BIOLOGICAL BULLETIN

OF THE

HDarine Biological Haborator?

WOODS HOLE, MASS.

lEMtorial Staff

GARY N. CALKINS — Columbia University.
E. G. CONKLIