the \e-ei.uion a- a m.-an- of reaching
the -nrface fcr air, of avoidii -unli-ht, and as breeding

places. The majority of such animal- plan- their into or

upon the plant-. Gross vegetation i- al-<> thickly covered \\ith
minute organisms \\hich afford Iced for many animal-.

It i- probable that the amount of rooted vegetation in i-olated

3O V.CTOR E. SHELFORD.

ponds may be taken as an index of plankton production. It
appears that this must be true on tin- l>a-i- of the conclusions
of Pond ('05) no matter what factor is of greatest importance in
controlling the quantity of plankton. Johnstone CoS) pointed
out that the plankton production follows Liebig's law of mini-
mum i. e., quantity is determined by the food substance present
in minimal quantity. If rooted vegetation is the controlling
factor a deficiency in one food substance in the soil would show
itself in the rooted vegetation and through this affect the plankton
production of the pond.

The question of the general application of the principle of
quantitative increase with age is important. It seems probable
that in all bodies of water with small outflow organisms increase
with age because, in addition to the effect of rooted vegetation,
inwash continuously brings food substances which are tied up
if not carried away by extensive outflow.

Experimental study of the quantitative problem is possible on
the basis of such a set of ponds as those at the head of Lake
Michigan. From such a set all the organisms can be trans-
planted and most of the conditions duplicated where closer control
would be possible than in the natural ponds. There appears
to be no difficulty in such experimental study except that it
requires extensive facilities and institution or government sup-
port. Such ponds as ours and such ponds as may be constructed
with them as a basis give promise of throwing more light on the
factors controlling the quantity of life than do the large and
complex bodies of water.

2. Economic. The writer has no practical knowledge of fish
culture and only the knowledge which has been acquired by reading
some of the characteristic literature. Apparently the economic
problems in fishes are concerned with questions of the preserva-
tion of fishes in natural waters, and their increase and main-
tenance against the removal tor tood, which makes them ot
economic importance. With these ends in view efforts have long
been made mainly to increase fish by increasing food suppK , to
care for fish during the critical reproductive season by artificial
hatching and pond culture, and to decrease enemies by de^t ruc-
tion of objectionable fish and fish parasites. The preservation

ECOLOGICAL SUCCESSION. 31

<>\ the fish environments has received little or no attention.
La\\- have been enacted to prevent the pollution of waters, but
the-e have been enforced but rarely.

In practice the importance of the breeding season ha- been

iv<ogni/ed by the culture worker- but appear- to have ivcehcd

little attention from the point of view of the pre-ervaiion or

culti\ation .! fish breeding places in the natural waters. Chirk

i<> i- one of the few who have empha-i/ed breeding Around-.

1 he main emphasis has been laid on nutrition Knauthe. '07.
Chap. 1\

( >ur data indicate that the breeding intere-t- and the feeding
interests of ,ij|| water food and gam.- ti-he- are ilistini'tly an-
onistic. I'.irge Clo) pointed out that \\here the quantity of
pl.inkion t and the fish food acconlingU great, the o\\geii

conteni i- |o\\ .it the bottom and the water accordingly un-uited
i" the production of certain of the be-t food ti-he- Knauthe

p. -,7' states that a large fish producm it\ in a pond i- com-
nioiiK indicated by large amount of gn-- \egetation, but says
also that thf general statement that Mich pond- are al\\a\- good
lroducer- ot ti-h cannot be made. Thi- indicate- that there
are Othei factors. lie makes no mention of breeding and d
not state the practice of pond owners as relating to the breeding.
In -landing and sluggish water, the problem of the balance I.e-
t \\eeii the lood supply and the fish prc-ent -eeiu- ivlathelv un-
important. Cilice feeding conditions of de-irable \,><\ li-lic- gn>\\
better \\ith time at the expense of the breeding condition-, the
major problem is that of the halam :/;/^ and lirccdin^

< <ni(liti(i.\. It appear* that such balance might be maintained

ea-il\ it u i had an adequate knoulrdge ot the environmental

lelatii'ii- ol the lish. Definite knowledge as to -pacing of nests
in natiin- -hould gi\'e data as to bn-eding area re|iiired per
ca|>ita b\ li-h. \\'ith such knowledge at hand, together \\ith the
e\i-ting knowledge of food habit-, it -hould not be difficult to
maintain adequate breeding area- a<ljai ent to good feeding areas
within our \\aters both public and prhate.

; v (/'c;/cn;/. \\'e have noted the aspects of the (|uantitali\ e
and eci.noiiiic problem- \\hich our data I'nable u- to disCUSS.
The remaining indication- of the reconnai ance are tho-e related
ictors governing distribution and metho<l- of -tnd\ .

32 VICTOR E. SHELFORD.

The study of factors governing distribution of fish and other
animals has never been reduced to an adequate working basis.
The problems are indeed complex, but the difficulty has arisen
in part from two causes, namely, (A) the lack of knowledge of
the activity which takes place within the- narrowest limits (Shel-
ford, 'ii 3 ), and (B} lack of recognition of the important factors
and features of the environment.

The conclusions of workers on distribution often seem to have
been to the effect that the food relations of fishes should stand
as first in importance, as factors of distribution. Hankinson
('io) states that the pond weed zone, the living and feeding place
of the fish of \Yalnu t Lake, is probably the most important
habitat. Our evidence on the same species points clearly to
the breeding grounds. Indeed much careful work must be done
before broad generalization should follow, but it is evident that
here as in birds (Merriam, '90; Adams, '08) and in the ti^'-r
beetles (Shelford, '07, 'ii 3 ) the breeding place and the breeding
activities are the most important. (Reighard, 'io, and cita-
tions.) Is variation in nest building real or only apparent
because we do not know the most important factors and seize
upon details wholly unessential to fish? What are the la\\>
governing the mores of species? Experimental work correlated
with field observations can answer these questions, and it is at
this point that contributions of lasting value can be made. The
first step in the necessary work of raising natural history from its
present state of vagary is to determine what activity takes place
within narrowest limits and which is least modifiable in as many
groups of animals as possible.

The second difficulty lack of recognition of the important and
unimportant in en\ inmments is one which we have emphasized
before.

The ecologist often uses vegetation as an index of conditions.
There is objection to this. Investigators have seen that the same
species of animals are not always associated \\itli a given species
of plant. Indeed, species of plants c.innot often and perhap>
usually be taken as an index, of the environmental conditions of
animals, especially in water, because species ol plants an- not
necessarily an index of conditions. Tin- physiological condition

ECOLOGICAL SUCCESSION. V,

ol plan 1 - i- the important thing and is commonly indicated In-
growth lorm (superficially hut not finally) which is the index of
internal physiological state induced by the surrounding condi-
tion-. IMant formation- arc tlu- expression of the condition- ot"
exi-tence for the plants of a definite area. The formation i- the
fundamental unit of the ecology of communities and carries with
it no consideration of species whatever. Identical or similar forma-
tion- ottcn <lo not ha\'e a single specie-- in common. A- \\ e lia\ e
pointed out I.efore, species are of importancr only in -o far as
their e< ..logical constitutions are specific character-. It i- not
s /-" n-li that we are to expect to be a ociated with species

of plant-, hut mores of fish with growth form in plant- or with
plain lormaiions. Furthermore, relation- to venation \\hich
are ot importance are to be expected primarily in connection
\\ ith breeding.

iihjrction to ilu- use of vegetation as an index of tndition-.
due ID mi-apprehension, is to be expected. Ho\\e\er. \\hen the
theoretic.il probabilities are understood, \\ e have not the data
in the case ol li-h, with which to determine \\ he t her or not ^roictii-

"i and more* are associated. The subject i- one for -pecial
experimental and observational investigation.

In connection with the problems of animal heha\ ior, (hi- point
o| \ie\\ Dpen- up a field wherein the role of the different cn\ iron-
menial c..iiditioii- in the control of behavior m.i\ he -! tidied in
naiuie .1- \\ell as in experiment. As a background |..r the -tud\

"t all a-pet 1- "I behavior the point of \ie\\ here pre-ented -eelll-

to < 'ii i dei ided advantages.

Comparative study ol behavior from thi- point of \ie\\ ha-
heen impracticable because of a lack ot kno\\led-e ot en\ iron-
mem-. I mil \\e can accjiiire a knouled-^e and a nomenclature
that -hall he ^eneralK" understood the \\orker mu-i \\rite exten-
sive de-criptioit- of the environment, and i- likeK to eni])ha-i/e
detail- \\lii-h are of little important

The acti\iiies of an animal (behavior are ..f greal economic
imporian.t. they determine distribution. The relation- ot the
hchdrior prohlcms and the distribution, the i/uuntitdtiir diid the eco-
nomic problems seem especially intimate, so that the investigation of
any one from this point of rie:c must contribute to all as well a- to

34 VICTOR E. SHELFORD.

bring about a better unification and organi/ation of biological
science as a whole.

V. SUMMARY OF TENTATIVE "CONCLUSIONS.

1. The quantity of bacteria, plankton, vegetation and large-
animals increases as a pond grows older.

2. Terrigenous bottom and oxygen content decrease as a pond
grows older.

3. The distribution and succession of fish are not determined
by kind of food; kind of food eaten is determined by the availa-
bility in localities suitable in other respects.

4. Fish are not necessarily present where food is quantitatively
greatest.

5. The food and game fishes here considered are closely as-
sociated with their breeding conditions to the neglect of depth
of water, food, etc.

6. Low oxygen content on breeding grounds is a sufficient
cause for their absence from the older ponds.

7. Conditions outside the breeding season are probably of
secondary importance in the success of fish in a given locality.

8. The food interests and breeding interests of the food and
game fish here considered are decidedly antagonistic. The for-
mer continually encroaches upon the latter.

9. Successful fish culture in ponds and small lakes depends
upon the maintenance of balance between the breeding and
feeding conditions.

10. Animal succession in ponds is due to an unused increment
of excretory and decomposition materials which causes an in-
crease in vegetation, a decrease in Oo, on the bottom and a general
change in surrounding conditions, all primarily affecting breeding.

i i . Succession of species is the result of stability of the mores
of species concerned; when mores are flexible species do not
succeed one another but continue with changes in behavior and
physiological characters.

HILL ZOOLOGICAL LABORATORIES,
UNIVERSITY OF CHICAGO,
August i, 191 1.

I i OLOGICAL SUCCESSION. ,^5

VI. ACKNOWLEDGE M- AND BIBLIOGRAPHY.
i. Acknowledgments. In the preparation of this paper the
a--istance of a number of persons has been nece>s ary. A number

I graduate students of the University have -tndied one or more

it" ilic ponds and have given me the use of their notes. The
following -liould be especially mentioiu-d: Mi-- Alma Bu<h,
pond - i \ Mr. \Y. J. Saunders, pond i; Mr. Max Rohde. pond
-a: Mr. B. K. Isely, Mr. \V. C. Alice, Mr. v S. Visher, Mr. G. D.
. \lli ii, Mr. I >. L. Gardner, made more general contribution-.. I
am indebted lo Dr. Chas. C. Adams for reading the manii-rript .

'Ihr following have rendered important service by identilyin^
the mail-rial of groups in which they are speriali-t-: Mr. G. D
1 nlli r. Plants; Dr. J. I'. Moore. Leeches; Mr. V . C. Raker.
M'lllu-' i I >r. ('. D. Marsh, Copepods; Mr. k. Sharpe. O>tra-
coda; Di. A. 1C. Ortmann, Crayfishes; Mi A. L. \\irkrl.
AmpliipixU; Dr. J.Ci. Needham, Aquatic insects; Dr. ( Drnelius
r,iiiii.( add!. Hies; Mr. W. J. Gerhard, Hemiptera; Dr. P. <.
I Iriiieinaim and Mr>. Hlva Class, Bacteria; Mrs. Kb a ( !lass and
Mi. \\ . C. . \lU-e, Gases; Mariner and Iloskin- ("mmeiiial
( 'hi-mi-i- . \\ iter analysis without charge.

2. BIBLIOGRAPHY.
Abbot, C. C.

'76 Note* "ii Some Fishes of the Delaware River. \<< ; :n..

pp. 828-32.

'70 Mini I ..viiiK Fishes. Am. Nat.. Vol. 1\".. pp. 385-"; i.
Ailums. C. C.

'08 I .^-cession of Birds. The Auk. Vol. XXV

BirRe. F. A.

ne and its Biological Significance. Vm \1

\" Sept.. pp. 5 33.

'07 ' >\ . . n I ' >lvcd in the \\'aters of \Vi-om-iii 1 .ik>-~. KI-IP"II \\i- i mn.

.0 I 1907. pp. 118-139.

'07 rhe Respiration of an Inland Lake. Tr.m- Ai 1-241,

"10 ' ed in the Waters of Wisconsii r.'ill. Bur.

H;'"N. \\\ III., pp. 1278-1294.

Birge, E. A., and Juday, C.

'11 1 'In- Inl.iiul Lakes of Wisconsin. fhe D the \\.n.i .n\<\

tliii ical Significance. \Vis. ' un<l N. Bull.

n. i. XML. Sc. Series, 7.
Colton, H. S

'08 : i n\ ii.niiiH-nt mi tin- (imwlh ui' I.vimi.i.i , , .hiiiiclla S

I'l'. Ac. I'lliluclrlph: . JS.

36 VICTOR E. SHELFORD.

Cowles, H. C.

'01 The Plant Societies of the Vicinity of Chii-.tyo. Bull. 2. The Geog. Soc. of

Chicago. Also the Bot. Gaz , Vol. 31, pp. 73-108 and i t>
'n The Causes of Vegetative Cycles. Bot. Gaz., Vol. 51. pp. IM 183.
Clark, F. N.

'10 A Plan for Promoting the Whitefish Production of the Great Lakes. Bull.

Bureau Fish.. Vol. XXVIII., 1908. pp. 637-642.
Clements, F. E.

'05 Research Methods in Ecology. Lincoln, Nebr.
Dachnowski, A.

'08 The Toxic Properties of Bog Water and Bog Soil. Botanical Gazette.

Vol. 46. p. 130.
Eycleshymer, A. G.

'01 Observations on the Breeding Habits of Ameiurus nebulosus. Am. Nat..

Vol. XXXV., pp. 911-18.
Forbes, S. A.

'78 The Food of Illinois Fishes. Bull. 111. State Lab. Nat. Hist.. Bull. no. 2,

pp. 71-89.
'80 The Food of Fishes Acanthopteri: On the Food of Voting Fishes, L. c..

Vol. I., no. 3, pp. 19-85.

'83 The Food of tin- Smaller Freshwater Fishes. L. c.. Bull. no. 6, pp. 65-94.
'88 Studies of the Food of Fresh Water Fishes. L. c.. Vol. II. Art. VII..

433-73-

and Richardson, R. E.

'08 The Fishes of Illinois. Nat. Hist. Stirv. of 111.. Ichthyology. Vol. III.
Gill, T.

'04 A Remarkable Genus of Fishes, the Umbras. Smithsonian Misc. Coll.,

Apr.. 1904, pp. 295-305.
'07 Parental Care among Freshwater Fishes. Smithsonian Rrpmt im 1905,

pp. 403-531-
Hankinson, T. L.

'07 Walnut Lake, Michigan. Biological Survey of Mich. Lansing, State

Board of Geological Survey, Rep. for 1907. pp. 157-288.
'10 Ecological Notes on the Fishes of Walnut Lake, Mich. Trans. Am. Fish

Soc., 1910, pp. 195-206.
Hay, O. P.

"94 Lampreys and Fishes of Indiana. Inch Dept. of Geol. and Nat. Resources

for 1894.
Henshall, J. A.

"81 Book of the Black Bass, Cincinnati. Clarke.
'03 Bass. Pike, Perch, and Others. N. V.. Macmillan.
Herrick, F. H.

'02 The Home Life of Wild Birds; a New Method of the Study of Photography

of Birds, New York.
Juday, C., and Wagner, Geo.

'08 Dissolved Oxygen as a Factor in the Distribution of Fishes. \Yi< Ac. Sr.

Arts and Lett., Vol. XVI.. pt. i.
Johnstone, James

'08 Conditions of Life in the Sea. Cambridge I'niv. Press.

ECOLOGICAL SUCCESSION. .^7

Jordan, D. S., and Evermann, B. W.

'02 . \nu-rican Food and Game Fi-lu--. X. V.
Kendall, W. C.

'02 The Habits of the Commercial Cat-Fishes. Bull. I". S. F. (".. p;> ,;QQ 415.
Knauthe, K.

'07 I ).i i <ser, Xeudamm.

Kofoid, C. A.

'03 I In- Plankton of the Illinois River. Part I.. (Juantitatix v Iiu f-tiu.uii>i>-

ami > .--I.. ial Results. Bull. 111. St. Lab. of N. II.. \".>1. II.. ArtirU- i i.
Lydell, D.

'02 Habits and Culture of the Black Bass. Bull.' - 1 i , pp \g 44.
Merriam, C. H.

"90 l<> ; a Biological Survey of the San Fraiu i- M init.iiii- ami tho

tin- Little Colo.. Ari/. L'. S. D. Agr., Bii>l. Surv \ \ I .nin.i. i.
Marsh, C. D.

'06 I li- I'lanktmi of Lake Winncbago and Green I.aki-. Wi-< HI-MI (.ml. anil

N. II ^ur\-y. Bull. XII.. Scientific Series, 3.
Marsh. M. C.

'10 N i hi* Dissolved Content of Waters in r I :- "ii l-'i-ln - Mull.

I > (Internal. Fisheries Congress), pp. *

Meek, S. E.

'08 List ol i Known to Occur in the Waters of Indiana. Hirmii.il !< p"i t

ni tin- ( I'imnissioner of Fisheries and Ciame of Iii'li.uui im i~^<- "S. pp.

i : ; 171 I :: 1 .iil.ipolN.

Meek, S. E.. and Hildebrand. S. F.

'10 s\m.| .if the Fishes Known to Occur Witliii.

I M mi of Natural History. Chicago. Publ. i i- 1 /mil. > i \"i
\ II \
Pfeffer, W.

'oo Plant Pli\ -i. .!!{>. Translation by E \vert, E. \\'.. < >ximd.
Pond, R H.

'05 I In I il Relation of Acpuatic Plants to tin- Sul>>ir.itum. I

.in. I I i Ii. lies Com. Rep. for 1903. pp. 483- ;
Reighard, J.

'05 Ii- its. Development, and Prop.i Blacl !' Hull.

i M i . K I :-li. Com.

'10 \liil. >tudying the Habita of Fishes with an Account "i tin I'.n-i-ilnii;

llal.it- ..i the Horned Dace. Bull. Bur. FMi.. \'..l. X\\ III . 1908. pp.

1 M I I .

Ryder, John A.

'86 I In 1 i. \ i li.pnu-nt of the Mud-minnow. \M Nat., V"l. -'... p
Shantz, H. L.

'06 A Stmly nt the Vegetation of the MI--.I K. - n 1 I 1 , ik. I'.iit

l 1 In- Bouteloua Formation, Bet. <.a/.. \'nl. i--. p 17.;.
Shelford. V. E.

"07 rirliiniii.ii \ Note on the Distribution t tin- 1 \^<-\ Bcrtlc-* <Cu-indela) and

it- Ri l.itinn in Plant Siu.r^inn. Bin]. Bull.. \'nl. 14. pp. <; i .
'10 Ecological Succession of Fish and its Bi : nu <>n Fi-h < "til tun-. 1 11. ~-t \.

IS \'nl. 1 I . pp. 1"* I"

V s VICTOR E. SHELFORD.

Ecological Succession. I. Stream Fishes and the Method of Physiographic
Analysis. Biol. Bull.. Vol. XXI.. pp. 9-35.
'n : Ecological Succession. II. Pond Fishes. Biol. Bull., Vol. XXI., pp. 127-

151-
'n 3 Physiological Animal Geography. Jour, of Morph. (Whitman Volume),

Vol. XXII.. pp. 551-617-
Smith, H. M.

'07 The Fishes of North Carolina. N. C. Geol and Economic Surv.. Vol. II.
Smith, H. M., and Harron, L. G.

'02 Breeding Habits of the Yellow Catfish. Bull. I'. S. Fish Com., Vol. XXII.
Titcomb, J. W.

'07 Aquatic Plants in Pond Culture. Report of the Com. of Fisheries, 1007.

NOTES, REVIEWS, ETC.

tically all nucleus. The corpuscles on the other hand have lost their
nuclei wholly. Between these extremes we have various stages/ of
chromatin reduction in the development of the specialized Metaeoan
tissues. The maturation divisions in ova and sperm, the bodily ex-
trusion of chromatin observed on the part of blood-cells, etc/ he re-
gards as illustrations of the process.

THE RESERVE OF FOOD IN TREES

Proton and\Phillips (Forest Quart., 1911) agree 7 with the com-
mon view that staV'h is the principal form in which reserve food is
stored in trees. Tney doubt that cellulose is abl/to act at all as a
reserve material. Tie maximum contained re/erve for deciduous
trees occurs about theVime the leaves fall, and during the next few
weeks there is a decidedVeduction in its amount. The sugar content
in trees remains pretty constant <.-xri.pt f"r ,'m incn-a-r in -pring (lur-
ing the unfolding of the btn

ALTERNATION OF ''. N M

Lewis ( Mot. Gaz., Mch.. iQrs^. by artificial plantings of tetra-
spores and carj>ospores of Polysw1(pnia and some other genera of
red algae gets experimental re:/ilts\upporting the general conclu-
sion that tetrasporcs produce only the sexual plants and carpospores
only the tetrasporic plants. /In no instance was an exception found
to the rule, although a considerable numbet of plantings developed
to maturity. Tetrasporcs from a given individual produced male
and female plants in approximately equal numbers. It is also con-
cluded that no greater growth vigor comes to th\ carpospores over
the tetraspores because of the double number of chromosomes con-
tained by them.

RELATION OF THE PROTOPLASM OF ADJACENT PROTOPLASTS

iy (Ann. Bot.. 1911) undertakes to throw light ofa the rela-
tion that exists between protoplasms of contiguous cells, b\ an ex-
amination of the relation between the parasite, Cuscuta, and r^s host.
She finds that there is no direct protoplasmic connection between the
cells of Cuscuta and the host, but that the phloem cells of the pa^a-
site haustoria apply themselves to the sieve plates of the phloem of

AMERICAN MICROSCOPICAL SOCIETV
ECOLOGICAL SUCCESSION OF PLANTS AND ANIMALS / 3

Shelford i Biol. I'.ull.. her.. KM i \ concludes a series of papers
dealing with the biological succession in ponds at the head of Lake
Michigan. The following are some of the conclusions reached by
the author as the result of this series of interesting studies :

1. The quantity of bacteria, plankton, vegetation, and large ani-
mals increases with the age of the pond.

2. Terrigenous bottom and oxygen content decrease with the
age of the pond.

3. Fish tend to adapt themselves to the type of food rather
than to become distributed or furnish successions in accordance with
the type of food. They are not necessarily most abundant where
food is greatest.

4. Small oxygen content of older ponds will account for ab-
sence of fish from them.

5. Conditions outside the breeding season are probably less im-
portant than those of this season in determining the success of fish.

6. The conditions most favorable to the normal feeding of fish
are not only different from those most favorable to breeding, but are
even antagonistic ; and the former tend to encroach on the latter.
ment, and the preservation of balance between the breeding condi-
tions and the adult life-conditions.

7. Animal succession in ponds is due to an unused increment
of excretory and decomposition products which causes increase in
vegetation ; a decrease in oxygen at the bottom ; and a general change
in the conditions affecting breeding.

8. Succession of particular species, rather than the continued
dominance of some when they once become dominant, results from
the inflexibility of their standards of demands in accordance with
the changing conditions.

CHROMATIC REDUCTION IN CELL DEVELOPMENT

ihde fZeit. Wiss. Zool.. 1911) undertakes to showL-tfeat~\a
marked clTafarteristic of the (lifferentiatiojr^^-rnSfunng of cells is
the reduction of chro^latSB^Hthe^uicleus. He suggests, as illus-
trative of trris j --a-'snes with bacte?hr-ftt_Qne end and the red blood-
cells of^fiammals at the other. The bacteriaTie-scajfiiders as prac-

THE CENTRAL NERVOUS SYSTEM IN TERATOPH-
I' HALM 1C AND TERAT()M< >KPHIC FORMS OF
PLANARIA DOROTOCEPHALA.

C. M. CHILD AND E. \'. M. McKIE.

The -tudv of the nervous system in tin- teratoplnhalmic and
teratomorphic forms of Planaria dorotocephahi \\a- undertaken
by i In- junior author of this paper at the senior author'- sugges-
tion. 1 h<- results of this study were accepted as a the-i- for the
M.i-i. ree by the Department of Zoology of tin- l'ni\i T-itv

o! ( hi. .1-0. Since the results of the work an- of ( -on-iderablc
inten-t .m<l since- Miss Mi-Kit- was prevented by \.iriou- cir-
ciim-i.iiice- from preparing the paper for publication, the -enior
author ha- undertaken, at her express request, to revise her manu-
3< ript lor publication and to add some figures from her -lide-; he
h.i- also added a section on the various method- b\ \\hich the
teratophthalmic and teratomorphic forms have been produced

and ha- e\ii-udrd sollU'what the scope ot the di-cn--ion ot the

results.

The primary object of the work was to determine the general
form and tin- degree of development of the cephalic part of the
central IK i \ oiis system in these abnormal form- as compared
uith normal animals. The observation- concern chiellv the
teraiomorphii form* since these repre-eiit a more extreme
departure I'mm the normal type and afford more definite and
-tiikin^ re-nit-.

The animal- for sectioning were anesthetized with \\eak alcohol
beloif iixation iii hot (Wilson's fluid or -nblimate. Sections were
cm ten micra in thickness. Frontal ami -au'ttal as well as trans-
verse sections \\i-re made, but all the h-uiv- are drawn from trans-
\ ei se sections -incc these show the e--ential feat ure- most clearK'.

Ml li-uie- of sections were drawn to the same -calc with
the camera. They are designed to -ho\\ . tir-t tin- -eiieral form
of the nerxou-, -\-tem and second, the general relation^ bet \\een
filler tracts and cell>. The cell- are repre-ented merel>- by small

39

4O C. M. CHILD AND E. V. M. Mi KI1 .

circles or ovals and the fiber tracts are filled in with dot-. except
where a distinct commissure or nerve is concerned; there the
direction of the fibers is indicated. Non-nervous structures are
not shown except in the case of the alimentary tract, which is
diagrammatically indicated where it is present in the sections
figured.

I. THE EXPERIMENTAL PRODUCTION OF TERATOPHTHAI.MK .\\n

TERAIOMORPFTC FORMS.

The senior author has given the names " teratophthalmic '
and "teratomorphic" to certain types of head which appear
under certain conditions in the regulation of pieces of Plannr'ui.
The teratophthalmic head (Child, 'ua, pp. 278-9; 'lie) is one
in which the eyes show some departure from the usual structure
or arrangement, but the head is otherwise normal in form. The
teratophthalmic forms may be divided into several groups ac-
cording to the character of the eyes, for these may be "abnormal "
in position, size or number or the pigment cups may show tlir
most various degrees of fusion (e. .(,'., Fig. 6 below).

The teratomorphic heads (Child, 'nr) represent a more ex-
treme departure from the norm. In these the abnormalities
involve not only the eyes but the shape of the head and the
position o! the auricles. The teratomorphic head usually pos-
sesses a single median eye and the auricular sense organs appear
on the front of the head, either separate (Figs. 10 and 16) or
more or less completely fused (Figs. 19 and 2^1. In the senior
author's earlier work on Plauaria the teratomorphic hca<l> \\ere
not separated from the teratophthalmic (Child, 'i \a), but as the
degree of experimental control in the production ot these lorms
increased it became desirable to set these peculiar forms apart
as a distinct group and to give them a name.

It is possible, as the senior author has shown in various papers
(Child, 'l la, 'l ic, 'lid), to control experimentally by a number
of different methods the production of these torms. In general
they are the result of conditions which decrease the rale ot the
dynamic processes below a certain level determined by existing
conditions which is necessary for the- production of normal ani-
mals. With the proper experimental conditions they can 1 >r

CENTRAL NERVOUS SYSTEM IN PLANARIA. 4!

produced from any region of tin- planarian body and from piece-
of any size above a certain minimum, which varies with region
ol tin- b'ldy, physiological condition, age. nutrition and external
condition-;. A regional factor does, however, exi-t Child. 'ii</>:
in piece- of a given length the more posterior the level within a
-ingle /ooid. the greater the frequency of the abnormal form-.
Thu- far it has been possible to control experimentally the
prodnciion of teratophthalmic and teratomorphic as \\e11 as
anophthalmic and headless forms (Child. 'lie) in the t'o'louing
ways first, under standard conditions of t-ni|)erature. ninritit>n.
eti . pi < es above a certain length with anterior end- at a certain
le\e| ,,| ili,- l.odv \vill produce normal whole-, -horter piece- will
prodiKe teratophthalmic forms and still shorter piece- terato-
morphic. .m< iphthalmic and headless form- a- the length de-
: 'Hid. in pieces of a given length from a gi\en \-< -ion.
under iinitinn conditions of temperature, nutrition, etc.. -tinml.t-
tion to motor activity increases the frequen<\ ol normal anim.t 1 -.
uliile lai 1: ot stimulation increases the frequency of teratophth.il-
mic, teratomorphic, anophthalmic and he, idle-- form-. Third,
in pie. es ol i Jvc-n length from a given re-ion of anim.il- of the
ie -j/e the frequency of abnormal and tioim.il form- varies
\\itli difference* in physiological age (C'hi'd. ' I I ; " and \\ith dif-
n. ex in nut rit ion. Fourth, in pieces ot -^i\ en length Irom a
given ie. ion of animals of the same size and a- ne.irK a- poi bit-
in the -ame | >li\ siological condition a \.uietv of external facior-.
-uch as lo\\ teni|)eratiire, metabolic produ. i- in tin water, dilute
al<-ohol. I'ther, chloretone, potassium c\anide. etc., \\ill increase
the fre(|uenc\ of abnormal forms and it i- po--ible to control to
a certain extent the type of abnormal form, both through the
length of the piece and the intensity of the experimental factor.
( >n the other hand, the frequency of normal form- in a -i\ en --t
of pieces can be increased by good nutrition, by high temperature
and probably also by certain stimulating drug-, 'luni-h as re-
gard- the-e la-t the results are complicated l.v the fact that in
main ca-e- the stimulating effect of drugs is of relatively -h..n
duration and i- fo!io\\ed by a depression.

Hut \\hate\ er may be the results of more ex: ended experiment,
the fact- alrcadv e-tabli-hed demon-irate that the normal and

42 C. M. CHILD- AND E. V. M. Me KIE.

the abnormal forms described represent differences in the dynamic
processes which are primarily purely quantii.itis e. The tem-
perature experiments illustrate this point very clearly. In a
given set of pieces higher temperatures increase the frequency of
normal, lower temperatures that of abnormal forms. The effect
of the anesthetics and the other external factors mentioned above
is probably also primarily quantitative.

In these cases then different morphological characteristics
appear as the result of primarily quantitative changes in the
dynamic processes in the organism. This fact is of considerable
theoretical importance, since it can mean nothing else than that
form, structure, localization, number and even presence or ab-
sence of parts may be determined by purely quantitative changes
in external factors, i. e., by changes which alter primarily the
rate and not the character of the dynamic processes.

Certain external characteristics of the head region of the ab-
normal forms, viz., the position and number of the eyes and
auricles, indicate that the cephalic ganglia of these forms must
show considerable departures from the norm. The question as
to how the form and structure of the central nervous system
may be altered by these quantitative changes in the dynamic
processes is one of interest from various points of view. The
data presented below give a partial answer to this question and
so form a contribution to our knowledge of the dynamics of
morphogenesis.

The method used for obtaining the teratophthalmic and tera-
tomorphic forms described in this paper was that of cutting
pieces of a certain length, determined by previous experiments
of the senior author, from the middle region of the body of large,
well fed worms and allowing them to undergo regulation at about
20 C. This method was used merely because it is the simplest.
Teratophthalmic and teratomorphic heads develop on pieces of
greater length from the middle 1 region of the body, /'. e., the
posterior region of the first zooid (Child, 'i\<i, 'iu;) than from
any other region. This makes it possible to use relatively long
pieces and the preparation and handling of the material is there-
fore less difficult. Of course the abnormal heads can be obtained
from still longer pieces if regulation occurs at low teinper.n mv>.

CENTRAL NERVOUS SYSTEM IN PLANARIA. 43

but the length of time necessary for regulation in such cases is a
disadvantage. Abnormal heads produced by the action of anes-
thetics and by various other conditions were not included within
i!i- scope of the present investigation. It is not improbable that
compari-on of the heads produced by different conditions will
-how more or less characteristic differences in the nervous system.
In all ca-es described the pieces were kept for at least two
ueek- after -ection. After this length of time the new head is
well de\ eloped and those cases in which the teratomorphic head
does ii"! UMII, iin teratomorphic but redifferentiates into a head
ot normal -hape have already undergone this further regulation
or -lio\\ unmistakable indications of it. The teratomorphic
head- \\liiili persist as such for two weeks at 20 ('. almost
-hou any further changes.

II I in MEAD OF THE NORMAL ANIMAL.
The form and the chief external features of the normal head
ot J'laiHiriti dorotocephala are .,ho\vn in Fig. i. The unpigmented
areas ! the eyes and the very slightly pigmented sensorj regions

ol iln- auricle- are indicated by dotted lines. Except t"i" these
the dor-al .-mtace of the head is usually rather deeply ami uni-
loimly pigmenied. I mm the ventral surface the outline- "t
the (ephalic ganglia and nerve cords are in li-iin. il\ \i-ible in
the li\ in:: animal.

I j 5 shou transverse sections ol the ner\ou- system

ai ilillerent levels of the head region. 1 i^. J i- I' mm a K-\el
al nit half uav between the eyes and the lip of the head and
four nerves extending to the anterior head region. Further
these ner\es break up and become less di-tin<t as
the\ are ili-nibuted. l ; ig. 3 shows the icanulia at about one
fourth ot the distance from the e\ e- to the tip. The\ consist
of hlier tracts including a few i ell- .md -urmnnded \>\ maii\
oilier-. At thi- level the chief fiber tract shou- indication- of a
beginning -eparation into right and left hal\e-. < )n e.ich side
of the chief tract is a small tract separated from it by cells:
the-e tuo -mall tracts are cros- .-ections of ner\e- \\hich pass
io the anterior re-ion- ol the head.

i i-me 4 shows th ^ Jia at the level of the eyes. The t\\o

44

C. M. CHILD AND E. V. M. MCK.IK.

masses lie some distance apart and are connected by "commissures.
The pigment cups of the eyes open laterally and the optic nerves
pass in a dorso-ventral direction. Between the eyes the median
anterior branch of the alimentary tract appears.

3

S

FIGS. 1-5.

Herr

Figure 5 is from a section at ihe level <>l tin- aurick>.
there is no distinct commissure between the ganglionic
It is quite possible, however, that they arc not mtiivly discon-

i I.NTRAL NKRYOUS ^\>TKM IN PLANARIA.

45

nected as they appear in the figure; some nerve fibers may extend
aero-, the space between them. The nerves passing to the
auri< les appear in the section and the alimentary tract lies on
i IK- dorsal side. Posterior to this level the nervous system con-
sists of the two main nerve cords, each composed of a fiber tract
including -ome cells and surrounded by other- ami ^i\inu ri-e
to in -r\e- and commissures at various level-.

III. TKRATOPHTHALMK Hi \i>-.

( t i In.- \aiious types of teratophthalmic heads only iln>-e \\hich
-hou partial t'u>ion of the optic pigment cup- were examined
Thf form- \\ith unef|ual or unsymmetrical I-NC> lonstitutr a
(littt-rc-nt type of teratophthalmia and n-'|tiin-a niori-
in\c-ii^ation: moreover, the partial !ti>i<>M- ot ilir eyes

9

FIGS. c>

lead thmui'li all possible stages to tin- -indr nirdian r\c o! tin-
ii-raioiiion>hic head.

l ; i-uir (> >ho\\- tin- outline of the- body and the condition of
tin- eyes in >iu- of the teratophthalmic heads >erti<mcd. The

46 C. M. CHILD AND E. V. M. McKIE.

two optic pigment cups are symmetrically situated but lie closer
together than in normal animals and are united by a continuous
band of pigment.

In Fig. 7 a transverse section of the nervous system from about
the posterior fourth of the preocular region is shown. It consists
of a single fiber tract surrounded by cells and without any trace
of division into right and left halves. Comparison with Fig. 3
which is from about the same level in the normal animal shows
a marked difference in form. Fig. 8 shows the- level of the eye-.
The difference between this and Fig. 4 from the normal animal
is striking. In Fig. 8 the fiber tract is partially divided into
right and left halves, but the two parts are close together instead
of being widely separated and connected by a long commissure
as in Fig. 4. In Fig. 9 a section at the level of the auricles is
drawn : much the same differences from the normal (Fig. 5) ap-
pear here. The two ganglionic masses are closely connected,
while in the normal animal they are widely separated.

The figures from this teratophthalmic head show one other
point of interest. The individual from which the sections were
made was much smaller than the full grown animal of Figs. 2-5.
Figs. 2-5 and 7-9 are drawn to the same scale and comparison
shows at once that the ganglia arc almost as large in the tera-
tophthalmic as in the normal animal . This is a general character-
istic of physiologically younger as compared with older and of
smaller as compared with larger animals. In the small young
animal the nervous system is always of relatively large size and in
small animals which result from the regulation of pieces the
same is true, except in the more extreme abnormal types, where
the nervous system is often small. Thus as regards the develop-
ment of the nervous system as well as its rate of metabolism
during development (Child, 'i \b) the animal formed by regulation
resembles a young animal.

In oilier teratophthalmic individuals with partially tused eyes
the general form of the ganglia was found to be murh the same
as in the case described and the degree of fusion or separation
of the ganglia corresponds rather closely with thedegreeof fusion
or separation of the eyes. In these forms then the eye- serve
to some extent as an index of the condition of the nervous -\ stem.

VTRAL NERVOUS SYSTEM IN PIANARIA.

47

IV. TERATOMORPHIC HEADS.

i 1 lu- first case to be described is shown in Fig. 10. Here the
auricles appear on the front of the head and extend anteriorly.
The anterior margin of the head between them is slightly rounded
instead <>t pointed as in the normal animal. In tin- median line
is a -in.'li- eye.

- tions of the head region of this animal are shown in Figs.
11-15 The eye is situated almost at the extreme .interior end

to

FIGS. 10-15.

"I i IK- \ .ionic mass and a few sect inn- anterior io it tin-
in-! -\nu- -\-tc-ni appears as in Fig. 11. Mere four filter tracts
Mirnumded by cells are visible and are e\identl\ nerves t the
anterior re-ion- of the head. The conditions ai the le\ el <>!'
the eye are -lnwn in Fig. 12. The -inije optic pigment cup
open- anteriorly instead of laterally, as the adjoining -ectinn-
on the -lide -ln>\\ . and it is farther from the dor-al -urla< e ot' tin-
head and more nearly imbedded in the u.uu-!ionie ma than in

C. M. CHILD AND E. V. M. McKIE.

the normal anima! (Fig. 4). The ganglionic mass itself is some-
what irregular in form and shows no trace of a division into
symmetrical right and left portions.

Figure 13 shows the condition of the ganglia six sections (sixty
micra) posterior to the eye. Here the fiber tracts show indica-
tions of a symmetrical arrangement, hut this arrangement is
widely different trom the normal. Three sections farther pos-
teriorly the fiber tracts are still more broken up, as shown in
Fig. 14. The level of this section is approximately the posterior
end of the anterior new tissue of the regenerated region. Fig. 15
shows a section sixty micra posterior to the level of Fig. 14, /. e.,
in the old tissue: here tin- nervous system appears in the usual
form of two ganglionic ventral cords, which, however, are much
less widely separated than in the- normal animal at this level.

2. The animal sectioned is shown in Fig. 16. The head is
much like that in Fig. 10, but the auricles are somewhat closer
together. A single median eye with a rather large pigment spot
is present.

Figures 17 and 18 show sections of the head. Fig. 17 is a

16

18

Fics. 16-18.

section a short distance anterior to the eyes, about the posterior
fourth of the preocular region. Fig. iX shows the level of tin-
eyes . The single pigment cup appears in the liguiv io open
ventrally, but the opening is actually antero-ventr.il in direction.
The ganglionic mass is distinctly double, /. c., more- like the

CENTRAL NERVOUS M>IKM IN I'l.ANAKIX.

49

normal than that of Fig. 12. Two nerves, one from each portion
of the ganglionic mass, pass to the optic cup.

He-re, as in the preceding case, the eye is situated near the
extreme anterior end of the ganglionic re-ion instead of a con-
-iderable distance posterior to it as in normal forms. Posterior
i" the eye tin- torm of the ganglionic mas- i-ontinues murh the
-ame a- in I i_ 18 to about the posterior end of the ne\\ iiue.
uliere the He lit and left portions become more distinetly -epa-
r.iied \\itli .1 commissure between them and then pa into the
tuo ner\ e e..rds.

In general form the nervous system is much le abnormal in
i hi- i hati in the preceding case. The chief difference- from the
norm are the anterior position of the eye on the ean-lionic ma--
and the partial !u-ion of the two ganglia for a considerable di--
tanee |i-ierior to the eye.

v \- indicated in Fig. i<), this case shows a someu hat extreme

2O

21

FlGS. IQ 22.

lorm of teratomorphism. The two auricle- are tn-ed at the lip
of the head, though the sensory areas are in large part separate.
A -ingle median eve is present as in the preceding <

I ig. 20 sho\\ - a t ransverse section < >t the IHT\ < m- -\ -tern .it the
level where the in t \ e- to the front of the head ari-e: rlii- is about

C. M. mil, I) A\l> I-;. V. M. M, KII-.

one tilth <>I the distance from the cyc> to the tip of the head.
Five fiber tracts unsymmetrically situated an- indicated in the
section.

In Fig. 21 the level of the eye is shown. The optic pigment
cup opens antero-ventrally and toward the left side and i- con-
nected by a nerve with the left side only of the ganglionic ma--.
The latter shows a distinct division into right and left halves.

Near the posterior end of the regenerated region the nervous
system possesses the form shown in Fig. 22 and a short distance
posterior to this level and in the- old tissue the two nerve conl-
become separate except for an occasional commissure.

In this case the eye, though median in portion, evidently
belongs to the left half of the ganglionic mass and the nervous
system is much less abnormal than in Case I. As in the other
cases, the eye is situated near the extreme anterior end of the
ganglion.

4. In this case (Fig. 23) the fusion of the auricles at the front

24-

26

FIGS. 2 ^

of the head is even more complete than in Case ,}, only the ba-e-
ot the sensory areas being separated. The eye is median and
apparently single and the pigment spot is of rather large si/e.
Anterior to the eye the ganglionic mass breaks up almost inline-

CENTRAL NERVOUS SYSTEM IN PLANARIA. 5!

diately into two nerves passing to the front of the head (Fig. 24).
In Fig. 25 it becomes evident that the apparently single eye is
actually double. One of the pigment cups lies slightly anterior
and \i-ntral to the other and somewhat to the left of it. The
opening <>f the more posterior and dorsal cup i- -een in Fig. 25,
while tli- other pigment cup appears here as a complete circle.
P.oth open antero-ventrally and toward the right. The gan-
Jionic ma i- not divided into right and left halves and the
optic nerve- arise from its median region. Both eyes are far
belo\\ i he dor-al surface of the body and the more ventral one
i- imbedded in the- ganglion.

The double nature of the eye is not apparent in the living
animal since the two pigment cups lie so close together and one
i- almo-i \entral to the other.

l'< -icrior to the level of the eyes the ganglion i- abnormal in
lorm to about the posterior end of the regenerated region. I
Jo. .1 -e< lion -li-htly anterior to the boundary bet \\een ne\\ and
old tissue, shous that in general form and arrangement of the
hi MI ILK i- i he ganglion at this level resembles that of Case i
Fig. 13). \ little farther posteriorly, in the old tissue, it
di\ide- into lijn and left j>ortions and gives rise to t\\o nerve
cords ot tin- n-nal form, but somewhat nearer together in their
anteHo] region than in normal forms.

.V Tin- case is like Case 4 in external appearam < IK 23
and al-o in the number and arrangement of the eyes, but -onie
diltereine- in the structure of the nervous system e\i-t. I ig. 28
-hou- the le\ el of origin of nerves to the front of the head, a l<-\ el
-liJnK anterior to the eyes. In Fig. 29 the eyes an seen to be
^lightly larger than in the preceding case, but other\\i-e -imilar
to it. Hoth open antero-ventrally and toward the ri^hi and one
lie- to the K It of, \-entral and slightly anterior to the other. The
gan^lionie ma-- is single, but larger than in ( a-e 4 at tin- level
I igs Jo and 25 and the optic nerves ari-e Irom it -omeuhat
to the ii-ht of the middle.

Posterior to the eyes the ganglion -ooii -lm\\ - di-tinct right
and left hal\e- but these are abnormal in -hape and each is
broki'ti iij) into a number of more or le-- di-tinct liber trai
\\hich. however, are apparently symmetrically ariMtt^ed in the

52 C. M. CHILD AND . V. M. McKIE.

right and left halves (Fig. 30). This peculiar arrangement con-
tinues to about the posterior end of the regenerated region and
then changes into the form shown in Fig. 31: posterior to this
the two ventral cords appear in the usual form.

These five cases give some idea of the variations in structure of
the eyes and the nervous system in the teratomorphic forms. A

FIGS. 28-31.

more extended investigation of these forms will undoubtedly
show other variations in structure and examination of anoph-
thalmic and headless forms will add still further data of interest.

V. DISCUSSION.

In all of the cases described, both the teratophthalmic and
the teratomorphic forms, the most conspicuous difference in the
nervous system as compared with the normal forms is the more
or less complete fusion in the median line of the two portions
of the ganglionic mass, or more correctly, their incomplete
separation.

It is evident that to some extent the condition of the eyes or
eye is an index of the condition of the nervous system. The
parallelism is, however, not complete: in Case I (Figs. 10-15),
for example, the nervous system is much more abnormal than
in Cases 2 (Figs. 16-18) and 3 (Figs. 19-22), though all three
possess a single median eye.

Moreover, the degree of fusion of the auricles does not corre-
spond exactly to the degree of fusion of the ganglia in all cases.

CENTRAL NERVOUS SYSTEM IN PLANARIA. V^

In Case I where the auricles are a considerable distance apart
(Fig. 10) the fusion of the ganglia (Figs. 11-15) * s more complete
and their structure is more abnormal than in Cases 2 (Figs. 16-
18) and 3 I L-. 19-22), where the auricles are nearer together.
In Cases 4 .ni'l 5. where Uvo eyes develop close together, far
from the -urface and in abnormal relations to each other and
where the auricle- are partially fused, the ganglionic region is
highly abnormal.

\\lien. ho\\e\er, we compare the teratophthalmic with the

lomorphic forms it is evident that a general parallelism be-

tueei! tin- external features of the head and the condition of the

nervous system does exist. So far as the observation- go at

in. tin- 111 r\ oiis system is always more abnormal in the

irraioini.rphii !<>rms than in the teratophthalmic form- \\ith

partialK tu-ed t

\ n gards the eyes themselves certain points are of inttn-t
In the normal and partially fused eyes the pigment cups open
laterally, \\hile in the teratomorphic forms they open anteriorly
or an 1 1 -i < i- \em rally. Moreover, the eyes are usually farther from
tin doi-s.il -urlace of the head in the teratomorphic form- than
in other-.

1 In -iiu'le <-ve of the teratomorphic head may be connected
\\ith both -ide- of the ganglionic mass (Fig. 18) or with only one

I i. j i and in case two optic cups arise in the teratomorphic
hea-l I =, and 2(>) they may both be connected with the -ame

part nt t ! . limi.

The pM-itioiiof the eyes or eye on the ganglion may differ more
or less ui'lek in the normal and abnormal forms. In the normal
animal Figs, i 5) the eyes lie dorsal to the posterior region of t hi'
^airjioii. \\liii-h divides a -hort distance behind them into the
t\\o nerve cords. In the teratophthalmic forms with partialK
fn-eil eyes a . cn-iilerable portion of the ganglion lie- anterior
to the eyes M^. 7) but the right and left sides of the nervon-
system remain united farther posteriorly than in tin- normal

I k-. 5 and T. In the teratomorphic forms the eye lies dor-al
to the extreme anterior portion of the ganglion and the t\\<> eonl-
do not become -eparated for a considerable di.-tance posterior
to it

54 C. M. CHILD AND E. V. M. McKIE.

In the development of the abnormal forms the eye undoubtedly
arises in connection with the central nervou^ system .1- it does
in the normal animals. The position and number of the eyes
must be determined primarily by the condition of the nervous
system, though other factors may play some part. Apparently
the median regions of the nervous system are more or less reduced
or fail to develop in the abnormal forms and the lateral region:-
consequently lie nearer together so that the eyes appear near
or in the median line. The approximation and fusion of the
auricles is also evidently due to the reduction or absence of the
median region of the head and this condition is undoubtedly
closely connected in one way or another with the condition of
the ganglia. There can be no doubt that the condition of the
nervous system is the most important factor in determining the
characteristic features of the teratophthalmic and teratomorphic
heads.

One of the most interesting points in connection with the
whole series of forms is the fact that in the teratomorphic forms
the whole length of the regenerated nervous system is abnormal
(Figs. 13 and 14, Fig. 26, Fig. 30). Not until the level of the old
tissue is reached do the two cords appear in their usual relations
and even there they are commonly nearer together than in
normal animals (Figs. 15, 22, 27, 31). This fact suggests that the
development of the regenerated portion of the nervous system is
in large measure independent of the already existing portion. 1 1
the development took place in the anterior direction from the
cut ends of the nerve cords in the old tissue, it is difficult to
understand how such structures as those shown in Figs. 13 and
14, 26 and 30 could arise near the old tissue. Unt if the develop-
ment of the regenerated part takes place independently of the
old part, the continuation of the abnormal structure back to the
level of the preexisting portion constitutes a less difficult problem.

According to recently published work of the senior author
(Child, 'ii</) the formation of a new whole from a headless pic i e
of Planaria consists essentially in the formation tirst of all ot a
new head region which then reorganizes the parts posterior to it
through correlation. The structure of the regenerated portion
of the nervous system in the teratomorphic lorms certainly oller>

CENTRAL NERVOUS SYSTEM IN PLANARIA. 55

mon- -upport to this conclusion than to that view which maintain-'
i hat the regenerating nervous system grow- out anteriorly from
the cut ends of the old nerve cords. Apparently in tln-r cases
,i neu central nervous system develops and is abnormal from
the lie-inning, but as its differentiation extend- posteriorly it
meet- the old nerve cords and unites with them. In this \\ ax-
il i- easy to account for the relatively sudden change in the
-iructure of the nervous system as we follow it posteriorly from
the ne\\ into the old tissue in some of the teratomorphic forms
Fij i } and Is, 26 and 27, 30 and 31). But ex'en the anterior

ions oi the old nerve cords undergo reorganization to a greater

or le-- extent under the influence of the nexv region anterior to
them. In bigs. 15, 27 and 31 they are nearer together and con-
m i ted 1.x larger commissures than they were ori-inallx- when
thex ii.nned a part of the posterior region of the In-t /ooid.
The structure of the nervous system in the posterior part of the

ni-rate. 1 tejon in such cases as Fig. 14 and Fig. $<> suggests
a breaking up into separate IUTVL-S , but posterior to the lex els of
iln se -' tions where the developing jjortion meets the o'd cords
there i- a return to something approaching normal structure. It
is possible that if the old cords were not present in such cases
the m u IHTXOUS system would extend posteriorly as a c,,n-i( Id-
able number of separated cords or nerves instead of in the lorm
eharai leii-tic of normal animals.

It \\as pointed out in Section I. that the teratophthalmic and
teratomorphic forms can be produced experinientallx bx de-

!-ing the rate of the dynamic processes in the piece b. l..\\ ,i
n Main rale necessary for the production of normal animals
\\hich i> it-elf not constant but dependent upon various condi-
tions. |'he-e abnormal forms then represent planarian morpho-

esis corresponding to certain rates ,.i reaction belo\\ the
" normal " rate for the existing conditions. The tact that change-
\\hich are primarily quantitative gixe rise to -uch dilleivnces in
structure a- those recorded is important. A- the rate <.t reaction
decreases we see certain parts, c. "... the preocnlar re-ion of tin-
head, decreasing in relative size and finally disappearing and
in the nervous system the bilateral structure of the ganglia
becomes lc-- and less distinct in consequence of the reduction and

56 C. M. CHILD AND E. V. M. McKIE.

disappearance of the median regions. Apparently \ve are jus-
tified in concluding that the reduction and di-.ippearance of
certain parts as the rate of reaction decreases is due one of two
alternatives: first, the reduced or absent part may represent a
relatively low rate of reaction in the normal animal and under
the experimental conditions the rate of the reaction which i>
essential for its formation approaches or falls below what m.iy
be called the morphbgenic threshold, i. e., it does not produce
the characteristic morphological effect. Second, a part may be
reduced or disappear under conditions which decrease the rate
of reaction, not because the reaction concerned in its formation is
directly affected by the experimental conditions, but because
its formation depends upon correlation with some other part
which is thus affected. It is probable, for example, that the
condition of the central nervous system in the abnormal forms
is largely, at least in the cephalic ganglia, a direct effect of the
experimental conditions, while the position, number and presence
or absence of the eyes and the degree of development of the

*

preocular region are to a considerable extent correlative effects.

But however we may account for the results it is a demon-
strated fact that the reduction and disappearance of parts of so
"essential" an organ as the central nervous system can be
brought about experimentally by quantitative changes in ex-
ternal or internal conditions. No absence of chromosomes or
determinants and no germinal variation is necessary for the
production of these abnormal forms, but only a decrease in the
rate of the dynamic processes in the piece, together with the
necessary correlative effects of such a decrease.

It is impossible to leave the subject without some reference
to the '"cyclopean" fish embryos which Stockard ('07, '09, '10)
has recently produced by means of magnesium chloride and
alcohol. The resemblance between these forms and the tera-
tophthalmic and teratomorphic forms of Planaria is striking. In
both cases organs which are normally bilaterally symmetrical
in position show various degrees of approach and in the extreme
types a single median organ develops in place of the two. An-
ophthalmic forms also occur in Planaria and under extreme ex-
perimental conditions completely headless forms also appe.ir.

CENTRAL NERVOUS SYSTEM IN PLANARIA. 57

Moreover, in Planar ia the auricles, like the eyes, show various
degrees of approximation and fusion and in the present paper
it has been shown th.it similar conditions appear in the cephalic
ganglia them-el\e-. In Planaria these monstrous forms can be
produced, not inrn-Iy by anesthetics, but by a variety of con.li-
tioii- tin essential efie.-t of which is a decrease in the rate of the
reaction-- in tln-li\iii^ -ystem. It seems probable that Stockard's
< \.-lope. m embryo- .uid the other intermediate forms between
.ui<l tin- normal animals are the result of a deriva-e in the
<>t reaction r.iilii-r than of any specific anesthetic effecl ot
either m.mne-inm -.ilt-, or alcohol. Moreover, thr double or
p.irti.illv double he. id- which Stockard obtained in some ca
are also p-.idiK .i> -counted for on a quantitative b.i-i-: a !>-
ised i. nr ot reaction means decreased correlation and thi-
... n< lit ion f.ivor-, ph\ -iological isolation of parts and repr.'.Iuciion,
;i- ihi- -enior autlior of tin- present paper has shown i-Ucwlu -n-

Child, 'n

1 IK- problem ol ilu- relation between morphogenesis and the
rate "t reaction in organisms is one of great important-, but it
h.i^ received i-omp.irati\-ely little attention. Current hypoth.
ol .li-\i-lo|uueiit .in.l inheritance scarcely consider the |)o-,i!,ilit\
ot altering tin- rhuructeristic morphological features of t!i organ-
ism bv i-liaiue> in the rate of reaction, but of the fact there < an

bi- no doubt.

\'I. StMMAKV.

1. Tin- ti i.ito|)hthalmic and teratomorphic forms of Planaria
tlnrot<>iff)liala can be produced experimentally by d.. ic.i-inu
the r.ite of the dynamic processes in the isolated pieces below
.1 certain \ari.ible le\el which is necessary for the product!.. n
o| not in.il I- 'i in-.

2. In the-e forms the cephalic region of the IHTVOII- system
.litter- more or le^s widely from that of normal .mim.iU. The
tu.i -.nulioiiic masses are always less completely M-p.ir.ited
than in the normal animals and often only a -in^le - m-lion
de\elop>. In the teratomorphic form- the ganglia are more
.ibnormal thai; in the teratophalmic form-.

V In the normal animals the cephalic ganglia extend a con-
-i.lei.tble (li-i.iiu-e anterior to the eyes and the l\\o -eparate

58 C. M. CHILD AND K. V. M. McKIE.

nerve cords arise near the level of the auricles. In the teratoph-
thalmic forms with partially fused eyes the eyes lie nearer
the anterior end of the gangliei and the right and left portions
are not separated at the level of the auricles. The eyes of the
teratomorphic forms are situated at the extreme anterior end
of the ganglionic mass.

4. The abnormal structure of the nervous system in the
teratophthalmic and teratomorphic forms continues posteriorly
through the regenerated anterior end to the- level of the old
tissue and even the nerve cords in the old tissue may he more or
U>s different from the normal. In some teratomorphic forms
the regenerated nervous tissue apparently begins to break up
into separate nerves a short distance posterior to the eyes, but
resumes the form of two nerve cords in the old tissue.

5. In the normal animal the optic pigment cups open laterally
and the same is true for the teratophthalmic forms with partially
fused eyes. In the teratomorphic forms the cup opens anteri-
orly or antero-ventrally and in some cases more or less to one
side. The single median eye may be connected by two nerves
with right or left portions of the ganglionic mass, or by a single
nerve with either one, or the optic nerve may arise from the
median region of the ganglion. The eyes are also farther from
the dorsal surface in the teratomorphic than in the normal forms
and are sometimes more or less completely imbedded in the

ganglionic mass.

HULL ZOOLOGICAL LABORATORY.
UNIVERSITY OF CHICAGO,
October, 191 1.

REFERENCES.
Child, C. M.

'na Studies on the Dynamics of Morphogenesis and Inheritance in Experi-
mental Reproduction. I. The Axial Gradient in Planaria dorotocephala
as a Limiting Factor in Regulation. Journ. Exp. Zool., Vol. X., No. 3,
191 i.

'nb A Study of Senescence and Rejuvenescence Based on Experiments with
Planaria dorotocephala. Arch. f. Entwickelungsmech., Bd. XXXI.. II
A, ion.

'lie Experimental Control of Morphogenesis in the Regulation of Planaria.
Biol. Bull., Vol. XX., No. 6, 1911.

"nd Studies on the Dynamics, etc. II. Physiological Dominance of Anti-iini
over Posterior Regions in the Regulation of Planaria dorotocephala. Journ.
Exp. Zool., Vol. XL. No. 3, 1911.

CENTRAL NERVOUS SYSTEM IN PLANARIA. 59

'ne Studies on the Dynamics, etc. III. The Formation of New Zooids in

Planaria and other Forms, Journ. Exp. Zool., Vol. XL, Xo. 3. ign.
"nf Die physiologische Isolation von Teilen des Organismus. Yortr. u. Aufs.

ii. EiHvvickelungsmech.. H. XL, 1911.
Stockard, C. R.

'07 The Artificial Production of a Single Median Cyclopean Eye in r the Fish

Embryo by Means of Sea Water Solutions of Magnesium Chloride. Arch.

: I ntwickelungsmech.. Bd. XXIIL, H. 2. 1907.
'09 Th' I )-.. -lupment of Artificially Produced Cyclopean Fish "Tin M

mi Embryo." Journ. Exp. Zool., Vol. VI.. N". j. 1909.
'10 Tin- Intlii'-nce of Alcohol and other Anesthetics mi Embrynnir

iiK-nt. Am. Juurn. Anal., \*ol. X., No. 3, 1910.

EVIDENCE ON THE ADAPTATION OF PARAM/ECIA
TO DIFFERENT ENVIRONMENTS.

LORAXDE LOSS WOODRUFF.

The fact being established that my pedigree culture of
Param&cium aurelia (I.) undoubtedly has unlimited power of
reproduction without conjugation or artificial stimulation, 1 a
culture of Paramcecium caudatum was started for comparison,
in order to determine if this animal would show throughout its
life history characteristics of specific value and also to determine
it it would continue to live and reproduce indefinitely without
conjugation or artificial stimulation.

The results with this culture led me to conclude, as did Jennings
and Hargitt, 2 that caudatum is a distinct species. This point I
have discussed in a previous paper. 3 The results in regard to
the second point arc briefly presented at this time.

The pedigree culture of Paramcecium caudatum (X.) was
started on May 14, 1910, and has been continued under observa-
tion to the present time, December i, i<)ii. The methods em-
ployed have already been described in detail in earlier papers on
pedigree cultures of Infusoria. It is only necessary to state here
that the culture was begun by placing a large "wild" individual
on a depression slide in about five drops of cuhuiv medium.
\Yhen this individual had divided twice, producing four animals,
each of these was placed on a separate slide, forming tin- four
lines of the culture. Thereafter (until June i, H)ii) a single
cell from each of the lines was isolated d.iily in fresh culture
medium and the number of divisions during the previous twenty-
four hour* was recorded.

In regard to the culture of I'arannci iiini aurelia (I.), which

1 L. L. Woodruff: "Two Thousand < < -m -i.itiiui^ of Paramacium." An'liir fiir
Prolistenkunde. Bd. 21, 3. 1911.

-II. S. Jennings and G. T. Harbin: "Characteristics of tin- Divei K.u-i-v ,,i
1'aramtrcium." Journ. Morph., Vol. 21. n<>. |. 1010.

3 L. I.. Woodruff: " Paramiccium aurelia and Paratmecittm ca utlatitin "
Morph., Vol. 22, no. 2, 10.11.

,..

ADAPTATION' OF PARAM.^ECIA TO ENVIRONMENTS. 6 1

served as a control and for comparison with the P. caudal um
culture, there are no results to be recorded which are not in
(in in- agreement with these already published. The culture has
kept on the even tenure of its way and is now, after over four and
"in- half years of daily observation, at tin- 2.7051!! generation,
and in every way in as normal morphological and physiological
londition as at the start. Given a favorable environment, thi-
race dearly has unlimited power of reproduction without con-

iin or artificial stimulation.

I In | .1 d^rec culture of Paramccc ium caudatiin: \ which \\a-
-ubi' - led 1 1 "in the start to the 500 th generations (twelve and one
halt montli-i to identically the same treatment and culture me-
dium a- i he I y . aitrclia culture, showed during the lir-t 350 genera
lion- K'ln months) essentially the same rate of repri.diieti.m

die <!ur<-!i(i culture. However, an examin.iiioii <>l the daia

I i and 2) shows that a slow decline in division rai.

in ai the -iart which finally resulted in a race of cell- PD--I- ini;
man\ "t i lie morphological and physiological characteri-tii - de-
scribed \<\ ('alkins 1 in liis careful study o!" pure line- of ihi-
species "i Paramacium. Alter about tin- 4501)1 ^eneiaiinn it
became iin leasingly ditVicult to keep the animal- ali\e mi the
slide- iii the culture medium which was supplied fiv-h daily.
llo\\e\er. the cells left over from the daily isolations, which were
alli'ued I" acru mil late in the old culture liquid, appeared healthv
and ii'iit'imed to reproduce slowly. It the-e \\eiv transferred

n in lre-h medium they would di\'ide a few tinii-- and then
<lie l-'inallv lhe\- would not live twenty-four hour- in the I'n -li
medium.

l'\ -ub-tiiuting from the "stock" in thi- \\a\\ the direct line-
"I ihe culture were kept replenished for nearK tlirc'e month-; bin
ImalK it uas evident that it was impo ible i" cmitinne the
culture b\ ilii- method, so that the exact number "t ^< in-iation-
could In- determined, and according . at the -,o"ih generation,
the method uas abandoned, and the animal- \\eiv thereafter
aiiicd iii -mall tlasks of old infusions. /. e., lhe\ were bred in a
i "inparati\ cl\ large volume of the same type of medium to \\ hich

1 (",. X. (\ilkin-: " Ilir I ilr Cycle of l'<ii.. .:i<ltltnm." \r,lr

En nismcn. Bd. 15, i, 190.'. "Death of th V Series ol

t onclusions." Jum. l-.\pfr. '/.. Vol. i, no. .;.

62

LORANDE LOSS WOODRUFF.

L

,03 200 300 o son
FIG. i. Paramacium caitdatnm (Culture X.). Graph of the rate of reproduction for the first 500 generations (May 14. t<;i<

June i, 1 911). See text.
The average rate of division of the four lines of the culture is again averaged for ten-day periods. The figures 100, -">... etc
present generations and are placed below the ten-day periods in which they were attained.

i

1

J

r

i

-

I

1 L. u

ADAPTATION OF I'AKAM.E( IA TO ENVIRONMENTS.

they had been previously subjected but a medium which was
from - \ ral days to several w<-k- old.

I'nder these conditions this culture of P. candatum now
Hoiiri-hc-, and it is continued by isolating a few cells every
feu \\rrks and inoculating with them another >mall flask of old
infusion. I nder these conditions it is impossible, of course, to
It !< inline with accuracy the rate of division or the number ol
j< IK r.ni'.n- attained to date, but the organi-m-. are apparently
in .1 normal physiological condition. H<\\e\er. it i> still im-

100

200

300

400

Fie, _. r.uimtriium faiiJatum (Culture X.). Graph of the rate of reproduc-
tion I>T tin- i. generations (May 14. 1910. to June i. i

1 In of division of the four lines of the cultun i- .1 d for

l>iii..'l- ,.i on :. The figures 100, 200. etc.. represent & -m -t.iti..n- ;uil are

IM-|I>\V tin- iiniiitlis in which they were attaim-'l.

]Hi^-,iliU- to keep them alive on slides in tin- n-^ulaiion li\e
<! trc-h ruhure medium, which has pr.-M-d so highly fax
ii 'i the nnn-liii culture.

\n\\ tin- i|iie-iin arises: Have the cells foniu-aied in the
larger \-olume <-t medium and so been "rejuvenated." ^ince I
have been unable to isolate the animals each day. I cannot pr<

thai nmiiiuaiii'ii ha- not occurred, for it i- ]>'>-- iMe thai one Of a

fe\\ pairs ha\e ennjugaifil unobserved ami ha\e given HM- to the

64 LORANDE LOSS WOODRUFF.

present generations. The only way to prove that conjugation
has not occurred is to make the conditions such that it is an im-
possibility for it to occur, ;'. c., by daily Isolations un<l record of
generations. Since the physiological condition of this pedigree
culture prohibited this after the 5OOth generation. 1 have adopted
the method employed by many investigators of problems of this
nature and allowed the Infusoria to accumulate in considerable
numbers. I have, however, in order to increase the accuracy of
the method, confined the cells in as small a volume of old infusion
as possible and have examined the flasks at frequent intervals
for signs of conjugation. 1 have never seen a single pair of con-
jugants in all the multitude of cells which I have examined, and
it seems highly improbable that conjugation has occurred. It
should be emphasized that, if conjugation has taken place, it has
not so altered the physiological condition of the cells that they
will live under the slide method of culture.

This culture, then, is apparently in as healthy a condition as at
the beginning of the work, but it has become so modified that
the animals are unable to exist in small volumes of fresh infusions.
This is a decidedly interesting result in the light of the work of
other investigators on Param&cium caudatum, since it shows that
a race of cells may exhibit all the signs of "senile degeneration"
at the end of a typical "cycle" of generations, and still may
appear healthy and exhibit a normal rate of reproduction when
put under other conditions which approximate what is probably
the usual environment of wild parama>cia.

In other words, this culture of P. caudatum substantiates the
conclusion of Calkins that, under the conditions of his experi-
ments, this organism may pass through a "cycle" which finally
terminates in death; but it further shows that this "cycle" is
probably an artificial one which is brought about by the sub-
jection of the race to an environment which is not suitable for
its prolonged existence. This culture also shows that pure lines
of different species of ParamdBcium (aiirclia and caudatum} are
adapted to different environmental conditions, in virw ol the tact
that the- race of 1* . aitrclia has thrived indefinitely on the same
culture medium which has proved increasingly unfavorable for
the race of P. caudatnm. It may be that I his is actually a specific

ADAPTATION OF PARAM.ECIA TO ENVIRONMENTS. 65

difference, but I believe that the fact that these two races belong
to different species is merely an incident and that it will be found
in be equally a variation of different pure races of the same species

.1- tin n -nils of Jennings clearly indicate. 1

CONCLUSIONS.

1. The discrepant results of various workers on the longevity
it" p.ti. i in. iii. i is in all probability due to variations in the cultural
driii. md- of the races isolated for study.

2. It is probable that most, if not all, normal indi\ idual.- ha\ e,
under suitable environmental conditions unlimited po\\rr ot
i. prodiK tion without conjugation or artificial stimulation.

SHIM ii ii' BIOLOGICAL LABORATORY.
V.M i 1 'NIVERSITY.

I' mi:!', llar^itt: loc. cit.. p. 538. Jennini:-: .\m<r. \atur, i '. \.>1. 45,

, i<n i

Vol. XXII. January, 1912. No. 2

BIOLOGICAL BULLETIN

()\ THE BEHAVIOR OF
TUBICOLOUS ANNKLIDS.

111.

(HAS. \V. HARGITT.

lii two earlier pa|>ers dealing with the general subject of the
. i\ i< 'i <il tube-dwelling annelids the writer endeavored t<> i;i\e
in -onie ilt i. ul .in account of experiments and observations made
upon several species of these worms available at Wood- 1 lole. and
in. Hi ni.ilK made reference to a few observations upon one f the
Viple- -pei ies, Protula protula. During a recent occupancy of
the sniiili-oiiiaM table at the Naples laboratory occa-ion wa-
taken i" i xiend these observations to several otlu-r -peeie-. and
to in. ike .1- i ritical a comparison as might be practieable ol tlic
I K h.i\ ic n M| i IK- la Her \v ilh that ol the \\oods HI >lr -] >< < ii -. 1 1
will IK- noted that in the present account less attention ha- been
vjv en to detail- of time reaction, various stimuli, etc., than before,
and that beha\ ior in relation to light has been enipha-i/ed. Thi-
\\.i- deemed the more important since it was upon ihe-e ^peeie-
th.it -nine o| the earlier work concerned with animal heliotropism
\\a- done. A- ma\ be recalled, my \\ork alread\ ment i< >m <!
'"'. '09 . did not tend to confirm these \ie\\-; ,unl it \\a- \\ith
a \ie\\ to te-t them \\\ a repetition of tin- experiment-^ that 1
underiook to i' \.miine the subject. In the follow in- a(enmt.
will be found the results and conclusions which my ob>er\ atioim
ha\ e -eemei 1 to warrant .

The following are the species which ha\e been n-ed: Protula
protnhi, IJydroides pcctinata (Serpula nncinat<D, Ponuitcxcrn^
tru/uctcr, and Spirographis spallanzanii. The experiment- were
carried on from January I to April 15, a period of three ,md one
half month-. Particular pains were taken to vary the experi-

67

68 < II VS. \V. HANI-TIT.

merits in every practicable way, and under a range of conditions
which would eliminate as fully as might IK errors of inference
based on limited experiments or fauhy en\ ironnu nt.il conditions.
Details on these points will be given in later section;, of the paper.

PROTULA PROTULA.

This annelid is a very familiar element of the fauna of the
Bay of Naples. Its large size, often 175 mm. in length by ,s s
mm. in diameter, its fantastically coiled tube, and the brilliant
orange-red gills which are splendidly displayed during expansion
conspire to make it a conspicuous object. The sensitiveness of
the creature to differences of light intensity, such as that involved
in the intervention of shadows, was one of the first aspects of
behavior to engage my interest many years ago, some brief
notice of which was made in my early paper ('06, pp. 311, 314).
These observations I have verified again and again during tin-
present series of experiments. Careful comparisons of many
specimens in their reactions reveal the fact of marked individual-
ity as expressed in the variability of behavior shown from day
to day. It is not necessary to go into details concerning this
point. What has been pointed out in the case of Ilydroidcs
dianthus ('09) is confirmed in the case of Protula. Certain speci-
mens were especially sensitive and extremely active in response,
while others would show the very opposite; and it was not un-
usual to find specimens which seemed totally indifferent to
shadow stimuli. Again, specimens might prove quite sensitise
at a given time and very indifferent at another. But let it be
noted that some specimens seemed normally to be highly sensi-
tive, while others seemed normally quite tin- opposite. Again,
the retraction aspects of behavior, that is, the time a gixen
specimen remained in the tube after a given contraction, \\.is
remarkably variable. In some cases the emergence was rela-
tively prompt, while in others it was extremely slow. In this
matter Prolnla differs materially from Ilydroidcs, whose retrac-
tion periods are usually and normally very briel. Protnla ohen
remained retracted for indefinite periods, otten tor one or t\\o
hours at a time, in marked contrast to Hydroides.

Tubular Aspects. The behavior of Protula as e\]>iv ed in the

ON THE BEHAVIOR OF TUBICOLOUS ANNELIDS.

6 9

form <>r aspects of the tubes is noteworthy. In my early paper
u.i- -lunvn a figure which made this very graphic. Xo less than
in tin- case of Hydroides the tubes of Protuhi show the record of
erratic behavior in very striking manner. i.Cf. '09, pp. 180-
[83. I Miring early life these tubes usually adlu-rr very Mruiisjy
and closely to the base of support; but in maturity they ohm
in< line in-.n- or less toward the vertical, though in a ratlur sinuous
fir - 1 lira I din-, i ion, or may even mil about rarh otlu-r and assume

l-ii.. i -li.,u . colony of Protttla. with Spirographis'w tin-

I li>- pi' : of the tubes is very m.ir kc-<l.

p. tit.

a do\\nu.ini a-|>ivt. This may be seen nm-t -trikin^K' in the
l.ii-r colonies i these creatures al\\a\^ ]>IVM-IH in tin- >how
aijiiaria ot' tin- lal IOIMI ory, where may .il>f> br M-i-n to biM dtCft
tlu- marked \ariability as to tubular brhaxior. Something of
tin- i^ ^iMpliicallN >ho\vn in 1'i--. I and ;-,. ropird li-oin |)lnto-

JO CHAS. W. HARGITT.

graphs taken by Dr. Sobotta, by whoso kind permission I am
able to present them here. As will be seen, the aspects of the
tubes and of the openings through which the creatures protruded
their heads are so extremely diversified as to seem to be abso-
lutely chaotic. If one may distinguish any tendency toward a
given aspect of position, still the departures are so numerous as
to render it almost certain that no single factor could have been
determinative. As in the case of Hydroides ('09, p. 180. etc.),
Pro! nla has left in ii> tubes a convincing record of the erratic
individuality of its behavior the significance of which is ex-
tremely important.

Autotomy. In this connection may be considered a feature of
behavior more or less unique, though not peculiar to Protula,
since it has been noted in several cases of Spirographis] namely,
that of autotomy, or the self-excision of certain organs of the
body. This was first observed in the case of Protula. A speci-
men of this worm was among the first to come under my observa-
tion, having come to my table almost the first day in the labora-
tory. It had been placed in a small aquarium jar on the table
for convenience of study. After finishing a given series of tests
the specimen was usually returned to the large aquarium. On
January 7 the specimen had been under observation and was
Ic-ft in the table jar which had a capacity of about four or five
liters, while I went out to lunch. This could hardly have been
more than an hour or so, but on ivturn I observed what seemed
strange the detached portion of about half of the gill mass lying
at the base of the tube. An examination of the gill failed to
reveal any signs of disease or other abnormality. My first im-
pression was that possibly the water had become "bad," yet
other living things, such as copepods, showed no signs of dis-
comfort. However the water was renewed several times during
the afternoon and the specimen finally left on the table over
night, as had been done several times before. The following
morning upon examining the jar I lound tin- oilier portion of the
gill in I he same detached condition, lying at I lu- base of t he tube,
but the specimen was deeply retracted within the tube. After
some time it came to the orifice and showed clearly that it was
entirely devoid of gill elements. It was now transferred to the

ON" THE BEHAVIOR OF TUBICOL* >l - ANNELIDS. ~l

large aquarium and left to see whether regeneration would occur
and if so at what rate. For a few days it frequently came to the
orifice and extended the mantle edge over the margin of the
tube and remained in that condition for some time. Later it
again withdrew deeply into the tube and did not .-how it-elf for
-e\eral da\-, indeed for some three weeks or more. Finally, on
l-Vbniary 13 it was once more seen to protrude it- head, but
then- was not the least sign of any regeneration It- appearance
m>\\ bee, mie more frequent and occasion was taken to test it by
-had' >u -, and to my -urpriM- it was found to react as promptly as
at tin- \vr\ first. These tests were made main time- on subse-
quent da\- with the almost uniformly prompt and po-iii\e re-
ai lion, but with the variations observed at first, /'. c.. -ometimc-
lc-- -harp, and thru more so, and occasionally not at all.

Several interesting questions arise- in this connection. I ir-i.

io i he regenerative capacity of Prptnla. For lu.uK tun

month- mil a sign of regeneration was distinguishable. 1 had

pie\iou-ly recorded ('06) tin- promptness with \\hieh Ilydronlcs

:ated ii- gill-, and similar records had also been made 1>\

/elen\ . I inalK on March 14 1 found undoubted e\ ideiice ol

n< raiion. and thi- went forward apparently rather rapidlx .

ti.i b\ April 10 the new gills had become quite om-picimu-

nearK a tilth as large as the originals, h mav be noted here

that Liter I had also a similar autotomy by another -pecimeii

ot /'"'//i/i/ and by two sjK'cinu'ns of Spirogra (this. In tin-'

there tdiild not be doubt as to any condition of water inducing

the auiimy, for the specimens continued to thri\e. as did

man\ otln-r- in the sanu- lank. Re-generation u.i- very pn>m]>t

and rapid in these cases.

\ -e.-i.nd |toint is in relation to such c.ne|.iti..n of funcii.m
.1- ui.uld enable tin- creature during lhi> IOIIL; ]>eri'd (' gill
depri\aii'>n to maintain the normal rc--piratm \ activity. It
n--|iilalii>n \\ere restricted to the gill- almie nf course it nui-t
ha\e i>eri-ln'd. This experiment sho\\ - learU that thi- func-
tion may be taken up by other organ- of the body without
serious inci>n\enieiuv. But the gills are also concerned in the
process i nutrition, acting as a medium for capture of prey.
How might this function have beeen Mippleim-nu-d.-' It has

72 CHAS. W. HARGITT.

been said that during this time the specimens remained rather
continuously within the tubes. Did they depend wholly upon
a reserve food supply?

It may not be possible to answer these queries fully, but of
the correlation of the skin in the function of respiration there
can be no serious doubt. In my earlier experiments on Hydroides
('06) it was found that when the gills w r ere excised to test their
relations to sensory reactions the creature did not seem to suffer
any serious inconvenience as to respiration. So in the case of
Protida, there was no evidence to the effect that respiration w r as
not normal during the long period of gilless life. Bounhiol
(1900) has reached similar conclusions from experiments on
Spirographis. He finds that respiration takes place through
both skin and gills, and that they supplement each other by
compensatory interaction. He finds also that it is apparently
easier for the gills to assume extra work than for the skin, and
that in excretion of COz the skin normally excretes about three
fourths of the entire amount.

In the third place, there is the interesting query as to the
sensory function. I have shown that for Hydroides light per-
ception is almost exclusively a function of the gills. In Protiila
this is not so certain. Its behavior in this respect is less easily
controlled, owing to the sulking disposition of the worm. But
it is quite certain that autotomy did not result in entire inhibi-
tion of reaction to shadows and it may not be improbable that
something of sensory compensation may obtain in this, as in the
respiratory activity; or possibly this sensory function may be
shared in part with sonic other head-organ, possibly the mantle
margin, which in normal lite is often extended over the orifice of
i he tube, hence in a position admirably adapted to such a function.

Concerning the entire matter of the significance of autotomy
little can be said. Such phenomena, similar in many respect^,
are well known among other animal groups, though not common
in any case, unless we may include phenomena of fission which
is a very familiar feature in many annelids; but this MVDIS to be
a wholly different problem. That it is spontaneous, hence not
attributable to the operation of gravity, contact, etc., seems
very evident.

ON THE BEHAVIOR OF TUBICOLOUS ANNELIDS. 73

POMATOCERAS TRIQUETER.

This, with an undetermined species of Serpnla, are tubicolous
annelids which much resemble in general aspect of size, structure
and behavior, Hydroides dianthus. Indeed in almost every par-
ticular they might have been substituted for the latter species
without marked changes of results. In general habitat the two
species arc much alike and often found growing on the same sub-
Mr.it urn. Pomatoceras is rather larger, and its tubes are charac-
teri/ed liv rather sharp triangularity with the dorsal angle form-
ing a sharp crest along the entire tube. Mo attempt will be made
to go into details as to matters of behavior, since as already
suggested, they show the same reaction phenomena as those
given in tin- accounts of Hydroides dianthus, and tin- growth
aspects are <|i' % as erratic. For the most part the tubes adhere
cloM-|\ to tin- substratum, and in some cases they a<lju-t thcin-
sel\e- \\itli -in h nicety to grooves or similar depressions that one
might gin -- ilit-\ wen- under the control of some such stimulus a-
thigmotaxi- or stereotropism. But when one com*- to examine
an\ i on-idt -rable number of specimens he soon finds that in
l>\ tar the larger number there is absolutely no Mich adjuM-
ment. The -ame is the case with Hydroides. Now and then a
s|)ci imi-ii mav In- found on a shell of Pecten in \\hich there i- a
verj tine illustration of such appearance, the creature ha\ing
kept \er\ i lo-rly and exactly in the radial groo\- ,,| th,- shell.
Hut on the same shell may now and then be found another
specimen \\liich has grown directly across the- grooves; and
ol course \>\ far the larger number have al>-"lute!v no -nnblance
oi -neh re-ponse. The conclusion is therefore forced upon one
that the operation of any such factor must be, i! not ulmlly nil,
\et ot onl\ incidental significance in behavior.

Again, in habitat one finds in the Mediterranean species the
same \\ide r.m-easin those of Woods Hole. 1 have dwelt upon
tlii- |)oint \\ith some emphasis in a lormer pa|>er ('oi), |>p.
i^j $). and need only refer to the matter in thi- connection by
\\a\- of further emphasizing a point which I regard of considerable
significance. The growth of these organisms indiscriminately
on a large \arietv of substrata, rock-, shells the latter both
dead and living nets, crabs, lobster^, etc., i- it-elf of no small

74 ' "AS. \V. IIARC.ITT.

import as to the negative influence of" such factor^ a< light,
gravity, etc., in relation to growth. This is further borne out.
by attention to the aspects of the several tubes which may com-
prise a given colony. In several such an actual count of the
growth direction was made. On a stone which contained 37
living specimens I found that 12 had a general upward direction;
15 had just as definite a downward course; and 10 had a hori-
zontal direction. Another colony growing on the inside of
an iron cup about 6X10 cm., made up of eleven specimens.
showed the following disposition as to direction: 4 upward, 5
directly down, and 2 horizontal. On the outside of the same cup
were eighteen specimens disposed as follows: 8 upward. 7 down-
ward, and .} horizontal. These plainly go to confirm the con-
clusions already drawn, that in the matter of orientation one is
utterly unable to discover the operation of any one or several
factors which are in any sense determining.

HYDROIDES PECTINATA.

This species and the one described in the following section,
Spirogaphis spallanzanii, were the ones used by Loeb in his well-
known experiments at Naples many years since, the results of
which, including also several species of hydroids, served as a
basis for his far-reaching theory of animal heliotropism, especially
as it relates to sessile animals. Naturally, therefore, more of
details will be expected in 'the following accounts than in the
preceding, and I shall endeavor to make explicit and definite
records both of methods and results.

Ilydroidcs pectinata (^Serpiila uncinata) is one of the most
common and abundant of the Naples annelids. I nlike Spiro-
graphis, it grows usually in immense colonies, aggregating hun-
dreds or perhaps thousands of individuals. Fig. 2 will give a
general idea as to their appearance in small colonies. The tubes
of a given colony form a mass of more or less parallel aspect, the
individuals apparently growing at approximately the same rate
and in the same general direction. When one casually examines
such a colony it \\otild seem to afford a typical illustration of
orientation due to some single constraining stimulus. Hut here
again, as in the case cited above as to stereo tropism, extended

ON THE BEHAVIOR OF TUHICOl.orS ANNELIDS.

75

ol>MT\ation brings to li^lit too many exception^ to any such
rule, and compels further inquiry. Like other species of 7/v-
<lroide.s this one secretes a calcareous tube, the shape of which
<1 -pi -nd- upon the mode of growth of tin- individual constructing
it. Hence in aspect, size, etc., these become permanent records
ol rvrryduy behavior, whether thi- be mf.-hanic.il, ecological, <,r
physiological in its nature. Several coloim- \\i-re put under

-Ih'ws three colonies of H\dri,: :.it<i whirh 1.

lixlit U'-t i. n in. nr than a month. As will be seen the curvings of the tubes

iinii h u- in \ \ + . I .

i \ .iiion c,irl\- in January arid urn- u--trd during a period of
mu<- than tluve months; to be more exact, lunn |.iim,ir\ .^ lo
April i:r. 'l"he\- were tested as to the po-^ibK- inlliu-nce of both
li.uht and i;ra\it\-. Loeb claims that in thi- -pccif- tlu- reaction
i- .|iiitr prompi. ' I noticrd that in tin- course of the nc\i day

76 CHAS. W. HARGITT.

the Scrpulidip, which like Spirographis presented only their gills
to the light, bent them strongly upward" ('90, "Gen. Phys.,"
Part I., p. 102), and he continues, "within six weeks the entire
block was covered with tubes which curved upward; not a single
individual had continued to grow in the original direction," and
presents a figure in illustration. There is apparent discrepancy
between the latter and the statement " not a single individual,
etc.," for in the figure about as many appear to "continue to
grow in the original direction" as have "curved upward." My
own experiments show a reasonable conformity to Loeb's figure,
but the ratio of tubes indicating reaction is very much smaller.
Figs. 2, a, b, c, are photograph reproductions, and may therefore
be taken at their face value, and they certainly fail to show any
such response as claimed above. For example, it w'as found by
actual count of a colony comprising hundreds of tubes which
had been under test for more than a month that only about
twenty tubes had definite curves toward the light, a similar
number had shown lateral curvatures, and a smaller number
had curved downward; but the larger number "had continued
to grow in the original direction." A smaller colony which had
been under test for twenty-five days under particularly favorable
light conditions showed a slightly larger response toward the
light; but here also the number was relatively small. Another
colony was placed in an aquarium which was covered on three
sides and above with a black hood. After a test of nearly two
months (January 23 to March 18), it was found by a careful
estimate from counting that at most only aboni Jo per cent.
showed any possible light reaction ; while by far the greater num-
ber either continued to grow in tin- original direction or showed
curvature laterally or downward. Tin- colony was submitted to
two others working at the laboratory, I)r. Butler, of University
College, Dublin, and 1 )r. S. k. Williams, of Miami University,
Ohio, both of whom made the per cent, ot light reaction much
lower than my own.

A very interesting and, 1 believe, significant leature of growth
in this species came to light during the observations, nainelv, it--
very erratic, or discontinuous character. Some individuals
showed a very prompt and rapid growth at lir>t and later its

ON THE BEHAVIOR OF TUBICOLOUS ANNELIDS. J7

cessation. In this process of rapid growth some show a bending
while others do not. Again, some bend toward the light, others
away from it, and still others continue in the original direction.
The point of importance here is not the bending or curving, but
Dimply the tube-extension. This extension is not, as I interpret
ii , an expression of growth at all, so far as the body mass of the
animal is concerned. Seldom are aquarium conditions especially
< niidiici\c to physiological growth. What then does such tube-
t -xten-ii MI IIH Mil.- 1 Isolated worms lying -ide by side, of essentially
similar age, state of vigor, under identical conditions, show the
most remarkable differences in relation to tin- matter of so-called
gro\\th. < >nr may in the course of a \\eek extend it- tube 3-5
mm.; another -hows not the slightest extension of its tube. < >n<-
mav extend it- tube in the line of the body axis, /. .. -traight, the
oilier -ho\\ .1 -harp curvature from the first. There has been
e.jiial access to food, air. liglit. Why ha- not gro\\th been the
-ame in direction and amount? As a matter o| laci it may In*
doubted \\ lieiher there has been any appiei i.iMe gn>\\ th. Indeed
' mav not these erratic phenomena expre ju-t the oppo-ite,
namely, 1m k of growth conditions, or some other ia< tors > onducive
to lointcirt.-' And if so then this erratic tube-exten-ion i- but
an expn---ioii of such discomfort, an expre ion o| the elloit-
iil th. . leatntv to seek better condition-, to n,i< h oiii , as it \\eiv,
in -earc h o| i.i.id, air, etc. Indeed, it' my interpretation be cor-
rect, the-e ( urvings are but the natural expre i< m- o| eitort-

at (...id-get tin-^ or respiration adjustments to tho-e particular

end- in\ol\ed in -nr\i\al or selection. In the light <>l this
intei pieiai i"ii the real factors in\o|\e<l in these aspects ot bc-
ha\ ior are intrinsic and not cxtrinsit . The lai ter are conditional
ami pa--i\ e; the trmer are indi\ idual. acti\ e. causative.

SPA] I \\/\II.

Tin- species lit't'ers most markedly from those already con-
sidered ill that it possesses a very flexible tube, hence i- capable
of considerable range of movement within a region mca-nivd
I >v tin- rad in- ot it- own length. It i- a large species at maturity,
averaging perhaps about 25 cm. in length, by about I cm. in
diameter. In im experiment- care \\.i- taken to ha\e -pccimens

C'HAS. \V. HARGITT.

of various sizes, and those actually used ranged from about 5 to
,}o cm. in length. No less care was exercised as to conditions
under which experiments were conducted. Three of the l.irge
aquaria supplied with running water were at my disposal during

Fir,. 3 is a portion of one of the- large show-aquaria containing Protulu and
Spirographis. The varied aspect of the latter is quite marked, as will be seen by
comparing specimens in various positions.

tin- period, and in addition l\vo special experimental a<|iiari.i of
smaller size, about 25 X 35 X 45 cm., also supplied with running

OX Till hi IIAVIOR OF TUBlM'i.nrs ANNELIDS. 7<)

, were employed for such experiments as called for a critical
control of light, etc. Of the large aquaria two were in a room
with north exposure, and hence with diffu-e light. Inn al\\a\-
adequaie for ordinary observation and experiment: while tin-
other was in a room with direct south exposure, In-mv with -un-
liyht ot almost any degree of intensity, modified by -hade- 01
screens. I In- iwo smaller aquaria \\eiv in a special room, the
light ol uhich was under easy control, and tin- aquaria them-cbe-
ea-il\ adjusted to any desired condition. a- to amount and direc-
iin o| li-ht. t-ic. A still further point i- \\orthy of notice At
all time- I had In-i- access to the latxe exhibition aquari.i, win-re
laro- number- ol these specimens \\eiv living under condition-
as iit-.irlv natural a> the long experience and painstaking -kill of
tho-e in <h.iro ha\'e been able to devise. 1 -hall ha\- O( .i-ion
io r.i. r to thi> in another connection.

I hiring tin- pio^ress of the experiment-- ^mie hlt\ ->|M-cimen-.

\\ i te . i\ ailal ile. .UK I tin- general healt h and \ i^or m.i\ In- intei "i < I

lioin tin- I. iet th.n in the three and one half month- not .1 >in-|-

-pi -i imeii died oi e\en showed signs of detei i. n ,n ion. except .1- a

in I. id i i re of the brilliance of coloration ma\ ha\e been indica-

ti\ e ol -mil. t are was taken to suppK 1 1 from time to time,

.iluio-t d.iil\. -uch as came in from plankton haul- \\hich \\eic
-up|ilied to tin- lo,.!))- quite regnlai'K . and this m.i\ ha\e con-
tributed to tin- . \. ellent conditions of health already alluded to.

.^[>iri>^nif>hi.\ -eem> to take ratln-r natnralK to the aquarium

environment and soon becomes quite at home so far as one may

judge I mm appeal .nice. >peiinien- reqmri- Irom I \\ o lo several

da\- lirmK \<> att.uh ihem-el\es to the bottom or sides of the

aquarium. Thi- i- accompli -lied b\ an adhe-i\ e se< r< lion ol the
\\orm \\hich i- di-char^ed through a -mall pore at the l<i\\er end
of tin- tube. Tin- time required for attachment max be \aried
\>\ having the bottom of the aquarium covered \\ith .1 la\er ol
>and or b\ |ilacinu fragments of rock in contact \\ith the ba-e o|
the tube-. While in most cases tin- -pecimen- attach tln-m-el\e-
\\ln-n-\er tlu\ hapj>en to be placed, which i- fortunate in -uch
experiment- a- tln^e under consideration, -till it not infn-queiuly
happen- that a specimen will go through various irau-lator\
movements before finally settling do\\n. It ma\ be noted that

8O II \S. \V. HAKC.II I .

these locomotor movements take place usually during the night.
This I have demonstrated by carefully marking ldc.it ions and
noting subsequent changes. At no time have 1 found evidence
of these movements during the day.

In general my experiments proceeded along the lines employed
by Loeb ('90, Arch.f. ges. Physiol., Bd. 47, p. 391), who>c objec-
tive aim was to establish the essential identity of heliotropism in
animals and plants, and his experiments were directed to that.
end. Incidentally it may be observed that he does hot hesitate
to claim "I think I have shown that the heliotropism of sessile
animals is essentially identical with the heliotropism of sessile
plants." And still later he asserts even more strongly, "It was
possible to show that heliotropism of animals agree in every point
with that of plants" ("Comp. Phys. of Brain," 1900, p. 181).
It may be doubted whether, in the light of present knowledge,
this would be seriously maintained. I shall not discuss the mat-
ter here further than to say that my own experiments were
undertaken with a very different aim, namely, that of ascertaining
the questions of fact, Are these organisms heliotropic? and
further, Do they exemplify, or conform to the mechanical concept
of behavior?

In the following account I shall present the matter under some
three distinct series. First, those experiments made in tin
aquaria located in a north room; second, those conducted in the
smaller- experimental aquaria; and third, those conducted in the
large aquarium located in a room with exposure to direct sunlight.

The first series began on January 6 with some six specimens.
To these additions were made from day to day, till on the i^th
I had twenty, which had been variously distributed in the two
large aquaria, some with the- heads directed away from the win-
dows, others directed at right angles, and still others facing the
windows. The aquaria were ol about the same si/e, probably 1.5
meters in length, by about 40 cm. in depth and width; the one
with its end toward the- window, the other with its side toward
the light. It was some time after specimens had become at-
tached before any sign of orientation was discernible. In the
aquarium (\o. II with the end directed louard the light there
were twelve specimens, in the other eight. The twelve had

ON THE HKII \\lok i >| it BICOLOUS \N\ELIDS. 8l

been distributed so that three should face toward each of the
-ide- ot the tank; i. e., three with head- directed toward light,
three auay from light, and six at right angles, three facing each
-ide. < )n January 26 all specimens weiv attached except one,
which tor some reason, perhaps injury, remained tree during the
entire course of the experiment, hence may he disregarded. A;
this date the following is the record of orientation. The three
facing tin- light continued in that position, one of which had
a imied a nearly erect attitude; the other t\\o had barely cle-
\ati-d tin- head to a degree sufficient to allo\\ tin- gill- to clear
ihe lioiiom (jf the tank when expanded. Four specimen- n<>\\
face tin- \\all, and all with barely sufficient up-bending to tree
iln- uilU Ir-.m the bottom. The laterally directed -pecimciis
toiitiiiued as at first, except thai one had made a distinct up-
Clirve, the hi-.id elevated to an angle of about .^5 degree-.

< MI I . -lu-iiary o the record of this tank is as follo\\-: < n speci-
mens facing light two are cursed upward, one neaiU \ertical,
peihap- 70, the other about 45, while the third remain-, a-
beloiv. and this in spite of the fact that direct light i- inter-
cepted li\ a tufaceous mass bearing tubes of I'rotulu, etc. The
tin -pi-< iini-ns facing the wall ha\e made con-idci able char
>IH- had rotated through an arc of about loo d< -gn -, \\\\ tai in-
ihe -id< . .uid \\iih head elevated about 35 d \notlier

ha- aU<> n>iatel to ne.irly right angles and cur\ed ujiuard 50
degrei The other t\\o continue unchanged. The specimen^
latn.ilK di -posed continue essentially as betc.it. excepl an up-
curve ol from 30 to 40 degrees. Then '"id for this aquarium

on I'ebinaix 25 is as follows: Five spe< inien- are imu la.
the \\all, three continue to face the light, uhilc the others con-
tinue e eniially as before.

I e lolliiuing reconls of the behaxi'T of the other aiinarium,
\\hich ma\ be called number two, are inteiv-iing. In thi- were
placet 1 rigln -jiecimens, two of which ueiv -i i-| tended head down-
ward, and in this position they attach them-el\e- ami continued
f<.r nian\ \\eeks. The others were locateil \\ith heads pre-
dominantly touard the wall, /'. <-., away tnun source of light,
onl\ one lacing light. In this tank but little -ign of light rea. tion
was ili-iingni-hable. The specimen originalK fating the light

82

CHAS. \V. HAKMTT.

later curved to the wall and remained in that position during
the entire time, xvhile one of the specimens placed lacing the wall
later curved toward the light >ide.

FIGS. .4, It, (.', /) illustrate certain aspects of a specimen which was suspended
head downward. At .1 is shown the first indication of change of position; a further
change is shown at B; this curvature has reached its limit in that format ('. ;m>l
continued thus for several days, oscillating somewhat from side to side, but with
no evidence of reaction to light. At D the sickle shape is converted into tin- loose
spiral, which likewise continued for some days essentially as shown in diagram;
as in the others, there was shifting and change but with no relation to

OX THE BEHAVIOR OF TUBICOLOUS ANNELIDS. 83

The most interest attaching to this experiment is the behavior
of the specimens suspended. For several days both remained
hanging downward. Finally one began to curve, and direct 1\
by a graduated process assumed the aspects shown in the dia-
gram figures A, B, C and D. In the entire course of the exprri-
incnt ilx-re was not the slightest indication of light response,
nor indeed was there more of a geotropic character. The final
altitude was that indicated in D.

The nher suspended specimen attached itself to the side of
thr M\( rtlou tube, and has continued head down, without appre-
eiable change of aspect, the tube remaining almost perfectly
-iraii;lii from first to last. Both specimens seemed equally at
, both equally active; but the one passed through the series
dl tubular i "iitortions, the other remained absolutely indifferent.
lin ideutally it may be remarked that specimens are often found
in nature attached to the under surface of bottom of boat- or
other -ubstrata, much like barnacles or other sessile organism-;
and hence it mu-t be admitted that there is nothing especially
unusual or unnatural in such an attitude. That the beha\ ior
of i he one difYered from that of the other is not more str.r
ill. in that differences likewise appear between others.

Sf>f< i<il Aquaria. The second series of experiments were con-
durt. -d in t\\o special aquaria, mentioned above, and were
piompted by two considerations First, the apparently negative
ehaiaet. r of the experiments began and carried forward in the
lar^e ai|iiaria. It had seemed as if one should have more prompt
and eoii\ in. ing results than appeared in the account just given.
"v . ou.lly. it was desirable to have aquaria of a size and adjti-t-
meiit \\hieh made possible ready and effective control at all
time-, \\ith such variation of tests as seemed desirable, lieu.-.-
the-e -mailer aquaria already described. They were set in a
room \\ho-e light and temperature were under easy control, and
\\.re t hem-elves of a size which enabled one to shift the position
at anv time it might seem desirable. It occurred to me that
po ibl\ the fact that in the first series the light had been ditfu-e
rather than direct might have resulted in thr -omeuhai ne^ati\e
beha\ior already noted. Again, it seemed de-irable to be able
ID rontrol both the direction and inu-n-ity of the li^ht Ac-

84 CHAS. \V. HARGITT.

cordingly the special aquaria were made use of, and the following
account is based entirely upon the behavior under tin- new con-
ditions. Two were used for the definite purpose of making of
one a control of the other. That is, given identical conditions
of temperature, food, etc., will the mere difference of direction
or intensity of light show itself in such measure as to warrant
conclusions?

This series was begun on January 15 with twelve specimens,
eight being placed in the experimental tank, and four in the
control tank. The bottoms of the aquaria were covered by a
layer of rather coarse, black sand to facilitate attachment, and
at the same time to render any access of light from the bottom
impossible. The test tank was covered on three sides and the
top with an opaque hood, painted black on the inside and so
adjusted as to render inspection easy without disturbing the
specimens. In this tank the eight specimens were placed with
heads facing away from the source of light. Similar disposition
was made of the four specimens of the control tank. In about
three days the specimens had apparently attached themselves,
and on January 19, four days after beginning, one specimen began
an upward curve. On the 2ist several had shown such reaction
and by the 25th several had curved upward to from 25 to 50
degrees. In the control tank similar responses began to appear.

On January 25, ten days after beginning, the record is as
follows: Of the eight specimens two have curved toward the light,
two are nearly vertical, two face toward the side, while two remain
as planted. Essentially the same condition obtains in the control
tank. One faces the window, one nearly vertical, and two as
originally located.

At the end of four weeks, February i i, three show apparent
light reaction, two are nearly vertical, two remain facing away
from light, and one shows an indifferent curve laterally. The
positions in the control tank remain as before. Repeating Loeb's
experiment at this point, I no\v rotated the aquaria through 180
degrees, so that everything was changed directly about. Con-
ditions went on as before, the test tank receiving light exclusively
from one end, the control receiving diffuse light from the room
as well as the direct light from tin- window. On February 25,

ON THE I;i;il\\!"K < ! TUBICOLOUS \N\KI.IDS. 85

or fifteen days after the aquaria had been rotated, the conditioti-
.in- ,i- follow-,: Five specimens now face the light, while three
I". u i the opposite direction. But of the five now facing the light
three were so placed in the readjustment made by the rotating.
or n Aer-ing of the tank, so that only two ha\e actually shown
a po--il>le light reaction. The three specimens \\hich had been
turned auav from the window by this reversal had not sho\\ n
i li' sliijiu-i response.

At i hi- i ime the acjuaria were again reversed, so that they came
back to the original positions. It should be noted that in the
control i. ink there had been no change induced by the re\er-al
ol tin pi -i i ion, the specimens all remaining as before.

Another a-|>ect of behavior may be stated in thi- connection.
n.imel\ . an actual downward curve of several specimen-. It was
on in-t notice thought that possibly this might be due to the
iin inn -in water, which happened to be in the region ot one such
llouever. it was later observed that other specimens quite
remoii- -houi-d the same thing, and on comparing similar con-
dition- in the exhibition aquaria it was found to have it- counter-
part there, 1 1' nee it may be regarded as only another cxpre--ii.n
\ the individuality of behavior which is more or less e\ident

iimler all i "minions.

The e\pi liments \\ith these special aquaria \\eiv continued to
M Mch J5, having thus been under operation for about ten weeks
i JannaiA i ;> to March 25), and have been in the present pi .-it ion
loi e\aitl\ one month. During this time there ha\e been inci-
dental -hillings on the part of various specimens, a bending this
\\a\ or that from time to time, but only to be re\-r-ed later, or
counterbalanced by opposite reactions of adjacent s|iecinien-.
The-e ha\ e been noted from lime to time during the o >ur-e of all
the experiment-, and are not to be considered as orienting re-
action-, but rather expressions of the individuality ot behasior
charactei i-t ic. a- I believe, of .ill grades ol animal behavior.
Tlu-\ correspond to what Jennings ha- de-ignated as trial rc-
(udons; and in the present instances probably relate to fo.>d-
seekinj or n-piration. These statements refer directly to
condition- in the darkened aquarium; but they are quite as
applicable to those of the control aquarium, and indeed, the

86 ('HAS. \V. II \Ki,iTT.

behavior of the- specimens in thi>, \vhile differing in various
details, have shown a striking similarity to that of specimens in
the former, as well as that of the first series in the large aquaria.
As remarked in the outset, the entire series of experiment-- have
involved no appreciable deterioration of the health or vigor of
the specimens. As an evidence of this may be mentioned tin-
fact that one very young specimen among those used in tin-
control tank showed an apparently continuous growth, haxing
nearly doubled its original size. The growth in this case seems
to have been real and normal, and not apparent as was the case
with Hydroides, mentioned in a previous section.

Third Series. Early in March it was found desirable to change
rooms in the laboratory, and I came into possession of one
admirably adapted to light experiments. Advantage was taken
of this circumstance to continue the experiments with Spiro-
graphis under light conditions which were exceptionally good.
In the room were two large aquaria, one of which I devoted
exclusively to this experiment. The aquarium was arranged with
its side facing the window and at a distance of about two meter-.
By covering the back, ends and top of the aquarium \\ith a
black opaque screen, and with windows also provided with ad-
justable shades, one was able to directly control the light condi-
tions at will, as to source, directness and intensity. The experi-
ment was begun with eight specimens, all of which \\en- placed
with heads facing away from the light, and two other- suspended
head down by attaching them to sides of the o\ erllow pipe, as
in the similar experiment in series I. Other specimens were
added a few days later making a total of twenty comprising the
experiments. As before some two to four days were required
for specimens to become attached to the aquarium. In tin-
present case to insure prompt and precise location several were
secured to a given place by putting over the terminal base o|
the tubes a small weight, such as a shell or rock fragment.
As before the first indication of reaction was the usual upward
curve of the oral end of the tubes, enabling the creature to I reel \
expand the gills. This reaction has little, it any, relation to
orientation movements, as it occurs usually in all cases and under
almost all conditions, whether in light or darkness.

ON THK BEHAVIOR OF TUBICOLOUS ANNELIDS. 87

On March 25, ten days after the specimens were installed, only
had assumed a nearly vertical aspect. Others showed
variou- phases of orientation, from ten to twenty, or thirty, or
fifty dej of elevation above the bottom.

< )n April i. the following is the record. Four -pecimen- with
:^ill- directed more or less toward the light; two with a vertical
attitude; three oriented at right angles to direction of light, and
facing darkest end of tank; nine remain oriented in original
po-ition, /. i\, facing away from light. The t\\o -u-pemle<l -peci-
ini 11- lieha\e almost exactly as in the previou- case; that i-. one
pcr!e< il\ unchanged and the other curved a\\a\ tn>m the pipe.
Tim- .tiler m-arlv two weeks half of the entire lot remain abso-

l-ii. i i- .in finl view of an experimental aquarium, tin- li.^lu o>min.n
iiv;lit -iik .it * ' n tin- i-ii;ht specimens shown only niu- i- :.n inv; tin- liKl
\<-i tii .il. the others facing the dark side of tank.

is

hilelv unrli.ui-.'d ; of the others onl\ ti\e ^Imu any \IT\ clear
rea( lion to p.^ibk- light stimulus. The experiment- o mi inued
under d.iih observation until April \2, a |)erio<l ot one month,
\\iili a tmal remnl as follows: Four -pet inien- -lm\\ a distinct
curvature t<>\\anl the light; nine show ju-i a- distinct inclination
a\\a\ tioni the light, in other \\oriU remain as ori-inalK fixed

except the slight curvature upward ; two are almost vertical ; the

other three occup\ portions at right an^le- t<> the line of light.

CHAS. \V. HAR(iITT.

The two suspended specimens continue as before, one absolutely
as at first, the other with a definite crescentic curvature, but
forty-five degrees away from light. Fig. 4 is from a photograph
taken ly Dr. S. \\. Williams and gives a good impression of the
orientation of such specimens as came within the view. It is
taken from the end in order to show the relation of the tubes
to light, which came directly from the right and into that side
of the aquarium. Of the twenty specimens only eight are shown,
and of these only one faces the light, one is almost vertical, the
other six incline very definitely toward the dark side of the
aquarium.

As will be seen, nothing especially new has developed beyond
what has been found in connection with the earlier series. How-
ever, since here the conditions of light, temperature, etc., have
been so ideal the results not only confirm those already given,
but render them more certain and conclusive. It seems quite
improbable that three series of experiments directed to a single
end should have given uniformly erroneous results; moreover,
it is equally improbable that any error of method should have
vitiated all three series, varied as these are shown to be, and
inspected as they were by several of my colleagues almost from
the beginning. Nor is it possible that the matter of season could
have been a modifying factor, for it coincided almost exactly
with that of Loeb's experiments. That light has been shown to
be a wholly negligible factor in relation to the behavior in (JIM -lion
has not at any time been claimed. That it has been shown to
have only a minor influence I believe the facts conspire to render
very certain.

But we are not yet done with the problem. In his original
account Loeb cited the behavior of Spirographis in the public
aquarium as tending to confirm his experimental results "for the
most part' I have studied the problem in this aquarium with
especial care during the entire course of my own experiments
and have found the behavior to confirm my experiments, as the
results will show. Let it be expressly understood that in these
large exhibition aquaria the best efforts of many years have been
directed to render them as nearly natural as it is possible to
have such limited portions of the sea; and the fact that some

OX THE BEHAVIOR OF TUBICOD >l- ANNELID-. 89

of their occupants have lived and thrived here for more than
t \\ enty-five years bears strong evidence to the measure of success
in the effort to render them natural. In the-e aquuriu $p:ro-
f>his seems to find a fairly congenial environment, and thri\c-
continuously in health for many months. For the sake of
exhibition advantages the specimens have been planted, or di--
po-ed in such ways as afford the display of the gorgeon-. tl>\\rr-
like yill- to the best advantage. Hence some are located on the
lloor of the aquaria, others on the back and end- where rocky
led^e- afford suitable bases for their support. It oujju aUo to
be -aid that in order to render these aquaria the be-i po--il>le
exhibition cages the illumination is chiefly, and in -ome case
uh'illv, Irom above; while the room itself is purposely kept dark.
ex< epi |..r the light which diffuses outward from the aquaria. It
becomes important that in reference to the problem before u-
thi >f the source and direction of light be borne in mind.

< in tin assumption of the compelling potency of light it \\ill be

< le.n that in the case under examination there should be a fairly
unilormlv vertical aspect of the various specimens, \\hate\er
m. iv ha\e been their original position. The following are the
l.it ' ! nn several attempts it was determined with approxi-
in, i; i racy that at this time there were about 150 specimen-

<//>// is in the aquarium. These were di-po>ed, as men-
tioned above, on the bottom, ends and back of the tank. < >! the
entire number about <)() were in more or less vertical at lit ml
\\ith upuard inclination, while 60 were otherwise inclined, tli.n i-,
the\ were horizontal or inclining downward. The ^n. ral t

well shown in Fig. ,v which is a photograph ol the
a(|iiarium made by Dr. Sobotta, by whose kind penni--ioii I am
able to use it in this connection. Of the 60 specimen- of ihU
ad\ erse aspect slightly more than half were horizontally di-po-rd,
\\liile the others, some 23 specimens, exhibited de< idedl\ <lo\\n-
\\ard inclination. The picture will afford excellent illn-t ration,
though not taken at the time my observations \\erc made.

I el n- now attempt to analyze these facts and their b.-arii

upon our problem. It may be stated at the out-tart that gra\ - ity

ha- little or no place in the behavior. I.oeb ha- so concluded

11 hi- experiments, and my own -o to confirm his venl

9O CHAS. W. HARGITT.

Both in experiments and in nature there seems to be no evidence
of its operation. Specimens attach themselves to the bottoms
of boats, to overhanging rocks, etc., and seem quite indifferent
to its influence. We may therefore proceed to consider the main
question at issue, namely, that of light.

Of the 90 specimens having a sub-vertical attitude about (>o
were on the bottom of the aquarium, which leaves 30 of this
class among those located on the back and end walls. In other
words, twice as many of the vertical specimens were located on
the bottom as on the sides. But let it be remembered that of the
total 150 specimens in the aquarium about 94 were planted on
the bottom while only 56 were located on the walls. Further,
it is to be noted that those located on the bottom must assume a
sufficient degree of elevation to afford a free expansion of the
gills; to those on the walls this is not essential. On the other
hand, of the 60 specimens which had assumed a horizontal, or
downward attitude about 25 were among the bottom specimens,
while the other 35 were among those attached to the walls.
Expressed in percentages we have the following: Of the whole
number about 60 per cent, showed a more or less vertical aspect,
while 40 per cent, showed otherwise, i. e., a downward inclination.
Of those planted on the bottom about 70 per cent, showed a
vertical tendency, and about 30 per cent, were inclined downward.
Of those on the walls about 65 per cent, inclined downward,
while 35 per cent, inclined toward the vertical.

Now, how shall one interpret these varying aspects? Accord-
ing to theory, " If the rays of light fall vertically from above into
the aquarium, Spirographis directs its tube vertically upward,
exactly as a stem grows vertically up into the air." In the case
before us the light comes vertically from above, yet a large per cent,
of the specimens fail to direct the tubes vertically upward. Of
wall specimens 65 per cent, incline downward, or are horizontal
in relation to light. Of those on the bottom the per cent, curving
downward is much smaller, but still too great to be explained
as merely incidental, or by the naive suggestion "Here, however,
where free-swimming forms easily disturb the orientation <>t
Spirographis, it is not so perfect as when all possible- disturbing
causes are avoided, as in an aquarium used only for such experi-

<>N THE BEHAVIOR OF TUBICOLOUS ANNELIDS. QI

ment." Unfortunately for such explanation "free-swimming
form-" are rarely present in this aquarium, the only specimens
during my observations being the slow ancTdelicate moving little
sea horse, Hippocampus, whose presence among the relatively
colossal Spirographis could hardly be of more influence as
,i disturbing factor than a few sparrows in an oak forest! In
fact -pre imrns of Hippocampus had been kept for weeks in one
nl' tin aquaria in which my special experiments \\ere being made
.ui'l \\mild frequently attach themselves by their delicate pre-
hrn-ilr tail- to the tubes of Spirographis but without the le.i-t
r\idm. e !" disturbance of any sort. One often finds the tubes
' it these .imielids more or less loaded with tunicate-, -p-nu' -
li\ilr<ii<l-. etc., but there was never any appreciable sign "I
di-iurbaiice therefrom so far as their orientation was concerned.
I think it must be rather obvious that the behavior exhibited
b\ these creatures under the sub-natural conditions of tli. -<
ma^iiiti' ' nt aquaria conforms in all essentials with that found in
tin- r\pi rimental tanks, and under both these tests then- seems
to be a fair equivalent of that to be observe* 1 in their nati\e

habitat.

CONCLUMV. REMARKS.

Tin Inn ^iiing account, especially when taken as a part ot the
more extended observations already repeatedly citr<l '06, '09
must make it more or less evident that so far from affording any
-iippun lo the sweeping assumption of the identity nl animal
and plain hrliotropism, based on the behavior of these organisms.
siKuid ,-sts, if indeed one might not say warrants, tin \ -r\

opposite. One might even go a step farther and saj that ii

us extremely doubtful whether the behavior of Hy<lriii<-s,
rnniti>, eras, Spirographis, or any of the tubio>lou> annelids may
l.r interpreted as an expression of tropi-m- at all. Without
-eeking in any way to discredit the possible role of light in rrlaii<m
to certain aspects of behavior, it may vet be fairly doubted
whether it sustains any such determining influence as has been
claimed by the exponents of the tropism hypoihesi-. Indeed the
tacts \\hich ha\e been passed in review show beyond rea- -liable
doubi that in relation to these organisms it can have but a sub-
ordinate and incidental place. It seems perfectly certain that

92 CHAS. W. HARC.I 1 1 .

there is not that degree of constancy, or character of reaction,
in orientation which would warrant a tropic interpretation of
any sort.

lint <m the other hand let it not he inferred that behavior is
chaotic or beyond scientific explanation. As I have elsewhere
pointed out, reactions and adjustments in relation to food-getting,
respiration, etc., are among the most fundamental of all phases of
behavior. These creatures must live, hence must have food ; but
they are sessile, and therefore must utilize such as may come
within reach. Furthermore, they must respire, and hence must
have room within which to expand the gills. All this implies
that such colonial species must of necessity frequently resort to
movements of readjustment directed to the above imperative
ends. In most of these creatures it so happens that one and the
same organ is involved in this dual function of food-taking and
respiration; a fact of some significance in simplifying or com-
plicating, according to condition, certain phases of behavior.
To the writer it seems probable to the point of certainty that
the aspects of behavior which have been under review are chiefly
but varied expressions of these common functions. In other
words, they are aspects of adjustment in the complex struggle
for existence varying modes in which each species has worked.
out its own special problem of life.

In the light of this mode of interpretation the complicated
serpentine torsions of the tubes of Hydroides and Pomatoccras
arc the most natural expressions of just such "trial movements"
as one might expect. Likewise the bending aspects of the
flexible tubes of Pot am ilia and Spirographis are not mysterious
enigmas over which students of behavior need array themselves
in warring camps, but rather the simple expressions of those
individual adjustments called for in the varying struggle of
life, to the interpretation of which Huxley would have found
necessary only "trained and organized common sense"

I am quite aware that to speak of individuality, or autonomy,
or spontaneity as factors involved in problems of animal behavior
may to some exponents of mechanism seem "no explanation,"
and of significance only to the psychologist. But as I ha\e
earlier pointed out, they are facts, and they bulk large in tin- sum

OX THE BEHAVIOR or TUHK < >!.< >l - ANNELIDS. O.}

int.il of animal economy and behavior. To ivco-ni/e them
as facts is not to imply thereby their explanation; but it does
imply that they are no less entitled to recognition ami cxplana-
t if in than any other classes of facts with which we have to deal.
I i~ an- sometimes characterized as "stubborn thins;-." they
ha\e ways of their own; they are tenacious fit" lilV; and -o<mcr or
later \\ill compel respectful attention and explanation. A- i-
well known, in his matchless account of the behavior ot earth-
\\onn- I'aruin <lid not hesitate to employ a terminology which
iiankK a->umed the presence in these creatures of nervou- and
p-\rlii( factors. While it may not be easy to pm\e that anne-
lid- ha\e a high degree of intelligence, on the other hand In \\lio

lys to prove that intelligence has no part whatever in their
beh.ixior \\ill hardly ha\ r e an easier problem.

\t no time has the writer questioned the important relation-,
ol ph\ -i .- hemieal factors to the phenomena of lit* .md beha\ i'r.
1 in tin i. hr has not questioned the possibility of the correla-
tion ot these phenomena under physical laics, mm h as ue n >
ni/e that |ihenomena of electricity and magneti-m and -ra\ i-
i.iiion .in conserved under other natural laws. Hut tin- b\
n.. mean- implies that these latter species of ener-\ h.i\e n-.i
i IK i o\\n ''fcial lau's, some of which are alread\ kno\\n \\hile
other- ha\e thus far defied definition and correlation, ^o. in
tlie in. nter under review, what he lias <|iie-tioiud i- the very
dillei.nt po-iilate, that knou'n properties of ch<-mi-ir\ of phy-ie-
in an\ ot their known interactions afford a<le(|ii.ite definition
and explanation of all the facts; or that kn:cn f>liysi,dl I:

ipplied b\ the sponsors of mechani-m, an- convincingl)
-iiiiuieiu. It is against the arrogant a>-umpiion that a i
ol" beha\ior. . .r an expression of emotion or atle. tion. i- never
explaiiu-il till cast into some physical or mathematical tormnla,
that prote-t ha- been iterated. In dire* -tin- attention to the
po^-ible interaction of well-known p-ychic factors in belia\ ior
tlu re i- merely the plea that similar re< -^nitif >n be -i\ en to them

to the lornier and, as suggested above, they be included in
the category of behavior calling for explanation. Ho\\e\.-r
independent or unrelated may appear certain of their expressions
it i- not a--uined that in any scientific sen-e they are mutually

94 CHAS. W. HARC.ITT.

exclusive, nor that the one class of phenomena are any less
related to causal antecedents than the other. But it is main-
tained that while in some cases these antecedents may be known,
and lend themselves to direction and control, on others they
are as yet absolutely unknown and more or less beyond pre-
diction or control. And furthermore, it is believed upon experi-
mental evidence that certain aspects of behavior may be more
or less variable under any given set of antecedents or conditions;
in other words, given stimuli do not always evoke the same response;
in fact, much of behavior is indeterminate in terms of existing
knowledge. But so far from implying a reactionary attitude
toward the value and importance of continued experimenta-
tion, the writer would hold the very opposite. It is well that
one pause now and then and take stock in science as well as in
business. That problems of behavior are complex beyond
earlier anticipation goes without saying. The same must be
admitted of every problem of biology. Only the biological
pessimist will find occasion to contemplate intellectual suicide
because he finds the dogmas of his science in process of revision!

LITERATURE CONSULTED.
Bounhiol

'oo Recherches exper. sur la rcspir. dcs annelides Etude du Spirographi-.

Compt. Rend., T. 132, 1900.
Hargitt, Chas. W.

'06 Experiments on the Behavior of Tubicolous Annelids. Jour. Exp. Zool.,

Vol. III., pp. 295-320.
"09 Further Observation on the Behavior of Tubicolous Annelids. Ibid., Vol.

VIL, pp. 157-187-

Harper, E. H.

'09 Tropic and Shock Reactions in Perichceta and Lunihriiit^. Jr. Conip.

Xeur. and Psych., Vol. XIX.. pp. 569-587.
Jennings, H. S.

'06 Behavior of the Lower Organisms. New York.
Loeb, J.

'90 \Veitere Untersuchungen uber d. Ili-lioin>pi>miis d. Tiere u. sciiu- t Ut-
einstimmirtig mit dem Heliotropismus drr PHanzcn. Arch. t'. d. gi--.ini.
Physiol., Bd. 47, p. 391.

'oo Comp. Physiol. Brain and Comp. Psych. New York.
Mast, S. O.

"ii Light and the Behavior of Organisms. New York.
Nagel, W. A.

'96 l)cr Lichtsin augenloser Tiere, cine biologische Studie. Jena.
Radl, Em.

'03 Untersuchungen ubcr den Photntropismii^ drr lirir.

THK DEVELOPMENT OF THK GONAD AM) GONO-
I)t;CTS IN TWO SPECII-> ' >F CHITONS.

ROSE M. HK;LEV AND HAROI D HEATH.

Tin- I. HIT development of the chitons ha- never been fully
iim-Mi'-.m-cl. and the fragmentary obscr\ MI i m- that have been
in. idi- relate .ilmost exclusively to immature form> in very ad-
Nam id stages. Accordingly we are at piv-ein alimt whollv
ignorant i.i tin- development of the principal systems of organ^
.nid ilu-ir homologies. Many of the more import, uu i|iie-timi-
relating to 'In se animals center in the formatinn of tin i-u-lmn,
.md ii \\.i- \\iih the hope of throwing some light <>n thi> >ul>iect
tli.it tin- piv-eiit work was undertaken.

111. iu< -prcies that form the basis of thi> investigation,
.hyiit-nnon ra ynwndi and Xutlallina tlwmasi, .\n- !.iirl\ almn-
il.mt li.ini- in certain localities along theci>.i-t <>i ( '.ilifnrni.i, and
n\\iiiv; tn tln-ir Miiall si/e are readily snt \< uinl. The tK i---\\ ini-
iniii^ young 1 were placed in aquaria together \\itli t'lMgnu-nt- uf
-ln-ll- . 'i!ns californica on which they Imallv -i tiled alter

partialK completing their metamorphosis. Tlu-v \\(ic then
traii-lenvd to small and protected tide pm.U \\hen- tln\ de\el-
"|ird mu inalU' and in se\ - eral instances were allowed t" re.n h the
-exualK mature condition. Precautions \\eiv taken to keep the
\ i ni nv; i >l eaeh -pe. ies in separate pools and it \\ a- I mi ml that they
travel essen dally the same developmental path tm- a Ion- period.
I >i-tiiiL;tii-liiiig characteristics accordingly appear late, in I'aet

roii-idt -ralil\ lievond the formation of the L;<'iia<l ami it- dm I-.
It U to In- understood therefore that while the tigure> are of
T. niynnuii the\ -erve equally \\ell tor A", tlnnintsi.

\\ a \ei\ earl\ stage the heart and peril ardial cavity are
dexeli.pe.l troni (clU, giving evidence of lu-ing derixed exclu-
>i\el\- from the -e. .-mlary mesolila-t pm-eiu ot 4/>i, whieh
lorm-> an irregular la\er mi the poMcro-dm--al -ide ot the larxa.
A rel.n : \el\- long period of tinu- then ensues, during which the

iii.u li.iliit- <>i tli> e /'./. In Bd. XXIX.. \i. ij.

95

9 6

ROSE M. HIGLKY AND HAROLD HI AMI.

other systems of the body develop to practically tin- same con-
dition as in tin- adult, before the gonad makes it> appearance.
When the primitive sex cells become recognizable they usually
form two groups attached to the anterior external surface of the
pericardium from which they appear to be proliferated. Very
soon, in rare instances at the time of their formation, these be-
come so closely appressed as to appear single though section^
show them to be distinct for a considerable time, frequently alter
the gonoducts have formed. Shortly after their appearance a
cavity forms within each group, and, with the growth of the
gonad, soon becomes more or less triangular. In later stages.

A

FIG. i. Gonad and ducts of Trachydermon raymondi. A, section through
animal about i mm. long, a, digestive tract; g, gonoduct connecting with gonad;
k. kidney; / liver; n, lateral nerve cord; s, shell. B, gonad (o) and ducts in matuie
animal, dorsal view. C, reconstruction of same stage a Fig. i. Gonad with
ducts ending blindly; kidney showing reno-pericardial and external openings.

generally about the point of development represented in the
figure, these cavities gradually fuse, commencing at the posterior
end of the gonad and progressing anteriorly. I n some individual-
a slight groove may persist on the ventral surlace between the
halves of the gonad for a considerable time, and in a lew case- .1
distinct cleft at the anterior end of tin- gland persists until the
animal is half grown.

The aorta holds the normal position on the dorsal surface
of the gonad, and there are slight evidences that a portion of
the blood it carries makes its way between the halves of the

DEVELOPMENT OF GONAD AND GONODUCTS IN CHITON-. <>7

organ as in the solenogastres. At all event- thefe are no signs
-.1 di-tinct branches penetrating the gland as in the later sta^

About the time of the fusion of the gonad cavities t when the
length ol the body is approx'mately I mm.', in a stage slightly
earlier than the one represented in Fig. 3, each gonoduct ari-c-
as a -lender evagination of the postero-latcral walls of each halt
of the reproductive gland. These grow rapidly, and in contact
\\iih the pericardia! wall proceed laterally and vcntrallv until
they come iii contact with the ectoderm of the mantle groove.
In the formation (A the outer opening the ectoderm cell- appear
inereK to -eparate: if an ectodermic diverticulum i- formed it i-
evidently \ery short and transitory.

In later -tages the proximal ends of the gonoduct- -hilt lor-
\\ard -li'Jitlv , and are attached to the dorsal side (l-'k. - <t the
'^onad close to tin- mid line. During this process ilu-ir \\all-
thii ken, and at the height of the breeding season there an -i^n-
ecretor) activ ity on the part of the component cell- especially
in the mi-hborho(Ml of the reproductive organ. Th of

boih of these species are held in the mantle cavity, and are lo.i-dv

1 it niiii 1 t' p'^ei her possibly by this secretion ol the o\ id in t .

Ilie i. nl\ niluT observations bearing on the development ol
the ^oimdin i- are those of Plate 1 who has made the claim lh.it
in the \oiiiii; <f Acanthopleura echinata, 15 mm. in length, the
gonad i- completely separated from the gonoduct s that , a- -lender
divert ieiila. are connected with the mantle cavity and are a<
cordinglj 'iodermii-. drained that this is the true state o|
a Hair- in .1 . ft hinatu it is unprofitable for the prc-ent to attempt
to correlate the two t\ pes of development when only three -pe. n -
..I .hitoii- have been examined on this point. Hov\ever, it i-
inti-re-tini; to note that in several species of California chiton-'
three millimeter- or less in length the gonad and it- duct- are
attached and open to the exterior. In some species, -nch as
h, hncuhiton iundalencnsis, the ducts are highly glandular and
it i- 1 1.. il.le, though it appears to us improbable, that this
-landulai -eciion is of ectodermic origin.

i/ ' -iippl. .1 (Fauna Chilensis, Vol. i

: Il.atli, Zoo/. Jahrh., Bd. 21. p. -29.

ASTEROPHILA, A \K\V GENUS OF PARASITIC GAS-
TROPODS. 1

JOSEPHINE RANDALL AND HAROLD HEATH.

During the dredging operations of the U. S. F. C. Str. Alba-
tross in the vicinty of Japan (summer of 1906) four specimens of
a starfish, Pedicellaster sp., were taken that had been parasi-
tized by a new genus of gastropods. All were dredged in the
sea of Japan off the coast of Corea at depths ranging from 150
fms. (sta. 4,867) to 163 fms. (sta. 4,861). In one host three
parasites occurred, while only one was present in each of the
other three, but in any event they occupied the ccelom in
the arm, and were attached by connective tissue strands to the
body wall in the vicinity of the ambulacral ridge. As noted
more particularly hereafter, this species is not put in communi-
cation with the exterior, the mouth and reproductive openings
communicating with the body cavity of the host. During the
time that the brood pouch is crowded with embryos the
pseudollapium becomes accordingly considerably distended and
tense, resulting in the inflation and consequent thinning of
the body wall of the host along the dorsal side of the arm
(PI. II., Fig. 2). Under such circumstances it is possible that
the body wall of the starfish finally ruptures, causing a diminu-
tion of the pressure on the pseudopallium which therefore dis-
charges the embryos into the surrounding medium. After
this process the break in the body wall is probably repaired,
as there are evidences that one of the larger individuals has
recently discharged its brood though there are no signs of a
rent in the starfish arm.

In every case the body resembles in form a kidney or thick-set
bean, and varies in size from i\\o to twenty millimeters, this last
extreme being due to some extent to the large number of embryos
and the fluid in which they float. The ovary and the embryos
themselves arc- light yellow in color due to the presence of yolk,

1 Published by permission of Hon. G. M. Bowers, Commissioner <>t I-'Nu-ries.

98

ASTEROPHILA. 99

the liver is of a light brownish shade while the other organs are
unpigmented and more or less translucent, especially in the case
"I i IK pseudopallium that in life is so thin and transparent that
tin- torm and movements of the larvae may be readily observed.
A- may be seen in PI. I., Fig. I, there are two openings into the
body, one the mouth, corresponding in position to the hilum of
,t In -.in while the reproductive opening is placed laterally up-
ward- of thirty degrees.

I nun \arious features of its organization it i- readily po-sible
to orient this animal and discover the axes of the body. As in
-ev.ral other parasitic gastropods the body is surrounded by a
pseudopallium that appears to be a development <>f the Miout or
adjacent regions of the body. Growing upward it ha- en\ eloped
ilir body completely save at one point, the reprodncti\e and
excretory pore. Considering the body proper, tin- foot i- seen
\i-t in the form of a small though broad wedgc--haped fold

I'l. I . 1 ig. I. /) covered with epithelium of greater thickne--
than thai surrounding the body generally. In section- I'l 11.,
Fig. i it is a fairly conspicuous object owing to its affinity ii
-tain-. Again, well-defined pedal ganglia and otocysts, located
in lose proximity to the foot and cerebral ganglia on the oppo-ite
-ide of the digestive tract, demonstrate the fact that the antero
iior axi- is the shorter of the roughly cllip-oid bo 1\ , and
that the transverse axis is accordingly the longer.

In tin- -|>ecies the degenerative processes ha\e a<l\aiurd to a
stage \\liere the mantle and mantle cavity ha\e largelv di--
appeared, and vet. though rudimentary, they maintain tin ii
i\ pi. al relations. Since the mantle fold is coniparaiixely narrou

I'l. 1.. Hg-. i. 2, t,') the cavity is accordingly -hallow, as the
mantle i- closely applied to the visceral mas>; ne\ . -rthele-- the
epithelial cells bounding the cavity are not only higher than tin >se
cUeuhere covering the body but they stain more inten-elv and
are ciliated. < Mi the left-hand side of the body the mantle bonlei
thicki-n- con-iderably, and forms a projecting riilge th.it cn-
tinues until the p.illial cavity itself di-appeai>. In the -malle-t
specimen the mantle and cavity are relatively larger and the
mantle fold is much more glandular, the gland cell- being large
and conspicuous

100 JOSEPHINE RANDALL AND HAROLD 1 1 LATH.

As noted in a preceding paragraph, the mouth opening is
borne on the summit of a low papilla in the mid line. In entire
specimens it is further distinguished from tin- opening into the
pseudopallium by occupying the center of a whitish area, up-
wards of 3 mm. in diameter in the largest specimens, canard by
the compact feltwork of circular and radiating muscles enveloping
what probably corresponds to the buccal tube. In the immediate
neighborhood of the mouth opening the canal is comparatively
slender, 0.28 mm. in diameter in large individuals, and is pro-
vided with a lining of simple columnar cells whose distal portions
contain small quantities of a faintly staining, vacuolated secre-
tion. Behind this point large numbers of small, irregularly dis-
tributed pyriform gland cells appear imbedded in the muscular
meshwork surrounding the digestive tract, and their darkly
staining ductules may be traced to intercellular openings in the
buccal or pharyngeal epithelium, whose extent is increased by two
symmetrically placed diverticula with short, stubby branches
(PI. I., Fig. 2) extending a short distance into the surrounding
muscle sheath. These paired glands probably correspond to the
ventral salivary glands of other molluscs as the buccal ganglia,
connected by a commissure, are located in their vicinity.

No trace of a radula exists.

The buccal-pharyngeal tube with its enveloping glands and
muscles, is relatively short, probably not over I mm. in length,
but it spans a well defined head cavity (PI. II., Fig. 3), which is a
portion of the h?emocele as in other molluscs. Curving gently
toward the ventral side of the body the tube leaves the sinus,
and now devoid of gland cells and with a comparatively thin
sheath of longitudinal and circular muscles, it passes back a
short distance into the body and unites with the main portion of
the digestive tract (/), a spacious cavity, lined with glandular
epithelium, occupying most of the visceral mass not held by the
gonad and its duct.

The pericardial cavity (Tl. I., Fig. 2) is situated on the anterior
surface- of the visceral mass on the right side. The contained
heart consists of a single auricle and ventricle, both of large size
and highly muscular. The first-named receives the blood from ,1
broad sinus, which on one hand passes from the liver surface

\>TEROPI1ILA. IOI

and tin.- neighborhood of the accessory reproductive glands in the
\ en t nil part of the visceral mass, and by means of another smaller
branch t.ikes the blood from the kidney. The aorta is very -hort
and leads directly into what may be termed the head cavity, the
large space surrounding the pharynx. From here numerous
1. ranch'- .\tend into the pseudopallium, liver and between the
oxarian lollieles. Of these the ones pa--ing through the pseudo-
palliuni probably function in the interchange of gases as there
are no trai es of ctenidia or branchia.

One nephridium I'l. I., Fig. 2, n) is pre-ent in the form of a
iil\ on i pressed sac covering the anterior surface of the
visceral ma on the right-hand side. Its inner wall- are often
pi<. \idrd \\ith lamelke or folds, of varying >i/-. projecting into
ihe rcniral lumen. The cells throughout are highly vacuolated
and contain \arying quantities of some granular secretion that
in >ome locaiions present the form of concrements. \\ V have
lut n unable to definitely locate any clearly defined ivno-peri-
aidial opening. As shown in 1M. 11.. Fig. 6, ;/, the kidnev
in\e-i- tin dorsal pericardia! wall but there are. so tar as we
lia\e seen, no modified cells indicating a nephro-tonie. Tlie ex-
ternal pore I'l. I., Fig. 2, e') is situated on the anicrior lace of
tin visceral mass immediately below the margin of the maiulr.

\\ liilc the ganglia are fairly well denned and distinct tin- nerve
liber- n M inble so closely the connective tU-ue and IHIIM le
bundle- tlii-on^li \\hich they make their way that it i- \ ery dif-
in nit to determine their course. The cerebral iIM. I., 1 i:^. I,D,
appan-nil\ a ociated with the pleural, are. in the type -periim-u,
-ilnated in an asymmetrical position, beinv; placed mi the ri-ht
>ide ot the pharynx. From this nerve ma-- connectives extend,
on each r-ide o| tlu- phar\ nx, to the pedal uan^lia. large, clo-ely
appn---i-d ^ioup> of ner\'e cells placed -\ninietricall\ \\ith refer-
ence to the mid ventral line. The cerebral uaujia likewi-e
originate buci\il connectives that, extending along the phar\u\.
unite \\ith ganglia imbedded in the >a!i\ar\ gland- on the dor-al
and \entral -ide. The buccal ganglia are further united by t\\o
c.pinmi ure- that lorm a collar about the pharynx. In the
neighborhood i.|' tl,. opening of the repn>diicti\ e -\-tem into
the pseudopallium there is a large ganglion. probabK the visceralj

IO2 JOSEPHINE RANDALL AND HAROLD HEATH.

that gives off a strong nerve which may be traced a short distance
posteriorly, and in the opposite direction a single connective
leads from it to the cerebro-pleural ganglia, as indicated in PI. I.,
Fig. i, v. At various points throughout the body it is possible
to discover nerve bundles, but in every case it has been impossible
to determine their origin.

In the three animals examined there is no sign of a testis,
though the seminal receptacle of one- individual contains a con-
siderable number of spermatozoa. These last named elements
possess almost spherical heads measuring approximately 0.004
mm. in diameter. Nuclei of somewhat similar appearance may
be detected here and there in the follicles of the reproductive
gland, but their close resemblance to those of the connective
tissue cells renders the determination uncertain. If self-fertiliza-
tion does not occur in this species it is difficult to understand the
method of sperm transfer especially in those examples where but
one parasite occurs in the host which completely envelops it.

The ovary, occupying fully half of the visceral mass, consists of
a large number of follicles united directly or indirectly with the
duct leading to the exterior. In a mature condition the central
portions of each follicle are packed with fully developed ova,
while numerous cells in the earlier stages of formation bound the
periphery. About the center of the visceral mass the common
chamber, communicating with the ovarian follicles, narrows
anteriorly and the short resulting tube, after a somewhat twisted
course, unites (elliptical stippled outline, PI. I., Fig. 2) with the
definite gonoduct leading to the exterior. This last-named canal
consists of three divisions corresponding to the albumen and
mucous glands and the seminal receptacle in oilier species of
gastropods. The canal from the ovary unites with the albumen
gland which extends posteriorly as a pouch of considerable size.
Its walls arc relatively thick, and are fashioned into a few promi-
nent folds, consisting of relatively slender cells, whose vacuolated
secretion stains lightly with Delafield's hsematoxylin. Slightly
anterior to the oviduct connection, a cone-shaped seminal
receptacle (PI. I., Fig. 2, r) is attached to the albumen secreting
section. Its epithelial lining is developed into a large number of
folds between which there are quantities of spermatozoa, that

ASTKROPH1LA. IO3

likewi-e occupy the main lumen and even extend in -mall quan-
tities some distance into the albunu-n gland. Anterior to the
-eminal receptacle the walls of the canal change abruptly, be-
coming thicker and the secretion stain- -o inten-cly that the
cell outline- and nuclei become almost completely <>l>literau-d
Thi <>f affairs exists between the seminal receptacle and a

poim -lijnly posterior to the external repn>ducti\ e opeiiin^.
Anterior to this region the duct presents the form of a roughly
i oni. .il sa< extending to a point opposite the foot. The wall- of
thi- |ioin h .ire similar to the darkly staining one- ju-t described
save that the secretion is more vacuolatcd and accordingly less
derpK -i. lined. The duct leading from this mueou> -envting,
in.iin i anal to the exterior is relatively short, thin-walled and
passes into the furrow at the right side of the body formed by
tin union oi the visceral mass with the pseudopallium.

In t\\o -pei imens whose pseudopallium contained fully 500
rmbr\o- the ovary held an e(|iial number of ova in a fully de-

\elop.d ( lition. Hence it is probable that during adult life

t In brood pouch is empty for short periods only.

The ;^i -nus may be defined as follows:

. 1 'ild new genus. Body globular, 2-2O mm. in diameter,

c. unpleieK enveloped in the pseudopallium. Foot and mantle
nidimentarx . liuccal-pharyiigcal tube, with salivary gland-,
opens into o unbilled stomach and digestive gland that otheru i-e
do not open to the exterior. Xo radula. Albumen and mucou-
gland- on reproductive canal highly developed, and seminal
i-.epiacle prominent. 1'arasitic in starfish Pedicellaster sp.,
>ea of Japan. Type of genus A.japonica.

.1 . ju/'onii'ti new species. \Yith characters of the genus.

IO4 JOSEPHINE RANDALL AND HAROLD IIFATH.

EXPLANATION OF FIGURES.
PLATE I.

FIG. i. Diagrammatic view of .4 slerophilajaponica, left side, with the greater
portion of the pseudopallium removed, a, albumen gland; c, cerebral ganglion;
/, papilla-like foot; g, mantle fold; /, digestive gland or liver; m, mucous gland;
oy, ovary; p, buccal tube and pharynx with salivary glands, buccal ganglia, con-
nectives and commissures; r, seminal receptacle under which is dotted outline of
duct from ovary; ur, urino-genital opening; i>, visceral ganglion.

FIG. 2. Diagram of anterior surface, e, e', openings of reproductive and ex-
cretory systems into pseudopallial space; g, mantle fold, the depth of the mantle
cavity indicated by broken line; h, heart; n, kidney; />, buccal tube; r, seminal
receptacle.

BIOLOGICAL BUUETIN VOL

PLATE I.

RANDALL AND HEATH

106 JOSEPHINE RANDALL AND II \ROI.D Ill-Alll.

PLATE II.

FK;. i. Anterior view of Aslerophila japonica with pscudopallium partially
removed.

FIG. 2. Arm of starfish containing parasite.

FIG. 3. Section through pharyngeal tube, showing salivary glands, buccal
ganglia and surrounding head sinus.

FIG. 4. Section through foot and visceral mass; along line w of PI. I., Fig. i.

FIG. 5. Section along line u, PI. I., Fig. i; the junction of the oesophagus and
stomach-intestine marked by an arrow.

FIG. 6. Section along line /, PI. I., Fig. i.

FIG. 7. Same along line s, PI. I., Fig. i.

FIG. 8. Diagram illustrating growth of pseudopallium in Aslerophila.

FIG. 9. Same, Ctenosculiim hawaiiense.

BIOLOGICAL BULLETIN, VOL XXII.

' ' -' ^'J

-M

'

.

t-?

- - * i

G

'

/

ITS- '

s_-

; \
.
-' ' . S ay '

'-'-

-

/ '

RANDALL AND HEATH.

A ' \SE OF YOLK FORMATION NOT rnNNKiTED
WITH IHI-: PRODUCTION < >F OVA.

OSCAR RIDDLE.

Tli.- -i-crclion or production of true yolk in situati. >n- other
th. in in ova or in the follicular cells which surround ova is nt
kiniuii so far as I am aware. 'Nurse" or "yolk" cell- have of
course long been known to exist in several group- .'I animal-;
the-e. however, are evidently the equivalents of follicular cell-
IT ol ova. It is therefore of some interest to record tin- lindin^
ol i rue \olk in spaces within the connective tissues which lie
i AN TH. illy to the follicular membrane of capsule- \\hirh had
pn-\ ion-Iv liberated ova.

I hese observations were made on the ovaries of the rinnion
fowl during the mid -summer season. During Jul) and .\umi-t ot
i In- ] in -i nt summer the writer had occasion to examine tin
ovaries t more than one hundred full-grown hens. Aimm^ ilu--e
iln i. \\ en- at least six or eight ovaries which showed unmi-takablv
iln- peculiar accumulation and jJacement of yolk \\hich i- In M
dr-t i il ii -d.

In oidi-r to be Mire that one is really di-.ilin- \\iih "extra-
o\ ul.ti " .md "extra-follicular" formation of yolk, and not nuT.-K
\\iili ,i mas(|ue of its usual source, it was i r\ \ deiermine

three things concerning the capsules within \\hich the \.>lk in
(|iif-iiou \\as found: (i) That an egg had been -mvly de\-i-lope<l
ami liberated from this capsule. (2) That tin- >pa<v in \\hich
tin- \ oik was found is quite separate and remo\ ( -d troin tin- space
lornuTK' occupied by the ovum, and like\\i~.- n nio\td from tin-
lollicnlar cells which surrounded the o\nin. ; I li.u ih
cumulated substance is true yolk. I belit-\e that tin- material
1 lia\ e i-xamined has enabled me satisfactorily to dcii-nnine each

ol these points.

The evidence that the yolk-containing capsiilc-> in .|iu-.tion
had prc-vioii-lv borne and liberated ova rests partially upon the
finding of capMilc^ showing all the intermediate stages between

107

IO8 OSCAR K1ODI 1 .

the recently broken capsules and the large. Hubby, often asym-
metrical, yolk-containing ones. Some of thor laitrr capsules
might be mistaken for resorbed ova, since they too have a closed
stigma; that is, the slit or splitting which occurs in the capsule
at the time of ovulation. and which allows the escape of tin- < >vum,
later heals together and the cavity of the follicle is once more
completely sealed. The chances for such confusion are further
increased by the fact that this central chamber may also oc-
casionally re-accumulate yolk.

It is possible nevertheless in favorable material to be quite
sure that the stigma has been broken and reunited a thickened,
accentuated, and often more or less ragged point of reunion
indicating this. Furthermore, a series of follicles in the same
ovary, showing the most recent ones still broken open, often
decides the matter at once with certainty. The capsule from
which a sample of yolk for analysis was taken was one of such a
scries. In this case there were nine yolk-containing capsules in
various stages of extra-ovular yolk-production; and in addition,
one other the newest follicle plainly recognized by its whole
appearance as a recently emptied one. This follicle, however,
showed the once broken lips of the stigma now nearly completely
grown together, but with its inner cavity as dean and free from
yolk as at the moment of ovulation. It is certain that the fol-
licles of this ovary had liberated ova, and that instead of degen-
erating thereafter these capsules quickly closed tin- breaches
formed in extruding the ova, and began the production of yolk
in their external walls.

It is easy to demonstrate that the yolk-filled spaces bulging
from the sides of the capsules have no open connection with the
central cavity of the capsule; that is to say, these spaces are not
connected with the former seat of yolk formation. Several times
I have made a slit in the scar or stigma and, finding the interior
clean and free from yolk, have tried by squeezing the various
bags of yolk lying in the external walls of the capsule to make
their yolk How into the central cavity. In no instance have 1
succeeded in thus finding any connection whatever between thr-.r
new yolk-containing cavities and the old cavity lormerly oc-
cupied by the egg. On the contrary, careful dirnioii> of

MILK. FORMATION. 109

these capsules show that the two spaces are always separated by a
rather thick wall; certainly much thicker than that which
separates the new yolk space from the exterior. This latter wall,
in fact, i- usually very thin. It consists, however, of an extremely
thin connective tissue layer in addition to the ovarial epithelium,
liv < .ireful handling the epithelium can be -tripped <>tY and the
thin la\T .-iiclosing the yolk space left intact.

"Ilit- \ery external position of the yolk -pa. es of \\hich there
ina\ ! .-r.il in a single capsule make- it evident that

in -in- 1. 1 i In cells of the old follicular membrane an- cn-a-ed in
tin- ]>n id in lion of yolk in this new and unusual -ite. The
|ini<lm lion of this yolk i> necessarily accomplished by the cells
which form the external theca a tissue from the former ovarian
stroma, \\lii-h in the late growth stages of the cap-ule of large
Mies a very thick, firm, essentially connect i\e ti ue
la\er i -in lii-ing possibly some scattered derivative- of the m-r-
niin.il epithelium -whose cells normally take no part in yolk
toriiiaiii.ii.

|u-t \\liai it. is that transforms these non-yolk-producing cells
into ell- ai ti\ely engaged in yolk production, ii \\mild lie nm-i
inii n sting to know. While confessing very complete ignorance
as to thi- i ause, it seems worth while to note that the cells ichich
here hike n(> n new function do so at the time v//r;/ the "normal"
tin * - do is to degenerate and be '/.

In thi- connection it should be stated that the true lollicular
e!l- iln-e \\hich ha\'e previously been engaged in pa--iir^ .-n
tin- . Dii-tiiueiit> of yolk to the egg are apparently the lea-t
liable i.t anv of the capsular cells to take part in any later yolk
production. < >nl\ occasionally in a group of cap-ule-, each of
\\hich mav In- producing yolk at one or more point- externally,
\\ill one tin, | th.it the follicular cells ha\- continued or rather
ha\e recommenced to produce yolk. What I ha\>- ob-er\ed
\\ould indicate that these follicle cells iu-\er in an\ case become

acti\e until after \ oik production ha- been initiated ill the more
exiernal la\er-; but of this latter point 1 ,1111 inn certain.

That the \ello\\Uh tluid enclo-ed in the-e \olk -|iac.-- i- true
\olk i- indicati'd b\ its microscopic appearance. The question
i- |io-iti\,-l\ .uul affirmatively answered by the chemical anal

no

' >S< \K

of a sample. 1.605 grams of such yolk were rolltvtrd from a
single one of the new yolk spaces; this was not all, but nearly all
of the contents of the cavity. In order to slum how closely
its chemical composition agrees with that of other tOnn< of true
yolk, I have added to the table the numbers resulting from the
analysis of four such samples of yolk. Reference to the table
readily shows the essential similarity of all these substances ; and
likewise a point or two of notable difference.

In Per. Cent, of Solids.

Analysis ol :

Leci-
thin.

Protein.

Neutral
Fat.

Total
Ash.

anic
Extractives.

H,<>.

Extra-follicular yolk .

IQ O^

26.21

4S 3Q

6.6l

2.6s

74.22

Central "yolk body" from in-
cubated hen's egg

10.68

28.87

46. os

7..4O

2.OO

37.13

Egg yolk, Jungle fowl

JQ.OO

7.O.47

46.74

1.7,0

I. SO

48.70

Contents of yolk-sac; 2r clays in-
cubation

17.62

-17.24

47.76

1.16

I. TO

S6.S2

Resorbecl ovum .

i<;.7o

7C.I8

A2.2S

1. 71

O '

67.20

It is true that I have selected for this comparison analyses
which most closely agree with the analysis of the "cxtra-fol-
licular" yolk. The high water content of the latter is of no con-
sequence; an analysis of "white" yolk from the hen having
yielded more than 80. per cent, of this constituent .

The high ash content, and very low protein content, do indi-
cate however a species of yolk not in all respects like that pro-
duced by the follicular cell and the ovum. In these two respects
this yolk stands as a rather bold extreme in a long series of
analyses of normal yolk. It can be said therefore that though
this substance is certainly "yolk," its peculiar origin stamps its
chemical composition with a specificity of its own.

The foregoing recital of the facts is perhaps hardly sufficient
to uncover at once to every reader one of the points of interest
in these findings; at any rate it is a point of interest to the writer.
I refer to the fact that in all of the hitherto known cases of
yolk formation the whole process of yolk building and stoi gi
appears so glaringly and profoundly ideological. The ovum pre-
pares and stores food for an embryo that is yet to form; a
follicular cell passes on this rich material only to an ovum \\liich
in turn accumulates for a promised organism that will arise and

FORMATION-. in

require the store; ovum and ten thousand follicular cells unite'
i" pn-p.ire and to hoard a pabulum for an organism whose
father exists as yet only in prophecy and in fortune; a "nur-e"
ci-11 arises in a distant part, migrate- with its supplie- and un-
erringly delivers all to the egg whose prospective accomplish-
ment only ran use or require them; or, again, 33 in some hydroids,
several adjacent ova laboriously produce a -olden store which

tlier with their own existence they place -acriticially upon the
abar of po-terity giving all to a more opulent neighbor, who
through the combined accumulations of main gatherer- ran the
more adequately and assuredly provide for the beginning of an
indi\ idual that is to be.

Nor i- Mich apparent teleology ab-ent from the\r\ chem-
ical composition of the material that is stored. The developing

mi-m requires above all else a store and -omve ,,i , -m -i
ami one notes that yolk the material actually stored i- richer
in lecithin and fat than is any other product n| the bo<l\ ; and
fuit her that t In -e constituents are the one- \\ hii h i arrv tar more
ener-\ pri unit of weight or volume than do an\ other-.

\\lien, ho \\ever, one turns to the sort of \..lk formation <!<-
-..i i beil in this paper, yolk formation which be-in- in -ubdiied
and a 1 1 .1 i, f.illii les, among cells largely "soinati/ed " and doomed
in certain de-i iteration; when one con-ider- the utter blimlne--
in\ol\ed in these ill-conditioned cell.- phr into a nio-i

active ]>roductioii of excessively rich loud-, ( ,nly to cast them
into the formless spaces of these spent cap-ule-. one ran n-ali/e
that the pr. .. ss of yolk building actually can be as grotesquely
ab-urd and inappropriate as it has else\\ here -eemed replete \\ ith

ill-i-tellt teleoliigy.

ORATORY OF EXPERIMENTAL TllKRAPEl'l l> -.

I i! u-irv OF CHICAGO.

i<.ii.

Vol. XXII. February, 1912. No.

BIOLOGICAL BULLETIN

Till. c-MoTIC AND SURFACE TENSION* PHI \OM1.\.\
OF LIVING ELEMENTS AND THEIR PHYSIO-
LOGICAL SIGNIFICANCE. 1

J. F. McCLENDON.

CONTENTS.

I. I lit: i; tion iij

1 1. ( Kinotic Phenomena in Plants 120

111 Phenomena u 7

i In plants I J7

le and Nerve \ _<,

i-l Movement i ; I

;>aKation of the Bio-electric Changes 136

1\ -is 139

\ |>crties of the Blood Corpuscles 142

VI V -i and Secretion

i >n through the Gut i

Relation of Aquatic Animals to the Medium i

;i of Lymph and Tissue Juice 152

u

VII. (Yll I>i\i-ion.. :

PREFACE.

ThU paper formed the basis for two lectures -ivm In-fore the
class in ph\-iology at Woods Hole, July 7 and 8. i)ii, although
I'uiii- in liinii.il lime, some parts were omitu-d. Since then
then- ha^ appeared a second edition of Hober's " Phy>ik.ili-clu-
Clu-inir clt-r /rile und Gewebe," which n-\ir\\> much of tin-
littT.iiurt cmi-itlered in this paper. H<>\\c\cr, owing to an
t-ntin-K- (lillncnt mode of presentation, it is hoped that the
pre-ent treat meiu of the subject might be helpful to many

ieral readers, some of whom would not read HoU-r's book.

1 l-iom tlii- 1- '.ml. i \ . 'i.'^ical Laboratory of Cornell l"nivfrsit>- Mnlical College,
.t\-.

114 J- F - MCCLENDON.

1 am indebted to several persons for suggestions, especially
to Dr. Ralph Lillie 1 and Professor B. M. Duggar.

I. INTRODUCTION.

The object of this paper is to bring the "vital" phenomena,
as far as possible, within the scope of physics and chemistry, and
not to elucidate physical and chemical processes. It should
therefore be borne in mind that the osmotic phenomena of
"dead" systems are not all satisfactorily explained.

The Vant Hoff-Arrhenius theory of osmosis concerns itself
with the number of particles, molecules and ions, in solution,
and is applicable to dilute solutions, in which the total volume of
the dissolved particles is negligible. However, in more con-
centrated solutions, the volume of the dissolved particles is of
the same importance as the volume of the molecules in gases, as
expressed in Van der Waal's equation. Also the dissolved
particles bind molecules of the solvent and so reduce the volume
of the free solvent.

That the molecules and ions of a dissolved substance bind
some molecules of the solvent, follows from the work of Jones
and his collaborators. 2 Compare also the work of Pickering. 3
Jones concludes that the larger the number of molecules of water
of crystallization, the greater the hydrating power of a substance
in aqueous solution. The number of molecules of water bound
by one molecule of the solute usually increases with dilution up
to a certain point (the boundary between concentrated and
dilute solutions, beyond which there is no heat of dilution).
The bond between ions and the solvent is also indicated by the
phenomenon known as "electrical transference." It an elec-
trolyte and a non-electrolyte be dissolved in water and an
electric current passed through the solution, water will be carried
along with the ions to the electrodes.

With these corrections, the Vant Hoff-Arrhenius theory
accounts for osmotic pressure, but does not show why main
substances exert no osmotic pressure, in other words, why no

J Cf. this journal, 1909. XVII., 188.

2 "Hydrates in Aqueous Solution," Pub. No. 8, Carnegie Ins. Wash., 1907.
' Whetam, "The Theory of Solution," 1902, Cambridge, p. 170.

TENSION PHENOMENA OF LIVING ELEMENTS. 115

membranes have been found that are impermeable to them.
(Ki-rt'.n -upposed that the substance, in order to diffuse, must
di--ol\e in the membrane. Kahlenberg and others consider a
solution as a chemical combination between solute and solvent,
and osmosis as a series of chemical reactions between the mem-
brane and the two solutions, continuing until equilibrium is
established. The essential points in the theory arc: that the
membrane is not a molecule sieve, but a substance with specific
properties and the chemical characters of the membrane and of
ilie di ol\ed substances affect osmosis.

Wiflard < iibbs found that the more a solute lower-, the surface
tension <-t a -olution, the more it tends to pass out of the solution,
mosis, or if this is prevented, to collect at the Mirt'ac.- of
the solution. This law has been extensively investigated ami
eoiiiirmed b\ I. Truube. For instance, in general, lipoid-Milnble
Mib-taiM c- lower the surf;tcc tension of water and tend to diiiu-e
on i ol ii. \\liereas electrolytes slightly raise the surfaee i en-ion
of water and attract water from the adjacent pha-e. < Kmo-i-
ma\ 01 < ur in opposite directions simultaneously. (ill>l>- and
I'raiibe -tate ih.it the greatest osmotic flow is from the -olm i, ,\\
of louei -in face tension to that of the higher, but tin- i- not
v.-ner.ill\ a<iepted. ( )smosis consists of two di-i in< i pr
Irom our -oluiion to the membrane, and from the membrane to

the -eeoinl -ollltioil.

In case the membrane consists of two or more ehenhrallv
dill'etviit membranes placed one on another, osmosis con-i-t- of a
series of Steps; and Hamburger 1 made double membrane-, through
\\hieh en tain substances diffuse more rapidly in one direction
than in the other.

Tranbe calls the bond between solute and solvent the "attrac-
tion pie me." In general, attraction pre--nre of ions incrc.i
\\ith valence. The less the attraction pressure of the solute, the
more it louers the surface tension and tends to pass out of the
solution. The presence of one solute louers tin- attraction
pressure "t another in the same solution, and the greater the
attraction prc-.-nrc of a solute the more it lowers that of another.
\\ e mi-lit e\pre>- this idea by saying that one substance takes

. 7.cit., lynS. XL, 443.

Il6 J. F. MCCLEXDOX.

part of the solvent away from the second and increases the con-
centration of the second substance. This may explain the effect
of a harmless substance in increasing the toxicity of a poison.
Schnerlcn 1 observed that a solution of phenol below the threshold
of toxicity for certain bacteria is rendered toxic by adding NaCl.
Stockard showed that the toxicity of pure solutions of salts on fish
eggs is increased by the addition of sugar, although the total
osmotic pressure of the mixture is less than that of the normal
medium. 2

Just as Traube's precipitation membranes are absolutely
impermeable to certain substances, so do living cells show this
selective permeability. For instance, the vacuole fluid or cell
sap of certain plant cells contains colored substances which do
not diffuse into the protoplasm surrounding the vacuoles. If a
cell be placed in a solution of the pigment, the protoplasm
remains colorless. If the protoplasm be squeezed out of the
cell into a solution of the pigment, it does not invariably become
stained. However, if the cell is injured in certain ways, or
dies from any cause, the pigment diffuses out of the vacuoles
into the protoplasm and thence into the surrounding medium.
\Ye might conclude that the protoplasm in general is imperme-
able to the color, but at death it becomes permeable. On the
other hand, Pfeffer 3 gives evidence for the existence of a mechani-
cal membrane on the surface of the cell and lining the vacuoles.
De Vries 4 placed cells into 10 per cent. KNOa solution colored
with eosin. The plasma membrane and granular plasm died
and stained long before any dye entered the vacuoles. How-
ever, the granular plasm may have absorbed all the dye, thus
preventing its entrance for some time, without the necessity of
any resistance of the vacuole membrane. Since protoplasm may
be squeezed out in the form of droplets and still appears to be
surrounded by membranes, Pfeffer concluded that the membrane
was formed by the contact of the protoplasm with the medium

i Arch. exp. Path.. 1X96, XXXVII.. 84.

- However the. NaCl in Schnerlen's and sugar in Stockard's experiment may
have increased the permeability to the toxic substances, as discussed in later
chapters.

" I'Han/cnphysiologie."

4 Jahrh. wiss. Bol.. 1885, XVI., 465.

TENSION PHENOMENA OF LIVING ELEMENTS. ll"J

or with cell sap. He supposed these membranes to be the semi-
permeable parts of the cell, and that they became altered at
death. Pfeffcr called this membrane on the cell surface the
"plasma membrane."

Whereas the nuclear membrane and certain vacuole mem-
branes are semipermeable, these are lacking in crythrocytes,
which arc then-fore good objects for testing the question whether
the protoplasm in general, or merely its surface, is semipermeal >le.
Hober 1 by two very ingenious but complicated methods, one
based on dielectric capacity, determined the electric conduc-
ti\it\ <>f the interior of the erythrocyte without rupture of the
pl.i-nia membrane. Since the conductivity of the interior
'about that of a .2 per cent. XaCl solution) was found to be
ni.mv times greater than that of the erythrocyte as a whole, the
membrane must be relatively impermeable to ions. There is
much other, but less direct, evidence that the semipermeability
resides in the plasma membrane, namely: the rapidity of change
in permeability of certain cells, the sudden innva-e in perme-
ability of a cell after swelling to a certain size idue presumably to
rupture of the plasma membrane), the ease with which mild
mechanical treatment increases the permeability, and the locali-
/ation of electric polarization at the cell surf

< Miim k -upposed these membranes to be of a fatty nature.
This idea \\as carried further by Overtoil, who considered the
plasma membrane to be composed, not of neutral lai-, but of
sub- 1. in. es of the class which are called "lipoids." \\ hich included
iion--aponif\ ing ether soluble extracts of or-an-, /. <-., choN-tcrm,
lecithin, cuofin, and rerebrin. He found 3 that all basic dyes
were ea-ilv absorbed by living cell.-, but not mo-i of the -ulphonic
acid dyes. This corresponded to their -olubility in melted
cholesterin. or solutions of lecithin and cholesterin, or particles
of lecithin, protagon or cerebrin. His argument is somewhat
\\eakened. hou ever, by the fact that cholesterin decomposes
on melting, and that if lecithin is allotted to absorb water its
s. .1\ cut potter changes.

\rtlt. f. .: /., 1910, CXXXIII., 237, and Eighth Internat. I'hysiol.

I'mmif . Vi.-mui. H)IO.

her. (/. Kn. Pm f. Akad. d. H U l'-rlin, 1888. B<1. XXXIN".

*Jahrh . Bo/., ij->n. X X X l\ . 669.

II> J. F. MCCLEXDOX.

Many of Overton's critics do not distinguish between lipoids
proper and a host of ether-soluble substances which are also
ca'lled lipoids, and of the data which they present we will con-
sider only that on lipoids proper. Ruhland 1 found that certain
dyes stain plant cells but are not soluble in solutions of cholesterin
(and vice versa). Robertson 2 observed that methyl green freed
from methyl violet was insoluble in a nearly saturated solution
of lecithin in benzol, whereas it stained living cells. Hober 3
obtained Ruhland's results, when using certain animal cells,
but found that certain nephric tubule cells absorb all dyes that
are not suspension colloids.

Faure-Fremiet, Mayer and Schaeffer 1 state that pure choles-
terin does not stain with any dyes (contrary to Overton), mala-
chite green (considered lipoid-insoluble by Ruhland and Hober)
stains lecithin, and Bismarck brown (considered lipoid-insoluble
by Ruhland) stains lecithin, cholesterin-oleate and cerebrin.
A mere trace of free fatty acid greatly affects the behavior of
lipoids toward stains.

Mathews 5 considers the absorption of dyes by cells as a chemical
process. Since basic dyes combine with albumin in alkaline
solution, lipoids in the membrane are not necessary for the ab-
sorption of such dyes.

Traube objected to Overton's hypothesis on the ground that
Overton's plasmolytic series is the same as found by Brown, who
used the membrane of the barley grain, 6 and the same as the
series of the attraction pressures of the substances in water.
But Traube admits in his later papers that the chemical character
of the membrane affects osmosis.

We may conclude that, although the plasma membrane of
some cells may be lipoid in character, this lias not been proven,
but, in general, it is more permeable the more the diffusing sub-
stance lowers the surface tension of water.

1 Juhrb. wiss. Bol., 1908, XLVI., i, and Ber. DeiUsch. hot. Gesellsch., 1909,
XXVI., 112.

2 Jour. Bio. Chem., 1908, LV., I.

3 Biochem. Zeit., 1909, XX., 55.

4 Arch, d' Anal. Mic., 1910, XII.. 19.

* Jour. Pharmacol. and Exp. Tlier., 1910, II., 201.

6 But this is not true of all seed coats. Atkins, Sci. Proc. Roy. Dublin Soc., XII.,
n. s.. No. 4, p. 35, observed that the membranes of the bean seed are freely pcrme-
at4l, semipermeable plasma membranes arising only after germination.

TENSION PHENOMENA OF LIVING ELI- Ml. NTS. IK)

Nathanson 1 supposed the plasma membrane to be a mosaic
of lipoids and "protoplasm," but it is evident that if the lipoid
portion is not continuous, it can not make the cell impermeable
to any -ubstance.

( /apck- states that lipoid solvents cause cytolysis when the
surface tension of the solution is reduced to .68, and concludes
from tlii- that the plasma membrane contains glycerine tri-oleate
sinci ii- emulsion reduces the surface tension of water to thi>
figure.

Tin- diffusion of water-soluble substances through swollen-
]il.n _els" or "sols" of gelatine, varies inversely with the

viscosity (Arrhenius). The great hysteresis of gelatine gels i-
t.il.en advantage of to show that diffusion depends on tin vis-
cositj ami not on the per cent, of gelatine, at the same temper-
aim

Tin absorption of water by a gelatine plate increases its per-
meability. and the temperature and therefore the presence of
MI! .-tain es which affect this swelling of gelatine affect its perme-
abilit\. Impregnation of colloidal membranes \viih bile salts,
al-oliol, ether, acetone or sugar changes (usually increases)
tin ir | .ei mt -ability. The effects of substances on tin- rate of
dilliisioii through gelatine plates, and on their swelling (viscositj
and tin -It ing point arc not always quite parallel. 4

In the substance added to the membrane is removable,

(he i lian^c in permeability becomes reversible, which is true
in regard to many of the substances mentioned above. Changes
in non-li\in- membranes are usually only partially reversible or
are irreversible. I >enaturalization of a colloid membrane by
lu-ai. heavy metals, or other coagulative agents which induce
chemical changes in the membrane, or the addition ot substances
which cannot be removed, produce irn -\vr-ible changes in
permeability.

That the permeability of the membrane- in living ti--ne is
increased at death is proven by a host of observation-. The
electric conductivity increases enormously at death. Contained

Jahrb , i r. Bot., 1903, XXXVIII.. 284; 190). XXX IV.. 601, and XI... 403-
/-'.-. deut ill. hot. Gesell., 1910. 28, 480.

r. Asher & Spiro's Ergeb. der Physiol.. 1908, VII., 99-
. loc. cit.

I2O J. F. MCCLENDON.

substances diffuse out, substances in the- medium (fixing fluids,
stains, etc.) diffuse in. There is a more general mixing of tissue
substances. Enzymes come in contact with proteids and
autolysis results.

Certain substances are known to increase the permeability
of membranes in tissues of the body. Thus ether, chloroform,
etc., increase the penetration of fixing fluids, and the exit of
contained substances, and the mixing of tissue substances.
In this way they increase autolysis.

II. OSMOTIC PHENOMENA IN PLANTS.

It is evident that water, salts, carbon dioxide and oxygen
can, at least occasionally, penetrate plant cells, as otherwise
no growth could occur. In case of the higher plants, the same
is true of sugars and other bodies. 1 Janse 2 found that so much
KNOs is absorbed by Spirogyra cells in 10 minutes, that it may
be easily detected microchemically with diphenylamin-sulphuric
acid.

Osterhout 3 grew seeds of Dianthus barbatus in distilled water.
The rate of growth during the several days of observation was
normal. In nature, calcium oxalate crystals are found in the
root hairs, but are not formed in the distilled water cultures,
showing that the Ca comes from the medium. If placed in
calcium solutions, crystals became large enough to see with the
polarizing microscope in four hours, showing permeability to Ca. 4

Nathanson 5 found that nitrates and other substances entered
the cell. Ruhland also observed penetration of salts.

Traube-Mengarini and Scala 6 conclude that salts enter plant
cells only through the partition walls. At these places there
appears an "acid reaction" (bluing of methyl violet). They

1 See Laurent in Livingstone, "The R6Ie of Diffusion and Osmotic Pressure in
Plants," 1903, p. 67.

1 Versl. en Medeel. der Konikl. Akad. van afdeel. Naturs., 3. Reeks, IV. part,
1888. p. 333-

3 Zeits. f. physik. Chem., 1909, LXX., 408.

4 But compare von Maycnberg, Jahrb. f. wiss. BoL, XXXVI., 381, who found
little penetration of salts into fungous hyphae. And see Demoussy, Comptcs Rcnilus,
CXXVIL, 970.

1 Jahrb. wiss. Bol., XXXVIII., -'84; XXXIX.. 601; XL., 403.
6 Biochcm. Zeit., 1909, XVII., 443.

TENSION PHENOMi \A OF LIVING ELEMENTS. 121

interpret this as showing that the anion of the salt unites with
an H ion of an amino group, forming a free acid, and the kation
of the -.dt unites with the protoplasm. It appears to me that
the ba-i> of this conclusion is very slight.

Permeability may be investigated by a study of plasmolysis,
which consists in the shrinkage of the surface protoplasm away
from tin (i llulose cell wall, due to the osmotic pressure of the
hypertonic -olution of a dissolved substance which does not
I it-iit -irate. A regaining of turgor by the cell while in the hyper-
tonie -oliitioii indicates slow penetration of the sultance. The
l>la-im>l\ ii. method was originated by Xageli, who aUo noted
thai a shrinkage resembling plasmolysis but accompanied by
omuard dillu-ioii of dissolved substances, occur- at death or
-e\ tic injury to the cell. 1

The plain cell is surrounded by an elastic cell \\all. The
internal "-m.itir pressure may be divided into three re-uliant-:
that aii-inu rounding up of the cell is called turgor, that re-
sulting in stretching of the cell wall is sometimes distinguished as
turgescence, and that resisting the surface ten-i.ni oi tin- cell,
" . ential pressure."

The plasmolytic experiments of I Wrie^- and other-' are
interpreted \>\- them as indicating a select i\e impermeability
"I tin- pla-m.i membrane to neutral salts.

In the plasmolytic experiments of Overtoil' all -alt- pla
moly/ed permanently. Non-electrolytes fell in four unmp-,
thus: (am -uijar, dextrose, manit, glycocoll > urea, ;_;lucenn>
( -i h\ K ne -ali < 'hi>l, acetamid> methyl-alcohol, acetmiitril, ethyl-
alt ..hoi. phenol, aniline, isobutyl-alcohol, isoamyl-alcohol, methyl
acetate, eth\l acetate, butyl aldehyde, acetmie. a< etald.>\im.
Diffusion of substances of homologous series in< reased \\iih molec-
ular \\ eiiLill t .

< >\ ei inn a-t er i. lined the permeability of plant cells to alkaloids

1 " PtUm/rnpliy-i..!. I iitersuchungcn." 1885.

.'. pliysikul. cVi.-wi.. iSSS. II., 415; 1889. III.. 103.

3 (I. l.i\'int;-t>iir. " The R61e of Diffusion and Osiniir Pressure in Plant-."
Cliu-.ii;.'. \<j>>i; J.in-'. />')/. Ccntlb., 1887, XXXII., 21; Duggar, Trans. Acad. Sc.
>/. Lout-*. 1906, XVI.. 473-

4 \'ifrlfljai.' ifr .V aturforschers. Gesell. in Zuriili, XL IV.. 88; Jahr. wiss.
Jit.. 1900. XXXH

122 J. F. MCCLKXDOX.

by their precipitation of the tannic acid in the cell sap. Most
alkaloids penetrate rapidly, but only in the form of the free
(undissociated) base produced by hydrolysis. Hence the pene-
tration (precipitation and toxic effect) may be prevented by
adding a little acid to the medium.

Pfeffer had shown that methylene blue is precipitated by tannic
acid in the cell sap of certain plants.

Some discussion has arisen as to whether the mechanism of
the entrance of dyes into plant cells is similar to that of alkaloids.
Overton showed that lipoid soluble basic dyes penetrate easily.
He at first supposed that only the free color base (undissociated)
is able to penetrate the cell. 1 Overton found, however, that
triphenylmethane and chinonimid dyes disprove his assumption,
showing that it is at least not general. This question was taken
up again by Harvey 2 who found that neutral red or methylene
blue, which stain Elodea leaves in tap water, do not do so if just
enough acid be added to the water to prevent any free color
base from forming.

He observed that, although these dyes are not precipitated
in the cell sap of this plant, they become more concentrated in
the cell sap than in the medium. Neutral red is bright red in
the cell sap, indicating that the reaction is acid (no free color
base is present). He supposes that the absence of any of the
dye in the form of the free color base prevents it from diffusing
out of the cell, hence it becomes more concentrated within than
without.

In using the plasmolytic method, if a cell does not recover
from plasmolysis in a solution" of a salt, it is said to be imperme-
able to that salt. However, the cell may recover, but may be
killed by penetration of the salt, and shrink again. It is possible
that Overton and others failed in some cases to note this transient
recovery. Contrary to Overton, Osterhout 3 found Spirogyra
permeable to alkali-salts and alkaline earth salts, but more

1 In this connection it is interesting to note that Robertson observed that free
color bases, and to a less extent free color acids, are much more soluble in fats
than are their salts. This is what we should expect, since the salts dissociate in
water, and ions are insoluble in fats.

* Science, 1910, n.s., XXX11., 565.

3 Science, 1911, n. s., XXXIV., 187 ; XXXV., 112.

TENSION PHENOMENA OF LIVING ELEMENTS. 123

easily to Xa than to Ca. It is plasmolyzed by .2. M CaCl- 2 and
not by the isosmotic .2<)M NaCl but by .$x.\f XaCl. .195. I/
CaClo and -375.1/ XaCl just failed to plasmolyze. On mixing
100 c.c. -375-^f XaCl with 10 c.c. .I95J/ CaCl 2 , thus decreasing
the osmotic pressure of the former, marked pla-mol\ -.-is occurred.
Thi- indicates that Ca decreases the permeability to Xa. 1 From
further work by the same author, not yet published, it appears
th.it N.i increases and Ca decreases the permeability of certain
marine plants. Also Fluri 2 obtained increase in permeability by
salt- of .ilnniinium, yttrium and lanthanum.

I >' \rir-, plasmolyzed cells of Tradescantia, containing blue
(i-ll sap, \\ith 4 percent. KXO 3 solution, then added nitric acid
until tin- color changed to red. The acid made the cells pcr-
i- \ K\U 3 for they regained their turgor and linally bur-i.

i- explain- the easy penetration of acids into n-11-. I'tViin
found ihat if red beet cells, petals of Pitlmonariu, Manu-n hair-
oi Track 'tuitia and other anthocyan-containing cell- are placed
in extivmelv dilute HC1 or H-jSO.,, they suddenK- turn red. in-
diiatinii iniinediate penetration of the acid. It allo\\cd t<> re-
main Inn a short time, the cells are not killed, and tin- color
( lian^c i- 1 1 -versed on returning the tissues to acid-free water.

I have n- pealed these experiments, using cells of n-d Uii,
ii d i.ibb.ige and red nectar glands of Vicia _,'/;'</, and tind that
mineral ai id> penetrate, but that (the lipoid soluble acetic acid
penetrates much nion- rapidly and also more easily alters the
jila^nia membrane, causing pigment to diffuse out, if not can-
tii>n-l\ applied. Alkalis also penetrate, but 'the Lipoid soluble)
ammonia penetrates much more rapidly than the oiheis. Am-
monia doi-> not so easily increase the permeability t<> the pi^mt-nt

a- dor- arctic acid.

Kuhlaiid 1 after staining root hairs of Triancn, etc., \\ith tin-
indicators, methyl orange and neutral re<l, found that mineral
acid- a- well as lipoid soluble acids penetrated.

1 Tin- \\oik .-I Ki.irney, Report 71, U. S. Dept. of Agriculture, indicates that
(a prevents tli<- pl.i-molytic and toxic effect of MK. Imt thU i

follou illi; <!i.Mth.

* Flora i - XCIX.. 81.

3 "(KiimtUchc Untersuchungen," Leipzig, 1877, p. 135.

*Jnlirb. u-iss. Bot.. 1908. XLVL. r.

124 .! F. MCCLENDON.

One defect in the plasmolytic method is the fact that the cel-
lulose cell wall, if not very thick, is elastic, and a slightly hyper-
tonic solution may cause the cell to decrease in volume without
pressing the protoplasm away from the cell wall. This source
of error may be eliminated by substituting calculations of the
volume of the cells (as necessary for animal cells) for observations
on plasmolysis.

It is well known that movement, and in many cases increase
in size of plants is due to changes in turgor of the cells. If we
exclude the turgor changes in aerial plants produced by variations
in the ratio of the water supply to the transpiration, turgor
changes may be due to changes in the osmotic pressure of the
external medium, or of the cell sap (due to metabolic changes)
or to changes in the permeability of the plasma membrane.
Lepeschkin 1 has confirmed Pfeffer in showing that changes in
permeability of stipule cells accompany (or immediately precede)
changes in turgor. By chemical analysis of the medium he has
shown that an outward diffusion of dissolved substances, from
the cells, accompanies loss of turgor, and by plasmolytic ex-
periments, that the permeability to certain substances increases.

It is interesting to note the force that may be exerted by such
changes in turgor. From measurements of the pull of a stamen
hair of Cynara scolumus or Centaurea jacea on loss of turgor fol-
lowing stimulation, it seems not improbable that the change in
turgor amounts to 2-4 atmospheres (Hober). This also indicates
the strength of the cell wall necessary to prevent rupture of the
plasma membrane'. The osmotic pressure of the juices pressed
out of plants varies from 3.5-9 atmospheres. 2 The pressing
out of the juices causes an error due to chemical changes; on
the other hand, in taking the freezing point or pieces of plant
tissues, an error arises from lowering of the freezing point by the
walls of the capillary spaces. Miiller-Thurgau 3 found the A
(corrected freezing point lowering) of plant tissues =.8-3.1.
Many plants respond to light by definite movements, produced

1 Her. deutsch. hot. Gesell.. XXVI. (a). 725.

2 DeVries, Pringsheime Jahrbuchcr wiss. Bot., 1884. XIV., 427; Pantanelli, ;'/</,/..
1904. XL., 303.

3 Landu'irtschaftl. Jahrb.. 1886, XV., 490.

TENSION PHENOMENA OF LIVING ELEMENTS. 1 25

by turgor changes in certain of their cells. Trondle 1 found that
light produced changes in permeability of these cells.

Changes in permeability may not only affect the turgor, but
also the assimilation and excretion, and consequently the
metabolism and growth of the cells. Chapin 2 observed that
( '), in certain doses is a stimulant to the growth, not only of
green pl;mt> but also of moulds. As only ,i few saprophytes can
decompo-e ('();, it is not probable thai its effect is nutritive.
A ^imilar simulating action of ether and various salts, even
such toxic ones as those cf zinc, was previously knoun. These
salt-, probably stimulate without penetratini; the cells, since
7.\\, lor in-taiice, is not a constituent of protoplasm. ' This
one to suppose that the initial effect of .ill of these
i- on tip- surface, changing the permeability of the cells.

Wachter limd that potassium decreases the p'-rmeabilitx of
onion i ells. SuL'ar diffused out of sections of Allinm cc{>n placeil
in distilled \\ater or hypotonic sugar solutions, but a trace of
potassjuin s.ilt entirely prohibited the difliisii.n. When the K
\\as n-ino\ed tin- diffusion recommenced.

Czapek 8 determined increase in permeability by the exosmosjs
ol tannin in cells of Echcveria leaves. Various inono\aleni al-
c.h..ls and k< tones, ether, ethyl urethan. di and tri acetin,
Na-oleate, ohic acid, lecithin and cholesterin all \\.\-\ caused
mosjs ,,; i.innin in concentrations (aqueous solutions) \\hich
had a siitlacc- tension of about 0.68. It \\ould appear therel'on-
that these substances, chiefly of the class ,,| indifferent narcotics,
alter ihe cells if tliey diffuse into them, or dilhise into certain
Structures -\\c\\ as the cell lipoids or the plasma membrane.
It seems more reasonable to suppose that the' plasma membrane
is the siructure affected, and the more the substance louers the
siirlace teii-ioii of water, the more it diffuses into the plasma
membrane, \\hen this membrane is altered, it allo\\~ escape
of tannin. Some substances such as chloral hydrate are el lei tive

; Bot., IQIO. XLVIII.. 171.

X

I "<-\>. " I )\ n.uiiii'- i.i' Living Maiti-i," pp. 73, 74.
1 J.ii-.i-i- XI. I., 165.

6 " t 'l.i-r rim- Mrtli.i.lt.- /in ilirekten Bestiininung der OberMchenspannung der
ri.i-iiKili.uit \"ii I'll.in/cii/clli-n." Jena, (".. Fi-clitT.

126 J. F. MCCLENDOX.

in less concentration, and probably affect the cell chemically
as well as physically.

Mineral acids caused exosmosis of tannin when the concen-
tration just exceeded 1/6,400 normal, and the effect is probably
due to H ions. At this same concentration Kahlenberg and True 1
found the growth of seedlings of Lupinus albns to cease. It
appears, therefore, that this cessation of growth is due to in-
creased permeability, causing decreased turgor of the cells.

Changes in permeability may also affect secretion (excretion).
The addition or formation of alcohol or acetates causes yeast and
other fungi to secrete (excrete) for a short time, various sub-
stances, especially enzymes which do not come out in a culture
medium lacking the reagent. 2 It appears that the alcohol or
acetates increase the permeability of the fungi to these substances.

My own experiments 3 indicate that pure MgClo solutions
increase the permeability of yeast. A certain per cent, of yeast
and dextrose in .3 molecular MgClo eliminated CO 2 more rapidly
than .$M NaCl or .^2^M CaCl 2 , all which have about the same
freezing points. Also, the CO 2 elimination was more rapid in
the magnesium solution than in a solution of the same concen-
tration of MgCl 2 with either of the other salts in addition, or in a
solution containing NaCl and CaCl 2 in the same concentrations
as in their respective pure solutions, or in a solution of all three
salts, or in tap or distilled water. In order to determine whether
the magnesium entered the cells I took two equal masses of com-
pressed yeast and agitated one in H 2 O and the other in a molecular
solution of MgCl 2 for 5 hours, i-hen washed each rapidly in H 2 O
by means of the centrifuge. The ash of the magnesium culture
= .048 gram, that of the control = .0466 gram. Evidently
the Mg did not enter the yeast to any great extent, and probably
acted on the surface, increasing the permeability.

Ewart 4 observed that after placing plant tissue in 2 per cent.
HC1 and washing in water its electric conductivity (ionic per-
meability) was increased. If one portion of the plant is stimu-
lated, the stimulus may be transmitted to other portions. ' In

1 Kahlenberg and True, Botanical Gazette, 1896, XXII., p. Si.

2 Zangger, "Asher and Spiro's Ergcb. d. Physiol.," 1908. VII., 144.
McClendon, Am. Jour. Physiol.. 1910, XXVII., p. 265.

4 " Protoplasmic Streaming in Plants," Oxford, 1903, p. 96.

TENSION PHENOMENA OF LIVING ELEMENTS. 12~

this way increase in electric conductivity was produced by stimu-
lation of a point outside the path of the current.

\Yhereas many plants are very sensitive to sudden and extreme
changes in osmotic pressure, Osterhout 1 found that certain marine
alga- thrived when subjected daily to a change from fresh water.
to si a water evaporated down until it crystallized out, and vice
versa. He does not state whether these algae survive extreme
plaHno]v,js : or whether they are so easily permeable to salts
as ii"t t. I..- plasmolyzed by the saturated sea water or burst
1>\ tin- fre-h water.

Foi rej 'il.u ion to slight changes in the osmotic pressure of the
m< dium. .i change in size of the cell altering the turgescence, or
ti nsion <>! the cell wall, is sufficient.

If / . iitia cells are placed in a hypotonic solution, they

begin to -\\ell. But soon crystals of calcium oxul.tu are formed
in the cell -ap, and in this way the turgor, due chielly to oxalic
.K ill. i- redinvd. 2 It would be interesting to know wh.it i- the
source <>f the Ca. Was it previously in combination with pn>-
teids?

Tin nmodation to a hypertonic medium take-, pla<< , ac

iin^ to van Rysselberghe, parth through ab-.o. pi inn of
Mil>-',c : the medium and partly through metabolic produc-

tion oi . -iiiitic substances, chiefly the transformation of >uuvh
into ox.ilic .1.

Ill BlO-ELECTRICAL PHENOMENA.

I. /;/ P hints.

( hai .< in permeability of the plasma membrane to ion- \\otild
necessarily cause electrical change due to its inlhu-iu on the
migration of ions. These electrical change- actually occur, and
may be ea-ily studied.

Stimulation or wounding in plants is accompanied by an elec-
tronegative variation of the affected suiface. This negative
region spreads in all directions over the surface, but the rate of

1 I *niv. of Cal. Pub.. Bot.. 1906. II.. 227.

Jic. Mem. d. 1'Acad. royale de Belgique, 1899. LVIII.. i.
1 Compare von Mayenberg Jahrb. f. wiss. Bot.. XXXVI., 381.

128 J. F. MCCLENDON.

propagation 1 is much slower than the similar process in muscle
or nerve. 2

Pfeffer 3 supposed that the plasma membrane is normally per-
meable to ions of only one sign. Since the normal cell surface
is positive in relation to the cell interior (cut surface) we may
conclude that the plasma membrane is normally more permeable
to kations (less permeable to anions). Just as the negative
variation of wounding is due to the removal or rupture of the
plasma membrane, so the negative variation of stimulation would,
on the membrane hypothesis, be due to increase in permeability
of the plasma membrane to the confined anions.

An alternative hypothesis is that these electrical changes
result from changes in metabolic activity. The production of an
electrolyte whose anion and kation have very different speeds
of migration (such as an acid or alkali) would cause electrical
changes. But how are we to account for changes in metabolic
activity? There exists varied evidence for changes in perme-
ability, and it is simpler to assume that changes in metabolic
activity and electrical changes are both the result of changes in
permeability.

Kunkel 4 tried to explain the vital electrical phenomena as the
result of the movement of fluids in the vessels of the tissues, but
bio-electrical changes may occur without such movement of
fluids (Burdon-Sanderson).

Kunkel observed in i882 5 that the movement of the leaf of

i

Mimosa pudica is accompanied by an "action current," or nega-
tive variation of one surface of the pulvinus. Similar results on
Dioruza leaves were obtained by Munk' 1 and specially studied
by Burdon-Sanderson. 7 It was stated above that Lepeschkin
had shown that the turgor changes in plants were accompanied
or i in mediately preceded by changes in permeability to certain
substances. The electrical phenomena suggest that the turgor

1 Which is in mimosa 600-1,000 times as fast as the geotropic impulse in a root.

2 Fitting, "Ashcr and Spiro's Ergeb. d. Physiol.," 1906, V., 155.
" Pflanzenphyaiologie."

*Arch.f. iL ges. Physiol., 1881, XXV., 342.

s See Wintcrstein's "Ilandbuch der vcrgleichenden Physiologic," III. (2), 2,
p. 214.

Arch. f. Anal. it. Physiol., 1876, XXX., 167.

" Proc. Roy. Soc. London, 1877, XXV., 441; Philos. Trans., 1888, < I XXIX.. 417-

TENSION PHENOMKNA OF LIVING ELEMENTS.

change is accompanied for immediately preceded) by increase in
permeability of the plasma membrane to anions. Burdon-
Sanderson states that, whereas the m< >\vment resulting from
turgor change begins 2.5 seconds after stimulation, the negative
variation reaches its maximum I second after simulation. This
max In- due to the mechanical inertia, or the time required for
the diffusion of substances.

It ua- sated in the preceding chapter that light change- the
permeability of the plasma membrane, and Waller 1 found cor-
responding electrical changes due to light, but not al\va\- in the
same direction in different plants. This inconstancy in direction
is probably due to the fact that light not only influences the
I -i i mi abilitx , but also the assimilation, and changes in a imi-
laiioii produce electric changes. This is supported by the fact-
thai i >IH rtoii- found that assimilation as well as electric chai
i- uios affected by the longer light rays.

2. In Muscle and Xerve. 3

i >s \\ald' proposed the hypothesis that the electric pin noniena
ot must le, nerve and the electric organs of fish (which mav n -at h
se\eral hum lud volts) are produced with the aid of semiper-
meable membranes. The alternative theory of Hermann, \\hich
\\oiild account for the current of injury by assuming the pn>-
oi some electrolyte (alkali?) in the wounded region. \\h"-'
- and kaiions have very different speeds, >eem- le , ]nb-
.ibl\ to be i he correct one.

\. c. .j-diii^ to the "membrane theory," the muscle or net \ e
eleineiii i-- surrounded by a semipermeablc membraiu- allo\\iu^
easier passage to kations than to anions. The kaii./n- ua in-
through the membrane are held back by the ne-ati\e field pro-
duced by the confined anions, but owing to their kinetic energy,
the kation- pa-s out far enough to give the outside of the cell
-in lace a i io-i t ive charge. Therefore an \ portion of the siirfai
that i- made freely permeable to anion> become- electronegative

Jota 1 l:\aiol.. iSgg-'oo, XXV., 18.

ntiilniti.ni a lYtude du mode de la production <!' l' l-i trii it. 'Ian- <-tres
\i\anii-. I r.i.,nt\ ,1,- 1'hfititnt Solvay. 1902, \'.

R 1 LHie, .1".-.;. .'ur. Physiol., ign. XXN'III.. 1.^7.

* /.fit. /-/IVV/A-. L'h,-IH.. I Soc.. \'I.. 71.

130 J. F. MCCLENDON.

in relation to the remainder of the surface. This negative
variation may be produced by artificially removing or altering a
portion of the membrane (producing the current of injury) or
as the result of normal stimulation, making it permeable to anions
(action current).

Bernstein resorted to mathematical proof of this hypothesis.
We \vill not here go into details, but the gist of the matter is
that if the process were as we have imagined it, the electromotive
force of the current of injury, or action current, should be pro-
portional to the absolute temperature. He found this to be
true for temperatures between o and 18, but between 18 and
32 the E.M.F. was found to be too small. The muscle was not
permanently injured by exposure to the higher temperatures
for the length of time necessary for the experiments. Bernstein
explained this discrepancy by the further assumption that at the
higher temperatures the plasma membrane became slightly
more permeable to anions. 1

Since the muscle contains a higher per cent, of potassium than
the blood plasma or lymph, it might be supposed that K ions
passed outward through the plasma membrane and gave the
surface of the muscle element the positive charge. But if this
were the case, the current of injury should be reversed by placing
the muscle in a solution containing potassium in greater concen-
tration than in the muscle. This reversal, how r ever, was shown
by Hober not to occur. Since lactic and carbonic acids are pro-
duced by muscle and diffuse out in increased amount on contrac-
tion, one might suppose H ions to give the muscle surface the
positive charge. This is only a guess (and a poor one, since un-
dissociated molecules of COz and lactic acid are lipoid-soluble)
but may be convenient until some better one is proposed. Per-
haps the carbonic acid combines with amphoteric proteids, which

1 This is similar to the conclusion reached by Biataszewicz, Bull. d. I' A cad. d'
Sc. d. Cracovie, Sc. Math. e. Nat., Oct., 1908, p. 783, in regard to the unfertilized
frog's egg. In order to explain his observation that the rate of swelling in tap
water increased 5 times for every 10 rise in temperature, he assumed that heat
increased the permeability to H-^O. This would seem to be the simplest explana-
tion, provided the swelling were not due to chemical production of osmotic sub-
stances: and since the A of the ripe ovarian egg is .48 but is reduced to .045 after
oviposition, Biochem: Zeit., 1909, XXII., 390, much if not all of the swelling is
probably due to the initial osmotic pressure of the egg interior.

TENSION PHENOMENA OF LIVING ELEMENTS.

tlu-n set free H + and HCOs~ ions, thus increa-in- the ionization
and therefore reducing the number of undissociated molecules,
which can escape. 1

Since Osterhout showed that certain electrolytes may alter
the permeability of cells, we might expect to find, on the membrane
hypothesis, an effect of salts on the electric polarization of
imi-rle. H ober 2 observed that a portion of the surface of a
mu-cle treated with certain salts, KC1 for instance, becomes
i -1. iro-nriMtive (more permeable to anions) whereas a portion
n-l \\ith Nal or LiCl becomes positive (still less permeable
i" anions than is the normal unstimulated muscle). The order
of eite< ii\ i ness of the ions is as follows: Li<Xa<Cs<XHj< Kb
<K ,m<l ( \S<NO 3 < I< Br<Cl <valerianate, bui\rate, pro-
pioiiai' ate, formate <SO4, tartrate. Similar ionic series

\\en- found by Overtoil, K. l.illie, Schwartz, Mathews, C.nii/iur,
I loin i , and Mayer in the effect of salts on the function.il at ii\ iiy
of inii-( It , nerve and cilia, but the exact relation of tlu--e phe-
nomena to permeability is not understood in every case. Pure
-ohnioii^ of salts of alkali metals may "inhibit" inu-cle by in-
permeability, but salts of alkali earth metals art -aid
to "inhibit" by decreasing permeability. . Mayer says that the
effect of -.ills on cilia is the reverse of that of muscle, but the
relation (.f ihis to permeability is not known. Since ion- aiit-i i
tin .ition state of hydrophile colloids in the same or ex-

actl\ n \ -i.-ed order, and the kation series is found in no other
knoun physico-chemical phenomena, it might be supposed th.it
tin semipermeable membranes of muscle are colloitl.il.

It -i t-ni- probable that sugar solutions inhibit the artivity of
nniM le l.v increasing the permeability, but since -u-.ir i- not an
t K i trol\ tt thi> question cannot be tested by electric method-.

A ne-ati\e variation of muscle may also be produced by the

ailed "h.emolytic" substances, but i> inv\ t -r-ible, whereas
that produced by salts may be reversible. In this connection it

1 RiMl'. Q. J. Exper. Physiol.. 1910, III., 171, suppo-< <1 ttio aninn to be i>n>t--in;
|II>\MAI-I H h.i- not been shown that proteids, or even aniino acids dilTu-r nut nn
stiniiil.ith.ti. I do not see that the speculation of Galcntti, Zeit.f. All gem. Physiol.,
1907, \ 1 at all explanatory.

r's Arch., 1910, CXXXI\'.. 311.

132 J. F. MCCLENDON.

is interesting to note that Overton 1 found the permeability of
muscle to be similar to that of plant cells.

It might appear to the reader that the membrane theory is
merely wild speculation. What proof have we that on injury
or during contraction the muscle is more permeable to any ion?

DuBois Reymond 2 and Hermann 3 explained the fact that living
muscle has a greater electric resistance than dead muscle on
the hypothesis that the resistance of living muscle is due to the
presence of membranes, which become more permeable at
death. They demonstrated the resistance of muscle tissue to
the passage of ions by the fact that electric polarization occurs
in muscle tissue on the pasage of an electric current. It seems
to me that Kodis 4 and Galeotti 5 take a step backward, in at-
tributing the decreased resistance of dead muscle to the liberation
of ions. Galeotti tried to support his view by determinations
of the freezing points of the living and dead muscle, but found
on the contrary that the change in electric conductivity of the
muscle did not correspond to the change in the osmotic pressure.

Du Bois Reymond 6 observed that the electric conductivity
of muscle changes on (during?) contraction and Galeotti 7 found
it to be greater on strong contraction than on weak contraction,
and least on fatigue-exhaustion or cold-anaesthesia. However,
the duration of a contraction is momentary (about 1/5 second for
frog's muscle) and it is not clear that these investigators measured
the conductivity accurately during such a brief period, in fact
they probably measured it after contraction. Therefore I
decided to repeat these experiments, using a method by which
I could measure the conductivity during the actual contraction
period, as well as in ilic unstimulntcd condition. 8

1 PJluger's Arch., 1902, XCII., 115.

2 " Untersuchungen iiber thierische Elcctricitat," 1849.

3 PJluger's Arch., 1872, V., 223, VI., 313.

4 Am. Jour. Physiol., 1901, V., 267.

*Zeil.f. Biol., n. f.. 1902, XXV., 289; 1903. XXVII.. 65.

6 Loc. cit.

' Loc. cit.

8 McClendon, American Journal of Physiology, 1912. XXIX., 302.

TENSION PHENOMI NA OF LIVING ELEMENTS.

Experimental.

Platinum electrodes, platinized with platinic chloride contain-
ing a little lead acetate, and of a form similar to those designed
by r.aleotti, were used. Galeotti stimulated tin- muscle through
the same electrodes used in measuring the electric conducmity,
by -\\iiching on a different electric current. Though it were

-iblc to throw a switch quickly enough to have the curivm
for inea-iireinent of conductivity pass through the muscle
during contraction, it would be necessary to use a string u.il-
\ -.mom! -it -r to take the reading, and this method would probably
noi b<- very accurate. A more accurate method is that of Kohl-
IMII-I li, in which a rapidly alternating current reduces polari/atioii
at the electrodes and in the tissue, but it is necessary to throu the
nni-cle into tetanus in order to have time for the reading. I
a< Miiipli-lu-d this by using the same current for stimulation and
inra-mvmein of conductivity. A very small induction coil wa-
luted \\itli a rheostat in the primary. Another rheostat in the

-e ilaiA could l>e thrown out of the circuit by a s\\itch. By

adjiMini; ihr rheostats, a current strong enough to be dis-
tiiu tlv heard in the telephone, yet too weak to stimulair the
nm-cle, \\a- obtained. By switching the resi-tance out of the

ondarj > in nit, the current could immediately be im-iva-rd so
a- to thiou the muscle into tetanus. Since the \Vhrai-tonr
britlg* \\a^ u->ed, the difference in current strength- had no <lii
effect "ii tht- readings. The conducti\ it\ increa-nl from o to
28 per cent, (being usually about 15 per cent.) on stimulation.
\Ye ha\c, then, evidence for the increase in prrnu-ability of
muscU- to ions during contraction, but what relation has this
to tin- mechanism of the contractile process? It has been su--

;rd by D'Arsonval, Ouincke, Imbert, P.t rn-t ( in, (ialeotti
ami others that the increased permeability to ion-, causes a dis-
appearance of the normal electrical polari/ation of the elements,
surface tension consequently increases, cau-ing them to
up (shorten). But what are the elements concerned:'
It would be confu-ing to assume them to be the fibers, as then the
function of the complicated internal structure would be mu\-
plained. They are probably not the sarcous elements (por-
tions of fiber bet \\een 2 Z-lines) as the rounding up of these ele-

134 J- F. MCCLENDOX.

ments would elongate the muscle. And even though contraction
were produced by inequality in surface tension, as assumed by
Macallum 1 the total surface change would be so small as not to
account for the energy liberated in contraction. In order to
avoid this last difficulty Bernstein made use of hypothetical
ellipsoids. These were surrounded by elastic material to account
for elongation of the muscle. 2

The great differences of potential (several hundred volts) that
may be produced by the electric organs of fish, is achieved by
the arrangement of the modified muscle plates in series. All of
the plates have the nerve termination on the same side. On
stimulation of the nerve, each plate becomes negative, first on
the nerve termination side, and thus the negative side of one
plate touches the positive side of the next plate. In this way
the direction of the current may be determined by studying the
anatomy of the innervation. This rule, discovered by Pacini,
finds an exception only in Mahpterurus, whose electric organ
is supposed by Fritsch to be derived, not from muscle but Irom
skin glands.

The electric fish are relatively immune to electric currents
passed through the medium. This is not merely an apparent
immunity due to the fish being out of the path of the current, or
the current being short circuited by sea water (in case of marine
fish). I have received severe shocks from a torpedo that was
entirely submerged in sea water.

3. Amoeboid Movement. 3

The normal unstimulated surface of plant and animal tissues
is electro-positive in relation to the cut or injured surface of
the cells. We have given reasons for assuming that this indicates
greater permeability of the plasma membrane to kations than to
anions, the latter accumulating in the cell interior, gives it a
negative charge.

There are two reasons for believing that this is true also of ihe
Amoeba:

1 Science, n. s., 1910, XXXII.. 822.

2 Meigs., Am. Jour. Physiol., 1910, XXVI., 191, supposes the rounding up of
muscle elements due to increased turgor.

3 McClendon, Arch. f. d. ges. Physiol., 1911, CXL., 271.

TENSION PHENOMENA OF LIVING ELEMENTS. 135

1. If a weak electric current is passed through water in which
an Amoeba is suspended, it is carried passively toward the anode,
indicating that it has a negative charge. This charge may be
due to confined anions.

2. If a stronger electric current is passed through an Anm'ba,
it begins to disintegrate first at that surface nearest the anode.
The disintegration is probably due to the accumulation of ions
retarded l>y the plasma membrane. The ions in the medium are

I- 1 pass around the Amoeba, but the contained ions must pass
i In- plasma membrane in order to migrate to the fleet rodes.
since tin- disintegration is toward the anode, it i- prol.ubly due
to anions which cannot get out of the Amoeba. Since no corre-
sponding disintegration begins toward the kathode, the plasma
membrane is probably more permeable to kations.

Tin- Mir face tension of the Amoeba is very low, ami apparently
iiKT'M-i - on strong stimulation (indicated by rounding up <>I the
ii'<i). \Ve saw that stimulation in plant and muscle alls
caused increased permeability to ions, and consequently dis-
appear. mee of the normal electrical polarization, and thercbv
can-iir^. im r eased surface tension. \\'e might conclude therefore
that the low surface tension of the Amoeba is caused b\ electric
pulari/ation, due to the production of some metabolic elect rol. te
\\ho-e anions cannot escape; and that strong stimulation causes
increased permeability and hence disappearance of the electrical
polarization.

This \\ould explain all negative tropisms of the Aniii:l><i. The
surface ten-ion of the portion most strongly stimulated is in-
. reased, an<l the Anucba Hows away from the stimulus.

In order to explain positive tropisms we would have to make
another assumption. If the stimulus did not act directly on the
plaMiia membrane, but penetrated the Anuclm and acted on the
protoplasm, and increased the production of the metabolic
product producing polarization of the plaMiia membrane, it
would thereby decrease the surface tension. The local decrease
in MII lace tension would cause the Amoeba to tlou toward the
source of the stimulus, just as the quicksilver drop in dilute
ll\( ' Hows toward potassium bichromate in Bernstein's experi-
ment .

136 J. F. MCCLENDOX.

All stimuli producing positive tropism would then have to
penetrate to a greater or less distance into the Amoeba. But the
same stimulus thus acting on the interior might, in greater
intensity, affect also the plasma membrane, increasing its
permeability and changing the positive to negative tropism.
Such a change of the sign of tropism has been observed.

Soap lowers the surface tension of fats and lipoids, and Ouincke,
Biitschli, Loeb, Robertson and others supposed that lowering
of the surface tension of living cells might be due to soap. How-
ever, I found that soap always causes negative tropism in Amoeba,
probably because it increases the permeability of the plasma
membrane.

4. The Propagation of the Bio-electric Changes.

On the hypothesis, that the electric phenomena in muscle and
nerve, as well as other animal and also plant tissues, is due to
change in permeability to ions, we might hope to explain the
wave-like propagation of these changes. Since extraneous
electric currents "stimulate" all tissues (presumably by in-
creasing permeability) thus causing them to produce additional
electric phenomena, it seems natural that these latter would be
self-propagating. It is probably the negative variation of nerve
which stimulates the muscle, and the negative variation of the
portion of the muscle fiber adjoining the nerve ending, which
stimulates the adjacent portions of the muscle. Ncrnst found
mathematical proof that electric stimulation is due to change in
ionic concentration at the semipermeable membranes.

I have found evidence that the negative variation (current of
injury) in plants, may strongly affect adjacent cells. If an
electric current of suitable density is passed through plant or
animal tissue, negatively charged colloids in the protoplasm
migrate toward the anode. I have observed this movement in
living cells, and the resulting displaced bodies in histological
sections. In certain cases there may be some doubt whether
the colloids moved toward the anode, or water toward the
kathode, but in others, easily distinguishable bodies such as
chromatin granules or threads moved toward the anode.

If the tip of a root be cut oil we observe a negative variation

TENSION PHENOMENA OF LIVING ELEMENTS.

of the cut surface. This produces an electric current through
the medium and surrounding tissue. The fact that the current
actually passes through adjacent cells is shown by a displacement
of their contained colloids, identical in appearance with the
displacement produced by the currents used in the abo\v experi-
ments. V-mec 1 apparently observed these changes but did not
correctly d< -< ribe or interpret them.

The fact that an electric current on inert MM- make Cumulates
inu-cle .it the kathode, and the fact that the muscle -nrface is
ix mil. illy po-itive in relation to the interior (cut surface), prob-
ably indicates that stimulation is produced by a rapid depolari.
/ation df the muscle surface.

It tlii- reasoning be applied to an individual contractile
element, \\e may assume that the current causes kations i<> leave
tin (HINT surface of the membrane, and other kation- to be
an i -at ltd i. ward the inner side of the membrane, and thus the
pnl.iri/ation disappears or may even be reversed. JUM how this
causes an increase in permeability of the membrane i- a matter
\\hich \\- \\ill leave to the future for discussion.

1 1 ha- b -i -i) supposed that the stimulated region acts as kathode
it. tin- adjacent portions, and these in turn act as katlnli < to
tin in \t portions and so the stimulus is propagated.

^i imnlai ii .11 of a part of the surface, causing it t > I eo >me nx >n-
pcinical.lt in ions, depolarizes the adjacent parts "f the -urlai e
n\\iii- tt> the fact that confined anions migrate through the
pcrmcablt n and neutrali/e the charge- t.| the kations on

adjacent parts of the impermeable region (see Fig. i). 1 <>r ihi-
iva-oii the increase in permeability is propagated.

I hi- e\|ilanation of the phenomenon in a single rleim-m hold-

lor a ti ue made up of many elements provided tin -t are in

contact, as illustrated by the accompanying Fig. 2. Thi- i-

]in.babl\- the mechanism of propagation of the negative variation

and "-tiniulus") in many plant and animal tis-ti>

This mechanism accounts for the movement of the negative
variation over a -nrface. But it may be possible for this electric
change to jump from one element, to another not touching it.
The observation- on the current of injury, cited above, show that

1 "Rci/li-itiini; u <1. rd/loitrrulrn Strukturen b. d. Pflanzen," Jena, 1901.

J. F. MCCLEXDOX.

increased permeability of part of a tissue surface, may cause
electric currents to flow through cells some distance from the
wound. These currents probably stimulate the cells through
which they pass, which in turn become permeable and produce
electric currents. This explains the propagation of stimuli

4-

Anions represented by minus sign, kations represented by plus sign. Arrows
denote the direction of migration of ions. The large, circle represents the plasma
membrane, the dotted line denoting the permeable and the continuous line, the
impermeable portion.

through loose tissues, and the structural changes, as observed
by Nemec.

The rate of propagation of the "wound stimulus" is very slow,
whereas that of propagation of the "stimulus" (negative vari-
ation) in sensitive plants is more rapid, and that of the nerve
impulse still more rapid. We have not, however, sufficient data
to show whether this is a mathematical objection to the hy-
pothesis.

The streaming movements in plants may be stopped by a
strong stimulus or "shock." This stimulus is usually propagated
in one or more directions. Ewart 1 states that the rate of propa-
gation at 1 8 in a single elongated cell of Nitella is 1-20 mm.

1 Loc. cit.

TENSION" PHENOMENA OF LIVING ELEMENTS.

139

per sec., but where it has to pass cell walls .001-. 03 mm. per sec.
However, the stoppage of the streaming was his criterion of the
presence of the stimulus, and probably the banking of the stream

+ -f + +

h
f-

h

1-

f
h

i~
+

T
-h

-f-

-1

FIG. 2.

I In- -|ii.iu-~ represent the plasma membranes <>: Fur further

explanation see Fig. i.

at din point, soon stopped the whole stn-am thn^ sinnil.uini!,
tin- pn 'i Motion of the stimulus.

\\ . \AK' ' i>IS.

If si ijnul.it ion consists in increase in prrmraln'lity, we should
r\P-t i in.i -tin tics to prevent this change. The object of this
c'liauu-r i- 10 prrsent evidence that may support or refute such a
hypothesis.

<>\ertoii i -1 .served that warm- and cold-blooded vertebra;
insects and entomostraca, require practically the same con-
centration of the ana-.-t Ill-tic for narcosis. Certain groups of

I4O J. F. MCCLENDOX.

worms require double, and protozoa and plants six times this
concentration. We might conclude from this that nerves (and
especially medullated nerves?) are more susceptible to narcosis
than arc other cells. All groups of worms contain nerves, but
Loeb has shown that certain worms may perform coordinated
movements after the nerves are cut, hence the higher concen-
tration of the narcotic required to quiet them. However it
should be remembered that over-stimulation causes rounding up
and quiescence of Amoeba and muscle may be paralyzed by
increasing the permeability. The growth of plants is increased
by a certain concentration of ether and retarded by a greater
concentration. It may be that true narcosis (decreased perme-
ability) of protozoa and plants cannot be produced by such
substances as ether, etc.

Vertebrate nerve tissues are rich in lipoids (which have similar
solubilities to neutral fats) and it is therefore significant that
Overton and also Meyer 1 found that the partition coefficient of
anaesthetic between olive oil and water corresponds to its anaes-
thetic power. Meyer 2 showed further, that with change of
temperature, the change in the partition coefficient between oil
and water, and the anaesthetic power of the substance were
parallel. Pohl, Frantz, Grehaut, and Archangelsky found that
chloroform, ether, alcohol, chloral-hydrate or acetone, became
more concentrated in the central nervous system than in other
tissues. This is probably due to the absorption of the narcotic
by the lipoids (especially the immense mass of myelin) in the
nerve tissues.

It it could be proven that the plasma membrane consists of
lipoids, this solubility of narcotics might be considered direct
evidence for or against the permeability hypothec-, but lacking
such proof we must first attack the subject from another side.

Hober 3 observed that ethyl-methane, phenyl-methane, chloral-
hydrate, chloroform and hypnon, in low concentration prevent
the production by salts, of the current of injury on muscle.
He showed that in lethal doses on the contrary these narcotics do

1 Arch. exp. Path. u. Pharm., 1889, XLII., 109.

2 Arch. exp. Path. u. Pharm., 1901, XLVI., 338.

3 Pfluger's Arch., 1907, CXX.. 492, 501, 508. Cf. R. Lillie, Am. Jour. P/y> /./..
1912, XXIX., 373.

TENSION' PHENOMENA OF I.IVIM, ELEMENTS.

not prevent but even produce a current of injury, in this \vay
explaining data which might otherwise seem to contradict the
first statement. Galeotti and Cristina 1 observed that ether,
ethyl-chlorid, and chloroform produce a current of injury on
's nm-cle.

\\'e may conclude, then, that anaesthetic-, in the concentration
prodiK ing narcosis, so change the pl,t-ma membrane as to
pri-Miu -.tits from making it permeable to anions. This i-
pn.bably a No true of nerve, since Hober found th.it cthyl-
intili UK in low concentration prevented the- -cii-iti/ing of
in i \ e \\ ith I\_M)<.

Hi'lit-r In- attempted to connect these facts with tin 1 lipoid
>olul)iliiy of narcotics. Moore and Roaf 2 h.nl ob-er\vd that
// i/utinlilics of such narcotics as chloroform, alcohol. ether.
"i brn/ul. precipitated lipoids extracted from organ- and -u -
prmlrd in \\ater. Hut Hober and Gordon"* found that colloidal
solnii.ui- i.i' Irdthin were not precipitated, but were made tran
j '.in-lit b\ i-ilu-r or chloroform in hi^h concentration. SimilarK,
( n.ld-( hmidt and 1'ribram 4 observed that lecithin .-n-])cnilrd in
Na< I -i.liitioii, which is dissolved by chloral hydrate, methane.
or ' '" linr. in high concentration, is precipitated by tin m in 1"\\
Concentration, < >u the other hand, Koch and Mcl...ii,
that hli.ral. hypnon, acetone, or pure ether, do nut change the
H/e i't ti'llnidal particle.s of lecithin (i. e., make them ta-ii-r or
more diiticult to salt out). Calugareanu 6 explain- the mech.m-
JMU of tlu- precipitation ol lipoids by anaesthetic- by tin incn
in >i/.- of the particles due to absorption of the an.e-thetic.

Tim- there seems to be a parallel difference bet \\een tin- action
of low and high concentrations of anaesthetic-, <> nni-clc ami
nerve, and the action of the same on lipoid -u-pi-n-ion-. but this
dm-- not hold true for all cases. Moore and l\oa! 7 conclude that
ana -thetii - are bound, not only by lipoids, but al-o by pn.teids,

1 , :ol.. IQIO, X.. I.

>ndon. 1904. LXXIII.. 38 . LXXVII., 86.
//. I^-ilrage, 1904, \'.. 432.

h. H. Ther., 1909. \'I.. i.
Jour. 1'lxirm. and Exp. Ther., 1910, II.. 9
. 1910, XXIX.. 96.

142 J. F. MCCLENDOX.

and their charactersitic action on the permeability of the living
cell may be due to their action on proteids. In other words, the
plasma membrane may be entirely proteid.

It is well known that during narcosis little or no oxygen is
absorbed by nerve tissue. Yerworn and his pupils assumed that
the narcotic directly suppressed oxidation. On the other hand
Mansfeld 1 supposed that the narcotic dissolving in a lipoid plasma
membrane made it less permeable to oxygen. It would be more
in harmony with the phenomena considered in previous chapters,
to suppose that the narcotic in low concentration decreased the
permeability of the plasma membrane to the anions and molecules
of some acid end product of oxidation, and thus stopped the
combustion. An objection to this hypothesis is made by War-
burg 2 who found that phenylurethan, which only slightly re-
duces oxidation in certain cells, fertilized eggs, delayed cell
division enormously. With greater concentration of the narcotic,
oxidation was greatly reduced.

V. OSMOTIC PROPERTIES OF THE BLOOD CORPUSCLES.

Hamburger and Bubonavik 3 have concluded that the ery-
throcytes are permeable to K, Na, Ca and Mg. However, the
opposite conclusion was reached by previous workers.

Gyrn'?, 4 Hedin, 8 Traube 6 and others observed that the ery-
throcytes are relatively impermeable to neutral salts (exc. NH.1
salts) amino acids, various sugars and hexite, slowly permeable
to erythrite, more permeable to glycerine, and easily permeable
to monovalent alcohols, aldehydes, ketones, esters, ether, and
urea. In general, it may be said that the erythrocyte is perme-
able to lipoid-soluble substances or those that lower the surface
tension of water. Such substances (for instance, ether) become
more concentrated in the corpuscle than in the serum. Saponin
becomes 120, and ammonia 880 times more concentrated in
corpuscle than in serum. 7

1 PJliiger's Arch., 1909, CXXIX., 69.

2 Zeit. physiol. Chem.. LXVI., 305.

3 Arch, internal, de Physiol., 1910. X., I.

' 1'flitger's Arch., 1896, LXIII., 86, and Koninkl. Akad. von Wetensch. Amsterdam,
1910, p. 347.

''PJliiger's Arch., 1897. LXVIII., 229; 1898, LXX.. 525.

Biochem. Zeit., 1908, X., 371.

7 Arrhcnius, Biochem. Zeit., iQofi, XI., 161.

TENSION PHENOMENA OF LIVING ELEMENTS. 143

The erythrocytes are practically impermeable to ions. Stewart 1
observed that they offered a great resistance to the electric
current. It is difficult to remove all of the serum from a mass of
ervthrocytes, but Bugarsky and Tangl, working independently
f.t" Mt -uart, obtained sediments of corpuscles having a conduc-
tivity of only 1/50 that of the serum. This indicates that the
corpuscles are practically impermeable to both classes of ions,
for it permeable to ions of one sign, they would probably not be
-mli good insulators. The electric conductivity of the a-h

made up to equal volume) of the corpuscles is about that of the
-eriim, although the osmotic pressure of the solution of ash of
tin- latter i- greater. 2

llrinr an increase in electric conductivity of tin- corpuscles

as \\ill !> considered below) indicates increased prrmrabilit\
ion- After the corpuscle becomes permeable to ion-, further
in. i. ase in conductivity might be due to liberation of ion- from
i . >ml 'iiiat i. nis with colloids in the interior. Howe\ IT many i< uis,
f. 'i in-tam . I'O^ cannot be liberated without incineration or other
rigorous treatment. Increase in conductivity of the blood by
lakii nts has been proven to be chiefly due to increased per-

nieabilit\ o! the corpuscles, since the conduct i\ ity of the serum
m MI -ho\\s so great an increase on the addition of the 1. iking
in. ami is usually diminished (by the hicmoglobin if the cor-
pn-cles are present.

The portion of the normal corpuscle presenting the greatest
resistance to the electric current is the surface layer, since I lol-r
ol.ser\ed thai the conductivity of 'the interior of tin corpn-

.1. i rmine.l by its dielectric value) is many time- greater than
thai of the corpuscle as a whole. Peskind 1 cau-ed bubbles of
nitrogen to form \\ i thin the corpuscle and observed that they were
retained by a superficial membrane. This may be the membrane
\\liicli n-i-ts the electric current.

The chemical composition of the corpuscle is suppo-ed to bear
.-ome relation to its permeability. Aside from the haemoglobin,
and the rather low water content (60 per cent.i the corpuscle
' S< .'-'.'. - . J.ni. a, 1897.

M.Mirr.m'l Ruaf. Biochcm. Jour., III., 155.
r -,.., \rch., 1910, CXXXIII.. 237.
1 .li. Jmtr. I'hysioL. VIII.

144 .! F. MCCLEXDON.

is composed of lecit-hin and cholestcrin with a little nucleo-
proteid. It is probable that these lipoids are chemically different
in different species of animals, since Lefmann 1 observed that the
lipoids of erythrocytes of the same species are not toxic-, whereas
those of another species may be very toxic.

The distribution of these substances in the corpuscle has not
been ascertained. Pascucci 2 supposed the corpuscle to be a bag
of proteid impregnated with lecithin and cholesterin and filled
with haemoglobin. He found that artificial lecithin-cholesterin
membranes were made more permeable to haemoglobin by the
laking agents, saponin, solanin and tetanus or cobra poison.
Dantwitz and Landsteiner suppose the lecithin to be in com-
bination with protein.

Hoppe-Seyler assumed the haemoglobin to be in combination
with lecithin in the corpuscle, and Bang 3 has shown that .lipoids
may lie fixed by haemoglobin. It seems evident that there does
not exist an aqueous solution of haemoglobin within the corpuscle,
since haemoglobin crystals may be made to form in Nectunis
corpuscles without extraction of water. Furthermore, Traube
and Goldenthal 4 find that haemoglobin has a haemolytic action,
and unless there exists some body within the corpuscle which
antagonizes this action (as serum does) a haemoglobin solution
could not be retained by the corpuscle. Probably all of the so-
called "stroma" constituents, not in combination with the hae-
moglobin, form the plasma membrane of the corpuscle.

Under certain conditions, the haemoglobin comes out of tin-
corpuscles, and the blood is said to be laked. Laking of "fixed"
corpuscles occurs only after the removal of the fixing reagent.
Thus, sublimate-fixed corpuscles may be laked by substances
which combine with mercury, such as potassium iodide, sodium
hyposulphite or even serum proteids. The fact that they may be
laked by heating in water is probably because the nucleo-histone
is not fixed by sublimate. This process is prevented by hypertonic
NaCl solution, presumably on account of its power to precipitate
nucleo-histone (Stewart). Formaldehyde-fixed corpuscles m.i\

1 Beitrdge chem. Physiol. it. Path., XI., 255.

2 Hofmeisler's Beilrdge, 1905. VI., 543, 552.
' Ergeb. d. Physiol., 1907, VI., 152.

4 Biochem. Zeit., 1908, X., 390.

TENSION PHENOMENA OF LIVING ELEMENTS. 145

be laked by ammoniacal water, at a temperature which must be
higher, the more thoroughly they have been fixed. Ammonia
combines with formaldehyde.

Sieuart 1 supposes that the haemoglobin must be liberated from
some compound before the blood can be laked. \Ye cannot say
th.it the corpuscle is always permeable to haemoglobin from within
mi tward. Ilouever the corpuscle probably is impermeable to it
from \\iihont inward, since it does not take up haemoglobin from
Union. .UK! alter the blood is laked the serum contains haemo-
globin in greater concentration than the "ghosts" do.

Ai .my rale, permeability to haemoglobin appears to be inde-
pendent of permeability to salts, since Rollett 2 found that hiking
by coiiden-er di-rhargcs may set free the haemoglobin \\ iilmut the
corpu-e|e becoming permeable to ions. Stewart 3 concluded tli.it
tin -.line i- inn- of hiking with sodium taurocholate even alter
i on-iilriin^ the depressing action of haemoglobin on tin- con-
din ti\ ii\ .

Stewart 4 and others had already shown that blood laked by
minimal appli< ations of such hiking agents as free/in^ and thaw-
iiu. Ueaii: oo), foreign serum, and autoly.-U -poniam-ous

l.iki iuse 1'iit a Alight increase in the permeability to ion-,

when-a- tin toiitiiuied application of some of these agents, or
e-l>e< i.ill\ -mil violent reagents as distilled water and saponin,
cause a marked increase in electric conductivity. On the other
liaiid if saponin is added to defibrinated blooil at o, the con-
ductivity "I ilie corpuscles to ions begins to inert a-e before any
ha nio-loliin escapes from the corpuscles.

The liberal ii>n "f the haemoglobin by some lakin^ agents ma\-
I'e <lue to i lu diiet t action of the reagent in breaking up ilie com-
ixiiind in \\liit-li ilie blood pigment exists, but is probably some-
times a -ei-oinlary effect, following increase in permeability to
electrolyt<

It ha- been shown that many laking agent.-, lipoid soKciu-,
saponin nn-atnrated fatty acids, soaps, and lurmolysins (con-
tainin- lipa-e are such as would alter lipoids physically or

i. ijnd l-.\f>fr. Therapeutics, 1909, I., 49.
/' ' . 1 XXX II., 199.

I .' . X.

1 Jour, i XXIV., Jii.

146 J. F. MCCLENDON.

chemically, whereas pressure, trituration, shaking, heat, condenser
discharges, freezing and thawing, water, drying and moistening,
salts (including bile salts), acids and alkalis, might act also on
proteids.

Since any treatment which causes great swelling 1 of the cor-
puscle leads to loss of haemoglobin, it is probable that stretching
or breaking of the surface film increases its permeability. But
laking may occur without swelling, and even crenated corpuscles
may be laked by sodium taurocholate.

Hober 2 observed that the relative action of ions in favoring
haemolysis is: salicylate>benzoate>I >NOs, Br>Cl>SO4 and
K > Rb > Cs > Na, Li. Since this is the order in which they affect
the aggregation state of colloids, their action is probably on the
aggregation state of the colloids of the corpuscle (proteids or lipoids
or their combinations).

The permeability of formaldehyde-fixed corpuscles to ions, is
greatly increased by extraction of the lipoids with ether, or by
treatment with substances such as saponin, which act on lipoids.
Since the proteids have been thoroughly fixed, it is evident that
they play no part in this process, though they may do so in the
non-fixed corpuscles.

The relation of lipoids outside of the corpuscles to ha?molysis
has been extensively investigated, and cannot be fully treated
here. Willstatter found that cholesterin combines with one of
the saponins, destroying its haemolytic power. Iscovesco 3 con-
cludes that cholesterin combines with soap, and prevents its
toxic action.

Changes in permeability of the corpuscles to ions were studied
chemically before the application of the electrolytic method.
Hamburger 4 and Limbeck 6 observed that when CO 2 is passed
through blood, chlorine passes from serum into corpuscles and
the alkalescence of tin- scrum is increased. On the other hand,
the distribution of sodium and potassium is not changed. 6

1 Roaf, 0- J- Exper. Physiol., III., 75, supposes this swelling to be due to ioniza-
tion and hence increased osmotic pressure of haemoglobin.

2 Biochem. Zeil., 1908, XIV., 209, and he. cil.

* Comptes Rendus, Soc. Biol., 1910, LXIX.. 566.
4 7,eit. f. Biol., 1891, XXVIII.. 405.

6 Arch. exp. Path., 1895, XXXV., 309.

* Giirber, Sitzungsber. physik.-med. Ges. Wurzlmrg, 1895.

TENSION PHI.NMMi NA OF LIVING ELEMENTS. 147

Koeppe 1 and Hober 2 explain this process in the following
manner: The lipoid-soluble CO 2 enters the corpuscle, and by
reacting with alkali albuminates in the protoplasm, gives off
more , i it ions than it does in the serum. During the presence of
CO 2 , the corpuscle is permeable to anions, and the CO 3 = or
HCO..;- ions pass back into the serum, brin^ e\chani;i <1 for Cl~
ion-, to equalize the electrical potential. Sodium bicarbonate
being mop- alkalescent than sodium chloride, the titratable
alkalinity ot the serum is increased.

Thi- explanation is supported by the follouiii'j tarts: \\'hen
CO Is passed through a suspension of erythrocyte- in cam- --ugar
solution tin- laiter does not become alkaline. If ( '< ' i- pa ed
through a mass of centrifuged erythocytes, which an- then added
to physiological salt solution, the latter become- more alkaline
than the M rum in Hamburger's experiment. Any -odium -.ill
mav be -ub-titiitc-d for serum, and its anions will pa into the
corpu-cle-.'' Also the number of ionic valence- pa in:; into the
coipu-clr i- constant, i. e., if sulphate is used only hall as many
ion- enter the corpuscles as when chloride or nitrate i- n-ed.
The pr< is reversed by removal of the O

Thi- -aim- phenomenon has been observed in lencoc\ it- by
\ an del ^i hroHT.

There -eeins to be some relation between ha-mol\-i- and
-liitinaii'iii of the corpuscles. Arrhenius 4 Mippo-cd thai ag-
glutination by acids is due to the coagulation ot the prou id- ot

the ei;\elnpe. Ho\Ve\XT, since aggllltillat i< HI i- folloued b\

precipitation, it seems probable that the loss of the negative

electric ih.iixe which tends to keep the corpu-cle in -n-peii>-iiiii
and causes ii to repel every other corpuscle, i- partly rc-pon-ible

I". 'I the |iheli< Uliena.

The fact that water-laking is preceded by agglutination mi'^ht
be explained if we assume that increa-e in permeability to ions
1 -.til- to lo-s of electric charge. The char-t may be due to the
charge- "ii the colloids of the corpuscle or to semi-permeability
to ion-. The corpuscle is very poorly permeable to ions, but may

\rch., 1897. LXVII.. 189.

- ; >:., 1904. CII., 196.

3 Il.unluirK'-r and van Lier, Engelmann's Arch., 1902, 492.

* Hi,: >;,;. /.,-ii.. 1907. VI.. 358.

148 J. F. MCCLEXDOX.

be slightly more permeable to some one ion than to others. If
this ion were more concentrated in the plasma or in the corpuscle,
the latter would become electrically charged, and a general in-
crease in ionic permeability would lead to a reduction or loss
of this charge. The loss of charge would favor their coming
in contact with one another and their precipitation, but their
cohesion is probably due to some other change, possibly the exit
of adhesive substances, on increase in permeability.

VI. ABSORPTION AND SECRETION.
i. Absorption through the Gut.

If a live vertebrate intestine be filled with one portion of a
physiological NaCl solution, and suspended in another portion
of the same solution, fluid will pass through the wall of the gut
from within outward. Cohnheim 1 found that holothurian gut
behaves in the same way toward sea water, and the absorption
stops if the gut is injured with chloroform or sodium fluoride.

It might be supposed that the hydrostatic pressure produced
by the contraction of the musculature, is the driving force of
absorption, but on the contrary, Reid 2 found that the wall of
the rabbit's intestine behaved in the same way when used as a
diaphragm.

Salt is absorbed by an intestine filled with a very hypotonic
solution of it, and water may be absorbed when the solution is
very hypertonic.

Blood salts enter the intestine when it is injured by an ex-
tremely hypertonic solution, or sodium fluoride, chinin or arsenic.

Grape sugar and sodium iodide may pass from without inwards
through the wall of a normal holothurian intestine.

Traube 3 claims that absorption is explained by his observation
that the surface tension of the contents of the gut is less than
that of the blood, but this does not apply to the experiments in
which an identical solution was placed on each surface of the
wall of the gut. Traube 4 found that the addition of a substance

1 Zeil. physiol. Chem., 1901, XXXIII., 9.
'-Jour. Physiol., 1901, XXVI.. 436.

3 Pfliiger's Arch., 1904, CV., 559. Cf. Iscovesco, Comptes Rendus, Soc. I
1911, I. XXI., 637.

4 Biochem. Zeit., 1910, XXIV., 323.

TENSION PHENOMENA OF LIVING ELEMENTS. 149

lowering the surface tension increased the absorption of NaCl
l>v the gut.

Absorption is probably due to irreciprocal permeability of the
wall of the gut. Hamburger showed that dead gut and even
artificial membranes showed irreciprocal permeability to certain
substances. These artificial membranes were of different com-
position on their opposite surfaces (parchment paper-chrome
albumin or parchment paper-collodion) and he assumed that
tin \\all of the gut is composed of two osmotically different lavers.
In reality tin-re maybe more than two such layers, and the plasma
membrane- of the individual cells of the gut may show irreciprocal
permeability.

TranU ' showed that the rate of absorption of a snlotance
by living v.ut is usually greater the more it lower- the Mir face
tension of water. The order of ions is: Cl >Br> I >N() 3 >SO I
Ill'<> : and K, Xa>Ca, Mg. The order of non-electrolytes,
. i' < online in Katzcnellcnbogen 2 is: glycocoll <urca<a( < t'>n< ,
maun t <erythrite<glycerine<acetamid l methylalcohol, propyl-

al< "h.il. am\ lalcohol.

Tin- rati <>f absorption through dead ox gut according to i led in 3
\^:\'n \( i >C1>SO 4 and K > Rb> Na>Li >Mg ami m.mnit
<er\ thtiti < glycerine < urethan < glycocoll < amylenh\ draie
< gKtnl < urea < propylalcohol < isolnitylalcohol < methyl-
alcdlmi. ethylalcohol.

Tin- action of poisons on absorption may be dm- in tin- alter-
atiiin i .t' tin plaMiia membranes of the individual celU. Ma\i-r-
hoi'< i and Mi in' state that even sugar in certain concentrations
inert -a-ed the permeability of the gut.

2. ' ^niotic Relation of Aquatic Animals to the Medium.

I redericq found that the salt content of tin- body fluids of

marine in\ .Ttebrates is about the same as that of sea water.

Henri and l.aloii 5 showed that the OMii"tic exchange bet \\eeii

cirl(.m llu'nl of M-a urchins and holothunaii- and medium i- chielly

/' \rch.. CXXXII.

- /' ; . i Xl\'., 522.

/' , . 1899, LXXX'III., 205.

1 Hi,;li,-ni. /.fit.. 1910, XXX'II.. .^76.
1 NViiitri-t.-iii. 11. < .', 2.

I5O J. F. MCCLENDOX.

water. If the sea water was diluted with ] { vol. of isotonic
cane sugar solution, the salt content of the coelom fluid is very
little lowered in 4 hours, and only traces of sugar appear in it.
The result is the same with isotonic urea (which easily pene-
trates most plasma membranes). But the salt content of the
blood of elasmobranchs and teleosts is about half that of the sea.

Botazzi and his colleagues observed that the osmotic pressure
of the blood of elasmobranchs is about equal to that of the
medium, the sails in the blood being supplemented by organic
substances, chiefly urea, of which there is 2-3 per cent.

If elasmobranchs are placed in concentrated sea water, the
osmotic pressure of the blood rises, but the ratio of urea to salts
remains the same. G. G. Scott found that changes in the density
and osmotic pressure of the blood of elasmobranchs accompany
changes in the salt content of the medium.

However, in marine teleosts as well as all fresh- water animals
which have been studied in this respect, both salinity and osmotic
pressure of the body fluids are very different from that of the
medium.

The osmotic pressure of the blood of marine teleosts is about
half that of the sea, but in fresh-water teleosts it is still less
(but much greater than the fresh water). This indicates that
there must be a change in the osmotic pressure of the blood as
the fish ascends a river. Greene 1 observed that it took salmon
30-40 days to pass the brackish water, in which time they were
acclimatized to fresh water. After being in fresh water 8-12
weeks, the osmotic pressure of the blood was reduced only 17.6
per cent. This reduction may be partly accounted for by the
absorption of the osmotic substances in the blood by the sexual
glands. In harmony with this view is the fact that the osmotic
pressure of the blood of the female was reduced much more than
that of the male. One salmon, that was very weak and probably
dying, showed 32 per cent, decrease in A of blood. Sumner 2
observed that changes in weight and salt content of marine tele-
osts accompany, but. are not proportional to changes in the
medium.

1 U. S. B. F.. 1904, XXIV.. 445; 1909, XXIX., 129; Jour. Ex[>. Zoo!., 1910.
IX.

2 Bull. U. S. B. F., 1905, XXV., 53, and Am. Jour. Physiol., 1907, XIX., 61.

TENSION PHENOMI \A OF LIVING I LI-MI. NTS.

Ovrrton observed that if the cloaca and mouth of a frog in
ire-h water are closed, the frog constantly increases in weight.
This can be prevented by the addition of .7 per cent. XaCl to
the medium. In a hypotonic solution water is constantly ab-
s<>rbed by the skin and excreted by the kidney-. Fischer's 1
experiment, in which ligature of the leg of i frog caused great
i low the ligature is probably to be explained 1>\ the
f.n i tin; v. ater was absorbed by the skin but could not reach the
kidne\~. -jure the blood circulation was stopped. In regard to
Fi-cher 1 - t \pl. mation, compare the results of Sidbury and < , ;
Sunmi i i .in. Inded that in the fish, the gills are the chief -eat of
\ch.uige.

It appear-, therefore, that osmosis occurs through tin iiiu-ii
mi-Hi unhiding gills), kidneys and gut simultaneously, and since

the contents of the gut and kidney tubules are not the -a me a- the
medium, u -hould not expect an osmotic equilibrium bet \\eui
the body tluids and the medium. Furthermore, all three . if these
membrane- m.i\- show irreciprocal permeability.

I n -li-\\.iter li-li and non-migratory marine fish an- killed b\
ii changes in the medium, even though it be \<r\ ^r.nlual.

I'.ert m. tiiii. lined that if fresh-water fish are placed in sea \\.iter.
ilie ^ill i .ipill.u ies contract and become blocked b\ the distorted
i "i 'liii-t le-. lii naked-skinned fishes, not only the v;ills are
.itterted. but \\.iter ma\' be lost from the tissues.

lien .UK! Simmer both agree that the salts in sea water cannot
IM replaced 1 .\ anv other substance, without can-in- tin- death
oi certain marine fishes. Mosso 3 claimed that \\hen -hark- are
placed in tn-li \\.uer, the gill capillaries become ><> 1 ilocke< 1 \\ith
laked corpu-cles that physiological salt solution coiuM ni be
1 1 -reed tlimu^h ilu-m. He observed that the difference- in the
resistance of ctiiain tish to changes in the -alt cmuent of the
medium, i-oi-re-]ioiided to differences in the iv-i-tance of their
bl..id it 11- to the luemolytic action of such changes. Simmer. 1
ho\\e\er. -late- that this blocking of gill capillarie- doe- ma
occur in -hark- or marine teleosts in fiv-h water.

1 Fischer, M. H., d -l.-nia," J. NViloy & Sons, i<.
x Exper. /';<>. and Mfdicine, 1911. \'II., m.j.

/. (,///'.. isyo. N.. 570.
4 /'- s >-nth Inttrnat. Zoo/. Congress, Bost"n,

152 J. F. MCCLENDOX.

Sumner showed that as the fish becomes enfeebled by the ab-
normal medium, it becomes more permeable to salts. 1 Whether
the direct action of the abnormal medium, or the blocking of
the gill capillaries, produce the increase in permeability, has not
been experimentally tested. However, the gills themselves would
not be asphyxiated by blocking of their capillaries, and it seems
probable that the change in permeability is due to the direct
action of the medium.

We may conclude therefore that the death of the fish results
from the osmotic exchange. This may be sufficient to cause death
while the fish still maintains its normal semi-permeability, or
death may occur only after increase in permeability, due to the
direct action of the medium on the osmotic membranes.

A similar increase in permeability may explain Wo. Ostwald's
observations on fresh-water Gammarus in pure salt solutions. 2 He
found that the ratio of the rapidity of death to the concentration
is about constant up to a certain point, above which it is much
greater. This critical concentration has nothing to do with the
osmotic pressure, since it is different for different salts. Perhaps
at this concentration the salt made the membranes more per-
meable.

Schiicking 3 found that nicotine and strychnine made the skin
of Aplysia more permeable to salts. Since cocain retarded
shrinkage in hypertonic solution, he supposed that the hydro-
static pressure produced by the muscles aided shrinkage. How-
ever the hydrostatic pressure is probably very small, and the
effect might have been due chiefly to an increase in permeability
to salts, produced by the cocain.

3. Secretion of Lymph and Tissue Juice,

Hober supposes the raising of the osmotic pressure by the kata-
bolism of the tissues, causes fluid to be drawn out of the blood-
vessels, and states that the lymph in the thoracic duct has a
greater osmotic pressure than the blood.

Traube states that the surface tension of transudates and

1 Cf. Greene, above.

2 PJliiger's Arch., 1905, CVI., 568.

* Arch. Anal. Physiol., Physiol. Abt.. 1902, 533.

TENSION PHENOMENA OF LIVING ELK MI-NTS. 153

exudate- is always greater than that of the blood. He cites a
case in which a transudate was caused to be absorbed by injecting
into it a substance which decreased its surface tension.

4. Excretion.

Milk and bile have about the same osmotic pressure as the
blood, but urine is almost dry in some animals: it i- u-ually
In I M rtonic in man but may be hypotonie.

Traube maintains that the surface tension of the normal urine
is alu r than that of the blood, and that thi- is the

dri\ iii;^ Ion e in excretion.

Houe\<r. ll<.licr and others suppose that tin- -iib-iam < -s to
1 may be formed into solid bodies in tin- tubule cells,
and ihroun out into the hi men.

It' lipoid-iii-oluble dyes arc fed to frogs, granule- in tlu- -11-
ol (i-riain -i ^mrnts of the kidney tubule are staim-d \\iih them.
Tin- <l\i i- not tirst excreted by the glomeruli and then ab-orbed
from tin lumen by the tubule cells, for if the vena Jacol >-. mi,
which -ii|>|>lies the tubules, is ligatured, no staining occurs, al-
though the renal arteries still supply the glomernli.

Ihi- -tained granules in the tubule cells are thrown out into
tin- lumen and pass into the bladder. The.-c granule- u-ually
di ol\ e to form a slimy substance in the urine, bin -ome oi them
max remain intact.

The circulation in mammalian kidneys cannot In- controlled
in the- -aim- \\ay, but after intravenous injection . >f , ( certain
lipoid-in-olublc- dye, no stain may be detected in the \\all- o|
the glomeruli, although the tubule cells are staineil. The -tain
in the lumen does not appear above the level of the -tained tubule
ci'lls. In the excretion of carmine, it may be found in -ranule-
in the tubule cells and lumen, similar to those, found in frog's
kidtiex 3.

h ha- been -upposed that urea is excreted by collecting in
the-e granule-, and passing out with them, but it would be e\en
sim;iK-r t> a--ume that some substance is excreted into the lumen,
\\hich combine- with urea and so lowers the concent ration of that
in solution, thus accelerating its excretion.

The chief recommendation for the granules is their valve-like

154 J- I r - MCCLENDON.

action, which would account for the secretion of urine against a
concentration gradient, but a simpler mechanism of such a process
is shown in Hamburger's double membranes.

The blood pressure may aid in the secretion of the water of
the urine, which is eliminated chiefly through the glomertili,
but its insignificance in the elimination of urea is shown by the
fact that after increasing the volume (and therefore pressure) of
rabbit's blood 70 per cent, by transfusion, the urea elimination
was not or only very slightly increased.

VII. CELL DIVISION.

Various hypotheses as to the cause of cell division have been
advanced by the morphologists. Hertwig, supposed that when
the ratio of nucleus to cytoplasm is less than normal, the cell
will divide. 1 Gerassimow 2 subjected cells of Spirogyra to low
temperatures and other abnormal conditions and obtained an
increased amount of chromatin in some of them. These cells
did not divide until the ratio of nucleus to cytoplasm was as
great as at the time of division of a normal cell.

I found that chromatin is not necessary for cell division. 3
After extracting the chromosomes from the starfish egg, I caused
it to divide. In this case the ratio of nucleus to cytoplasm was
zero; however the cell did not continue to divide indefinitely.

There is no easy method of determining the ratio of nucleus
to cytoplasm. Some cells contain large vacuoles \vhose contents
are not considered as cytoplasm. Eggs contain fat drops and
granules compounded of protein and lipoids. These are not
considered as cytoplasm by all investigators. If the granules and
oil are included as cytoplasm, the ratio of nucleus to cytoplasm
is very small, and yet the egg cell does not divide unless "stimu-
lated" by the sperm or some other means.

k. Lillie 4 observed that chemical substances, which in low
concentration cause the Arbacia egg to divide, in high concen-
tration cause outward diffusion of the red pigment (echinochrome)
and compared this to the laking of erythrocytes.

1 lie is not confirmed by Conklin, Jour. Expcr. Zoo!.. 1912, XII.. i.
- Bull. Soc. Imp. Nat., Moskau, I<;<M. No. i.

3 McClendon, Arch. f. Entwicklungsmcch., 1908, XXVI, 662.

4 Bioi.. BULL., 1909, XVII., 188.

TENSION PHENOMENA OF L1VINC, KLI.MENTS. 155

This is made more striking by the fact, mentioned first by
Loci), that haimolytic agents are effective in artificial partheno-
ui -nesis. R. Lillio observed that pure solutions of sodium salts
caused the egg to divide, the order of effectiyene-s of unions being
'Br<ClO 3 <XO 3 <CXS<I. He also found that these salts
could !>e inhibited by others (CaCl;, MgCl 2 ), as i- characteristic
of the antagonistic effects of salts in physiological phenomena,
and tin- pr.-i ipitation of colloids.

1 found that the sea urchin's egg contains fatty sul Mance-, and
ielati\el\ large amounts of lecithin probably in combination with
pioirid-.. I found that Toxopneustes eggs freed from tin.- jclly-
likc ci.\crin^s, contained about 10 per cent. lecithin 'alcohol
extract p|it. with acetone) and about 2 per cent, of an extract
-olul'le in alcohol or acetone and containing rosette-. <>t tat -like
taU. This extract blackened strongly with o-mic tein>\ide
and -lter\e-ced on adding dry Xa-carbonate in \\atcr, Hun
emul-ilied. probably it contained unsaturated fatt> aciil.

\<.i<ling to a private communication by Mathews, tin-
of i he -taiii-h contains lecithin and an unsaturated tatty acid,
but no ( ho]rMerin. In this last characteristic ii diller- markedly
from the eiAthrocyte. There is no way of determining \\hether
tin -i -ub-.t.inces enter into the composition of the plasma
membrane, but the facts are presented in order to indicate the
pi '--ibiliiies.

\\ e haM'seeii that the exit of haemoglobin i> probably nm due
t. in. i, MM. I permeability to this substance. It is po--ible ihat
the same i- true of echinochrome. I found that the echiuochroine
in the e;cu -hows a continuous spectrum, when-a- that extra< led
in \arioii- ways shows characteristic band^. It ma\ po-^iMy
In held b\ chemical combination in the e.

llo\\-\er I found. other evidence for increase in permeability of
the sea urchin's egg coincident with be^innin^ d-\< lopment :'

t. 1 eiiili/cd eggs are caused to shrink more quickly than un-
t\-i tili/i .1 eggs, with isotonic sugar solution. Presumably the
fertili/ed euv;s are more permeable to the substances exerting
the internal osmotic pressure.

j. Tin- electric conductivity of the euy increases about ' \ \\hen

\1, Cl. -in Ion. Attifr. Jour. Physiol., 1910. XX\ II., 240.

156 J. F. MCCLEXDOX.

it is fertilized or made parthcnogenetic with acetic acid, indicating
increased permeability to ions.

Lyon and Shackell 1 and Harvey 2 observed that methylene
blue and neutral red enter fertilized eggs more quickly than
unfertilized eggs. Harvey supposed that only the free color
base (undissociated) entered, since the addition of a little acid
to the sea water prevented the staining of the eggs.

Mathews 3 considered the penetration of stains into the egg
as a chemical process (the stain forming a salt combination with
the lecithin or proteins of the egg surface).

Harvey observed, further, that NaOH penetrates fertilized
more easily than unfertilized eggs, but the eggs are killed by the
alkali.

The fact that the unfertilized frog's egg continues to swell for
a long time in water (Biataszewitz) whereas the osmotic pressure
of the fertilized frog's egg is quickly reduced to equal that of the
medium (Backmann and Runnstrom) indicates increase in perme-
ability to osmotic substances on fertilization. In this connection
it is interesting to note that Bataillon, 4 Brachet, and myself 5
caused the unfertilized frog's egg to rotate normally and segment
merely by pricking it.

1 1 has been supposed by various observers that the "forma-
tion" of the fertilization membrane in very closely related to the
segmentation of the egg. Loeb observed that the sea urchin's
egg may develop without the formation of a fertilization mem-
brane, and I have confirmed this observation, and shown that
it is very probably wrong to suppose that this is a case of failure
in "pushing out" of the membrane. Apparently "membrane
formation" is not essential for the segmentation of the egg,
although by furnishing protection it may insure the development
of the embryo.

Loeb postulated that an osmotically active colloid exists in
the unfertilized egg, but is so covered with lipoids that it does
not absorb water until it is squeezed out or otherwise exposed

1 Science, 1910, XXXII., 250.
= Ibid., p. 565.

3 Jour. Phurmacol. and Exp. Ther., 1910, II., 201.

4 Arch. Zoo/. Exper., 1910 (5), VI., 101.

8 McClcntlon, Amer. Jour. Physiol,, 1912, XXIX.. 298.

TENSION PHENOMENA OF LIVING ELEMENTS. 157

at the surface of the egg, at the beginning of development (when
it fills the so-called " perivitelline space"). I observed that this
substance bears a positive charge (is basic) since it migrates
toward the kathode when an electric current is passed through
water containing the fertilized egg.

The unfertilized egg is imbedded in a mass of jelly which is
probably inucin. This jelly bears a negative charge (is acid)
since it combines with color bases.

When the positively charged colloid is exposed at the Mirface
ise in permeability) and come> in contact \\ith the
ne^.imeh (harmed jelly, the two mum. illy precipitate at their
MII" t' ntuct, thus forming the fertilization membrane.

Mm it .ill "t" the jelly is washed off of the egg before ihe l.n u-r is
(.iu-i.1 i" develop, no fertilization membrane i- (Wined .1- I
h,i\r <ib-er\ed) because no two oppositely charged colloid^ are
bronchi in < nMct, but the basic colloid may with difficulty be
Bi 'ii as .1 n fra< live layer, which has been mi-taken for a p->'>rlv
developed "fertilization membrane."

flu oba i \ .it ion of Lyon 1 makes it appear that catal.i-e conu->
out of fertili/ed more quickly than unfertilized v. -. prokil>l\ due
t" in- 1 1 .i-ed permeability.

l.\i.n "b-crved that C().j came out of fertili/ed more <|ui kl\
th.iii iiiilVi tili/ed eggs, and ( ). Warburg. l."cb .uul IU\M!!
ob-iTM-d that oxygen is absorbed more rapidly by the lonui-r.
\\- miiilit a-k: I >oe> increased permeability allow increased
o\i<l.i!i"U, or is increased oxidation the primarx cause "f the
incrc.iM-il respiration?

The p-rineability change is the simplest : xplaiiatimi, and in
\\liat other \\a\- could oxidation be inciv.i-rd J I .neb -uppo-i-d
tin- >pcnu carried an oxidase into the egg. 3 ISut QO addition . ,t
oxidate i- concerned in artificial parthenogenesis, and I .orb

Mined that the oxidase (or other en/\ me, kina-r.-'i i- held in
the egg prriphery and cannot penetrate the e;^^ interior until
the permeability is increased.

In addition to oxygen, oxidase, and escape of CO 2 , hydro\yl

im.J . 1909, IV.. 199.

\1.< Mini, .n an.l M iti. hell. Jour. Biol. C u. X.. .

* In tin- .piiiuvti-'ii it K interesting to note that M.IMII.U. '/.fil. j>hy>i>- <
iyi". 1 \\ I 265. i.iili-il to tinl inoir iron in s|n-rm than in sea water.

158 J. F. MCCLEXDON.

ions are necessary for the rapid oxidation of the sea urchin egg
(Loeb), and Harvey showed that the unfertilized egg is practi-
cally impermeable to OH ions of low concentration. The
increased permeability allows hydroxyl ions in the sea water to
penetrate the egg, as shown by Harvey, and, since the sea is
always alkaline, this may explain the increased oxidation.

Asters always develop in the egg before segmentation. In the
normal egg these have some relation to the division of the nucleus,
but even if a nucleus is not present, 1 have observed that the
cytoplasm constricts along a line on the surface farthest removed
from the centers of the asters.

The constriction of the cytoplasm is probably due to a band of
increased surface tension (or to decreased surface tension at
the poles). This might be caused by local increase in perme-
ability to ions, causing decreased polarization, at the equator
(or increased polarization at the poles, clue to increased pro-
duction of the polarizing electrolyte in the asters).

The same reasons that were given for assuming that the surface
of the Amceba is electrically polarized, hold good for the egg.
The first change is probably a general increase in surface tension,
indicated by rounding up of the egg. Later this may become
localized from internal causes and result in cleavage.

Hyde 1 observed local changes in electric polarization of
Fundulus eggs during cleavage, indicating that surface tension
changes and cleavage are due to this cause.

It has been objected that the segmentation of the egg is not a
typical case of cell division, since the egg cell is "wound up"
and ready for some "stimulus" to set it going, whereas tissue
cells must "grow" or "rest" after each division before dividing
again.

1 1 may be true that growth is prerequisite to division, but
this cannot be formulated quantitatively. In the spore-forma-
tion of certain organisms, a cell may divide in a relatively short
time into myriads of almost ultra-microscopic cells.

Hertwig may be right, in general, in assuming that the relative
growth of nucleus and cytoplasm influences division, but the
difficulties in proving this have been indicated, and this cannot

1 Am. Jour. Physiol., XII.. 241.

TENSION FHEM'Mi-.NA OF LIVING ELEMENTS. 159

be e\pn---ed in chemical terms. It is generally supposed that
nucleic acid is a more abundant constituent of the nucleus than
of the cytoplasm, but much evidence ha- appeared for believing
that iti- often present in considerable quantities in the cytoplasm.
l.o-l> -upposed that the segmentation of the sea urchin egg is
accompanied by an "autocatalytic" svntlu--i- of nucleic acid,
HIM -I- tin- nuclei increased in number. But Ma-in- 1 and more
recentlj ^hackell 2 by chemical analysis found as much nucleic
.11 id in the unsegmented egg or i-cell st.ro as in the Ma-tula
-i.i.

I IM n i- -omc indirect evidence that increase in permeability
ina\ e.m-e an increased division rate of tis-ue cell-. Though
ii-ll Drouth may inlluence division, it is probable that permc-
abiliix inlliienccs growth.

Y.iriou- "-tiniuli" cause increased proliferation of cells of the
-< Tinin.il layer of the skin. It is commonly known that mechani-
,il -tiimili increase growth of the skin.

I'.tinlian! 1 i.-her observed that Sudan 111. or Scharlack \\
cause iinii-.i^ed proliferation of the epidermi-. \\lu-n iln d\e
Is inje< i< d under the skin of a rabbit tin- skin ^r<>\\ - ti.uard the

dye.

I m-r fi'imd that gradual increase of temperature caused a
( -.in --ponding increase in proliferation of ti ne ci-ll- due io
incre.iM-d chemical reaction and inflammation <>t the ii--n> .
Hut \\hen a certain temperature was reached a Midden jump in
i he incre.i-e in proliferation was observed wiihoui .1 corresponding
iuciva-e in inflammation. This is similar to ihe plieiioineiiini
-ten in iinlei lili/ed eggs, where a rise in temperature l.e\nnd a
cei lain point causes segmentation.

It h.i- .ilso been observed that electrical Mimulation max
can-e increased proliferation of tissue cell-.

\ll of these changes (electrical, thermal, or mecliaiiic.il -limu-
laiion. or tn-atment with lipoid soluble .-ub-tances) cause in-

1 /.,-it. /'.\MO/. Chem.. 1910, LXXVII.. 161.

. eji i . n. s., XXXIX'.. 573.

1 \Vliii-h an- practically insoluble in water but soluble in fats and lipoids and, as
1 II.IM- iili<i-i\t-il. -lightly in lipoid-protein combinatii'-

4 S.-r v. I >uni;rrn u. Werner, "Das Wesen Bosartigen Gesch\\ iil-tf," I.cip/i.i;,
. p. 65.

160 J. F. MCCLENDON.

creased permeability and segmentation of the sea urchin's egg.
Therefore, from analogy, we may conclude that increase in
permeability may cause tissue cells to divide.

The "wound stimulus" to regeneration of tissue may also
cause increased permeability of the cells.

In a preceding chapter it was shown that the "current of
injury" produced by the negative electric potential of a wounded
surface is common to animal and plant tissues. The wounded
cell acts as an electric generator and a current flows through
neighboring cells.

I observed that if a current is passed through living tissue,
which is subsequently fixed and stained, basophile substances
will be found displaced toward the anode. In sections of tissue
adjacent to a wound the extent of the current is indicated by the
displacement of basophile granules. The current affects first
the cells in contact with the wounded cells, then extends in some
directions more than others. Electric currents ("currents of
growth") continue for many days after the wound has healed.

Since electric currents cause sea-urchin eggs and tissue cells
to divide and proliferate, probably these bio-electric currents
constitute the so-called "formative stimulus" of regeneration.

Embryonic cells, cells of germinal regions, and cancer cells
are distinguished by their great power of proliferation, or rapid
division. It is probable that the plasma membranes of these
cells are more permeable than those of other tissue cells in the
same medium or under the same conditions.

Cancers have been produced by the action of X-rays (electric
pulsations) on the skin. The cells in the skin were so changed
that they proliferated more rapidly. Similarly, electric changes
have been observed to start the egg cell to rapid proliferation.
There is probably some irreversible change in the permeability
of these cells, which does not, however, make the plasma mem-
brane incapable of subsequent reversible changes in perme-
ability (i. e., the change is unlike what occurs at death of the cell).

The suggestion that cancer cells are more permeable than
tissue cells in general may possibly be of therapeutic importance.
Loeb has shown that fertilized eggs are more sensitive than un-
fertilized eggs to various toxic substances (probably

TENSION PHKN'iMI NA OF LIVING ELEMENTS. l6l

these sub-tances enter the fertilized eggs more easily). The
same explanation may po--ibly he applied to the effect of sugar
on ten. tin living cells. The- unfertilized eggs of the frog, petro-
niyzon, sea urchin and annelid have been caused to segment,
by placing them in sugar Dilution-. Mayerhofer and Stein 1 ob-
. <'! that sugar in certain concentrations increased the per-
meahility of the gut to certain -.ill-, ami in this condition the
ytn was more easily injured by the diffusion of substances.

similarly Mockard obser\ ed that sugar increased the toxicity
"t pun- -oliitions of salts on the Fundulus egg. Morgan and
stiH k.nd -hi.ued that ih\< \\a- not dm- to the inversion of sugar
or to thr o-motic pressure, and suppn-ed that the sugar might
combine chemically with the -alt. It >cems probable that the
ii increased the permeal.ilitx t" -.ilt. The fact that sugar
in Ire-h \\ater is toxic whena- the -amc amount of sugar in
tin- nimal medium (sea water Uimt toxic or less toxic, indicates
that tin- -alts within the l-'nndidit are the same as those

out-idi- in sea water), and iiicrea-e in permeability to them
d'<-- nut lead to diffusion while they remain in sea water, but
diltii>ioii takes place in fresh \\ateiv

It it In -liiiun that cancer cell- are MHHV permeable,' substances
Mia\ he toiiml which kill cam er cell- mure easily than tissue
i (II- as i \pl. lined below.

\\liena- a certain men .i-e in j HTMU -ability of the cell seems
in cau-e di\i-imi, a very :^reat iih in permeability causes

death h. i nicK -is, cytoK -i-. hai terii il\ <i- ' . It has been shown
that certain l\ -ins are spt cilic t'm- certain cells, probably because
the pla-iiia Mieinhraiie- of ilu-e cells differ chemically.

I In- fertili/ed e-- is more easily c\ t|\ /od than the unfertilized
i tain -lib-lances. It then-tore appears that the more
peimeable the cell is in the beginning, the more easily is the
IH i ineahiliiN brought to the point \\hit-hcausescytolysis.

1 It nee ii i- pmhahle that certain substances may be found by
\\ hich cancer cell- tan 1 e nn >\ e ea-ily cytolvzcd than normal tissue
cells.

n. /.fit.. 1910, \\V1I.. 376.

- BIOL. r.i M .. [907, xiii.. -

3 In tin .il -IIU.H I have -tiown that no diffusion take3 place in fresh

\\.iu-r. Am fr. Jour, i \.XIX.. 295.

1 62 J. F. MCCLENDOX.

It has been shown that narcosis is accompanied by decreased
permeability. On the other hand, certain forms of inhibition
of muscle are accompanied by an increase in permeability.
May certain cells be inhibited in proliferation by an increase
in permeability, too great for cell division but not great enough
for cytolysis? The great oxidation rate in eggs inhibited in
cleavage by very hypertonic solutions as determined by Warburg,
seem to indicate this.

It has been shown that certain tissue cells inhibit the pro-
liferation of others. In the healing of wounds, the epidermis
inhibits the growth of connective tissue. If a wound remains
uncovered by epidermis for a relatively long time, processes of
connective tissue may grow outward, but this is prevented by
the growth or transplantation of epidermis over the wound.

Perhaps the proliferation of the connective tissue is due to
abnormal "stimuli" (bio-electric currents, diffusion of sub-
stances) such as cause proliferation in regenerating tissue gen-
erally. The presence of epidermis over the wound might protect
the connective tissue from these "stimuli."

The foregoing facts and the speculations based on them may
not be of far-reaching importance in themselves, but they suggest
lines of research, which if followed, it is hoped, will add a great
deal to cell physiology and pathology and be an aid to the under-
standing of many problems in therapeutics.

THE LARVA OF SARCOPHAGA, A PARASITE OF

CISTUDO CAROLINA AND THE HISTOLOGY

OF ITS RESPIRATORY APPARATI -

\VM. A. KEPNER.
UNIVERSITY OF VIRGINIA.

The ^tiulcnt of zoology is early impressed with the intcii-ive
ni.iniuT in which animal life has penetrated everv available
-pace. Even so remote and strange place as the poison ijand-
o| i In- rattle-snake have been entered, these glands furnishing
ample ] mil i -id and oxygen supply fora little nematode ih.it make-
i In in it- habitat. In the example of this Sarcophaga we find
a lly that lias entered the- nucha of the "box-turtle" a region
oi i he body where its larva will not be exposed to serious pre tire
bet \\een parts of the "turtle's" body and where it will al-o be five
from tin attacks of the appendages and mandibles of tin- ho-t.
The occurrence of this parasite in Cistndo was first ob-i\id
b\ I'ai kard ('82). Packard described and figured it as in u^'rid
larva. Tim-, so far as I have been able to determine, arose the
b.e-i- for believing that a "bot-fly" infested a reptile. Aldrich
in n terring to Packard's paper states that perhap- it is
not an o -trid. Shar[>e in the Cambridge Natural lli-tory says
that <1 -trididie may occur in the reptiles. Wheeler ('90 next

>nU the occurrence of the- dipteron lar\^e on tin- nucha of
iirolina. He succeeded in getting the larvae to pupau-
and iii rearing imagines from the pupae. These adult Hie- proved
to belon- to the genus Sarcophaga and not to be cestrid llie-;.
Thn> there appears to remain no evidence of a " bot-fly " infesting
a ivpiile.

In Octobi-r, H)io, a female specimen of Cistudo Carolina was
brought into my laboratory. It was kept through the winter
in a sink. January, 1911, a student called my attention to what
lu- called a "growth" in the nucha of the ri^ht >ide. This, h<>\\ -
ever, proved to be an insect larva. Two days later the lar\a

iped from the perforation made in the skin of the host.

163

1 64 W M. A. KEPNER.

Subsequently two other larvae left the excavated region of the
nucha. These specimens were preserved in alcohol. May 14,
the fourth and most vigorous larva dropped from the host.
This larva was placed upon soil in a box where it burrowed into
the earth and formed an oval, dark brown pupa. This pupa
has not yielded an imago, so that I have been unable to cor-
roborate Wheeler's diagnosis as based upon the adult fly.

Except for some details which are readily overlooked in pre-
served specimens, such as Packard had, the larvae I found closely
resemble the figures and descriptions given by Packard. With
the living material which I had at my service, I was able to see

|

details which make these larvae correspond more closely to the
following description of larvae of Sarcophagidae than to that of
(Estrididae larvae. Brauer ('83) says that the larvae of Sarco-
pluigidae "are rounded, thinner anteriorly and amphipneustic.
The antennae are short, thick, cylindrical, divergent, wart-like
tubercles, each with two ocellus-like chitinous rings at the tip.
The mouth hooklets are distinct, strongly curved and separated
from each other. The abdominal segments are distinctly dif-
ferentiated by transverse swellings and are each provided with a
girdle of spines. The hind stigma-plate is situated in a deep
cavity, which is formed by the last segment alone. The anal
swelling is two-pointed. The puparium is oval. 1 " Thus I am
led to infer that I have the same kind of larva that Packard had
figured and described and am able to corroborate Wheeler's
statement that this is not a "bot-fly" larva but a sarcophagid
larva.

Apart from this I have been interested in certain details that
no one has recorded for this particular sarcophagid. Figure
I represents the dorsal aspect of the larva magnified ten diameters.
Each segment is seen to bear a band of spines. The antennae
are seen from the ventral side (Fig. 6, ant.} together with the
strongly curved, (list i net mandibles (Fig. 6). On the ventral
side of the posterior segment there is a trilobed disc armed with
stout spines (Fig. 3 and Fig. 5, d). This may function as a
sucking disc. The posterior rn<l of the last segment is divided

1 This translation of Brauer's description was taken from Williston's "N'oitli
American Diptera," 3d ed., page 349, by Dr. J. M. Aldrich.

THE LARVA OF SARCOPHAGA. 165

into a wide, dorsal lobe and a narrow, projecting, ventral lobe.
Between these two lobes is a deep recess into which the anus and
po-n -rior -ligmata open. The posterior stigmata are guarded
by a large -tigma-plate which has two lobes. Each lobe bears
three spatulate chitinous bars (Fig. 4, c.p.} which articulate with
six similar bars on the ventral lobe of the segment (Fig. 4, c'. /?'.).
The -hape and relation of these dorsal and ventral rhitim>u> bars
i<> ea<h "i her are such that I am led to believe that thev function
.1- pn -In -unle structures; the lower lobe of tin- se-ment pressing
ii- bar- .t'_,iinst the bars of the stigmatic plate can lay hold of
the \\all of the excavated region in the skin of the ho-t and thus
anrhor tin- larva. The most striking feature to which attention
not 1'i-t-n called is the presence of two anterior stigmata

Fig. 1 These stigmata are fan-shaped struct tin-- \\hicli

bear -i \euirrii or eighteen papilhe along their let initial edge

I i. I.). In a specimen cleared with xylol each ot" tl

niata can be seen to lead directly into a large lateral tr.u lua.
"1 1m- the\ are provided with an air-breathing apparatus though
they li\e in a thick fluid of suppurated matter which make-
liable the i losing of one or more of these iracheal opening or
ma\ nei e--it.ite the temporary closing of one of them. In this
connection it is interesting to find a transverse tra< heal commissure
posterior to the anterior stigmata and another transverse tracheal
( ommi mv anterior to the posterior stigmata. These commis-
sures enable both tracheal trunks to get air though for any reason
some of the -ligmata may be closed. Thu- the chief tracheal
-\ stem consists of a pair of anterior and a pair of posterior stig-
mata and two lateral tracheal trunks which are connected by
mean- of an anterior and a posterior tracheal commi ure.

Nothing unusual has been noted concerning the histologv of
the tracheal trunks and posterior stigmata. The histology of
the anterior stigmata has, however, attracted my attention.
These fan-shaped structures are for the mo-t part proliferated
masses "f cuticle. The anterior half of the stigma projects
be\ond the contour of the body as a stigmatic process. The
posterior half lies beneath the surface of the body and is covered
by an epithelium which represents the hypodermis modified as
tracheal epithelium ( Fig. 8, te.}. l-'mm the posterior margin of

166 WM. A. KEENER.

the stigmatic process there is a cuticular and hypodermal in-
vagination which extends to near the base of the stigma as a
retaining thread (Fig. 8, inv.). This retaining thread of cuticle
and hypodermal epithelium is seen in transverse section at -inv.
in Fig. 9. The entire stigma represents a modified region of
hypodermis and cuticle. On the mesial side of the stigma near
the base of its anterior third the hypodermis becomes very
pronounced, the cells becoming very large and columnar. These
cells, so far as their form is concerned, are the most conspicuous
tracheal cells (Fig. 7, te.). From them slender processes go into
the cuticular mass of the stigmatic process. These processes
and the position of these cells suggest that they not only help to
elaborate the cuticular substances of the stigmatic process but
that, also, they may be able to move the stigmatic process.
Within the mesial wall of the stigmatic process no cytoplasm ex-
tends except that of these cellular processes; within the lateral
wall of the stigmatic process scattered hypodermal cells are
found. There is thus an indifferent cellular supply to the tracheal
process of the stigma. Indeed the entire stigma is for the most
part a cuticular structure. The cuticle of the general surface
of the body is distinctly two-layered. The outer layer is the
deeper and in hacmotoxylin stains the more deeply. The inner
layer is clearly a softer substance and does not stain deeply.
These two strata are involved in the formation of the anterior
stigma. The inner layer, except for becoming more abundant
in the stigma, is not modified. Figure 7 at c and Fig. 8
show this layer of the cuticle passing over into that of the
stigma. The outer layer of cuticle, however, is thinner over
the stigmatic process than over the general surface of the body.
When it reaches the tips of the papillae it is invaginated and passes
as a series of converging tubules to the bases of the papillae where
the tubules unite to form a large tube whose lining is confluent
with the lining of the tracheal trunk. The cuticular lining of
the tracheal trunk also presents a deeply staining layer and a
layer that does not readily stain (Fig. 11, //.), thus resembling
the cuticle, of which I believe it represents a modified region.
The inner denser layer of this tracheal lining gives rise to spiral
taenidia as shown in Fig. u at /. When this denser layer passes

THE LARVA OF SARCOPHAGA. 1 67

into that of the stigma very minute slender processes arise from
it into the lumen: these processes branch and rebranch to form
a reticulated layer which takes the place of the taenidia of the
trachea (Fig. 10, r.). This reticulated layer is increased until the
entire lumen is rilled with a reticulated mass or plug (Figs. 8,
7 md 9, rp.). At the base of each papilla the reticulated plug
branches and continues to near the tip of the papilla where there
is a -mall chamber into which the branch of the reticulated plug
sends its terminal filaments (text-figure i). Thus we find

KrP!

i Lou il section of a papilla of the anterior atigma, showing the termi-

n.il i li.iiiil>i-i mi< which filaments of the reticulated phi. t. X 1.500.

the entitle, trachea! lining and the cuticular mas- of the -ti-ma
lo he t\\o-l,i\ercd. In all three places the non-staining I i\er is
little iiin.lilied ; but in the tracheal lining the deeply staining layer
i- iiiinliiied to form the taenidia, and in the tracheal pmce-- it
I" ..... i - .1 reticulated plug.

The lar\.e of blow-lly and house fly have likc\\i-e pn 'thoracic
siigniatir processes with finger-like papilla-. The-e in turn,
a. onlini; to de Meijere ('02), have reticulated plni;- which he
rail- "felt -chambers" (Feltkammern). What does -uch histo-
lo-ji-.il -tructure mean? We see the cuticular hair- mianlin- the
>tumata ot ants or other insects and we interpret them as being
<le\i< . ^ to protect the trachea from foreign bodie-. lint hen- \ve
ha\c in ]lace of protecting hairs an exten-i\e. Imelv reiiculated
phi:; \\hich resembles the cotton plug ot a bacterial culture tube
.1- though it were constructed for the purpose of protecting the
trachea from microscopically minute bodies. Tin- lar\a feeds
upon the suppurated fluid found within the excavated region
of the nucha of the host, hence while the larva is feeding these
bacteria can hardlv be of service, for the anterior end of its body

1 68 WM. A. KEPNER.

is bathed in the suppurated mass. However, when about to
pupate the larva reverses its position with reference to the sup-
purated mass, and lies with its anterior end directed towards or
through the opening in the skin of the turtle. The larva is then
in a position to breathe air through the anterior stigmata. At
the same time the larva during the three or four days spent in
emerging from the host, frequently retreats into the excavated
cavity when disturbed, thus its anterior end may repeatedly be-
come contaminated with the bacteria of the suppurated mass.
I think, therefore, that the anterior stigmata are chiefly functional
during the two or three days spent by the larva in passing from
the turtle to the ground and that the reticulated plug is a bacterial
screen protecting the trachea from infection threatened by the
repeated retreat of the larva into the excavated cavity when it
lies with its posterior end at or within the suppurated mass.
If this conjecture concerning the time and character of the func-
tioning of the anterior stigmata is not warranted, I believe that
I am justified in agreeing with Hewitt ('08), that the anterior
stigmata of this character are functional at some stage in the
life of the larva.

THE LARVA OF SARCOPHAGA. 169

LITERATURE.

Aldrich, J. M.

'05 uc of the American Diptcra. Smithsonian M ' tions, 4''.

Braucr, F.

"83 I 1 iuRcr des kaisorlichcn Museums zu Wien: III. Sy-tnn.iti

I'litim auf Grundlagc dcr diptorcn larvan. etc. U-nk-. In. .1. r k

. math-naturwisse Classe. Bd. 47. s. i-ioo.
Brues.

'oo '1 i.i. li.-.il I lil.itations. Biol. Bui.. Vol. i. p. 6.
Hewitt, L. Gordon
'08 The Structure. Development, and Bionomics of II

.n. Part 99. Quart. Journ. Micr. Sc.. \'ol. 5.-. I'.nt IV.
Krancher. O.

'81 I >- : Bau 5l . ma l>ci den Insecten. Zeits. wiss. Z<-1.. H<1 | sos.

dc Meijere, J. C. H.

'02 i eba ML rVothorakalstlgmen der Dipterenpuppen. /.-! Jahrb. \i
M.I. \\'.. 8.623.

Packard, A. S.

'8j M.'t M. I .irvu- in a Turtle's Neck. Amer. Nat.. \'ol. 16.
Scheiber, S. H.

"62 \'i-iKl'-i< hi-ndc Anatomie und Physiologie dcr Oestril<-n-l.iivi-n. !< -jui.i-
tioi n. Sitzb. Akad. \Viss. Wien. Math-ii.tiurw. < I . M.I. -\>. 9. ~.

Wheeler, Wm. M.

'90 I'lir supposed Bot-Fly Parasite of the " Box-Tuitl.-.'

WM. A. KEPNER.

EXPLANATION OF PLATE I.

FIG. i. Dorsal aspect of larva. 5*., stigma. X 10.

FIG. 2. Lateral aspect of anterior end of larva, mo., mouth; m., mandible;
st., stigma. X 100.

FIG. 3. Ventral aspect of posterior segment, d., tri-lobed disc with stout
spines. X 25.

FIG. 4. Ventral aspect of posterior segment. The ventral lobe is laid back so as
to expose its six chitinous bars c'p'., and the two-lobed stigma-plate with its six
chitinous bars cp. X 25.

FIG. 5. Lateral spect of posterior end of larva, d., tri-lobed disc; cp., chitonous
bar of stigma-plate. X 10.

FIG. 6. Ventral aspect of anterior end of larva, m., mandible; mo., mouth;
st., stigma; ant., antenna. X 25.

FIG. 7. Transverse section through base of tracheal process at level indicated
by arrow 7 on Fig. 8. c., cuticle; rp., reticulated plug; h., hypodermis; te., tracheal
epithelium. X 250.

BIOLOGICAL BULLETIN VOL. IXII.

PLATE I.

St.. ...

c .

WH >[fNC*.

WNf. A. KEENER.

EXPLANATION OF PLATE II.

FIG. 8. Reconstructed drawing of anterior stigma, h., hypodermis; te.,
tracheal epithelium; /., tsenidia; c., cuticle; rp., reticulated plug; inv., invagination
of cuticle. X 200.

FIG. 9. Transverse section of trachea through level indicated by arrow 9.
It shows the secondary invagination with its cuticular core inv., rp., reticulated
plug; te., tracheal epithelium. X 500.

FIG. 10. Part of trachea in the transitional zone between the reticulated plug
and the tsenidia of the trachea, r., reticulated chitin arising from the denser layer
of chitin; te., traceal epithelium. X 1,500.

FIG. n. Part of wall of trachea. /., tsenidia; te., tracheal epithelium; tl.,
tracheal lining. X 1,500.

BIOLOGICAL BULLETIN VOL. Jin
A

V

i

I

8

9.

_

'

* M > . I CM u

KARLY DEVELOPMENT OF GRAFFILLAGEMELLIPARA

-A SI PPOSED CASE OF POLYKMBRYt "NY. 1

J. THOMAS PATTERSON.

I. INTRODUCTION.

In th< Bl "k-' Memorial Volume of the Journal of I-'..\-f>eri-

:<il /"".'";; v, Vol. o,, 1910, Professor Edwin Linum rep--n- the
li-.o\.T\ <-t a very interesting viviparous rhabdoccele commensal

with i In i unmon ribbed mussel, Modiolus plicatnlns, found along
tin- All. mti< coast. Dr. Linton refers this worm to the ^eiui>
unl on account of its peculiar method of producing
eml-iAo- in pairs, designates it by the name Gra' nielli {tarn.

So I'.ir aa ui know the only other statement in the litiT.utiiv ih.it
could l-i- interpreted as referring to this interesting turbellari.ni

-nnd in .1 -horl j);i[x?r by Nicoll, '06, entitli-d "Notes
I I- in. it... I. I' twites of the Cockle (Cardium eti id

( Mytilu\ filn'.

Ni "II ' in his Fig. 7) what he calls a trem.ii'>dr -\n<\^ .

i tin- li\t r i if the cockle, but it is quite clear fmm Linnm'-
\\ciik ih.it In i- in error in calling this specimen .1 sporocyst.
\\li.n he in .ill probability had was a specimen <-i .1 species <>t
tnrl ifll. iri. in ( l<i-ely related to if not identical \\ith (/'. -i-nifllifHini.
1 hi- i- e\i< l-iii IP-MI the fact that his figure >h.-u- the pn-M-n. r
nl" p.iiifd i-iiil>r\os, as well as a pharynx, \\hich .ilmu- umild
elude the case IP-ID the categorx' of sporoi\-i-.

l.iin-iM'- |M|n-r gives an account of the more m-n.T.il t"r,itin
nl" tin- \\onn. I nit leaves several important questions unanswered,
.iMioii;^ uhiih may be mentioned the fnllt>\\ iiu : i ll.-u i> iln
\t-lk df]io^ititl in the ova? (2) How do the -pi TIM- reach the
"sperm-sac"? ; Is the species protandnm >. J 4 \\IKTC. m- the
3 lVnili/rtl. J 51 Finally, and mo>t importani of all, How do
tin i\\o cmlir\ os in each capsule ari-

In p-^.ird to thi- last point, Lintt-n -nggests that \\v may have a

I'K-in tin- M.uiii' ':i'l tin- /<">l"i;i'a tin-

I ni\ ntril-iH: 109.

173

174 J- THOMAS PATTERSON.

case of polyembryony. It was this suggestion that induced me
to undertake a study of certain phases of the development of
Graffi.Ua; and this not only because of my interest in the general
subject of polyembryony, but also for the reason that an oppor-
tunity seemed to be offered to work out the details of this peculiar
phenomenon. Furthermore, if a true gemelliparous develop-
ment really did exist in so simple a fashion in a relatively low
organism like Graffilla, it might be possible to modify experi-
mentally the process and thus to be able to get at some of the
factors underlying it.

While the results obtained from these studies have proved dis-
appointing, at least so far as the main object for which the in-
vestigation was undertaken, yet they are of a character such as
to warrant record, especially as they answer satisfactorily some of
the questions raised above. Furthermore, we have as yet only a
very few papers dealing with the development of rhabdocceles,
and consequently there is need of contributions along this line.
Methods. Various methods for preserving the material have
been used, but the most successful fixing fluid has been found to
be Benda's modification of Flcmming's strong solution. Speci-
mens fixed for two hours in this fluid give beautiful results for
cytological study, especially when followed by iron-haematoxylin
stains. Bouin's fluid also gave good preparations, but is much
less certain in its results. In making whole mounts the speci-
mens are placed under slight pressure and killed over a gentle
flame, and then fixed in a corrosive-sublimate solution. If
followed by borax carmine such material gives very clear figures
of many structures. However, I find the same "indefiniteness"
I about the reproductive organs as noted by Linton, especially in
| regard to the ducts, so that one can not rely upon mounts for
' one's interpretation of the conditions of these structures.

Notes on the Habits. Linton states that G. gemeUipara lives
on the gills of Modiolits, but there is some evidence that they
inhabit the kidney. This is brought out in the following experi-
ment. Two dozen specimens of Modiolus from a lot yielding no
Graffilloe from the gills were opened, care being taken not to injure
any of the tissues, and thoroughly washed out in water. No
parasite was found. The kidneys of these same individuals

EARLY DEVELOPMENT OF GRAFFILLA. 1 75

leased out and the specimens again washed in water, with
the rr-nlt that thirty-eight Graffillcc were -ecuivd. I'ndoubtcdlv
many individual parasites escape from tin- kidiu-y of the ho-t and
latrr found in the mantle cavity and on the pill-, and this
would .Kiount for their discovery there by Linton. Further-
nion- tin- method ordinarily employed in opening the niolli.
\\oiild in e--arily result in injuring the kidney, and thu- |u rinit
the < -( ape of the parasite from that organ. Th< iinnit

mentioned above would seem to indicate clearly that G. ^fnn-I-
li()nr,i i- a true endoparasite, but the experiment \\a- performed
.it tin- c|o-i- of the season and the opportunity \\a- IK>I ofieivd
to -i -i ilt- the <|iiestion conclusively, as that could onl\ In done by
in. tkin. I'ul dissections of the individual mollu-i -. \\ V

-hould expect to find this species of parasite in the kidney or li\ er
oi i IK ho-t -ince all of the other species of the genii ''ilia

are found in the same organs of the various mollu-

The In -i M ,i>on of the year in which to secure ( llifmni

.n \\<H,d- ||..|r is during August, from the loth to ihe joih of
ilie nioiiili. ^I'ecimens may be obtained prior to \\i\--, but iliev
.in u^u.ilK iininobile individuals which contain nuinerou-, \oiin-
thai lia\ e libel. iti-d themselves from their capsule- and . m- -\\ini-
iniii'c about in the mescnehyme. Such material i- valuable lor
obtaining \ei\ \ounganimals. On July 5. 1911, several of these
e\h. ui-ied niothcrs wiTi' placed separatel\- in lian-iii- drop- of
the iluid taken from the mantle cavity of Moiliolns. The cover
sli|> tioin \\hich the drop was suspended wa- placed abo\e the
cavity "t .1 hollow ground slide and sealed with \a-eliin-. In
ihi- \\a\- the -[leciniens could easily be studied under the ini-
croscope, < >n the following day it was noted that mo-i of the

\-oiing ha<l ruptured the wall of the mother and were -\\inuniiiii
about in the drop. In one case the escape of tin- youn^ \\a^
actualk ol,M-r\ed. \'oung animals secured by this method can
be kepi ali\e \\iihoiit much trouble for about two days, and un-
doubted lv \\ould li\'e longer if proper care \\ere taken. I lo\vc\ er,
it \\a- found unnecessary to obtain material for -tudy in this
\\.iv atier toii\-ci-ht hours, for the washings of Modinlns yield
manv young -pecimens that correspond in -i/e to ihe-e two-day
old \\ orm-.

1 76 J. THOMAS PATTERSON.

An interesting periodicity in the reproduction of G. gemellipara
occurs at Woods Hole. From the 2oth to the 25th of June (191 1),
shortly after the writer arrived there, specimens were secured
in considerable numbers, but from this date until about the loth
of August it was extremely difficult to obtain material, although
molluscs from many different regions were examined. From an
entire bucketful of the Modiolus not more than a dozen would
be secured, and these were either very young, sexually immature
animals, or very large individuals which were about on the point
of undergoing degeneration and freeing their young. About the
middle of August, both in 1910 and 1911, Graffillce were secured
without difficulty, but from the 25th of the month until the I2th
of September, when I left Woods Hole, they were extremely
scarce. From this it would seem that there are two summer
periods of rapid multiplication, one in June and the other in
August; and possibly a third period occurs in October. Linton
reports that Coe found Graffilla in abundance at New Haven
during the month of October.

At no time does one find G. gemellipara in such numbers as
reported by some of the writers on the other species of the genus.
Jameson, '97, states that from four to several dozen individuals
of G. buccinicola, which is parasitic in the kidneys of Bnccinum
undatiim and Fusus antiquus, are found in every specimen of the
two molluscs.

II. STRUCTURE OF THE REPRODUCTIVE ORGANS.

The reproductive organs of this Graffilla arc difficult to make
out, both on account of the viviparous method of reproduction as
well as on account of the variability in the development of the
different parts. G. gemellipara, like certain other members of
the genus, exhibits successive hermaphroditism, but the case is
not so extreme as that described for G. buccinicola by Jameson,
'97. The male organs develop first and upon reaching their
maturity at a comparatively early period in the post-natal life,
in part atrophy, and are then followed by the development of
the female organs.

The male organs consist of the following parts: (i) a pair of
testes which lie just posterior to the pharynx, one- on each side

EARLY DEVELOPMENT OF GRAFFILI.A. 1 77

of the median line somewhat below the central axis of the animal
Fig i 2) two very delicate, short sperm ducts which place
i In- gonads in communication with the seminal vesicle; (3) a
seminal vesicle, which is a rather large pear-shaped sac situated
just In-low the genital pore; and finally, a plug-like peni> arising
troin tin- pointed, ventrally directed end of the seminal vesicle.
In OIK- ot the clearest specimens secured earli -perm duct i- -mi
to ari-i- troin the posterior median corner of the te-ti- ,m<l to pass
inuard to the anterior face of the seminal vesicle, inn tin- the
latii-r .it .ilioiit the di\'iding line between it- upper, bullion-
|iortioii and the smaller lower part. The peni- \\lu-n contracted
i- c\ti mel\ difficult to make out, and since in mounted prepa-
ration-, thi- condition is almost invariably met with, not m.my of
tin- ili i, til- o| the- organ were studied. The penis \\ln-n extended
ol course |irotrudes into the common atrium, \\hich in turn
communicates with the exterior by means of the small genital

1 1 he pore lies in the median ventral line at a point -i main I

ibont one third the distance from the anterior end of tin- l>od\.

In large individuals the testes are seldom found, ami \\hrn
|irr-fin .in- mi i e degenerating fragments. Tin- seminal \e-icle,
ho\\e\er. p.i-i-t- .it least until a late period of tin \ n i-t -naial
life, but iii many animals becomes reduced in si/e. The penis
also degi derates sooner or later. During this perio<l of degen-
eration o| the male organs the female reprodin ti\ i -titntnie-
gtadualK make their appearance. One occasionally meet- \\ith
specimen- in which the transition from tin "male" to the
"female" -tate is seen, and from such individuals nio-t of the
important ]n.ints concerning the female organs can be made out.

In the t\ pical " female" condition the seminal \e-i< le i- alua\ -
pn-i nt. though as stated above it mn\ become greatly n-din ed
in >i/e, and the atrium with its genital pore still per-i-t-. Ju>t
back of the -eminal \~esicle and (loyally the atrium gives ri-e
to a small di\ erticiilum, which both from it- po-iti<,n and char-
acter -ugge-t- its homology with the receptaculum >emini> of
the other members of this genus, although in the two clearest
cases coming under the observations of the writer tin- \e~icle
contained no spermatozoa (Fig. 2). If this interpretation is
correct then the receptaculum seinini- i- in this species clearly a
degenerate -tructure.

1 78 J. THOMAS PATTERSON.

Posteriorly the atrium is directly continuous with an enlarged,
rather thick-walled uterus, which in turn gives rise to a duct-like
structure that extends backwards and upwards (Fig. 5, ).
At the point where these two parts join, the uterus receives the
small ducts of the many unicellular shell-glands (Fig. 1,5).

Towards its distal end the uterus bifurcates, sending a branch
to each of the bilaterally arranged ovaries (Fig. 3). The bi-
furcated part of the uterus serves as a receptacle for spermatozoa
a condition that is not entirely unique for this species and
also performs the function of insemination. On account of the
backward and upward course taken by the uterus, the two distal
parts come to lie just below the ventral surface of the intestine,
at a place slightly posterior to the middle point of the body

(Fig. 5)-

The development of the uterus has not been studied and I
can not therefore state with certainty the exact nature of this
organ. Slightly posterior to the point where the proximal and
distal parts join the duct is frequently very indefinite and difficult
to trace. This, together with the fact that small yolk cells are
frequently found within its cavity (Figs. 4, 5) has led the writer
to believe that the distal part of the uterus is the product of
fusion between the ducts coming from the reproductive glands
and therefore should probably be called the oviduct.

The female reproductive glands consist of a paired " germarium "
and a paired "vitellarium," the two glands on each side being so
closely associated that the compound structure might properly be
termed a "germ-vitellarium." The ovarian portion occupies
the anterior part of the body, while the yolk glands occupy the
posterior half mainly.

The clearest idea of the relation of these various parts to each
other and to the reproductive ducts can best be gained in a study
of horizontal sections which pass just below the ventral side of
the intestine. In such sections the ovary on each side is seen to
begin slightly anterior to the seminal vesicle, and tx> increase
gradually in diameter in passing backwards until it reaches the
region occupied by the distal end of the uterus. Here it spreads
as a fan-like structure, with ilu- inner margins of the ova con-
verging to meet the tip of the uterus (Fig. 4). In composition

EARLY DEVELOPMENT OF GRAFFILLA. 1 79

tin- ovary is made up of flattened cells, and one might compare it
i rouleau of coins of gradually increa-ing size, the smallest
In-ill.: !'' ,tted at the anterior end. The larger cells of the ovary
arc produced by the absorption of nutritive material- t'mm the
\iiellim- cells, in a manner that will be <K --crilied in tin- IH \t

ion.

The vitellarium is an extensive organ, and in the posterior half
of i li<- body almost completely envelops tin- inte-tine Fig. 6 .
In tin- c.irly stages of its formation the cells are very similar in
iliM-i 1,1" tin- ovary, and even in the definitive condition their
inn Ic-i II.IM- the characteristics of ovarian nuclei. Tin- <>\.iti.m
and \itelline cells are in very close association at the middle
die body, and for some little distance anterir t<> thi-
tin . ,\.n\ i- overlaid by the yolk cells.

111. MARLY DEVELOPMENT.

i \ittr it ion oj the Ova and the Formation of the . :f>sulc.

In oidiT to be able to understand clearly the mamu r in which
tin >\ i .ire nourished and the egg-capsule is formed it i- necessary
in i .ill .mention to the characteristic condition in GraffUla <>t the
duple\it\ of embryos in each capsule In all of the "Id.-r sta
the t\\<> embryos are surrounded by a very thin t ran -parent niein-
br.me m -hell inside of which the two ciliated indi\ idnal- m.i\
nm\e .il.niii each other with considerable case. In !
or indci d in any stage of segmentation, this thin -hell in the
Mil* if the word does not exist, though the nut'-nin^t

ii.n nl the yolk is of a consistency such that it servee the
]>ni|M'-e nt a shell, and out of this surface l.t\er the true -hell
di'iil'tle dillerentiates. During the cleava. u i- seen

th.u .1 mi-idt rable mass of yolk surround-- the t\\.. enibrxn-
I i( [9 The two embryos may be either clu-i in-. iln-r. \\iih
onl\ .1 \ir\- thin intervening layer of yolk. i>r \\idtl\ separated
and situated at the extreme opposite end- it the c.ip-nle I igs.
In either e\ - ent the most pertinent <|Uc-ii<m thai mie i .in
rai-e i- hnw the two embryos have come t<> e\i-t \\ithin the same

\ i ilk ni.i-s.

\- \\e have pointed out in the preceding -ec(i>n. the ovaries
are at their posterior ends somewhat clo-ely .iiiprovimated on

ISO J. THOMAS PATTERSON.

the ventral side of the intestine, and are intimately associated
with i he yolk glands, being surrounded on the dorsal and posterior
aspects by them. In a longitudinal section of almost any indi-
vidual in the egg-producing stage one can observe that the ova are
at their upper margins absorbing yolk from these glands, and while
the nutritive process may involve the ova of one half of the ovary,
yet it is much more conspicuous in the posterior third of that organ
(Fig. 9). At the extreme end of the ovary the absorption goes
on with great rapidity, the ova soon becoming gorged with nu-
tritive material. In consequence of this rapid growth certain
retrogressive changes involving the cell membranes separating
contiguous ova frequently make their appearance. As a result
two cr even more nuclei may come to lie within a common yolk
mass, which occupies the extreme tip of the ovary (Figs. 9, 10).
In other words, a syncytium is formed here. In the vast majority
of cases only two ova are involved so that the usual picture dis-
played in this region represents a binucleated yolk mass (Fig. 15).
It should be noted here that in this peculiar method of nu-
trition we have a mechanism alone adequate to account fully for
the reason why two embryos are habitually borne within a single
capsule. Just why two should appear is difficult to answer.
As a matter of fact, however, tw r o are not always present, for as
Linton has pointed out capsules are sometimes seen with three
embryos, and a few cases were noted by him in which only one
embryo is surrounded by the envelope. Furthermore, in the
figure of Nicoll referred to above, two capsules containing three
embryos each are clearly shown. In my own material several
cases of "triplets," including one with undivided eggs, have been
observed, as well as several with one embryo each. \Yhilc in
the light of these facts the twin condition in Graffilhi loses much
of its apparent significance, yet its appearance in the great major-
ity of cases made it necessary to undertake a careful study of
the histogenesis of the ovary in order to see if any mechanism,
other than that of the breaking down of intervening membranes,
could be discovered that would explain a potency to gemellipa-
rous reproduction on ihc part of that organ. At first it srrmrd
probable that a binucleated ovum was produced somewhere in
the oogonial history. A diligent search in the ovary fails to

EARLY DEVELOPMENT OF GRAFFILLA. l8l

n-\cal any binucleated ova, except of course at the extreme tip,
nor has the slightest evidence been secured of nuclear divisions
either mitotic or amitotic throughout the entire length of a fully
matured ovary. We are therefore forced to the conclusion that
what \\r have described in connection with the absorption of
yolk furnishes the key to the twin condition in Gnifilla. It can
ii"i IK- argued that the breaking down of the membranes is only
apparent and therefore an artifact produced by reagent-. I'or
it ha- been observed in preparations made from material pn -
-er\e<| in a dozen different fixing fluids, and t".ll..\\ed by as many
di Herein -tain-. However, not in all ca-e- do the two contiguous
ova l"-e their intervening membranes, but some become com-
pletek -iirroimc|ed by vitelline cells, which through a process
<!' di-inii ^ration form the yolk mass of the definitive capsule
3. 7, 16). In such cases the two ova d> not lo-r iheir "in-
di\ idnalii\ ." and a subsequent reorganization of ne\\ meinbrane-
alioin ihe i\\o nuclei will not lake place. Considerable evidence
ha- been se< nred which indicates that the-e tuo method- of
cap-ule formation are but the extremes of one and the -aim-

pi 01 ess.

Throughout the entire history of yolk ab-orption main inter-
e-iing (han^es, involving both the nuclei:- and c\ topla-m, are
seen, but \\e can not deal with all of them In r < >ur attention
mu-t then-lore be directed to those that stem- to u- to be mo-t
impi ii i ant .

In I'ig. 14 is represented a pair of nuclei lying within a single,
niembraiu The lower of these is immediately -urroumled 1\ a
la\er ot tinelv granular protoplasm, about which one can trace
another \et\- delicate, but neverthele distinct, membrane.
Thi- condition has been observed in a number of ova, and may
begin before the binucleated stage- is reached, that i-, in ova
-ituaud from two to six cells from the tip of the ovary. I have
not been able to demonstrate the universality of thi- membrane,
and I am therefore inclined to regard it a- the intra-cellular or
intra-\ itelline membrane that i- sometimes laid down about
the ovarian nucleus. It may be that in Graffilla it marks the
beginning of the segregation of the protoplasm from the yolk,
and is then-fore the first step in the reorganization of a cell about
each of the nuclei in the capsule.

1 82 J. THOMAS PATTERSON.

In Fig. 9 is seen the last trace of the intracellular membrane
in a binucleated mass that is about ready to be freed from the
ovary. It is possible of course that the faint line about the
large nucleus is not an intra-cellular membrane, but only the
original cell-wall which has become much attenuated through
the absorption of yolk by the ovum. This figure is of further
interest in that it demonstrates with remarkable clearness the
manner in which the yolk is absorbed by the ova. At the ex-
treme end of the ovary the process is at its height, and one can
actually observe the configuration of the streams of food material
extending from the vitelline cells to the larger nucleus. This is
particularly true in the pseudopodial-like structure in the upper
median portion of the figure. On the extreme right, near the.
section of the tip of the second nucleus, the yolk cells are directly
open to the ova. It is not quite clear as to what extent the yolk
cells participate in the formation of the mass of yolk surrounding
the eggs, aside from merely giving up their nutritive materials;
but that they do assist in this formation is abundantly proved
in those capsules the yolk contents of which show many degen-
erating nuclei of vitelline cells. In some cases these fading
nuclei form a complete row just below the surface of the capsule.

Some hall dozen cases have been found in which the ovum
apparently does not become surrounded by any considerable
amount of yolk, but after absorbing a small amount of food
material is set free from the ovary. These single naked eggs
float about in the parenchyma and probably never succeed in
producing embryos (Fig. 13).

Some time prior to the liberation of the ova from the ovary and
the yolk-gland, the ovarian nuclei undergo marked changes.
During all of the preceding oogonial history the nucleus possesses
that characteristic coarse network of chromatin extending
throughout the nucleoplasm, and a very large, deeply staining
nucleolus (Fig. 9); but during the last stages of yolk absorption
the chromatin network becomes more or less indistinct (Fig. 7),
finally disappearing altogether, and in its stead a finely granular
condition of the chromatin appears. At the same time the
nucleolus stains less intensely and M>OII becomes very irregular
in < nitline (Fig. 10).

EARLY DEVELOPMENT OF GRAFFILLA. 183

It i- necessary to mention only briefly the manner in which
the "o\ ulation" takes place. By the time the absorption of
yolk ha- reached the point seen in the case of tin- <>\ a on the ridit
of Fig. 9 the formative capsule may be -aid to be practically
independent of any ovarian connections, and it only remains for
the cap-ule to be freed from the vitellarium. Ilo\\e\er, its
aM.nlinient with the yolk glands per-i-t- f.-r some time after
thi-. even indeed until the two eggs reorganized, if reorganization
i- IM-I essary. In Fig. 10 is a capsule just about ivadv to be set
free into tin- parenchyma; most of the yolk cell- ha\<- \ieldcd
11 1 > i heir I'ooii i ntents to the capsule, and the region immediately
-in roimdin- its upper margin shows only delicate -trand- con-
ii \\ith a few of the remaining nurse- cell-. Shortly
in.: this period tin- strands are severed and the capsule
round- up. and as the whole structure is pu-hed about in the
parent h\ ma by the movements of the mother \\onn ill

Illldel w . i ( |e\ eli ipmellt .

I p io ihe present we have been using the term "capsule" to
mean the \\hole yolk mass surrounding the t\v<> eggs; and \\e
mu-i now consider brielly the formation of the thin cap-ule <T
-In II, 1 iv \\ hich we mean the membrane containing the i u ciliated
embryos ! the later stages. Since the eggs with their illicular
la\i i ..t \olk do not enter the uterus, it is not probable that an\
ol i In- - , reiioiis from the unicellular shell-gland- reach the
and thu- lake part in the formation of the -hell, as mt ur- in
the i ase "I "\ i | i.i rous forms. I have not followed all <>t the -tep-
in the li.nnalioii of the shell, but it has been ob-er\ed that a-
dexelopmeiu proceeds the outermost layer of the \.>lk, \\hich .it
lii-t i- very plastic and yields readily to any ob-tnn i ii.n in the
pareiich\ma. gradually becomes more re-i-iant. linally taking
on (he thin elastic character met with in all of the advanced
stages. It is probable that the shell is in part the product of
the parenchyma.

It remain- to say a word about the "rcorgani/ation " of cells
in those cases in which the membrane in part or completely
di-appears from the two ova. Kven in the extreme cases it is
doubtful whether the cytoplasmic part of the cell becomes in-
di-criminaicly a ociated with the yolk portion of the cap-ule.

1 84 J- THOMAS PATTERSON.

This part of the study has furnished many difficulties, because
of the fact that the capsule at this particular stage is very plastic
and hard to fix properly. Only a few cases of good fixation have
been secured: and in one of the clearest of these the nuclei are
seen to be surrounded by a finely granular protoplasm, about
which a membrane must later be secreted.

2. The Aborting Spindle. The study of maturation and fer-
tili/ation was made- difficult by the presence of a spindle which
appeared in the egg some time before the egg capsule was set
free into the parenchyma. On account of its large size the spindle
was at first taken to be that of the first cleavage, but inasmuch
as the first division of the fertilized egg results in cutting off a
small micromere, it soon became evident that this interpretation
was incorrect. Furthermore, in the eggs in which the large
spindle appeared the most diligent search failed to reveal any
polar bodies. \Yhen this fact once became fully established
it was evident that we had in Graffilla a display of that remarkable
phenomenon of a "disappearing" or "aborting" spindle, first
discovered by Selenka, '81, and to our knowledge of which
Wheeler, Gardiner, and others have contributed.

Selenka's discovery was made in connection with his work on
the polyclad Thysanozoon Diesingli. He describes the aborting
spindle as appearing in the uterine eggs. After the egg has
reached its full growth, the germinal vesicle begins to make prepa-
rations to divide in the typical manner; the chromatin forms a
spireme, the achromatic spindle with its two centrosomes appears,
and the chromosomes pass into the equatorial-plate position.
At this point the process stops, and the nucleus returns to a
resting condition. Subsequently the egg throws off two polar
bodies, is fertilized, and develops in the normal manner. Inas-
much as the yolk granules are evenly distributed throughout the
egg at the beginning of this peculiar phenomenon and are col-
lected about the astral centers at its close, Selenka supposes that
the function of the aborting spindle is to mass the granules at
the center of the egg. But this interpretation fails to explain
the appearance of the spindle in those eggs in which a collecting
of the granules about the astral centers does not take place, as
both Lang and Wheeler have observed.

EARLY DEVELOPMENT OF GRAFFILLA. 185

Lang, '84, next noted the aborting spindle in several polyclad
eggs, and figures it in the uterine egg of Thysanozoon Brocchii.

\\ "heeler, '94, describes briefly the appearance of the uterine
^pindle in the eggs of Planocera inqiiilina, a polyclad inhabiting
the br.iiichi.il chamber of Sycotypus canalicalalns, but does not
attempt to work out the details of the process. He also noted
the spindle in the eggs of the acoelan Polycfnvrus cuitdutn*.

< .ardiner. '95 and '98, working on the latter specie- came to
tin i MM lu-ion that the aborting spindle is abnormal, ivpre-enting
tin- hr-t clea\ai;e spindle of eggs retained too long in the HUTU-
o| .in animal kept under abnormal conditions. His point does
not -eem to be well taken, as Surface, '07, ha- -hown in hi- work
on I'lniiK, ,-r<i.

I lie la-t reference to the aborting spindle thai \\e may note
i- tint ot I.. \ on ( iraff, '82, in his monograph on the Khabdocielida.
Von (.rail, although making no reference to tin- -pindle in the
text. ili.uK h-ures one in the uterine eggs of .1 f>li<uio*ti>nui
<//. md Cyptomorpha saliens.

In our -pei ies, (/. gemettipara, the aborting -pimlle appear- in

the < e time before the freeing of the egg-capsule from the

\iiellaiinm. The spindle is really anticipated Ion- belon- .ill ot
the \ulk i- laid down about the two eggs, as can be -> n in 1 iu. to.
In man\ ie-pcvts the spindle is truly remarkable. \\ t ,\ ,,\\\\ on
,n ' omit ot its great size, but also for the iva^m that tie,|iieiitl\
the clii-onio-onu> do not appear upon it. < >ne ot the dearest
cases that ha- come under my observation i- -ho\\n in 1 i-. 17.
Thi- i- an e-pei'ially well preserved egg, \et one can not detecl
the ^li^hte-t tiare of chromosomes in the cell. HOU.-MT, it i-
probable that the chromatin is represented by some of the central
-pimlle liber-, \\hich are (juite thick but do not take the -tain
\\ell. Thi- i- most certainly the case in some eggs in which MT\
delicate i hroinatin threads among the spindle fibers can with
(lillicnlty be made out.

- -int-iime- the chromatin is in the form of chromo-<>me.-, which
ho\\c\ er are not located on the spindle. In Fii;. I S is -ho\\ n -nch
a case. Hen- the large conspicuou- ^pindle i- it-ill" free from
chromatin, but among the astral ra\> of one end are four chro-
mo-omr-, \\ hich are of intcre-t not onh bei ause of their pi-i-uliar

1 86 J. THOMAS PATTERSON.

position, but also because they are apparently bivalent. They
are not tetrads in shape, as in the characteristic condition of the
first maturation, yet that they are the egg chromosomes and
not those of the sperm is evidenced by the fact that the sperm is
located in another part of the ovum.

The peculiar behavior of this karyokinetic figure is not con-
fined to the chromatin; the centrosomes frequently present unique
conditions. It is not uncommon to find the centrosome at one
or both ends of the spindle undergoing division, but this would
not be striking since in many germ cells, both male and female,
a precocious division occurs were it not for the fact that at
one end the axis of the two centrosomes is at right angles to
that of the spindle, while at the other end it is simply a continu-
ation of the spindle axis. The precocious division of the cen-
trosome frequently results in the formation of a double aster.

I have not been able to follow with certainty all of the sub-
sequent steps in the. history of this spindle, but the end result
in all cases would seem to be a return to a sort of resting stage
on the part of the nucleus. It differs from the corresponding
stage of Thysanozoon Diesingii, in that the nucleus instead of
being a large vesicle, appears in the form of four vesicles, one
for each chromosome (Fig. 19). These may be more or less
grouped together or widely separated, but they later come to-
gether and_ fuse, producing a lobulated nucleus which retains
this condition until the onset of maturation (Fig. 21). It will
be seen from this rather brief account that the only function
which one might assign to the aborting spindle in G. gemellipara
is that of scattering the chromosomes in the form of vesicles;
but since these are later collected together into a single vesicle
before maturation, it is difficult to attach any real significance to
this whole peculiar phenomenon. Inasmuch as several odd con-
ditions have been observed, both in the centrosomes and the
chromosomes, it is not at all improbable that the aborting spindle
is an abnormal display. But it can not be the result of placing
the animals under unfavorable conditions because the spindles
are found in worms killed immediately upon their removal from
the moll use.

It should be pointed out here that Graffilla is not a favorable

EARLY DEVELOPMENT OF GRAFFILLA. IS;

form in which to work out the history and significance of the
aborting spindle, for owing to the viviparous mode of repro-
<lu< -lion prevailing in this species it is quite impossible to secure
a complete series of stages showing the different steps. One
(,ui not he at all certain that it occurs in every egg, though the
frequency at which it is met would indicate that it did. Ne\er-
thele , it would seem that some rather important function
-hould he aligned to the aborting spindle; for it- appearance in
-ome do/en different species of flat worms niu-t exclude it from
the category of abnormal behavior. It i- therefore hoped that
an oppori unity may be offered to work out it- history in detail
in a !a\orahle form, such as one of the o\ipan>u- species from
\\hich a -eries of stages can be secured from the uteru-.

J, / '.inatiini. By insemination is usually meant the act
ot introducing the spermatozoa into the egg. In ('>raflUla the
process curs during the last stages of yolk ah-orption \\hilc
tin- formative capsule is still attached to the ovary, and consists
in the introduction of spermatozoa into tin- cap-ule. The in-
M ininaiin- organ is the modified, or bifurcated part <>l the uteru-.
In 1 is -hown a beautiful case. The -ection passes through

the di-tal end of the uterus, and the left-hand lohe of ihat or^an,
tilled \\ith -pennatozoa, is in direct contact with the hinucleated
i ap-ule. Any number of similar fign n he demon-iraied

in the preparations, so that no doubt can exisl regarding ihe
interpretation which we have placed upon Mich picture-. It
\\milil -eem that the uterus took an acti\e part in the pn-
of in-eminaiiin. Linton reports an ob-er\ aiim \\hich points
to the -ame < ' inclusion.

This method of insemination must necessarily permit a numher
f -permaii>/oa to get into the capsule, hut o\\in- to their -mall
-i/e thev are soon lost among the yolk granules, so that an
enunu-ration of them is impossible. S- > far as one can tell the
-perm- di. not at first invade the immeiliate neighborhood of
the two nuclei, but remain in the peripheral portion of the cap-
Mile, and later penetrate the egg- a -lion time before the begin-
ning of maturation.

4. Maturation. As in the case of all ova accompanied by the
proce of feriili/ation. those of Gratfilhi throw off t\\o polar

1 88 J. THOMAS PATTERSON.

bodies. The first maturation follows immediately upon the
fusing of the chromosome vesicles produced by the aborting
spindle, and at the time it occurs the sperm is already present
in the egg (Fig. 21). The demonstration of maturation as taking
place simultaneously in the two eggs within the same capsule
is the most cogent proof we can offer against the idea that this
animal exhibits polyembryony; because if this is a fact, each egg
must subsequently be fertilized before it could develop, and that
would at once remove the case from the category of polyembry-
ony; and even though no other proof could be offered, such as
we have given in connection with the section on the formation
of the capsule, this would be sufficient to establish our main
contention. As a matter of fact we have found two very clear
cases in which each of the two eggs is undergoing maturescence.

The egg in one of these shows the first maturation spindle in
the anaphase (Fig. 20). The spindle is extremely large and has
at each end a large aster with very conspicuous centrospheres,
in the lower of which is a single centrosome and in the upper of
which are tw r o centrosomes. The sperm head, already showing
signs of its transformation into a pronucleus, lies near the lower
aster. Between the upper pole of the spindle and the egg-
membrane is a clear space due to a depression in the egg at this
point. In a slightly later stage the egg elongates in the direction
of the long axis of the spindle, taking on an appearance much
like that of a pear, with the smaller end representing the animal
pole. A very large polar body is then cut off, and the mate to
this egg fortunately shows this process going on (Fig. 24). Since
the first cleavage division results in producing a micromere of
about the same size, opportunity is afforded for confusing this
cell with tin- first polar body, but. the difference can easily be
told if the chromosomes arc in a condition that allows their
enumeration to be- made.

In the second case one of the eggs (Fig. 22, on the left) sho\\>
the maturation spindle in prophase with four distinct tetrad^,
and the other cell a polar view, in which only three chromosomes
appear. I have been unable to find a totinh tetrad, and I theiv-
fore assume that it must have been destroyed by the knite.

Several eggs showing the first polar body just extruded have

EABLY DEVELOPMENT OF GRAFFILLA. I Si)

been found. In a tr\v <>i these the egg nucleus is in a resting
condition, thus indicating that the second division may not
follow immediately upon the first. However, I have not yet
succeeded in finding the spindle of the second polar body divi-
sion, but th.it a second polar body is thrown off is clearly shown
in ai It -.1-1 one case (Fig. 25). Here the constriction of the second
polar body has just been completed, while the first polar body
liaxini: undergone division is in the process of disintegration.
The rapid disappearance of the polar bodies immcdiatrK after
the\ are vi\-ii off has added to the difficulty of studying their
formation, a-, well as to the study of the formation of tin- lir-t
micromere.

iVrli.ip- ihc most striking feature of maturation on druffilhi
is i In- large size of the first polar body. This i- not surprising :
foi it i- noi uncommon for a large polar body to bi- :J\en off in

tin i tain Hat worms. It was in the egg of a turbellarian,

Prostth , that Francotte, '97, discovered the intere-tin^

lai i that tin- first polar body may be nearly as lar^c as the <
it-ell, ami may occasionally be fertilized and d-\elop into a
.small iM-trnla, after having first formed a small polar body like
the MI i .n< 1 c tin- .it the egg.

5. l-i-rtilizutiini und the First Cleavage. Fcrtili/ation follou-
alnio-t immrdialfly upon the throwing oil of the -,-,-,, D,| p.ilar
bil\ . I ha\e found no exceptions to the rule that only one
.spn -niat i/oon i-nters the egg. The sperm pem-tratr- the
in thr vegetative hemisphere (Figs. 20, 21 , 24). and passes io\\ard
tin- center \\ln-re it remains while the polar bodie- an- bcin-
^i\rii ott. I 'In- -piTin nucleus tlu-n nm\ i-> to a point mar the
animal ]>oK- \\here the copulation of the t\\o pi.nmclei occurs
(Fig. 26).

The fir^t i lr,i\a-e is unequal and iv-nlt-, in cutting off a micro-
mere at the animal pole. Any numhi-r ot lir-t cK-axa-f -|iindlr-
ha\e been observed, and they are all characteri/ed by having
eii;ht chromo-omi-s. and by having centnsomes which are much
more conspicuous than those ot the maturation spindle--. In
this a- in all of the subsequent early cleavages, the nuclei enter
into a "rest" st.i^r immediately after the completion of the
di\ isji in : and in-trad of forming a single vesicle, the chromosomes

I9O J. THOMAS PATTERSON.

more or less retain their individuality, thus producing a number
of small vesicles, some of which may, ho\ve\vr, IUM- together
(Fig. 8).

IV. SOME GENERAL CONCLUSIONS.

We find no evidence in Graffilla that the two embryos commonly
found within a capsule are the product of a single fertilixed egg.
On the contrary, it is clear that they spring from two ova, which
have become enclosed within a common envelope. In tlii-
respect our species does not present anything unusual; for while
it is the rule among the rhabdocceles to have one embryo in a
capsule, yet there are a number of well-known exceptions to
this. In his excellent monograph on the turbellaria Von Graff,
'08, has recently given a list (p. 2338) of these exceptions, which
are as follows: Gyratrix hermaphroditus, Provortex, Collastoma,
Umagilla, Polycystis, Fecampia, and Monocells lineata, each has
two embryos in a capsule; Anoplodium, 1-2; Prorhynchus stag-
nalis, 1-3; P. balticus, 6; Graffilla, 2-3; Promesostoma marmo-
ratuw, 4-7; Dalyellia truncata, millportiana and viridis, 4-12;
Plagiostomum mttatum and girardi, 10-12; and finally, Syndesmis,
2-13. All of this goes to show that the facts which we have
brought forward concerning the method of reproduction in
G. gemellipara arc entirely in harmony with what is known to
occur in the other turbellaria. Even the manner in which the
two ova become surrounded by nurse cells within the reproductive
glands presents nothing new (unless it be in those cases in which
the ova for a while lose their individuality). Furthermore, the
habit of directly freeing the ova, with their nurse cells, into the
mesenchyme is also seen in such forms as Dalyellia viridis and
Olisthanella obtusa. In most forms in which two or more eggs
are enclosed within a capsule the ova become surrounded by a
common follicle (A nurse cells before they pass to the uterus,
where the shell or true capsule is usually secreted.

Some of the rather rare conditions seen in G. gemellipara are
the indefiniteness of the reproductive ducts, the rudimnnarv
state of the reccptaculum scminis, the failure of tin- i-gg^ to
enter the uterus, and consequently the probable secretion of tin-
shell by the mesenchyme. But all of these conditions arc in-
cident to the viviparous mode of reproduction. Lin ton

EARLY DEVELOPMENT OF GRAFFILLA.

that this viviparity may be seasonal and parallel with the pro-
duction of summer eggs, as is known to be the case in some of
the Me-o-i,,mata. Certain facts in Graffilla might seem to
indicate that what we have de-cribed are the conditions peculiar
to a period r,f s Um mer egg production. Thus the thin shell i- a
di-tinctive characteristic of a typical >uniiner egg (Siibitanei<
and the well de\ eloped unicellular >hell--lands suggest at lea-t
that these organs could function later, if the species entered
ii|on .1 period . ,f winter egg production (I >auen -ier . However,
in the ab-eiice ot any proof that winter eggs are produced, and
in the li'Jit ot the fact that several of the female reproductive
ins s| 1( ,\v a rudimentary or degenerate condition, we are
inclined to belie\e that what we have described is the e\clu-i\e
method of reproduction in this species. The pr--ence ot shell-
vi 1. 1 nd-, ot a rudimentary receptaculum semini-, and of an indefi-
nite uterus and ducts, instead of indicating that t he species could
Liter produce \\ inter eggs, may and probably do, sj-nifv the i 1
relationship of this species to the other member-, of the -enus in
\\hich ilii--e -tructures are functional.

( )f the halt do/en species of Graffilla described iii the literature,

'llifxirii appears tO COme Closest, in its gent r.il arrangement

of or-. m~, to d. Miiricicohi. It also shows -oine -imilarity to

Ha iniiolii. but dilters primarily from the latter in ha\in^ the

lital poie situated further back on the bod\ .

In conclusion, we should like to point out. as a result of our
studies on this animal, the necessity of exercising mv.u precaution
in concluding that a ^iven species exhibit s p, ,|\ embr\ on\ . I n-
( lou bit i IK the phenomenon of polyembryonj will, in the future, be
found to be much more extensive than we ha\e Mi-pecied; but
before coming to anv delinite conclusions, the in\e-ii-aior should
trace the de\elo].inent back to the fertili/eil .

LITERATI ki

Bbhmig, L.

'86 1 ntct-iu luini;' n uln-r rhabdocoele Turbellarien. I. Das genus Graffilla.

/fit. mi \\ ISB. /.>1.. H.I. 43.
Francotte

'97 R- In !. In - ~iir la maturation i-lu-/ l.> I'nly<-l;ile<. Mem. Cour. A
Belg.

IQ2 J. THOMAS PATTERSON.

Gardiner, E. G.

'95 Early Development of Polychcerus caudatus Mark. Journal of Mor-
phology, Vol. 1 1 .
Gardiner, E. G.

'98 The Growth of the Ovum, Formation of the Polar Bodies, and Fertilization

in Polycharrus caudatits. Journal of Morphology, Vol. 15.
v. Graff, L.

'82 Monographic der Turbellarien, I. Rhabdocoele, Leipzig.
'08 Turbcllaria. Bronn's Tier-reichs, Leipzig.
Hallez, P.

'87 Embryogenie des Dendrocoeles d'eau douce. Paris.
Jameson, H. L.

'97 Additional Notes on the Turbcllaria of the L. M. B. C. District. Proc.

and Trans, of the Liverpool Biol. Society, Vol. 9, pp. 160-178.
v. Ihering, H.

'80 Graffilla muricola, eine parasitische Rhabdocoele. Zeit. fur Wiss. Zool.,

Bd. 34-
Lang, A.

'84 Die Polycladen, Monographic. Fauna und Flora des Golfes von Naepel.,

Bd. ii.
Linton, Edwin

'10 On a New Rhabdocoele Commensal with Modiolus plicalttlus. Journal

of Experimental Zoology, Vol. 9.
Nicoll, W.

'06 Xotcs on Trematode Parasites of the Cockle and Mussel. Annals and

Magazine of Natural History, Ser. 7, Vol. 17.
Schmidt, F.

'86 Graffilla braunii n. sp. Archiv fur Naturgesch., Bd. i.
Selenka, E.

'81 Ueber eine eigentiimliche Art der Kernmetamorphose. Biol. Central-

blatt, Bd. i.
Surface, F. M.

'07 Tin- Early Development of a Polyclad, Planocera inquilina Win. Proc.

Acad. Nat. Sci. of Phil., Dec., 1907.
Wheeler, W. M.

'94 Planocera inquilina, A Polyclad Inhabiting the Branchial Chamber of
Sycolypus canaliculatus Gill. Journal of Morphology, Vol. 9.

194 J- THOMAS PATTERSON.

PLATE I.

FIG. i. Horizontal section of a young specimen, showing the testes (<). seminal
vesicle (sv) which contains sperms, uterus (K), unicellular shell-elands (s), and the
germ-vitellarium (v). X 222.

FIG. 2. Anterior half of a slightly oblique section from an adult individual.
The uterus shows a distinct, but small diverticulum (sr) which in all probability
corresponds to the receptaculum seminis of the other members of the genus. Note
that the testes have disappeared. X 222.

FIG. 3. Horizontal section passing just below the intestine of a sexually ma-
tured individual. The section passes through the distal or bifurcated region of
the uterus (u), which contains spermatozoa, o, ovary; c, capsule containing two
eggs, one of which is giving off the first polar body; v, vitellarium. X 222.

FIG. 4. Horizontal section of another sexually matured animal, but which
passes at a slightly lower level than the preceding. It shows clearly the bifurcated
region of the uterus; and also the relationship existing between the uterus, ovary
and vitellarium. X 222.

FIG. 5. A longitudinal median section (slightly schrmuti/i-d) of a rather old
individual. It shows an advanced stage of the "female" condition, m, mouth;
pit, pharynx; oe, oesophagus; a, atrium; g, genital port-; />, penis; sv, seminal ve-
s, unicellular shell-glands; u, uterus; v, vitelline cell in uterus; c, capsules containing
embryos; /', intestine. X 117.

BIOLOGICAL BULLETIN, VOL. XXH

PLATE i.

r- -e

^_ _ j

\ / f

) '

m

V C

J. T. PATTERSON.

J. THOMAS PATTERSON.

PLATE II.

FIG. 6. Transverse section taken through the region of the tip of the uterus.
X 381.

FIG. 7. Two ova that are beginning to be surrounded by vitelline cells pre-
paratory to the formation of a capsiile. X 784.

FIG. 8. The two-celled stage, showing a micromere and a macromere. X 740.

BIOLOGICAL BULLETIN, VOL. XXII.

PLATE ,1.

v

J T. PATTERSON.

198 J. THOMAS PATTERSON.

PLATE III.

FIG. 9. The posterior half of an ovary which shows the process of yolk ab-
sorption. On the right a capsule is being formed about two nuclei. X 650.

FIG. 10. A later stage in the same part of another ovary. Note that the two
nuclei are immediately surrounded by a finely granular protoplasm. X 650.

FIGS, ii and 12. Two eggs from the same capsule. This represents the con-
dition shortly after the disappearance of the aborting spindle. The nucleus is in
the form of faintly staining vesicles which in part are fused together. X 812.

BIOLOGICAL BULLETIN, VOL. XXII

ft:

, ' ','/}iir '

.;,-; /,- '

10

-'o

12

2OO J. THOMAS PATTERSON.

PLATE IV.

FIG. 13. Two naked ova that have not become surrounded by a capsule.
Such eggs apparently float about in the parenchyma, but probably never produce
embryos. X 543-

FIG. 14. A binucleated capsule in which the lower nucleus is surrouned by an
intravitelline membrane. X 798.

FIG. 15. A binucleated capsule. X 543.

FIG. 16. Two ova completely surrounded by a follicular layer of vitelline cells.
Only a part of one of the eggs is seen in the section. X 543.

FIG. 17. A typical case of an aborting spindle. Note that chromosomes are
absent from the spindle. X 798.

FIG. 1 8. Another example of aborting spindle, in which the chromosomes are
located among the rays at one end. X 798.

BIOLOGICAL BUILEIIN, VOL XXII.

PLATE IV

13

5

14

"':-": .

- .

5ft

. ..

.-;;

.: : .v "

16

v

*

17

J T PATTERSON.

2O2 J. THOMAS PATTERSON.

PLATE V.

FIG. 19. This shows a capsule about to be set free into the parenchyma. The
eggs exhibit the condition which immediately follows the disappearance of the
aborting spindle. Each egg has four chromosome-vesicles, and in the one on the
left the centrosome is present. Lying just above this newly formed capsule is
another in the process of formation. Only one of the ova shows in the section, and
in it the centrosome has divided and the aster is present, thus anticipating the forth-
coming aborting spindle. X 993.

FIG. 20. The anaphase stage of the first polar spindle. X 2,394.

BIOLOGICAL BULLE'IN, VOL.

PLATE V.

I -

'/I

20

J. T. PATTERSON.

2O4 J- THOMAS PATTERSON.

PLATE VI.

FIG. 21. An ovum shortly before the formation of the first polar body. The
nucleus is the product of the fusion of the chromosome-vesicles of a stage like that
in Fig. 19. The section passes through but one of the two ova in the capsule. In
most of the capsules of this period the protoplasm of the eggs contracts in the re-
agents more than does the surrounding vitelline material, thus producing a clear
space between the two materials. X 543.

FIG. 22. A capsule in which both eggs are undergoing maturation at the same
time. X 543-

FIG. 23. Two of the tetrads from the preceding figure. X 2,394.

FIG. 24. The cutting off of the first polar body. This egg is a mate to the one
shown in Fig. 20. X 798.

FIG. 25. This stage shows the close of maturation. The first polar body has
undergone division and is disintegrating. X 543.

FIG. 26. Fertilization stage. X 543.

BIOLOGICAL BULLETIN, VOL. XXII.

PLATE VI

\

21

23

22

24

25

:

,

J. T. PATTERSON

Vol. XXII. March, 1912. No.

BIOLOGICAL BULLETIN

"STRAINS" IN IIVDATINA SENTA

D. U. WHITNEY.

In .i lormer paper results of experiments upon t\\<> \
the roiiier Hydatina scuta were given in regard t<> tin- production
ot OIK- hundred generations of females without the appearance
..I males in either race. These experiments have In en extended
further fur al>uiit seventeen months and as they are <-onrlnde.!
it seems desirable to record the results obtained partly a< a
iilii niation of the earlier conclusions and partly because they
turni \ idence which shows that there exists diffrreiit races '>r
strain- < >r lines within this particular species of Hydatina scntn.

In the furmer paper it was shown how readily male --prodiu -inu
females could be produced in newly made dilute unconkrd
hoi -e manure cultures and also how readily the male-pro-
ducing lemales could be repressed in newly made concentrated

ked hoi-e manure cultures.

In the piv-eiit paper the parallel history of three r, ( ces of
rutiler- .1, H, and C is given. Races B and C are the same
races ii|iun \\hich the former conclusions were based while i
.1 i- .in additional one. Races A and B are si-ter rate-, both
ha\ in- tle\ eloped from one fertilized egg while race ( ' is unrelated
to races .1 and B except in as far as all three races came from
the s.inie general culture of rotifers which was originally collected
at ( irantuood. New Jersey, in 1906.

Race- .1 and B were always conducted in a parallel -, ries but
race < ' \\as not put into the parallel series until it \va- in the 301 h

Deration. During this early period of the three race- before
they \\erc all put into the parallel series the food \\a- from mis-
cellaneous protozoa cultures of various ages made in dilute un-
cooked hor-e manure media. The summary ..f the early history

205

D. D. WH1 PNEY.

of these three races before they were all conducted in the parallel
series is recorded in Table T. The percentage of male-producing
females of races A and B are practically equivalent, while that
of race C is much lower.

TABLE 1.

Showing the number of female- and male-producers in the three races .4, B,
and C, from their origin to the time at which parallel records were taken. Female-
producers are designated 9 9 , male-producers c? 9

Race.

Genera-
tions.

No. of

9 9-

No. of

<?9.

Per Cent, of

c?9.

Time.

Food.

A

I-I44 I,l88 iSl

13.22 +

Oct. 6, 1908, to

Dilute uncooked

Aug. 31, 1909.

horse manure

B

I-l

1,224

167

12.00 +

Oct. 6, 1908, to

media, 7-28 days

Aug. 31, 1909.

old. Miscella-

C

i- 35

2IO

10

4-54 +

June 16, 1909, to

neous protozoa

Aug. 31, 1909.

growing in them.

September 3, 1909, these three races A, B, and C were started
in a parallel series under as identical external conditions as
possible. At the beginning of this parallel series the generations
were renumbered and the beginning generation of each of the
races ifi this series is called No. I. Ten young females from each

TAHLI-: II.

K:

No. of No. of

Per Cent, of

Generation. $ Q . d" 9 .

cF9.

1

ime.

Food.

I

9 3

25

June

16.

1909.

June

18,

1909.

2

20 o

O

June

18.

1909.

June

20,

1909.

3

16

4

2O

June

21,

1909.

hH

June

22.

1909.

w

3

4

8

2

20

June

24.

1909.

rt
EH

June

26,

1909.

en

Partial

rt
V

summary. 53 9

14-51 +

summary. 53 9

14

5H

E

&

5-34 15" D

Juno

26,

i'J

Aug.

30,

1909.

35

7

i

12.

5 Aug.

30,

Sept.

i,

1909.

Total

summary. 210

10 |

54 +

Detailed history of race C throughout the first 35 generations, wliirh i
summarized in Table I.

"STRAINS" ix HYDATINA SIM \. 207

generation of each race were isolated at the same time and each
female placed in a Syracuse watch glass and allowed to mature
and to product- daughter females. Then this pn>cc was re-
peated tor .^45 generations. All the females at each isolation
urn- placed in the same quantity of tap water to which \\a-
added the -ame amount of food culture that was taken from one
food jar. The watch glasses in which the rotifers lived al\\a\-
\\en- in thn-f -tacks side by side at room temperature. IVac-
ticallv all external influences were as identical as it was po ible
to make them.

Tin- del tiled observations are given in Table III. in parallel
column- and the -ummary is given in Table IV.

At ill' <nd of Table I., races A and B, which up t<> tin- time
were l< d on \arious protozoa cultures, were practically identical
in re-ard to the percentages of male-producing females in each
. but at the beginning of Tables III. and IV. when the t \\ o
races \\ere -ubjected to. uncooked concentrated food culture
media a de< ided change occurred. Race .1 retained and e\en
ede. I its former rate of production of male-producing female-,
but in ra.e li the rate was very perceptibly lowered. K
louered -liuhilv its rate of male-producing females. Thi-
runvd during the first 50 generations. From the 57th to the
^I5ih generation in races A and 6' and to the end of race />'. t lu-
ll ;o -iii -ration, concentrated cooked food media \\a- u-cd ami
cau-eil a decided lowering of the production of male-producing
female- in all race-. In race t' this was reduced to zero, in i
/>' to less than I per cent., and in race -1 to about 3.5 per cent.

Tin- confirm- the earlier results in showing that it i- po--ible
1>\ external conditions to repress entirely the production of male-
pr.'du. in. females in some races of this rotifer for a long period
of time. In race (' the male-producing female- were repre ed
for Hi-rations and then reappeared when the food media

\\a- made too dilute accidentally.

If the-e three races were exactly alike in their power to produce
male-producin- females and all were subjected to the -ame ex-
ternal conditions they ought to produce -uch male-pro<hn in-
female- at the same rate. Howe\ er, as the <ibo\e ob-er\ ation-
-ho\\ that the rates of production of male-la\ in- leniale- vary

208

D. D. WHITNEY.

TABLE III.

Showing number of female- and male-producers in a parallel series of 345 gen-
erations in the three races A , B, and C. Generations 1-56 show the detailed results
when the three races were fed upon concentrated uncooked food media and gen-
erations 57-345 show the detailed results when the same three races were fed
upon concentrated cooked media.

Generation.

Race A.

Race B.

Race C.

No. of

9 9-

No. of

c? 1 ?.

No. of

9 9.

No. of

c?9.

No. of

9 9.

No. of

c?9.

I

5

6

7 o

2

8

2

IO

o

10

3

9

IO

9

4

9

o

9

o

9

5

9

o

8

9

6

9

o

9

9

o

7

9

o

9

9

I

8

10

o

9

o

IO

o

9

8

2

IO

o

10

10

8

2 IO O

9

1

1 1

9

I 10

IO

12

10

O

IO O

IO

13

10

O

I O . O

10

M

7

3

10

9

I

15

9

I

9

1

10

16

7

3

8

I

10

17

10

o

IO

10

o

18

9

I

10

10

o

19

10

o

9

9

I

20

9

I

10

IO

o

21

8

2

IO

8

2

22

8

2

9

I

10

o

23

9

I

8 i

IO

24

8

2

IO O

10

o

25

9

I

IO

10

26

9

I

9 i

9

I

27

S

5

IO O

10

28

10

o

1C) O

9

29

10

O' M)

9 i

30

8

2

10

IO O

31

5

5

9 i

IO O

32

10

8 ' 2

10

33

10

IO O

10

34

4

6

IO O

10

35

6

4

Q O III i>

36

6

3

10

IO O

37

9

I IO O

10

38

8

2

IO O

10

39

9

I

IO O

9

I

:'

i

9

6 4

6

4

41

7

3 "i o

10

8

2 IO

IO

43 7

3 '" o

10

o

44 7

3

IO O

9

I

45 7

i <i <;

o

46

7

3

10 10

47

7

3

9 i

10

"STRAINS" IN HYDATINA SKNTA.

TABLE III. Continued.

Race A.

Race B.

Race C.

Deration.
No. of

No. of

No. of of

No. of

\ i. of

9 9-

cf 9.

99.

c?9.

9 9.

?9.

7

3

o

.

I

49

6

3

9

I

IO

o

7

3

I

IO

51

7

3

IO

o

IO

o

7

2

IO

id

o

IO

O

IO

o

54

9

1

IO

o

IO

o

IO

O

10

9

I

56

6

4

I

-

2

57

10

O

10

IO

IO

O

IO

(}

IO

o

2

10 .1

10

o

4

6

IO O

IO

i.i 8

2

IO O

10

62

4

6

IO O

10

63

)

y

IO O

10

o

64

to

o

10 o

IO

o

65

10

O

IO O

10

66

9

i

1C) o

10

o

6?

8

2

10 o

10

o

68

10

O

10 O

IO

o

69

10

IO O

IO

70

10

O

IO

o

10

o

7i

5

3

8

10

o

72

IO

o

10

o

IO

o

7J

10

o

10

o

IO

o

74 10

10

10

IO

o

IO O

IO

o

76

10

o

10

10

77

10

o

10 o

IO

o

78

10

o

10 o

10

o

79

10

o

10

10

o

10

o

10

10

o

M

IO

o

IO O

IO

o

1

IO O

10

o

I,,

o

10 O

IO

o

10

o

10

IO

10

o

IO O

IO

o

10

o

IO O

10

o

10

o

10

o

IO

o

10

o

IO

IO

o

IO

IO O

10

90

IO

o

10

10

91

9

I

10

10

o

pa

10

10

9

o

93

10

10

o

IO

<i\ i"

o

IO O

IO

o

95 10

o

IO

IO

o

96 i"

10

10

o

97

IO

IO

IO

o

98

10

o

10

o

99 1 10

o

! 10

1 10

< lurnti.ited cooked horse manure media began to be used.

210

D. D. WHITM.N .

TABLK III. Continued.

Generation.

Race A .

Race B.

U.i. e C.

No. of

9 9.

No. of

c?9.

No. of

9 9.

No. of

c?9.

No. of

9 9.

No. of

cT9.

100 9

I

IO

IO

o

101 10

10

IO

o

IO2

10

o

IO

IO

103

IO O

10

o

10

o

104

10 O 10

IO

o

105

9

I IO

10

1 06

IO

o

IO

IO

o

IO7

10

o

IO

o

10

108

10

o

10

IO O

109

10

10

10

o

no

IO

O IO

o

IO

o

1 1 1

10

o 10

o

6

o

112

10

o

10

o

10

H3

IO O

10

O IO O

114

IO

o

IO

O 10 O

"5

IO

o

10

o

10

116

10

10

IO

117

10

10

o

IO

o

118

IO O

IO

o

10

o

119

IO O

IO

o

IO

120

10

10

10

o

121

10

IO

o

IO

o

122

10

o

10

o

10

o

123

10

o

IO

10

o

124

IO

10

IO

125

10

10

o

IO

o

126

IO

o

10

o

10

o

127

10

o

10

IO

128

10

10

o

IO

129

10

10

o

10

o

130

10

10

o

10

o

131

10

10

o

10

132

10

o

10

IO

o

133

IO

IO

o

10

o

134

10

IO

o

10

135

10

O lo

IO O

136

10

10

o

IO O

137

10

10

o

IO O

138

10

o

IO

10

139

10

10

o

IO

o

I4O

10 o

10

10

o

141

10

10

10

142

IO

o

10

10

143

10

o

10

10

o

144

10

IO

IO

145 M>

10

10

1 46 lo

o

10

o

10

147

IO

10

10

148

10

o

10

o

10

o

149

ID

o

IO

I)

10

ISO

Id

O lo O

ID

151

10

O 10

o 10

152

10

10

IO

153

IO o in o lo

STRAIN'S IN HYDATINA SENTA.

21 1

TAHLK III. Continued.

e .-J.

Race B.

aeration.

No. of

No. of

No. of

No

No

?.

c?9.

99-

<7 9.

9 9-

c?9.

[0

o

10

10

I ii

I O

o

IO

i

In O

IO

IO

is?

pi O

IO

IO

In O

IO

10 O

159

!

IO

o

10

160

pi O

10

IO ii

161

I'l O

10

10

162

In

10

10 O

163

10

o

IO

o

IO

i6 4

to

(J

10

o

IO O

10

10

o

10

y

1

IO

o

IO O

167

9

1

10

o

10 O

111

o

10

IO O

169

IO

IO

o

10 O

IT"

10

o

10

171

10

IO

o

IO

IO

10

10 il

173

10 o

IO

10

10

o

IO

o

IO O

IO

IO

o

10

'7''

10

IO

10

'77

IO

o

10 O

10

I O

10

179

10

IO

o

10

10 o

IO

10

181

10

ID

o

IO

IO

IO

o

10

o

IO

10

o

10

o

IO

IO

10

o

10

o

IO

o

10

10

o

ID

o

10

10

10

o

10

10

IO

111

o

10

o

IO

IO

8

IO

10

o

191

10 i'

10

10

o

10

o

10 o

2

o

2

o

I

194

1

o

I

I

195

4

o

10

o

10 O

[96

6

8

o

3 o

197

10

IO

o

10

198

10

o

IO

IO

199

10

o

IO

IO O

10

10

III

10

10

o .

In O

202

10 o

IO

Hi

9 i

10

o

III

9

1

10

111

o

10

o

10

in O

10

o

I II

Hi

IO

o

10

IO O

212

D. D. WHITNEY.

TABLE III. Continued.

Generation.

Race.-/.

Race B.

Race C.

No. of

9 9.

No. of

cT9.

No. of

99.

No. of

d 1 9.

No. of

9 9.

No of

c?9.

208

10

10

o

10

209

IO

o

IO

10

o

2IO

10

10

10

o

21 I

10

o

10

10

212

10

o

9

1

10

o

213

10

IO

IO

214

10

10

o

10

215

IO

o

IO

o

10

216

8

8

8

217

9

I

10

o

10

o

218

10

10

to

219

IO

o

10 O

10

o

22O

10

o

10

10

221

10

IO

o

10

o

222

5

o

4

IO

223

10

o

IO

o

IO

224

IO

IO

10

225

IO

10

o

10

o

226

IO

o

9

I

10

227

I

I

I

o

228

10

IO

o

IO

229

IO

o

10

10

230

IO

o

4

6

10

231

10

10

10

232

10

5

10

233

10

o

10

o

10

o

234

7

4

9

o

235

i

2

10

236

4

I

3

237

10

o

I

10

o

2 3 8

IO

o

4

o

IO

239

10

9

240

10

o

Died.

10

241

3

10

o

242

4

IO

243

10

IO

244

8

2

IO

245

10

o

10

o

246

IO

10

o

247

10

10

248

10

IO

249

IO

o

10

250

IO

10

o

251

IO

o

10

252

9 o

10

o

253

10

10

254

9

9

255

10

10

o

256

6

o

9

o

257

10

IO

258

10

IO

o

259 ",

IO

o

10

o

10

261 10

' 10

STRAINS IX HYDATIXA SEXTA.

TABLE III. Continued.

Race A. Race B.

Xo. of Xo. of No. of

No.

\

d"9. 99. (79. 99.

9.

7

IO

90 10

o

90 10

IO O 1 'I

o

O in

o

O IO

IO

IO

o

IO

10

o

-'71 I O O

IO

o

in O

IO

10

10

7 3

10

o

IO O

IO

o

IO O IO

10 10

o

278 If)

IO

II

10

1 II

8 o

IO

II

9 1 10

8 2 10

1

8 a

IO

If)

10

II

4 6

10

5 4

10

II

9 1

IO

II

8 2

10

II

6 o

10

7 o

7

-;! 6 1

9

1

8 o

2

9

III

o

IO

pi

II

4 o

9

o

-/ 40 9

i o v

o

4

o

4 5

4

II

8 o

-

70 10

II

10 10

o

9 o

o

10 1"

II

9 o

o

IO

-

o

80 9

o

40 7

309 60 i"

7 10

o

.ill 10 1"

o

3 5 o

5

o

313 5 o

5

.i 1 4 5

5

315 3 S

o

214

D. D. Will I \1 V.

TAHLK III. Concluded.

Generation.

Race A.

Race B.

K.I.

No. ot

9 9.

; C.

No. of

99.

No. of

d"9.

No. of

9 9.

No. of

<79.

No. of

d"9.

316

4

I

5

317

5

5

318

5

o

.

5

319

10

o

10

32O S 2

10

321

5

5

322

5

5

323

4

I

5

324

5

5

325

5 o

IO

326

9 i

10

327

6 2

IO

o

328

3

o

10

329

9

I

10

o

330

IO

o

10

331

7

3

10

o

332

IO

o

10

333

8

2

IO

o

334

6

4

10

o

335 8

10

o

336

8 2

10

337

8 2

IO

o

338

9

I

10

o

339

IO

O

10

o

340

10

o

10

341

10

o

10

342

IO

10

343

4

2

IO

o

344

9

10

345

5

I

10

o

346

6

8

2'

347

8

o

10

348

5

9

I

349

3

1)

9

I

in the three strains, A, B, and C, when all external conditions
are identical the only conclusion that can he drawn is that the
three strains differ at least in regard to this single characteristic.
Perhaps they all may be potentially alike in their capacity to
produce male-laying females but some races may be more easily
effected than other races by the influence which causes male-
producing females to be produced.

Whenever races A and B were put into newly made diluted
uncooked culture media in battery jars great numbers of fertilized
eggs were produced in U>ih races. From general observations
they seemed to be produced in equal numbers, thus seeming to

1 Food media was diluted accidentally.

"STRAINS" IN HVDATINA SHNTA. 215

form evidence that these two races were potentially alike in their
power to produce male-producing females but when conditions
were unfavorable they differed, as shown in the parallel series, in
their re-p. ,ii-iveiiess to the influences that so acted upon the
females as to cau-e them to produce male-producing daughter
fem. ili--. Ho\ve\er. when race C was put into newly made un-

ked culture media in battery jars very few fenili/ed eggs
were prodm ed, thus K-eming to show either that thi- race C was
potentially different from the other two races in it- rapacity t
produce male-producing females or that it was not as ea-ily acted
upon by iln- male-producing female influences as were races .1
and /v \oi\\hh-tanding this fact that race C produced very
le\\ male- producing f. -males when put into battery jar- containing
dilute iiiiiooki-d horse manure media it should be stated that in
the e.irK lii.~tor\ of race C' it had as high a percentage of male-
producin- female- in the first four generations as \\a- found in
either race .1 or B. Table- II. The race was isolated from a

. r.il cnli ure jar in \\hich an abundance of males were appearing
at the time i >l i-o|ation. Beginning with generation ii\e verj
le\\ in. ile- appeared thereafter. This early history sh.\\ - that
race Cat one time was as potential in its power to produce male-
producipv: it-males as races A and B, but whether it later lost
thi- power ( -r never was again subjected to as favorable intlnem es
for the production of male-producing females it i- impo--ib!e
to state. \\ ii.ii> \<r may be the true explanation of thi- di\-r-
gence in the male-pnKlucing female rates of the three races it
-iiteK indicate- a dillerence in the races either in their capacity
to product- male-producing females or their responsiveness to
tin- inlluenct - thai cause male-producing femak-- to U- produ. .-d.
I'niinett coinliidid that he found "sex -train- ' in Ilydntiiid
/ \\hich differed in their power to prodnc.- male- and even
i on.-ludt-il that he found some strains that produced no male-.
It is verj po--ible that such maleless strain- \\ire n-alh like
rat c ' in the above experiments. From ob-<-r\ atioiis and experi-
ments publi-hed in an earlier paper ('07 . it was -lu>\vn that no
pun- female -trains could be found. The re-ult- of the pie-.nt
experiment- o>i roborate this earlier conclusion. However, the
evidence at that time showed no strains of anv kind but the

216

D. D. WHITM.V

:

ss

'S

'3

w

8

3

>

""" S
]S4

i

8 5

v
u

o

3

C
=

- s
3

o

1C

U)

"" o

_

~

o 8

l- O

Ifi

E S

3 a/

i

o

> i u->

S S | ?>

CJ rt N *rv

aa 5

8^-

I- ',".

5^-a

"C """' C 'X

a? q

c 3 f

-5 5 s

^- w .3 ^

towo and other
protozoa,
ooked concentrated

1 0} U

z '-. =
c

O 3
u o

e 7 s "
1*1

C M PH

J.S2

of Polytoma.

D

U

6

H

.

o ^

o o\

Ov

- o

O IO

- .

V

c ^

-^ M

O v

c

Ov

S* -^

^

" C

- o

M rt

ft

^

C

o

^,

>

J*

+

o

V

n

.

0-

o

u-

- o.
r". - <-*

00

o

jg-b

<- .

~2

-

o

~5

10

(S'^

ci

6

S

A

1"

O

00

"oV

N ^
M

c

ZO

w

O

1 O

Tf

o

U XQ

ro

-t

V

00

ro

-

V

u. .

u

o*-

O

N

l-b

o

1-4

Ov

u. .

in

in

-

Tt

o

Z 01 "

M"

n

c

_0

>/-,

>o

00

to

I

i

(N

1

J

10

I/I

STRAINS IX HYDATIXA SI.MA. 21 7

pre-ent evidence collected from observations extending over a
period of about three years and including 300-500 partheno-

etic ^em-ration-, -hows very clearly that strains exist which
differ in their percenta^- of male-producing females.

Moreover, tin- t\\o -ister strains A and B which developed
from the -amc fertili/ed egg differed in their longevity. Strain
/! died out from general exhaustion in the 384th partheiio-enetic

ration, whik- -train .-1 is still alive in the 5041)1 partheno-
n< -ration, although in a very weak and exhausted con-
dition.

Shull lia> compared Mime of the New York strains of Ilydntina

; \\itli a -train from Baltimore and has found a divided
ilitlereiii in the rate of production of males in the tw Miain-.
II' 'It is safe to say, therefore, that we have lure t\\o

pun- line- 1 1 1,1 1 1 1; -m one another in a fairly constant maim- i .

and the din. is an internal one."

SUMMARY.

i. The prudiietioii of male-producing females can In- p.irtly
or \\holK i-epreNM-d hy external conditions in parthenogenetic
iai es ot Ily<!nlinn *i-nta.

j. The jiartlu lie strains are shown to he distinct i

cause mi'ltr identical external conditions they differ in their
po\\er t.. produi-i- male-producing females. This may indicate
that the\ differ in their potentiality of producing male-pn>ducin-
female- or that they differ in degree of respond cne to the
inllueiice- \\hich cause male-producing females to \n- ]>roducei|.
The latter alternative seems more probable.

j. The tuo -i-ier parthenogenetic strains developing from one
fertili/ed egg dilh red in their longevity. One li\ed almut a year
longer and pro.lnced over one hundred IIIOR generations than
the other.

1>I 'I ' ".li \l I \Hi>K.VTORY.

WESLEYAN I'NIVERSITY.
MIDIM.KTOWN. CONN..
January 3. 1912.

21 ^ I). D. WHITM N .

LITERATI-RE CITED.

Punnett, R. C.

'06 Sex-determination in Hydatina, with Some Remarks on Parthenogenesis.

Proc. Roy. Soc., B, Vol. 78, pp. 223-231.
Shull, A. F.

'n Studies in the Life Cycle of Hydatina senta. Jour. Exp. Zool., Vol. 10.
Whitney, D. D.

'07 Determination of Sex in Hydatina senta. Jour. Exp. Zool., Vol. 5.
'10 The Influence of External Conditions upon the Life-cycle of Hydatina
senta. Science, N. S., Vol. 32, No. 819, Sept. 9.

-I PERN1 MI.KAKY CHROMOSOMES, AND SVXAPSIS

i\ < 1:1 i Hopmi.rs (sr

N. M. STEVENS.

'I'll' i ( '- i<!i:!>liilus which I have iist-d in this siudy,

I ha\e ii"! liccii able to identify. The material s t -eiiis to be
homogeneous, .m<l i- the only species of this gemi- that I have
seen about Br\ n M.iur. The insects were found, usually in
pairs, in their bnrro\\s under stones, and wen- i < >llected in
< ><tober and November, i<m> and i<)ii. They are not abundant.
and onlv 7 m.di - \\ ured in 1910, and 5 in i<ii i .

METHODS.
In ill- h individual a few aceto-carmine preparations

\\ITC in. id. . .nid tin- rcniainder of the testes fixed in
mercuro-nitric iluid <>r in Remming's stronger (-liro

'ic niixiiiM I In- \><-[ results were obtained t'rmn -i- ( |i..n<
"I I I* ininiii:' in.iit-ri.il. < ut 10/1 thick and stained \\ith thionin.

^i 11 K\I \i i KAKV CHROMOSOMES.

In one "i tin- I<H insects one, and in another \\\ -uprr-
A cliniin..-.!!!!!-- were found. Tlu-si- rhrunn -i >nn-- .ire
n.idilv di-iiii'<iii-ln-d li"in the other chromic-urn - b\- their
Miialh r -i/e ,unl i rralir brhavior.

The -perm. no: 1 ' mia! chromosomes, like tho-e of Stenopelmatus,

are diltn nil to count, because they do not form a llat jil.ite ,u
an\ stage, bin lie .it -oinewhat dilTerent K-\eU and o\eilap.
The number i-> probabk ,7, exclusive <M -upernuiiierai ;

In the !ir-t -|>- i niatocytes there an i^ bi\alent- an<l the
uni\-.ilent A" 1-1 . \\lu-n no supernumeraries are preM-m Fig. i .
Fig. 2 -lio\\- i ^ bi\.iK-nts, the unpaired chronio^oiiu- A", and 2
-upernuiiu-rar-i The odd chromosome A' i^ u-uall\- found

at one pole of tin- >pindle when the other chromosome-, are in
metapliase Fig ; r but it not infrequently lags behind the others
in the anapha-e 1 'ijr. 4) and is enclosed in a separate membrane

219

22O

N, M. STEVENS.

3

--/- x

8

FIGS. 1-2. Mctaphasc of first maturation mitosis, showing i8+A' and i8+X
+ 25. (Mag. 1,500 for all figures.)

FIGS. 3-5. Mctaphase and anaphascs showing position of X.

FIGS. 60 and b. Young sister spermatocytes of the second order.

FIGS. 7-12. Variations in position and division of the two supernumeraries
in first spermatocytes.

SUPERNUMERARY CHROMOSOMES IX CKUTHOPHII 221

in the tclophase (Fig. 5). Figs. 6a and 6b are sister spermatocv
of the second order, showing X in a separate vesicle: this does
not happen in by any means one half of the second spermatocytes,
X l.rin:/. I ^-hould say, more often included within the same
nm Ic-ar mem! T.ui'' with tin- other chromosomes.

1 i 7 is from the individual which had one supernumerary,
am! I )'<-. 8 to 14 from tin- < >ne that had two. 1 ilso >h-nv-

a less ii-ual p"-iiioii of A", near the equatorial plate. As in the
/'/ 5l the supernumeraries mayor max- not

divide in the t"ir-t maturation mitosis, and they may, when un-
di\i > eiiher pole of the spindle. Their position in the

met.ipha--e does not nec.-^.irily indicate whether they will di\ id.
or in-!. 1 In- di'ti-niiiniiig factor is probably the attachmen;
-pindle ill MI OIK- n both poles. The supernumeraries

in Figs. 7, >> and ') would not divide in this mitosis, but the
piv-en. i.iry in an equatorial plate or between

daii;JiiiT pi. HI ~ d. - essarily assure its division. In I

i" one i- di\i<l<-d. tlie other undivided, while in Fie-. 11 and u
both are divided. In the telophases shown in 1 i^-. i.> and 14,
both -np< rnumerai : divided in one case and neither in the

othi-t. I 'hi p(--ible c i.mbinations of A' and the 2 -npernume-
raries in the ^pcrmaiid- are A", A" + is, X + 25, o, 15 and
loiiroi the possibilities are shown in Figs. 15 to [8.

As to the origin of these supernumeraries, there is little evidence
in tin- material. In MctapodlHS Wilson ('09) disco\end the
pmbable i.ii^in of the -npeniumeraries in an irregular ^erond
>pei -maiiH -\ te miti.~i- in which both " idiochromosomes " \\ent
! the -ami- IM.IC ot the >pindle, and therefore to the same -penna-
lid. The supernumeraries are thus shown \ be duplicate- ..i
the -mailer " idii .chromosome" in Mctapodius, or in one case of
.in " ":-chroMio-i niie" *io). In neither Diahrotica nor in Ccntlio-
philn* i- there a -mailer mate for the A' chromosome pre-ent.
The three supernumeraries which have been observed in (.'cntho-
philus are of about the same size; considerably less than mie half,
and app.irciitU about one fourth the size of A'. The behavior
of the -II|H iniimeraries in growth and re-t -tai^e- of the nucleus
indicate-; their probable relationship to A', and their behavior
in mito-i-. dividing only once, either in the lir-t or the -econd

222

X. M. STEVENS.

19

21

23

24 26

FIGS. 13-14. Telophascs showing the supernumeraries divided (13) or un-
divided (1-4).

FIGS. 15-18. Spermated nuclei showing variations as to presence or absence of
X and the two supernumeraries.

FIGS. 19-20. Anaphases showing unusual position of X.

SUPERNUMERARY CHROMOSOMES IN CEUTHOPHELUS. 223

maturation mitosis, shows that they are univalent. In Diabrotica
soror I have considerable evidence that the supernumeraries owe
their origin to a transverse and a longitudinal division of X
('12 . tii'l it seems probable that those of Ccnthophilus have had
-'milar origin. I have occasionally found ca-e- where A"
"ied about to divide late in the first maturation mitosis
I md 20), but I have as yet no evidence of a tran-\ <

di\ i-ion.

SYNAPSIS.

The. material which was collected in K)ii with the hope of
at on the origin of the Bupernumerari< \ed

to be l'.i\orable for a study of synapsis, or, as I -hould prefer to
call the phenomenon, conjugation of the chronio-onu -. T!

' - uere all fixed in Flemming and stained with either thionin
or iron-haematoxylin. Thionin gave the clearest figures.

In the re-ting nuclei of the spermatogonia the chromo-ome-
are eii her \ i-iblr as separate individuals as in Fig. 21, or are m >re
(.1 1 upleiely resolved into rather fine spireme thread- a-

in I -. In sonic follicles one sees only such -pennatogonial

nuclei .1- in Fig. 21; in others the various cyst- -h\\ \arimis
Dilution into spireme threads. The former condi-
tion I should attribute to more rapid division of the -permato-
'^onia, the time between mitoses being insufficient for complete
H -olntion.

In the \oimgcst spermatocytes, distinguished from the -p.-rma-
>nia by the condensed condition of A', the -pin -me thread-
are -i milar to those of the spermatogonia, per hap- a lit tie coarser.
They are finely granular and more or less nodular. There is no
contra, tion, or synizesis, stage and no complete pol.iri/atioii of
loop- to form a perfect bouquet stage. The -pirem- threads are
n-uallv irregularly but rather evenly distributed through the
nucleus as in I ig. 2.}, which also shows A' in characteristic position
a-ain-t the nuclear membrane. Fig. 2.\ -ho\\ - an extreme and

1 i j. Spermatogonial nuclei, slmwiMi; 1 mm --..Ivr. I . liro-

1 i.. . ; Spcrmatocyte nucleus before syna;

l-'ii.. .'i Similar stage showing partial pnlari/.iticn \ rhr<>i:

I i. 16. Stages in parasynapsi-.

224

N. M. STEVENS.

a b

34

30

36

33

35

FIG. 27. Double spireme stage.
FIGS. 28-29. Prophase stages.

FIG. 300-1. Various prophase (a-g) and metaphase (h, i) forms of the bivalent
chromosomes.

FIGS. 31-32. Chromosomes in metaphase of the first maturation mitosis.

SUPERNUMERARY CH K< >M< >-< >M 1 - IN i I I 1 HOPI11L! S. 225

unusual case of polarization, but here only nine polarized loops are
-ent and the other spireme thread- extended in various diiec-
tion- through the nucleus. In some follicles there is a perfectly
sharp divi-ion-line between cysts containing spermatocvte nuclei
of the chaiacter of that shown in Fic r . 2.^. and cysts in which the
-pin-mi- i- about twice as thick (Fig. 27). There is no evidence
of ti -I..-;., nap-is, and also no evidence of any such longitudinal
contraction of the threads as would be required to give tin- thicker
-pin-iii'- of I ig. 27. In other follicles there come in bet \\eeii
tli -, other cysts in which as a rule the- chromatin

thn-.id- are \-ry irregularly distributed; i. e., thru- i- a con-
cpicnoii- ab-i-mv of the rather even spacing of tin- thn-ad- ol>-
-T\i-d in Figs 22 and 23. Careful inspection of -uch cysts
reveals varioi .'-sin pairing, conjugations, or -ynap-i- "I the

chroiiMtin tlin-ads (Figs. 25 and 26). In tin- sunc cyst "tic
ofii n find- \arious stages, only one or two pair- and the other
thread- -iirJe as in Fig. 25, and all stages up to one in which
all ot the threads arc in pairs. Occasionally some of the pair- in
a inn leu- -how as much polarization as is indicated in I ig. J<>,
I'lit in the -ame nucleus other pairs are differently di-iril.ui. !.
I Inn i- in \ IT -uch complete polarization as i- sin >wn in ^ >m< . >t
tin tii'iin-- of ( ,n'i;oire Cio), the Schreiner- '"i 'n

and oilu-i^. The occurrence of various stage- of p.u\i-\ nap-is
in the -.inn- cyst, the substantial agreement in -i/e of the chn.-
m.itin thn-.id- through the stages indicated in I -L-. 2; 2'>, and
the -pirenir of double size in the next stage, pn-elude the p"
bilit\ of iiiirrpreiing the paired condition seen in I i--. 25 and 2<>
as a longitudinal split. Homologous chronu in some way

("in. together, anil gradually twist up into a tighter and ti^hii T
ro|u--like -Hand. A casual comparison of cysts in th'
sho\\ n in I i-. j; with the earlier stages (Figs. 2,^ to ji.i, u-ing low
po\\er- of the microscope, simply gives the impression that here
\\e ha\e a -pitvme, or sections of a -pireine, twice as thick a^ in
the |n-\ious stages, but study of such nuclei with /ei-- 1.5 nun.
and oc. 12 n-veals the double and twisted condition of the strands

Ei Anaphasc showing segregation of homologous chromosomes and

Ion.nitmlin.il splitting.

1- 1.-. .^4-36. Second maturation initosis -howin^ diinnr|ihi-m in nunilx-r (18
and ii; 1 .tnd i-i|ii;U ional divisiiin.

226 X. M. STEVENS.

in practically every nucleus, indicating that the paired threads
are at no time so thoroughly fused as even apparently to lose their
identity. This is further indicated by the earliest prophase
stage (Fig. 28) where the paired strands begin to untwist. The
following prophase stages consist of further untwisting and longi-
tudinal contraction of the paired homologous chromosomes. In
the synapsis stage (Figs. 25 and 26) it is impossible to tell whether
threads of equal length form the pairs, but in the prophase pairs
this is perfectly evident (Figs. 29 and 30). The untwisting and
contraction frequently proceed at different rates in different
pairs in the same nucleus and in different nuclei of the same cyst,
so that one can easily compare the various stages of the process
and be perfectly sure that the untwisting is continuous. There is
no secondary fusion of paired threads such as frequently occurs
in cases where a precocious longitudinal split appears in a
telosynaptic bivalent and then closes up before the rings and
crosses are formed (see Blattilla germanica, Stevens, '05). Fig.
30 shows various stages in the formation of the definitive chromo-
somes of the first maturation mitosis from the parasynaptic
threads of a stage a little later than that of Fig. 28. The paired
chromosomes untwist and contract simultaneously. Some re-
main united at one end (b and d) while in other cases union of a
pair at one or both ends is a secondary phenomenon and may
even occur after the spindle has formed (a and c}. That there is
much variation in the form and size of the 18 bivalents in meta-
phase is shown in Figs. 3, 7, 8, 9, 30, 31, 32 and 33. The most
frequent forms are rings, E's and crosses, though one or more
pairs of straight rods may be found in nearly every spindle.
Most of the chromosomes are attached to the spindle fibers at
or near the middle of each univalent member of the pair, so that
the separated chromosomes pass to the poles of the spindle in
the form of V's (Figs. 31 to 33). In the case of the double rods
the fibers are attached at the ends. Many of the chromosomes
are partly or wholly split longitudinally in the anaphase (Fig. 33).
There are of course two kinds of second spermatocyte equatorial
plates containing 18 and 19 chromosomes respectively (Figs. 34
and 35), X appearing in the form of a large V (Fig. 35) in one
half of the cells. Division of the chromosomes is here longi-
tudinal as seen in Fig. 36.

-I IM.KM Ml KARY CHROMOSOMES IN CHUTHOPHILUS. 22

-- i

In Ceuthophilus the first maturation mitosis is therefore a
segregating division of the previously paired and united homolo-
gous univalent chromosomes, while the second mitosis is as clearly
an equational division of all of the univalent chromosomes
in< hiding A".

DISCUSSION.

In an earli< T -tudy ('05) of the spermatogenesis of t\v<> other
species of ( >nln>pirra, Blattella (Blatta] ennanicti and Stcnopcl-
nnitns (-p. I >und what seemed to be good evidence. <>t" telo-
synapsia '05, PI. II., Figs. 55, 56, 58, 59, 62, 63, 64, and PI. III.,
to 115). That material I have reviewed and compared
with th< ' ''kilns preparations, and I find no such evidence

it" parasynapsis in fit her of them. Naturally I expected to find
tel"-v n.tp-i-, in Ceuthophilus, and was surprised on working back-
ward t'n .in the maturation mitoses to find no evidence of telo-
synapsis outside of the late prophases, and abundant evidence of
parasynapsis in the young spermatocytes at a st.'.-^c \\lure s\n-
i/e-i- i- frequently found in other material.

A ivc.-nt review of the literature on conjugation of chromo-
somes ha> only strengthened my previous conviction, based on
in\ o\\ n i \prricnce with the spermatogenesis of a variety of
forms that tin- phenomenon is one which vari- aly in

ditii n-nt L;roiii of organisms, and even in dinVn-nt species of
the ^anic - nus, or different sexes of the same '//a,

Stevens '03, '05; Btifo, King '07, '08). Indeed I should not be
Mirpri>-l it the range of variation should prove to i-\t.-nd troni
. ases \\lu-re there is nothing that could be called conjugation,
but nu-n-ly such a pairing, without contact even, as will secure
ition of homologous maternal and patrmul chromosomes
to different daughter cells, through (b) an intermediate condition
of telosynapsis and less intimate parasynapsis, to (c) cases where
hi Min )1< >;^ ii c hromosomes are so completely fused in parasynapsis
that it is impossible to tell whether the resulting chromosomes
\\ hich are segregated in mitosis are identical with those that went
into synap>is or not; and the variation may extend to cases
which may give further support to Janssens' chiasma theory
('09) or to Morgan's modification of it t'i O in which homologous
chromosomes are supposed to be twisted tightly together in

228 N. M. STEVENS.

parasynapsis and split across the twists in preparation for mitosis,
giving daughter chromsomes which contain both maternal and
paternal chromatin.

In Ceuthophilus the parasynapsis stage of Fig. 27 is intimate
enough and long enough to favor the supposition that it is a
true conjugation involving exchange of material particles or of
chemical substances (genes), but there is no evidence of any
splitting of Morgan's chiasma type. All of the evidence indicates
that homologous paternal and maternal chromosomes twist
together in parasynapsis and untwist in the prophase of the
first maturation mitosis. In the flies and mosquitoes (Stevens,
'08, '10, f n) we have examples of even more pronounced para-
synapsis than in Ceuthophilus, but so far as I have seen, the
indications are that the chromosome pairs twist up in synapsis
and untwist in prophase much as in Ceuthophilus; i. e., an op-
portunity for interchange of genes between homologous maternal
and paternal chromosomes is furnished by the observed phe-
nomena of parasynapsis in these forms, but no evidence of such a
chiasma type of splitting after synapsis as is suggested by Morgan
('IT) to account for the results of his breeding experiments with
Drosophila. Such an exchange of parts of chromosomes as that
described by Janssens ('09) might of course occur without being
detected, at almost any point in the process of twisting or un-
twisting of the pairs, since the time element is not determinate
in fixed preparations.

Moreover, it seems to me that, in view of the great range of
variation in the phenomena of conjugation and segregation of the
chromosome in the maturation of germ cells, cytological evidence
from one form cannot safely be taken to serve as the basis of a
theory or hypothesis to account for the experimental results on
another form, but cytological and experimental work on the same
form must go hand in hand, in order that any safe conclusions
may be drawn from the results.

There seems to be no question but that synapsis, or conjugation
of the chromosomes is the most difficult phenomenon connected
with the maturation of the germ cells, to interpret correctly, and
doubtless earlier parasynaptic stages have been overlooked 'in
some cases where telosynapsis alone has been described in con-

-I IMKNUMKK \KY CHROMOSOMES IN CEl I Hnl'HILUS.

nection with the mitotic stages ot maturation, but it seems to me
quite- unlikely that synapsis in all organisms follows one method;
and, moreover, I believe that the variations in method of synapsis
and intimacy of union of homologous chromosomes in different
forms will be found to he directly connected with variations in
method- i,f inherit. tin e >t" unit characters, especially in relation
to inter, li.m..- or Lick of interchange of maternal and paternal
genes. If thi- i- true \\e should expect to lind more cases of

>f unit characters where telosynapsis or no
real svnap urs. If parasynapsis is an adaptation to secure

in!- T< i hould expect to find cas< < ot tclvnap-

sis folloucd IA parasynapsis, as indicated, but nor certainly
d.-iiion-traied in the guinea-pig (Stevens, 'l I, Figs. 9, 10, 11
In my -nnli< - on spermatogenesis of the Coleoptera 1*05, '06,
. I found evidence of telosynapsis in several cases and no
c\ iilt-iK of para^\ nap^is, but this was only an incidental mat ter
at i In- time, an. I . ; interest merely in relation to the s< 'ion

oi uliolr i In - in the maturation mito-e<. It is ni\ in-

trntion to nine all of my Coleoptera and I Mptrra material

\\itli refer ih- (juestions whether para-vnap^U occur-

in the Coleoptera, and whether the Diptera >ho\\ .m\ evidence

ot Janssens 1 ehia-ma types of synapsis.

i I;E.

J.i : 2.

LITERATURE CITED.
Agar, W. E.

1 1 i i - -;is of Ltpidosirtn parad-

I I.
Gregoire, V.

'10 I le Maturation dans les deux K gn< . L'un H 11.- du

IM. tiquo. La Cellule, XXVI.

Janssens, F. A.
'09 >|ii-iin. dans les Batrachiens. 1

ictations des cindses de inatuiatimi. La C.-llulc, XX\ .
King, H. D.

'07 Ih<- >|..! 11.. i!. 'genesis of Bufo . Am. Jmirn. of Anat., \"1I.

'08 I'ln- ' - of Bufo lentiginosus. Journ. of M>rpli., XIX.

Morgan, T. H.

'n An Attempt to Analyze the Constitution of the Chromosomes on the H.

ni Si-x-limited Inheritance in Drosophilti. Journ. Exp. Zool., XI.
Schreiner, A. and K. E.

'04 Dif Reiiiin.u-teilungen bei den Wirln Itinon. Ein Beitrag zur l-'iai;.' nach

.ktiiin. Anat. An/.. XX1\".

230 N. M. STEVENS.

Stevens, N. M.

'03 On the Oogenesis and Spermatogenesis of Sagitta. Zool. Jahrb., XVIII.

'05 Further Studies on the Oogenesis of Sagitta. Ibid., XXI.

"05 Studies in Spermatogenesis, I.

'06 Studies in Spermatogenesis, II. Carnegie Inst., Pub. 36, Parts I. and II.

'08 A Study of the Germ Cells of certain Diptera with Reference to the Hetero-

chromosomes and the Phenomena of Synapsis. Journ. Exp. Zool., V.
'08 The Chromosomes in Diabrotica viltata, Diabrotica soror and Diabrolica

1 2-punctata. A Contribution to the Literature on Heterochromosomes and

Sex Determination. Ibid.

'09 Further Studies on the Chromosomes of the Coleoptera. Ibid., VI.
'10 The Chromosomes in the Germ Cells of Culex. Ibid., VIII.
'n Further Studies on Heterochromosomes in Mosquitoes. Biol. Bull., XX.
'n Heterochromosomes in the Guinea-pigs. Ibid., XXI.
'12 Further Observations on Supernumerary Chromosomes and Sex Ratios

in Diabrotica soror. Biol. Bull.
Wilson, E. B.

'09 Studies on Chromosomes. V. The Chromosomes of Metapodius, a Con-
tribution to the Hypothesis of the Genetic Continuity of the Chromosomes.

Journ. Exp. Zool., VL
'10 Studies on Chromosomes. VI. A New Type of Chromosome Combination

in Metapodius. Ibid., IX.

I I Kllll.k IM-:KVATIONS ON St'PERM MF.RARY

( IIK< M< >-< >MES. AND SEX RATIOS IN

MABROTICA SOROR.

X. M. STEVENS.
^i Tl KM MKKARY CHROMOSOMES.

lii iln- -iiniiiHT iif 1910 while I was enjoying the privile.
.nnl ho-pii,ility uf ihf Marine Biological Laboratory at I. a Julia.
< 'alifornia. I \<>k advantage of the opportunity to study the
malt 'o-rin cell- of Diabrotica soror from a new locality, Ilaxini;
pn-\ iou-lv >und suj>ernumerary chromosomes \ar\in^

in uuiiil.iT fi .in one to five in about 50 per cent, of the male
individuals of Diabrotica soror at Mountain View. < 'aliti nia.
and /'. a 12-punctata at Bryn Mawr, I'a., I \\a inten --led

\\ln-ther sujK-rnuineraries would be found in ilu- >aine
pr-'piinimi in a third locality.

Pestes IK-HI a hundred individuals were studii-d in aceto-
(aimiiH pn paratimi-. The greater part of the mati-rial \\a-
ciillrct-d in a corn-field in the open country between 1. a Julia
and tin new laboratory which is two miles north t tin- town.
A h u \\in- "litainrd from a rose-garden in La Joll.i .iiid one lot
o| os m.,1, -, .mil females was collecteil for me \>\ Mi-- M\nU-
|<'liii^<ni on (..MI in a gardi-n at National < itv, ju^t -omh of
San l>i(^n. lndi\ idual records were kept for earh loi. Imt the

i'iiditi.iii- \\iih respect to number of supernumeraries pn\ed to
l-e aliout the same for the three collecting Around-..

To m\ .-nrpii-e I found supernumeraries scarce. In the tir-i
J5 male- examinrd, Ji had no supernumeraries and 4 one; while
out <.f tin- lu-t 25 examined the same >ummer at Mountain
\'ie\\ 15 had no supernumerary, 7 one and ; two; and in the
tir-t 25 at Mountain View in 1909, there \\.-n i ^ with no super-
numerary, ') with one, 2 with two and i \\iih three. In the
I. a Jolla material the 89th individual was n-.irhed before a
of two supernumeraries was met with, and in the first 100
male> ;() had no supernumerary, 20 one, and I two. The follow-
er

232

X. M. STEVENS.

ing table shows the per cent, of supernumeraries in the two
species different years and in different localities.

Number of Supernumeraries.

o

i

2

3

4

S

D.

r>.

sor., Mt. V., '07, June 23-Aug. 7.
12 p., B-M., '07, Oct. 3-9

51
48

35

-17

II
I ?

2
-i

I
I

D.
D.
D.
D.

s., Mt. V., '09, July lO-Aug. 12. .
s., Mt. V., '09, Aug. 2i-Sept. 15.
s., Mt. V., '10, July 28-Sept. i.. ,
s.. La J., '10, June 17-July 4. ...

43
46
52
79

44
38
29
20

10
IO

16
I

3

4
3

2

As I had never seen any signs of degeneration of the super-
numeraries, the natural interpretation of their infrequency at
La Jolla would seem to be either that they had originated here
more recently, or that they had originally appeared in fewer
individuals in this locality.

The behavior of supernumeraries in all cases where they have
been shown to occur at once classes them with the hetero-
chromosomes, and in Metapodius Wilson ('09) has shown that
they have probably originated in an irregular second maturation
mitosis in which both idiochromosomes went to one pole of the
spindle instead of separating. He therefore regards the super-
numeraries in Metapodius as duplicates of the smaller idiochro-
mosome. In 1908 I suggested that there might be two varieties of
Diabrotica soror and also of D. 12-punctala, one having only the
odd heterochromosome and the other an unequal pair, and that
hybridization might have given rise to the supernumeraries
with their peculiar behavior, dividing sometimes in one some-
times in the other maturation mitosis. I have, however, been
able to find no evidence in favor of this view. In 1910 I studied
carefully the testesof many individuals where no supernumeraries
were present, seeking some clue to the origin of these chromo-
somes.

As a rule the odd chromosome X appears near one pole of the
spindle in the metaphase of the first maturation mitosis, but I
had always noticed that occasionally X is in or near the equa-
torial plate, and in some individuals this is quite common. At
La Jolla I found two spindles in which X was between the
daughter plates in the anaphase, and stretched out lengthwise
(Figs, i and 2). In one of these cases (Fig. 2) X was split so

-I I'l.KMMI KARV CHROMOSOMES IN 1MAHK'I1> 233

f

f

3

*.

ntt

8

B

9

1 ; An.iphases of first maturation mit~i-. -Imwin^ al>mniiial po-ition

ami tian-vfi-i- .livision of X. (Mag. 1,500 tW all ti^iin

I- i -i.l /'. M -c of first inatuiatiiui init"-i-. -li"\\ini; t\\" -upi-rnu-

;ii' . iii.il in size.

Fi< utlxT stages from same tr-ii- -ln.\sin.i; l-lia\iii m A", ami the

Hipernui

234 N - M - STEVENS.

that it was certain that it was in a position such that it might
divide transversely, but I was not able to find any cases of actual
transverse division of X. Later at Mountain View I did find
two anaphases where X appeared to have divided transversely
and unequally (Figs. 3 and 5) and one in which X was caught in
the cell plate between the daughter cells (Fig. 4). Now the
supernumeraries are usually very uniform in size and certainly
less than one half the sixe of X. I have one individual noted as
having an unusually large supernumerary, about one-half as
large as X, and a few cases where an unusually small one occurs.
One of the latter cases is shown in '08, PI. III., Figs. 76 to 78.
From the evidence now at hand I should infer that the probable
origin of the supernumeraries in the Diabroticas has been an
occasional transverse division of X followed by a longitudinal
division of the two parts. Evidently the transverse division has
usually been an equal one, but that it may be unequal is show r n by
Figs. 3 and 5, and the rather rare occurrence of unusually large
and unusually small supernumeraries. Figs. 6 to 10 are from a
male captured at Mountain View, July 29, 1910. Here we
have a large and a small supernumerary in the same individual.
In the metaphase (6a and 6b) X and the two supernumeraries
were all near one pole of the spindle, while in Figs. 8 and 9 the
supernumeraries are at opposite poles and in Fig. 9, X is near
the equatorial plate. In Fig. 10, X and both supernumeraries
have gone undivided to one second spermatocyte. No cases of
the division of either supernumerary in the first maturation
mitosis were found in this individual.

In Metapodius Wilson found no somatic variations correspond-
ing to the variation in the number of supernumeraries. In fact
the insects with X alone, X and Y, or X, Y and I to 6 super-
numeraries are described as indistinguishable. These speciec of
Diabrotica are very variable in size, and in regard to size and
fusion of the 12 black spots on the elytra, but as I showed in 1908
there is no significant correlation between these somatic varia-
tions and the presence or absence or number of supernumeraries
('08, Tables I. and II., and p. 465, text). In Metapodius the
indications are that the chromosome Y is of no hereditary value,
and the supernumeraries, as duplicates of Y would not be ex-

SUPERNUMERARY CHROMOSOMES IN DIABROTICA.

pected to affect the somatic characteristics of the insects. If,
however, i he supernumeraries of Diabrotica come originally from
different regi< >ns - >f A', there would seem to be no reason why they
should not bear functional genes for sex and other characters.
The in. ile alv. 'in. tins A" so far as my experience goes (over

700 male- , but one would suppose, if the supernumeraries are
functional in heredity, that one A' and a supernumerary might
frequently deii-niiim- the development of a female, and if so
there -hould be males without A', but with a supernumerary
in its place, h may. of course, be true that the abnormal di\ i-i-ui
V pp-dii- in. -up. inumeraries in itself indicates a de-.-iu-rati-
or non-functional condition of that particular A' chromosome,
and ther. its progeny the resulting stipernuinera<

This \\uiild tall in line with Schleip's ('ll) s ' :n in regard

i" the rejected A" chromosome in the spcrmato-em -sis of the
lieniiaphiodi ration of .-1 n^iostomum nigru .. that it

had ahead\ bec< une non-functional at an earlier stage, \\ hence its
later beha\ior. It is exceedingly desirable that the female sex
elU of il <>ticas should be studied, but I ha\e not been

able i ni\' favorable mitoses in the adults, or to secure

lar\ .e oi pupa- I r oin the soil or roots of plants on u hit h they live.
N-\eral attempts to breed them have given no iv-uli-,.

Tin u males each having one supernumerary \\ere

studied from the point of view of tin- division of the super-
numerarie- in the first maturation mitosis. All anaph.tses and
metapha-es in each preparation were examined and all cases
\\lure it \\.is possible to determine the position and beha\ior of
the sii].ernnmerar\' reconled. In the metapha-i- the super-
numerary \\as in the equatorial plate in 5} i p< r cent, of <oi
cases and out of the j)late nearer one pole of the spindle in

45.0 per cent. Apparently the supernumeraries, \\lnn they
di\ide. do ~o later than the bivalent chromosome-., so all ana-
phases \\ere examined on this point. In 55.6 per cent, of the
anapha-es found in the 12 testes, the supernumerary was ],e-
t \\een the daughter plates, and in 44 percent, it was di\ filing or
di\ filed. Here the 56.6 per cent, corresponds closely with the

54.1 per cent, in the equatorial plato in the metaphase, and the
44.4 per cent, outside of the daughter plates in anaph.ises comes

236

X. M. STKYKXS.

very near the 45.9 per cent, out of the equatorial plates in
metaphase. The division of supernumeraries or their failure to
divide in the first maturation mitosis seems to be a matter of
chance, depending on their position in the spindle in the prophase
and on the attachment of spindle fibers from one or from both
poles of the spindle. In Fig. n both supernumeraries are con-
nected by fibers with both poles, in Fig. 12 the 5-chromosome is
connected with both poles and is about to divide, and in Fig. 13
one 5-chromosome is connected with both poles and will later
divide, while the other will go undivided to the upper pole of the
spindle and therefore to one second spermatocyte. The behavior
of the other chromosomes indicates a more or less definite attach-
ment point for the spindle fibers, near the middle of the chromo-

11

12

FIG. ii. Spindle showing two supernumeraries (s), each attached to spindle
fibers from both poles. Mag. 2,000.

FIG. 12. Anaphase showing supernumerary (s) about to divide.

FIG. 13. Metaphase showing X, a supernumerary (5) attached to one spindle
fiber, and another (5) attached to two. m = mitochondria.

some in both spermatogonial and spermatocyte mitosis (Figs,
ii and 13). The supernumeraries seem to be able to make
connections with both poles in most cases if they are in or near
the equatorial plate in late prophase stages.

SEX RATIOS.

The sex ratios in Diabrotica soror and Diabrotica 12-pitnctata
have shown very peculiar variations. In studying the male
germ cells of D. soror in 1907 I made no note of the number of
females found in random collections, but in dissecting D. 12-

SUPERNUMERARY CHROMOSOMES IN DIABROTICA.

punctata in October, 1907, I found more than two males to one
female, in one lot 58 males to 25 females. In 1909 the number
of males and females was noted for each lot dissected. Between
July lo and August 12, 107 males and 102 females were counted
in random collections from two neighboring gardens, but it was
noticed that the latios in the two garden- were quite different.
In garden A there were 58 males to 26 females; in garden B,
4<> males to 76 females. A second lot from garden A collected
betueeii August 21 and September 15 gave 101 male- to 24
female-. The percentage of females in garden .1. first lot. was
30.9, -ecnnd lot 19.2, average 23.9, and for garden B 60.8.
At I ..i Jll.i in 1910 the ratios ran more evenly.

cf 9

la, June 17 and 18

ioii.il < iiy. June 22

I a Julia. June 28 14

l"ll.i. July i i;

I ..i J"ll.i. July 4

ig

i j i

Ai Moiini.iin View again the ratios were peculiar. I i\-
rand' mi < - -llertions in Garden A gave loo male- to jo tnnale-, and
i\\o oilier later collections 76 males to 18 female-. ( >nl\ a few
\\i M . ollrcted from garden B giving 12 male-, to d female-.. In
i<n i mixed lots from both gardens gave more male- than female-.
(I 147. These were recorded incidentally while fixing a lot of
testes lor sections. By referring to the table on p.rj< 232, it will
be seen that the numbei of supernumeraries run- about the -aim-
for the lir-t 100 in 1909, about one half of which came from each
warden .1 51 and B 49), and for the second ion, .ill of which
came from garden .-1. It therefore seems unlikely that the -nper-
mimcrarie- have anything to do with the dineivn. . in sex ratio-
iu the t\\o gardens. The soil in garden .1 is harder in -nmmer
more adobe in it and less thoroughly cultivated than B. T\\<-
possibilities are suggested in this connection: (a) The male- may
be more successful in pupating and e-c.iping from the hard -oil
than the females or (b) few of either sex may emerge from the
hard -nil in garden A, and the males may be better livers and
so come in larger numbers from other neighboring gardens. The
latter i- re-aided as more probable.

238 N. M. STEVENS.

The Bryn Mawr Diabroticas of 1907 were all collected on a
large clump of golden rod in a pasture that had not been culti-
vated for many years, and they may have come out of the ground
in the immediate neighborhood or from more recently cultivated
fields near by.

These erratic sex ratios are probably merely another example
of the interference of external conditions in what would otherwise
be an equality of sexes, or in other words a shifting of normally
equal sex ratios, or partial exclusion of one sex by peculiarities
in the environment. The collections were all random in the
sense that all the individuals that could be found were collected

each time.

BRYN MAWR COLLEGE,
January 3, 1912.

LITERATURE CITED.
Schleip, W.

'n Uber die Chromatinverhaltnisse bei Angiosiomum (Rhabdonema) nigro-

venosiim. Ber. d. Naturf. Gesell. Freiberg i/B, XIX.

'n Das Verhalten des Chromatins bei Angiosiomum (Rliabdom-ma) nigro-
venosum. Ein Beitrag zur Kenntnis der Beziehungen zwischen Chromatin
und Geschlechtsbestimmung. Arch. f. Zellforsch., VII.
Stevens, N. M.

"08 The Chromosomes in Diabrotica vitlala, Diabrotica soror and Diabrotica
12-punctata. A Contribution to the Literature of Heterochromosomes and
Sex Determination. Journ. Exp. Zool., \ .
Wilson, E. B.

'09 Studies on Chromosomes, V. The Chromosomes of Melapodhis, a Con-
tribution to the Hypothesis of the Genetic Continuity of Chromosomes.
Journ. Exp. Zool., VI.

THK RELATION OF THE FIRST CLEAVAGE PLANE
To I Hi; ENTRANCE POINT OF THE SPERM.*

ERNEST E. JIM

I hiring the summer of IQII at tin.- Marine Biological Labora-
tory under the direction of ProK r Frank R. Lillie, I was cii-

''! in tlu- study of the eggs of AY/r/\ of certain cvtolo-ical

problem- tin- results of which will appear later. The question

ot the relation of the entrance-point of the ^perm and the tir-t

plane occurred to me. A verv pretty method made

-ible in .t satisfactory- fashion the determination of this relation
the result- ol which this paper embodies. I here take this oppor-
tunity to express my thanks and sense of gratitude to l'n>te !
Lillie for hi- ins[>iring interest in the work of which this is a part.

MATERIAL AND M i i in >i>-.

Tin of Xereis when shed are irregular i n -h.ipe din- i<>

I ire nre \\hile in the body of the female. They soon till oni in
the >ea \\ater, measuring about IOO /z eqnatoriallv and -oine\\ hat
less in a polar direction. There is, ln\\e\er. i dial of

indi\ idnal ~i/e variation in the j i\ en female Tli

are almo-i transparent, colored a pale ^reen liy numerous deuto-
plasin spherules distributed throughout the endoplasm; around
the eijuator i- .m irregular double -irdle of 14 to jj i.il drops
I ig i hi polar view the lar-e -erininal ve-icle appears to be
in die center of ih. h i-. lio\\ t -\cr. s|j-htl\- i-lnn^ated in

the pol.ir diri-ciion. The pnlaritx' of the o\ is, therefore,

expressed by tin- polar tlatteninu already mentioned, the position
of tin- oil dro|>. and the form of the nucleus.

A- ha> been >ln'wn (^Lillie, 'in there are not two membranes
in the unfertili/ed egg of Nereis, but rather a single vitellinc
membrane external to the radially striated cortical layer ("zona
radiata," \\ilson) of the egg. The ov<>cvte remains thus with

*A11 ilrawin.ys. oi living t-v;u~. made with the aid of a camera luci'la.

240

ERNEST E. JUST.

nucleus intact until inseminated or otherwise stimulated as for
instance, by squirting forcibly through a pipette.

Two or three minutes after insemination, a jelly is rhythmically
extruded from the cortical protoplasm. In ten minutes the
germinal vesicle breaks down, development is initiated.

Males and females captured in the evening while swimming at
the surface of Eel Pond were kept in separate dishes until morning
when they were transferred to fresh clean sea-water. To get an
abundance of eggs and of sperm for an experiment, it was merely
necessary to cut open a female and a male. The cut animals

FIG. i. Egg of Nereis at time of insemination; polar view.
FIG. 2. Maturation stage; second membrane formed; oil drops at vegetative
pole.

were removed from the dishes at once; moreover, every other
precaution was taken to avoid abnormalities superinduced
through toxic influences, mechanical shock, etc. In several
watch glasses of sea water in which India ink had been ground
up eggs were put together with a minute quantity of sea water
containing very few spermatozoa. The time of insemination \\ .1^
noted and the numbered dishes observed to the second cleavage.
This method was varied somewhat as I shall later note.

Kinged slides also were used; eggs placed on these in SIM water
and ink were inseminated. Sometimes a cover slip was placed
on the eggs. Finally, for the later observations a very few eggs
were placed on slides and the cover slips supported with glass rods.

CLKAV.U1K PLANK AM) ENTRANCE POINT OF SPERM. 24!

OBSERVATION-.
Outline of Development to First Clc<:

Eggs in sea-water in which India ink has been previously
ground up show clearly the formation of the jelly, the formation
of the fertilixation cone, and the entrance of the spermatozoon.

A -in-.ik i if ink points like a dagger or an exclamation point to
the entrance cone above which on tin- membrane the spermato-
/oon is .in. iched (Fig. 4). This "exclamation point" i- an aid
quickly to drtennine in a large number of eggs the relation of
the -perm entrance-point. The ^perma to/nun enter- the e^ .n
any point . ->. . also Lillie. 'i I .

FIG. |. Firel clea>

I 1 !'' i. l "in- and iii'lu.it
line Hi.uk- li"iiii' My.

t"iiiiati"n. 15 ininutc~ .itu-r iii-rminut i"ii. ()utci

This ink "exclamation p.-int." or ">perm indicator" as I shall
call it. i- a very in ten-Mil):; and striking formation worthy of
detailed smdv. With me. lunvever. the interest lay not SO much
in the lonnation of this indicator as in it- a\ ailability to help
ans\\cr the (itie-iion: \\hat is the relation of the sperm entrance

242 ERNEST E. JUST.

point to the first cleavage plane? I here, therefore, give only
as much of an outline of its formation and of the development of
the egg to the time of first cleavage as will suffice to render
intelligible the subsequent record of observations.

Almost at the moment the spermatozoon touches the egg
membrane, the contents of the cortical layer begin to flow out
as a viscid transparent substance of the same refractive index
as water, leaving only radiating lines across the space (perivitel-
line space) between protoplasm and membrane which represent
the walls of the emptied alveoli. This jelly in its flow carries
the ink from the periphery of the egg so that between each egg
and the surrounding ink is a clear space. This outflow of jelly
may last for fifteen minutes. The jelly forms about the egg a
layer everywhere continuous except along the tail of the sperm
which thus forms a canal that increases in length as the jelly
area widens.

Below the spermatozoon, the protoplasm of the egg begins to
form a cone at thirteen to fifteen minutes after insemination
which gradually increases in height until it reaches the membrane
and then slowly retrogresses. With this retrogression, the mem-
brane at this point sinks; in this depression lies the sperm.
During this behavior, as the jelly area widens, the canal in the
jelly in which the tail of the sperm lies fills in with particles of
ink. This process is a gradual one, the indicator reaching its
maximum of development fifteen to twenty minutes after in-
semination. The indicator, therefore, is formed along the tail
of the sperm and points to the entrance-point of the sperm.

Twenty minutes after insemination, the spermatozoon may be
seen attached to the membrane at the end of the indicator. The
perivitelline space now becomes slight. The egg "assumes an
amoeboid appearance" (Wilson), changing its shape and becoming
very irregular. The sperm cannot be seen readily (Fig. 5).
About forty minutes after insemination the egg becomes spherical
again. The sperm is easily visible on the membrane which is
more widely separated from the protoplasm by the perivitelline
space.

This condition is of short duration for the egg begins another
series of changes. The membrane appears everywhere equi-

CLEAVAGE PLANE AND ENTRANCE POINT OF SPERM. 2-J.^

distant from the egg except at the point of sperm attachment
where it is nearer the membrane. Then gradually to the right
and left of the point of sperm attachment the perivitelline space
becomes greater; the egg elongates along a line through the
point of sperm attachment (Fig. 6). \Yith the disappearance of
the sperm head within the egg (about fifty minutes after in-

5

l-i-.. 5. Alter retraction of cone; membraru -itly

in tin- I'Ki;.

1-n. ' 1 wo minutes before sperm is

this elongated appearance is lo- 7 : the egg

out. The egg flattens at tin.- animal ]><>|r i Fig. 8) and
the polar bodies are given off from a clear apparently yolk-!
region of the flattened pole (Fig. 2). Some little time later the
tir-a iK.ivage furrow appears and the egg is divided unequally

I ig. 3).

Tlu- t'l'-rrvations on the relation of this cleavage to the en-
trance-point of tin- spi-rm \\ill be considered under three heads
corresponding to the methods used.

^44

KRXEST E. JUST.

Watch Glass Series.

A female was opened at 9:58, a male at 10:00. In five watch
glasses of india ink ground up in sea-water eggs and sperm were
mixed at intervals of two minutes. At 10:10 a few eggs were

FIG. 7. Just after disappearance of sperm within the egg.

inseminated in the ink solution on an uncovered slide (no. 6).
About two minutes after an insemination the jelly began to form;
in fifteen minutes the sperm indicator was well developed. Eggs

FIG. 8. First polar body forming.

inseminated at 10:15 in a watch glass (no. 7) were washed at
10:30: that is, when the indicator had reached its maximum of
development.

CLEAVAGE PLANE AND ENTRANCE POINT OF SPERM. 245

The dishes (nos. I to 5) and the slide (no. 6) were examined as
the first cleavage furrow appeared. In 95 per cent, of the eggs
the first cleavage plane passed through the point of sperm
entrance (Fig. 9). Dish no. 7 showed, on the other hand, that in
only 50 per cent, of the eggs the first cle.ixa-e furrow passed
through the point of sperm entrance.

FIG. g. First cl<\i\

At 2:45 p.m. of the same day, eggs \\rtv in>eininate<l in watch
no. 8. Examination revealed that tin- fir>t Hea\a".e plane

through the point of entrance in No prr cent. ot ej
I jgs transferred from india ink ami sea-water to dean fiv-h
MM-u.iit-r twenty to thirty minute^ alter insemination >li<>\\r<l
per i nt. of first cleavages through the point of entrance.
A Mimmarv of the results of Kxperimein - I to s i-, as iollo\\>:

i > iiiM-iiiin.itr<l in watch ;. i \v.i-ln-'l. -hn\vi-il t'u-t cleavage tlin>ui;li

i-nti.uii e p< MII! . -i 5 PIT cent.

6 iiiM-iiiin.iU'd on slide glass, not wa-li>-l. -li-iwf-l tit-t thnni^h -n-

ti.iiu i- point . <>5 per cent.

\ - iii-.-niiii.ucd in watch glass. wasln-<l, ln>\vi-il tir-t cli M\ a.i;<- thn'iii;h <-ntrance
p. -int. >n p-r cent.

ii.it.i-il in \\.itrh .ula--. nut \va?>lK-i|. slmwi-il tir-t cleavage thruuKh en-
ti.in. < p..int, 80 per cent.

No i) in-cniinated in wat. h glass, ti.ui-i. rn-<l to slide, slmwc-'l tir-t rlrava.ui- through
rntr.mre point, 60 p-r o-nt.

That tin' ink i> not to\ic to the eggs and, ilietefore, does not
inhiliit cleavage I was able to prove by inseminating at the same
time t\\o di-lir- of eggs, one with ink and one without; develop-

246 ERNEST E. JUST.

ment in both went on at the same rate and in prrlVrtly normal
fashion. I concluded, therefore, that it was not necessary to
wash the eggs. Also, I found later that the eggs \\ere often too
greatly crowded and that it was hard to make counts unless the
eggs were in a single layer. A trial made with very few eggs
unwashed in four watch glasses gave the following result (actual
numbers are given) :

FIRST CLEAVAGE PLANE.

Through Not Through

Number. Entrance Point. Entrance Point.

1 8 2

2 16 4

3 10 I

4 12 3

To what extent the eggs might rotate in the jelly was yet to be
determined. It was absolutely necessary that the relation of the
indicator and the sperm entrance-point remain constant; other-
wise, the indicator would prove a very pretty but useless phe-
nomenon. Could it be possible for two spermatozoa to reach the
egg and the indicator to form along one sperm and not the other?
How would such an egg cleave? These points were next to be
determined.

I found, first, that the position of the indicator could be altered
through tilting the watch glass, for the eggs would rotate in the
jelly especially when they lay on the side. I found later that
the eggs are most liable to rotation after the sperm has dis-
appeared. This might easily prove a serious source of error.
Secondly, I demonstrated in several experiments that polyspermic
eggs are not apt to cleave. (Professor Lillie has obtained the
same results.) But with fairly dilute sperm and sea water,
polyspermy, which merely cuts down the number of cleaving
eggs, may be avoided.

In this connection it will be interesting to note the results
obtained with old eggs and sperm. On July 30 eggs from a
female captured in the evening of July 28 were used with fresh
sperm of a male captured in the evening of July 29. These eggs
proved very susceptible to polyspermy. This proved true in
other trials. These eggs if they segmented at all showed sixty
per cent, of first cleavages through the entrance-point of the

CLEAVAGE PLANE AM) ENTRANCE POINT OF SPERM. 247

sperm. In general, eggs that have stood in sea water for some
time after leaving the female, show a low per cent, of cleavages
through the entrance-point. Five hours after leaving the female
eggs fail to develop on insemination.

These results seem to indicate that the first cleavage tend- to
pa-s through the sperm entrance point i. e., through the point at
the end of the indicator if the e--- lie fresh, undisturbed ami
fertilized with a single sperm. Why then do some first cleavages
fail to pass through this point? During this time a number of
experiments made by day and often at ni^hi immediately alter
the rapture- of the animals sho\\ed e eniialK the same propor-
tion-.

Ringed Slide

It was stated above, it will be remembered, that the egg tends
to lie with either pole uppermost. It. however, the eggs are
not disturbed those that settle on the side uill so remain. The
- are accessible to sperm at any point it not under pre tire
it no time in this study they were, Hie first cleavage always
rnt^ through the animal pole near the polar bodie-. <b\ioii-ly
then, the question of the relation of tin- hr-t cleavage plane to
the entrance-spot of the sperm cannot be -ettled by the 1 a\
o! iho-e eggs in which the spermato/oa enter either at the point
In-low which the polar bodies are extruded or at a -p-i i
lr< mi ihi- point.

In the next trial with very few eggs on ringed -.li.lt-, tho-e e-gs
in \\hirh the sperm indicator pointed either to the polar bodies
or to a point 180 from the polar bodir- \\ere not counted. This
trial re-ulied as follows:

FIRST ( ii \\ AI.M I'l A

Number. Entrance I'mm. MI.

1 U 4

2 4

3 20 9

()(her experiment-- with ringed -lide- -ho\\.d ab-uit the same

proportions.

l-'or fear that the rin-ed -lides were toxic owing to the vaseline
n-ed the\ \\ere abamloned and slides with cover >lips >up|)Oried

248

HRNKST E. JUST.

by glass rods as well as the open watch glass were used throughout
the next series of observations.

Slides ii4th Glass Support for Cover Slip.

Four or five eggs on a slide were watched continuously through
the first cleavage, the indicator used merely to point out quickly
the point on the membrane where the sperm was attached. Very
few sperm were used in these observations, obtained through
diluting several times the water which contained them. These
observations were repeatedly made at night and at different
times during the day. Some of the eggs failed to show the indi-
cator and to develop. In all that segmented, the first cleavage
plane passed either directly through the entrance-point of the sperm

FIG. 10. First cleavage.

or a degree or so from it, with the indicator parallel to the cleavage
furrow (Fig. 10). It is possible, as stated above, to keep the
spermatozoon in view after the amceboid stage until it disappears
within the egg. The middle piece is left without. With the aid
of the middle piece, the character of the membrane at the en-
trance point (Fig. 7), and the oil drops near, it is possible ab-
solutely to hold in view the exact spot at which the sperm was
engulfed.

At intervals of two to three minutes, seven slides with very
few eggs on each were prepared. Sperm was added and after
a minute the eggs covered and every precaution taken to avoid

CLI-.\\ \(,I PLANE AND ENTRANCK POINT OF SPERM. 249

di-turbance. In the sixty eggs counted the first cleavage furr\\
passed through the sperm entrance-point in every case. In some
cases the indicator appeared to be at right angles to the furro\v
but in all such it proved to be aboic the egg and ended in the
cleavage plane (Fig. n). This was Sunday, August 20. The
laboratory was quiet, the temperature conditions favorable.
The results of Au.uu-i 23, 24 and 2~ an- Hinilar. I "amera sketches

Fit;. 1 1. First i

\\ t ! made of these eggs. Often I a-ked .111 in\ esl igator, who did
not know the purport of the experiments, t<> make the sketches;
the indicator without doubt was above the e.^ and pointed to
the elt-avagc furrow.

DISCUSSION.

Tin- first cleavage plane usually coinride- \\iih tin- median
plane of the future animal in the IV according to Koux,

\e\\port, Pfliiger and Morgan. In the squid' I so, according

it ' \\ .n.i-e, the first cleavage plane falls in with the median plane
of the embryo. Agassiz and Whitman (,'^41 nou-d a like co-
intideiH e in the teleost egg; and \"an Beneden and Julin, Castle
'96) and C'onklin ('05) found that the first cleavage plane marks
the Ion- axis of the embryo in the a-< -idian egg. 1

:.liiiK to Harper, tin- -pt-rni .ntrr< tin- pixoin'- i-i;.c pn-vim< ti> tlic ogg's
riui.in.r into the oviduct. Il- In-li.-vc-; that th.- -pi-mi iiiu-t <-iuer as soon as the
,li~r i~ i-\p, .-<-,! through rupture of the tollicular wall. In the p I..-MII

i-ntiano- i- nu-ir ir I' ized. According to his figure, th( ige plane

in.ik.- an .ui^l.- ..i 45 with the lonp axis of the embry> A- we know from oilier
j-ches, the Iniiv; axis "I the embryo is similarly placed in tlie egg.

25O ERNEST E. JUST.

But there are other eggs in which the future median plane does
not fall in the plane of the first cleavage. In Xereis (Wilson,
'92) the second cleavage plane, although it does not divide the
animal into "equal halves," coincides with the long axis. So in
Crepidjila, the first cleavage plane is at right angles to the future
median plane (Conklin, '97). In the newt (Jordan, '93) the
case is the same. In Chatopterus (Lillie, '06) the axis of the
first cleavage spindle lies in the longitudinal axis of the embryo.

There is a third group of eggs in which coincidence \vith any
cleavage plane is wanting This is true of the egg of Amia (Whit-
man and Eycleshymer, '97), of the toadfish (Clapp, '91), and of
certain amphibians (Jordan and Eycleshymer, '94), to name a
few. And yet in most of these eggs the symmetry and the
bilaterality of the cleavage may be sharply marked.

In the frog's egg the first cleavage plane usually and the median
plane of the embryo always (Ran a fused) pass through the en-
trance point of the sperm (Roux, '85; Schulze, '99; Brachet).

In the egg of Toxopneustes (Wilson, '95) the first cleavage plane
passes through the entrance-point of the sperm, "in the great
majority of cases, at least." This plane of cleavage coincides
with the transverse diameter of the embryo (Driesch).

In the ascidian egg, the belief of Castle ('96) is that the first
cleavage plane cuts through the entrance-point of the sperm.
Conklin ('05) says that there is no question but that the first
cleavage plane is through the copulation path of the germ nuclei.
And indeed his figures show very beautifully that this is actually
the case.

If now we grant that in the egg of the frog and of Toxopneustes
as in the egg of Nereis and of the ascidian the first cleavage plane is
determined by the copulation-path, or the entrance-point, of the
sperm we have this interesting conclusion : The first cleavage plane
in eggs whose cleavages have different values and different rela-
tions to the future long axes of the embryos is determined by the
entrance of the sperm. While the sperm entrance determines
the first cleavage, the first cleavage does not in all of these forms
coincide with the median plane of the future animal.

Since in the egg of Nereis the sperm may enter at any point and
since the first cleavage plane passes through this point, the struc-

CLEAVAGE PLANE AND ENTRANCE POINT OF SPERM. 25!

turc of the o\ < -\ tc of Xcreis at the time of insemination must be
the same in all meridians. This, I K-lu-ve, has an important
bearing on theories of germinal areas in tin- cytoplasm, of pre-
localization, and of precocious segregation. The determination
of bilaterality follows fertilization.

I.ITEKATrKK < I I FD.

Agassiz and Whitman, C. O.

'84 On tip I ) -. < lopment !' laui. I-'i-h Eggs I'n-limmaiy N.>'.

I': . Acad. An -.XX.

Van Beneden et Julin, J.

'84 I. a -I-UIIH nt.aion chez les A-cidii-: rapp"i; .ni-.ition de

I. live. Archive de Biologic, V.
Castle, W. E.

'96 The Early Embryology of dona Flnninv: 1 Hull. Mus.

' mp. Zool., XXVII.
Clapp, C. M.

'91 Some Points in the Development of the I ...id-Fi-h. .|.>m. M..iph.. \'.
Conklin, E. G.

'97 I lie Embryology of Crepidulti. Jmir. Mr|>li.. XIII.

"04 The Organization and Cell Lineage of 1 I l^m. A>al. x

<if Phil.. XIII.
Harper. E. H.
'04 r*he Fertilization and Early Development of thi r i Am..I"iii.

it., III.
Jordan, E. O.

'93 I ![ II. i 1 n>; and Development <>t t J"iu. M-irpli.. \'1II.

Jordan, E. O., and Eycleshymer, A. C.

'94 The Cleavage of Amphibian Ova.
Lillie, F. R.

'06 Ol >ns and Experiments com c-niinv; tin- Kli-iiinituiy I'lii-nnim-n

Development in Chu l-m I \ / .1.. III.

"ii ^! idii "i Fertilization in A I rtical ( h.mui-- in ili> 1

II. I Fertilization. Jour. M.'ipli.. XXII.

Moszowski. Max.

'02 i i Einfluss der SC!I\M ; luinu uml Eih.iltu

bilateralen Symmetric des F h. Mik. An.it.. IX.

Newport, G.

'51. '53. '54 *^ n lnt -' Iinpri'Kiuiti'Mi ni the ( >vum in tin- Amphibia. Phil.

l\ -.>c. London.
Roux, W.

'85 Britriige zur Enlwickelu: .uiik dr- Embr\n. Nr. {. 1 ebei lic

.innuing der Hauptrichtungen -I- I ; licnibryn im Ki uml ulit-r die
erste lli.-il-,. i roschei<. !'.: -lauor iirztl. V-it-> hr.

'87 Nr. .}. I 1 iiimiinii ilT Mrdianebene des Froschembryo durch de

pulatinii-iii-htuiiK di_< Eikcrm-s und ilt-s Sp.-i inakn in -. Anhiv Mikr.
An.it.. XXIX

252 ERNEST E. JUST.

Schultze, O.

'99 Ueber das erste Auftreten cler bilateralen Symmetric im Yerlauf der Ent-

wicklung. Archiv Mik. Anat., LV.
Watase, S.

'90 Studies on Cephalopods. I. Cleavage of the Ovum. Jour. Murph., IV.
Whitman, C. O., and Eycleshymer, A. C.

'97 The Egg of Amia and its Cleavage. Jour. Morph., XIII.
Wilson, E. B.

'92 The Cell Lineage of Nereis. Jour. Morph., VI.
Wilson, E. B., and Mathews, A. P.

'95 Maturation, Fertilization, and Polarity of the Echinoderm Egg. Jour.
Morph., X.

PALMEN'S ORGAN AND ITS FUNCTION IN NYMPHS

OF THE EPHEMERID/E, HEPTA'.IAIA INTER-

PUNCTATA (SAY) AND ECDYURUS MACULI-

PENNIS ( WALSH K

J. E. \VODSEDALEK.
INTRODUCTORY RKMARKS.

Our knowledge concerning the tracheal s\ ~u in in tin- Ephem-
eii-I.e .I. acs baclc to the time of Swammerdamm (1752 , luit the
existence of this interesting modification, Palmcn's or-an, found
only in the tracheal system of this group of insects, was not
knoun until comparatively recent times. Swammerdamm in his
"I'.iUI der N.itur" gives a large figure (Plate XIV. . -howin^ in
ome detail the internal anatomy of a may-tly nymph. Inn the
I'.ilmeii's organ and even the four tracheal tulu -. diivnlv leading
to it , if | in-sent in that species, apparently e>eaped his ob-en aiion.
Thi-> oini-sioii was no doubt due to an imperfeet di i < -lion ; for,
upon do-ely observing his representation of the air tul.es in the
he.id of the nymph he figures, one can detect .1 nnje projeetioii
le.idiiu fn>m the main tracheal tube on the left, \vhieh eorre-ponds
Mnne\\ h.it to one of the four tubes normally leading to \\\\< -n-.m ;
the other three tubes and the organ itself \\cre doubtle^> de-
Btroyed in his preparation, and hence not represented in his
figure.

The presence of this chitinous structure \\.i- tir>t noted In
1 >r. J. A. I'.ilmen (1877), after whom the or^an is named, and in
his work he says: "Die vier im Schc-itel zusammenstossenden
Ae>te bildeii in ilirem Kreuzpunkt eincn rundliehen, aus c<m-
eentri^chen ( "hitinschichten besteheiiden Korper. dessen Medeut-
un.u iih nirht kenne." On Plate I. (Fig. 7) he gives .1 Imure of
the head and thorax of the nymph of Clocon diptcntni L., showing
tlu- loeat ion of this organ in its relation to the four traeheal tubes
of the head, without making any attempt to dexrilu- it. He
makes the statement that the tracheal >> stem is essentially the

253

254

J. E. WODSEDALEK.

same in the twenty-three species which he examined. It is not
entirely safe, however, to infer from this that the prominence of
Palmen's organ is essentially similar in these various species.

The species upon which the present study is based are Hepta-
genia interpundata and Ecdyurus maculipennis. These two forms
are very closely allied, not only in matters concerning this organ,
but also in their natural habits and general behavior, and the
present paper will concern itself with nymphs of //. inter pun data,
unless otherwise specified.

FIG. A. Head of H. inlerpunctata nymph. Basal joint of antennae only drawn.
The brain is drawn, dotted, under the three ocelli just posterior to which is shown
the Palmen's organ and the four tracheal tubes leading into it.

Fig. A shows the relative position of the organs in the head
of a nymph. Palmen's organ together with the tracheal tubes
leading to it can be readily seen through the chitinous covering,
especially in the newly moulted specimens, by placing them under
a binocular. It is symmetrically located between the two large
compound eyes and a little posterior to the brain. Fig. I shows
the organ in its relation to the entire tracheal system of the head.
It has been the fortune of the writer to be able to make a perfect
dissection of the system, the first time merely through an acci-
dent. Upon pkicing a specimen which had been dead for some
time under a binocular almost the entire tracheal system of the
insect became visible through the transparent chitinous covering.
The muscles and all the other soft tissues had sufficiently de-
composed to form a sort of liquid mass. The thin hypodermal
walls surrounding the air tubes too had disintegrated, and prac-
tically all that remained in good condition was the exoskeleton
and its internal continuation, the tracheal system. The location

PALMEN S ORGAN IN HEPTAG1-.NIA AND ECDYURUS. 255

and arrangement of the more important parts of the system were
carefully noted and a diagram indicating the relative position of
the main tubes was sketched. The external covering was care-
fully broken between the pro- and mesothorax and a gentle pull
on the anterior edge of the head removed it, fully exposing the
air tubes which remained in position. The macerated mass was
carefully washed off and the tracheal system being completely
filled with air presented the most beautiful silvery effect a-ain-t
a dark back-ground. Even the very finest branches remained,
but no attempt was made to include them in the figure.

DESCRIPTION OF PALMEN'S ORGAN.

Gross (1903) attempts to describe the organ in c<>n-i(lcr,il >!
detail; this description and his ideas in general an IK-I fully
corroborated by the results of my studies. He says in part:
" Reconstruiren wir jetzt aus den besprocheaen Schnittbildern
ganze Organ, so erhalten wir folgendes Gesammtbild, Kin
K'irper, der im Liingsschnitt kurz elliptisch, im < >IKT chnii i
ungerfahr kreisfdrmig ist, setzt sich aus 14 concentrischen, an-
xaitem Chitin bestehenclen Schalen /u-ammm, die an ilnvr
Innentluche mit feinen Hiirchcn dicht U--rut -iml. I >as Kllip-
s<>il i-t aberkein vollkommen geschlosseiie-. Yielmehr i-t es v>n
vier Seiten her [sehr] tief ausgehdhlt. Das uan/c i-t in den Kn-n/-
nn^-punkt von 4 im Scheitel des Hinterkopfs /n-ainmm'ivtti-n-
dm Tracheeniistcn eingeschaltet, und xuar so, da-.- die Luft
/\\i-elien den Schalen trei circuliren kann, \veiin aiich (lurch <li<-
grosse Xahl der Hiirchen einigermaassen behindi-rt. < ".an/ ahn-
lich i^estaltct wie bei Ephemera vulyitd L. fand irh da- I'alinm'-
selie ( )r^an noch bci Baetis rhodani Piet.. Ht'()ttr^cnia sulphured
Miill.. ferner bei den Larvcn einer Cacnis sp. un<l eiiur Chiro-
les sp. Einige geringe Abweirhnn-en in di-r Cn--talt bei
/>'<.-(V/.v rhodani konnte ich nicht hinreii IK-IK 1 ^ciiau fe-istellen,
nn -it- hier zu besprechen, da ich inir nieht lienii^md Material
1'. ~i haticii konnte."

The 1'almen's organ in both II. inter punctata and E. niacnli-
f>cnnis, i- not composed of coneentrie shells nor are there any
hair- pp sent on the inner surface of the scales which Gross
Ic-cribes and pictures in allied -prcie-. \\V11 prcparc'l -tides

256 J. E. WODSEDALEK.

of cross sections show that the organ is not perforated with air
passages but is a continuous mass of chitin in which the differ-
entiations are due mainly to variation in density of color. Sec-
tions of adult specimens weie also made and carefully examined,
but no difference in the structure of the organ could be detected.
Dr. Gross has no doubt mistaken the clearer areas or concentric
layers for air passages and the darker layers for separate solid
areas forming the scales from which the hairs protrude.

Fig. 4 shows the external dorsal appearance of the organ and
its relation to the four tracheal tubes, the entire structure being
enveloped by the hypodermal layer; Fig. 5 is a horizontal section
of the same. The description of the organ can be best understood
by studying it in connection with its development and growth.
It is a well understood fact that the tracheal system in insects is
formed by the invagination of the ectodermal layer. As to the
origin of Palmen's organ I am not at all certain for embryonic
material has thus far in this study not been available. The
appearance of the structure of the central portion of the organ
suggests that, during the process of the development of the
tracheal system, the four large tubes leading to the organ (Fig. i)
come together at a common point; here the blunt ends of the
invaginated portions, the tracheae, surrounded by the hypo-
dermis, fuse and secrete this common center. From the various
cross sections of which Fig. 8 is typical, it can be inferred that the
two posterior tubes come together first and that a portion of the
center is secreted before it is met by the two anterior tubes.

In the many sections of //. inter pnnctata and E. niacnlipennis,
which were examined, the center of the organ does not show any
ring-like structure, but is an irregular mass which is apparently
M-nvted Ix-forc the hr\.i c ists the firsl limii'j <>i its trachea]
system. At the time of this first ecdysis which is accompanied
by the shedding of the inner lining of the air tubes, this central
mass is larger than the openings in any of the four tubes and hence
the impossibility of its being cast out of the body. Shortly
after the casting of the inner lining of the trachea?, the hypo-
dermal cells surrounding the tubes undoubtedly begin to secrete
the new chitinous wall. The hypodermal layer surrounding the
central mass, the beginning of the Palmen's organ, is continuous

PALMEN S OKtiAN IN HKIMAt,! MA AND ECDYURl - 257

with the layer surrounding the air tubes and apparently begins
active secretion at about the same time. The different con-
spicuous rings which are shown ( Fig?. 5-9) are sections through
the concentric layers of the organ and are directly correlated
with the various moults. Further evidence of this correlation
is obvious from the fact that the- number of rings is directly in
proportion to the size of the insects thcin-elves. An examination
of the sections figured show- thai tin- hypodermal cell- surround-
ing the organ are much larger than those enclosing the trachea-,
and hence, the greater the secretion <>f these larger cells; I nun this
results the greater thickness of tlu- chitinous layers of the on;. in
as compared with that of their continuations, tin- walls of the
t radical tubes. Coincident with tin- increase of volume of the

in, the cells surrounding it must necessarily multiply as they
are pushed outward. Thus, by means of succe i\v periodic
BC< ret ions the Palmen's organ is built up; the old layers of the

m arc not cast off as are the walls of the tradieal -y>tem,
\\ith which they are continuous.

' .ross in commenting on the function of Palmcn'.-, or-an says:
"l>h glaube deshalb, dass fur das rath-elhafte < )r^an keine
KrlJarung gefunden werden kann oluie Beriicksichtigung de-
\ir\en. Xehmen wir aber an, dass dicker \\irklieh /.\\ dnn

in gdiort, so kann dieses nichts anden-- >ein aU t in Sim

m. Da es aber, wenn auch zicmlich direct miter der Ilypo-
di TinU von dieser nur durch wenig Feitk(")rper getrennt doch
jedeii 1 alls im Imiemdes Korpers der Thii n gelegen i-t. k.mn es
\ on alien uns %-on andern Thiergnippen bekannteii Sinne-func-
lionen nur denen eines Gleichgewichtssinnes dienen." I p to the
pn-si'iit study no experimental work on the ori;an has lui n at-
tempted with the view of obtaining evident--- ,nl- iis func-
lion. < '.ross also says: "Man konnte nieim-n. der Heueis fiir die
Kichti^keit der \ - on mir ver>uchten I'eutun^ (K-- < )r^an- \\>
sich \ idleicht durch zvveckmassig anuc-tellte X'ersudie i-rlirin^en.
1 >a- ei-M -heint mir aber ziemlich an icht>los. Es \\lire ja gewiss
nicht unmoglich, das recht oberll;ichlidi gelegene Organ zu zer-
storon, nachdem man vorher seine Lage so genau festgestellt hat,
ila-s man sie schon von an en am lelu-nden Thier angeben kann.
. \lier it h fiirchte, dieses Experiment wird nicht viel helfen. Stellt

258 J. E. WODSEDALEK.

sich nach clem operativen Eingriff irgend eine Aenderung dcr
Flugweise ein, so kann diese auch durch die Verletzung an und
fiir sich bewirkt sein. \\~ir wissen aus der experimentellen
Gehirnphysiologie der Vertebraten zur Geniige, in welche schwere
Irrthiimer man geraten kann, wenn man die Yerletzung oder
Zerstorung eines Organs oder Organtheils als reinen Versuch
betrachtet. Wahrend man aber bei einem Wirbelthier wohl
warten kann, bis die storenden Nebeneffecte des operativen
Eingriffs verschwunden sind, so scheint mir das bein einer
' Eintagsfliege ' kaum moglich zu sein. Selbst ein nicht zur
Begaltung gelangtes Exemplar diirfte in der Gefangenschaft nur
zu bald eingehen. Auch wiirden die Thiere wohl kaum den Hoch-
zeitsflug aufnehmen, wenn man sie nicht in die ihnen zusagende,
natiirliche Umgebung und unter Artgenossen bringt. Thut man
dies aber, so wiirden einem die Versuchsthiere gar zu leicht
entschliipfen, nachdem sie einmal aufgestiegen sind. Ebenso
wenig Erfolg verspreche ich mir von dem Versuch, die Function
des Organs durch Verkleben der in die Kopftracheen fiihrenden
Stigmen festzustellen."

REMOVAL OF THE ORGAN.

Experimental work on the removal of the organ did, as Gross
said, at first seem impossible. It is needless to say that the
task was very tedious and at the outset far from encouraging,
this was mainly due to the small size of the organ and its close
proximity to the brain. At first the cauterizing method was
used but without satisfactory results, then two very fine platinum
needles which were attached to the two wires leading from a
galvanic battery were employed. The apparatus was provided
with a resistance box so that the voltage could be varied at will.
In this method the end of one needle was turned into a small
loop through which the sharp point of the other was inserted,
thereby completing the current, heating the sharp point intended
for the operation, and at the same time, greatly facilitating the
necessary steady manipulation of the outfit. The hot point
of the needle would be brought directly over the organ and then
a rapid insertion and withdrawal of the point of contact followed.
It was impossible at each attempt to destroy the organ owing

PALMEN S ORGAN IX HEPTAGENIA AND ECDYURl '-. 259

to its natural instability. A few specimens from which the
organ had been thus entirely removed, lived a sufficient length
of time to enable studies of the behavior of the individuals, and
of the regeneration of some of the destroyed part -

Becoming more thoroughly familiar with tin- structure and
exact position of the organ in its relation t< the vital pans ot the
head, a more simple method wa- de\i-ed. By mean- <>f two
very fine and sharp-pointed needles a small slit can be made
through the chitin above the organ and thru, in-crtin- a needle
at each side between the posterior and anterior trarlu-al tubes
leading to the organ, it can with some pra< -tier, be entirely
removed; this treatment apparently causes tin- in-ect- but little
pain. The four tracheal tubes were usually separated near the
organ though sometimes they would break off near their juncture
with the main longitudinal trachea?. In special for -tudics

of regeneration of the organ, the four tubes were broken off
at their immediate attachment to the organ or at various drlinite
di-tances from it. This was possible by pressing the two point-
I the needles on either side of the place where the break was
desired. Bleeding was very rare and usually the edges of the
chitinous slit were brought so close together that the detection
ot the wound was rendered almost impossible.

After treatment by this method the activity of the nymphs
\\hcn placed back into the water did not set-in to lie impaired
by the operation, and the wounds healed over within a leu
days. By this method not only was the remo\ al of the organ
a mvd. but mortality was reduced to a minimum. In one set
of experiments forty-nine out of fifty specimen- operated on
li\cd for more than two months after the operation. It might
be -aid in this connection that no regeneration of the organ
takes place. The ends of the broken tube- heal over within
t\\o or three weeks and with the e\eeption of a few small air
tubes which grow out from the blunt ends of the four tubes,
during the same time, no further growth was observed in any
of the specimens as long as four months after the organs had
been removed. Fig. 3 is drawn from a nymph in which the
traehe.e were broken off at their point of contact with Palmen'-
,m, they almost touched but no regeneration of the organ

260 J. E. WODSEDALEK.

took place, nor was there a union formed between the different
tracheae Fig. 2 is of a specimen in which the trachea? were
broken at quite a distance from the organ; again, no growth
beyond the covering over of the broken ends and the formation
of a few small tubules took place.

COMPARISON OF THE BEHAVIOR OF NORMAL AND OPERATED
SPECIMENS IN RELATION TO THE FUNCTION OF
PALMEN'S ORGAN.

In my previous papers (Wodsedalek, 'n and '12), the behavior
of H, interpunctata nymphs has been discussed in considerable
detail, and hence only the more important phases of the behavior
of this insect which directly concern this problem will be cited
here. The nymphs are decidedly negative in their phototactic
response in all gradations of light, varying from ordinary day-
light to very intense electric illumination. Their thigmotactic
propensity, or tendency to come in contact with and cling to
objects, is especially pronounced. In their natural environ-
ment the nymphs are never seen swimming freely about in the
water, even when observed in their favorite places in which
they occur in great abundance. In their natural habitat they
are always found clinging to the under surfaces of small rocks,
and this same position is regularly assumed by all normal ones
in the aquaria of the laboratory. When a stone, to which the
specimens are attached is inverted in the water, the insects
soon make for its under side, many of them doing this as the
stone is being turned over. This is also true of normal specimens
in the dark-room, and hence it is obvious that this tendency of
the nymphs to cling to the lower surfaces of rocks, with their
dorsal side downward, is not due entirely to their negative reac-
tion to light. It is unquestionably due, in part, to a definite
power of orientation independent of phototaxis.

Specimens from which the Pill men's organ was removed
react to light in practically the same way as do the normal
specimens. Their thigmotactic inclinations, too, do not seem
to be impaired. However, when the insects are taken into a
very shaded or a dark-room the difference in orientation becomes
quite obvious. When a stone to which the insects are attached is

PALMEX'S ORGAN IX HEPTAGENIA AND ECDYURUS. 26l

inverted in the water, or when the specimens are dropped on a
stone in the water in a dark-room they remain on the upper
surface or on the sides of the rock for a considerably longer time
than do the normal individuals By the removal of the organ
the nymphs have no doubt lost, to some extent, their usual
keen sense of orientation, for under such conditions they would
!'< seen on the top, sides or any part of the rock for hours, days,
and even weeks after the operation had been performed. The
same was true of every lot experimented with. It was also
noticed, with several lots of operated specimen-, that the tend-
ency to remain on almost any purl of the stone was gradually
dimini-hed and that after -everal week- and in some rases- about
two months there were comparatively few individuals l( , he seen
on the upper surface, regardless of the fact that in some -pedal
experiments the stone would be in\ cried at e\ cry >b-er\ ation
with the view of bringing more specimen- to the upper surface
\\ith little disturbance. This growing partiality to the lower
-nit". ice of the stone does not lessen the significance oi their tornu-r
behavior, for, from my studies on the po\\er of the tormation
of a--odations in the nymphs of II. inter (nnntata \ \\ '< 1-edalek,
*I2) it was found that they gradually formed -e\eral i\\n>< \
a ' iaiions. The associations formed in the- rimeiu- were

in connection with their thi^motactic inclination-, which were
in ^ivat part responsible for the ^radual de. i ! the number

i on lop, and the gradual diminishing of ihe time the various
indi\ iduals required to retreat to tiie l..\\er surfa

In another paper (\Vodsedalek, 'u) on the natural history
and general behavior uf these insects I ha\e di-cu ed their
thigmotaxis in considerable detail. It was le.irned from a -imple
experiment that their thigmotactic ])ropen-ities are. be-,t -ati-lied
when their dorsal as well as their \entral -urface- are in contact
\\ith some object. "XYhen several specimen- are placed in an
aquarium they mass together into clusters where they remain for
hours, and if recently collected, even days. As soon as a rock or
any other object is placed in the water the loose forms swim toward
it, while con-iderable time often elapses before the masses are
broken up. Two long bricks were placed one over the other in a
basin of water and between them were placed small pebbles

262 J. E. WODSEDALEK.

varying in size so that the space gradually varied in thickness from
one end to the other. Then a large number of specimens were
put in the water and after a short time it was found that nearly
all of the specimens were attached to the lower surface of the
upper brick with their dorsal sides downward, and a large major-
ity of the specimens were in that part of the wedge-shaped space
where their backs came in contact with the brick below." The
operated specimens in their wandering about over the surface
of the stone accidently came into such a place where their backs
came in contact with the floor of the basin. This stimulus
naturally appealed to their thigmotactic propensity and hence
the greater tendency to remain on that portion of the rock. It
seems only natural, therefore, that an association would be
formed between this more satisfactory environment and the
lower surface of the stone. It is not altogether improbable how-
ever, that such a habit had already been partially formed before
the operation took place.

Further evidence for the fact that this thigmotaxis is largely
responsible for the gradual disappearance of the insects from the
upper surface, is apparent from the results obtained in some
checking experiments. In those experiments the stone was sus-
pended in the water so that the backs of the nymphs could not
come in contact with other objects. The results were surprising
and all remaining doubts as to the function of the Palmen's organ
in the nymphs were resolved. As long as the experiment was
continued the specimens remained quite evenly scattered over
the entire surface of the suspended stone. A similar experiment
was tried with the normal specimens, also in the dark chamber,
and practically all of the specimens remained exclusively on the
lower surface. It is only natural, then, to conclude that Pal-
men's organ has a great deal to do with the orientation of these
insects. That this unusual behavior is not due to the shock the
insects receive from the operation was proven by the fact that
when other parts of the head and body were destroyed no com-
parable results in behavior took place.

Although the foregoing results are thoroughly convincing as
to the function of the organ in these nymphs, further results of
observations on behavior relative to the role of the organ may

PALMEN'S ORGAN IN HEPTAGKNIA AND ECDYURUS. 263

be cited. When the specimens are collected and dropped into a
dish of water many of the individuals fall to the bottom with
their ventral sides upward. This toppling over is even more
obvious when the specimens are placed in a dish of water in \ir
a light. In their attempts to get away from the light and
repeatedly clawing at the opposite end of the <li-h tin- spedniens
become exhausted and very frequently when the clavvin- move-
ments cease the apparently lifeless individual* fall to the 1 >< >\[\\\,
dorsal side downward. This period of re>t t >rr< '-ponds some-
what to the death-feigning instinct of the insect. By \ i-onm-ly
stirring up the specimens or throwing them into water havii
temperature to which the specimens are not aivn-tonied. or into
relatively strong chemical solutions of various son-, a- adds.
salts, alcohol, etc., practically all of the specimen- tall into this
momentary, rather stiff, inactive state and slowly <! -< -ml to the
bottom of the dish. In so doing almost all of the specimens
topple over and fall down head-first, ventral side up ami on the
average, at an angle of about 45 degrees. It ini^lit al-o In-
mentioned here that nymphs which are found dead in the aquaria
lie almost invariably with their ventral side up. ( >n the oilier
hand, the turning over is under similar conditions far le-- frequent
among the specimens from which the organ had been remo\ed.
It two groups of freshly killed specimens are taken, all <>l which
have been cleaned and their appendage- arranged, the one i^roup
normal in every way, the other having the I'.ilnn-n'- organ re-
move.d. we find by allowing them to de-rend through a d
jar of water that almost invariably the li-rnier topple o\cr and
-ettle on the bottom ventral side up, while the latter eqnallv as
frequently reach the bottom and remain there \\iih thdr \entral
side downward.

CONCLUDING RIMAK;

The results of the foregoing experiments show conclusively
that the organ, as small as it is, plays a very important role in
the behavior of the nymphs upon which these experiments v,
performed. This is doubtless due to the weight of the chitinous
ma-> whose pressure seems, to a large extent, to control certain
orientation of the in-ects. Gross (1903) gives a figure of the

264 J. E. WODSEDALEK.

cross section of the head of a may-fly showing the position of
Palmen's organ in relation to the other parts, and in his discussion
says, "Unter dem Palmen'schen Organ verlauft namlich bei
alien 5 von mir untersuchten Ephemeridenspecies ein starker,
vom Gehirn kommender Nervenstrang. Seine Lagebeziehungen
ergeben sich aus Fig. B, die einen Medianschnitt durch den Kopf
einer Ephemera vulgata bei schwacher Vergrosserung darstellt.
Der erwahnte Nerv (np) verlauft in der Medianlinie vom Gehirn
(g) nach hinten unter dem Palmen'schen Organ (/?) hindurch und
heftet sich hinter ihm an der Korperwand an. In einem Theil
seines Verlaufs liegt er direct auf dem Nervus recurrens (nr) des
unpaaren sympathischen Nervensystems."

Careful examination of many nymphs showed no evidence of
the presence of the two large nerves which Gross speaks of as
present in the imaginal species which he examined ; this was also
true of the adult specimens which I examined. It appears from
his discussion of the subject and from his figure (page 98), that
what he speaks of as nerves may possibly be the two muscles
which play an important part in the movement of the head.
The posterior attachment of these muscles to the exoskeleton
evidently corresponds to the attachment of the large nerves he
misrepresents . In my preparations very thin sections were made,
but no signs of nerves extending directly from the brain to the
organ were detected. Taking the structure and function of the
organ into consideration we should not expect the presence of
such nerves. A mass of rather loose tissue exists between the
organ and the brain, and the two are loosely united by means of
connective tissue. It is the writer's opinion that the chitinous
organ being so loosely supported by the four tracheal tubes
exerts a pressure on the surrounding tissues, whereby the dis-
turbing stimulus reaches the central nervous system. The ob-
servations mentioned on the descent of nymphs in various con-
ditions, through the water, particularly the death-feigning and
the dead individuals, seems to indicate that the orientation is
also, in part, a self-directing process, that is, by the presence of
the organ the nymph is swerved into position a matter of
physical equilibrium.

Gross' theory that the organ functions only in the adult speci-

PALMEN'S ORGAN ix HEPTAGEXIA AND ECDYURUS. 265

mens seems quite untenable. Aside from the results of my
experimental work arises another question. Why should this
structure occur and persist in very small nymphs, and grow in
relative proportion during the comparatively long nymphal stage
of two, and in some cases three years, for the purpose of becoming
functional only after the nymph metamorphoses into its short-
lived adult stage, when all the other modifications which are of
a direct advantage to the adults develop during tin- comparath -ly
short time immediately preceding the- transformation:'

The extent of the functions of this organ in the adults thus tar
remains unknown. Miall (1895) in >[>.. ikin- of the Kphenicrid.i'
gives the following quotation: "The n-cently eim-r^rd tl\ , " says
\><- Geer, "settles on trees, plant-, \\alls etc., lu-.ir the water
wliich harbored the larva. II- re it iiv- it-df by the hooks of
tin- feet, usually with the head downuanU. and rests until the
List or sub-imaginal moult is at hand." Mv o|,- ( r\ ation
tin- behavior of adult may-flies are to some extent in accord \\ith
i he. -i- of De Geer, however, no theory a> to tin- probable function
I the organ in the adults can be propounded, unless it can be
-upported by reliable results of experimental work. A I.
nuinber of nearly lull grown nymphs from \\hich the 1'aliiH n's
in had been removed are now in the aquaria \\ith tin- \i<-\\ of
making a study of their behavior, \\hen they emero- .1- adults in
comparison with that of the normal indi\ idual>.

Among the twenty-three species in which 1'almen 1^77 noted

t lie i ire-nice of this organ, there are se\ i al tin- -u immini; fi inns,
and at this time, it is difficult to say HIM \\hat |>art 1'alim-n's
m |>la\-s in those forms during their life history as \cr\- little
i> known of their natural habits.

1 am greatly indebted to ProlV--or \\illiain S. Mar-hall for
su-v;-ting this problem, and also, lor his help and earnest interest
in the progress of the work.

/< < M K.ICAL LABORATORY.

UNIVERSITY OF \\'IM "\-i\.

266 J. E. WODSEDALEK.

REFERENCES.
Berger, E.

'78 Untersuchungen iiber den Bau des Gehirns und der Retina der Arthropoden

Arb. zool. Inst. Wien, V., i.
Borner, C.

'08 Die Tracheenkiemen der Ephemeriden. Zool. Anz., XXXIII.
Brehm.

'92 Thierleben, 3. Aufl., Leipzig u. Wien.
Dewitz, H.

'90 Einige Beobachtungen, betreffend das geschlossene Tracheensystem bei

Insectlarven. Zool. Anz., XIII.
Eaton, A. E.

'88 A Revisional Monograph of the Recent Ephemerid;e or Mayflies. Trans.

Linn. Soc. London, (2), V., 3.
Gerstacker, A.

'66 Die Arthropoden. Bronn Class. ( Ordn. Thierreich, V. 5, Berlin u. Heidel-
berg.
Gross, J.

'03 Uber das Palmen'sche Organ der Ephemeriden. Zool. Jahrb., XIX.

(Anat.).
Hiibner, O.

'02 Neue Versuche aus dem Gebiet der Regeneration und ihre Beziehungen zu

Anpassungserscheinungen. Zool. Jahrb., XV.
Joly, N.

'78 Etudes sur les metamorphoses et 1'embryogenie des Ephemerines, et

specialement sur celles de la Palingenia virgo. Seance.
Kolbe, H.

'93 Einfiihrung in die Kenntniss der Insecten. Berlin.
Leydig, F.

'57 Lehrbuch der Histologie. Frankfurt a./M.
Lubbock, J.

'63 On the Development of Chloeon( (Ephemera) dimidiatum. Trans. Linn.

Soc., Vol. XXIV.
Miall, L. C.

'95 Aquatic Insects. London.
Needham, J. G.

'05 May Flies and Midges of New York. N. Y. State Museum, Bulletin 86.
Oppenheim, S.

'08 Segmentregeneration bei Ephemeriden-Larven. Zool. Anz., XXX I II.
Palmen, J. A.

'77 Zur Morphologic des Tracheensystems. Leipzig.
Pictet, F. J.

^

"43 Famille Des Ephemerines. Paris.
Stein, F.

'47 Yerglcichende Anatomic und Physiologic der Insecten. Monographieen

bearbeitet, i, Berlin.
Swammerdam, J.
1752 Bibel der Natur.
Vayssiere, M. A.

'82 L'organisation des Larves des Ephcmerines. Paris.

PALMEN'S ORGAN IN HEPTAGENIA AND ECDYIKI s. 267

Wodsedalek, J. E.

'n Phototactic Reactions and Their Reversal in the May-fly Nymph? //.

interpunclala. Biological Bulletin. XXI.

'12 Formation of Associations in the May-fly Nymphs H. intfrpundata. Jour-
nal of Animal Behavior, Vol. 2, no. i.

'12 Natural History- and General Behavior of the Ephemerida- Nymph* II- f-
nia inlerpunct.it<i. Annals of the Entomological Society of America,
Vol. 5, No. i.
Zimmer, C.

'98 Die Facettenaugen der Epln-nn-riden. Z. wiss. Zool., \" .. '

268 J. E. WODSEDALEK.

EXPLANATION OF FIGURES.
PLATE I.

All drawings (except Figs. 2 and 3) made with a camera-lucida. X240.

FIG. i. Palmen's organ in its relation to the tracheal system in the head of the
nymph H. inter punctata. X 60.

FIG. 2. Sketch drawn from a specimen which had the organ removed and the
four tracheae broken off near their juncture with the main longitudinal tubes.

FIG. 3. Sketch drawn from a nymph in which the tracheae were severed at
their point of contact with the organ.

BIOLOGICAL BULLETIN, VOL. XXII.

PLATE I.

J. E. WOMEDALEK.

2JO J. K. WODSEDALEK.

PLATE II.

FIG. 4. Representation of the entire organ surrounded by hypoderrnis, as it
appears in a mounted specimen. Circular bands can be seen, especially at the
edges of the organ, owing to the fact that we look at the vertical portion of each
deeply colored part. In this view the large light areas appear at the entrances of
the tracheal tubes; this is due to the fact that we look through a comparatively
thin portion of the chitin in those regions owing to the direct extension of the cavities
of the tubes into the organ. The darker areas appear as such because of their
thickness; each is a concentric mass around the organ and forms the partition be-
tween the cavities.

FIG. 5. A horizontal section almost directly through the center of the two
posterior tubes and a little above the center of the two anterior ones. It is only
natural, therefore, that the two posterior tracheae should lead to the solid central
mass. The entrances of the two anterior ones are not in the same plane with that
of the posterior pair and therefore the innermost portion of their cavities are not
represented in this section. The gradually increasing diameter of each cavity is
understood when we recall the development of the organ and the tubes leading to it.

FIG. 6. A horizontal section through the ventral projection of the organ which
is apparent in Figs. 7 and 8. The central part of this figure appears clear because
the section was quite thin and the cut parallel with the light portion of one of the
concentric layers.

BIOLOGICA. BULLETIN, VOL. XXII.

PLATE II.

J. E. WOMEDAIEK

2J2 J. E. WODSEDALEK.

PLATE III.

FIG. 7. A transverse section cut near the center of the posterior pair of cavities.
FIG. 8. An oblique transverse section cut through the front part of the posterior
tube cavities and through the tips of the anterior cavities.

FIG. 9. A still more anterior view, only the cross sections of the two deeper
portions of the anterior cavities being in evidence.

I

BIOLOGICAL BULLETIN, VOL. XXII.

PLATE III.

J. E. *ODSEDAlE>v.

Vol. XXII. . April, 1912. No. j

BIOLOGICAL BULLETIN

THE EFFECTS OF SOME AMIDO-ACIDS OX THE

DEVELOPMENT OF THE E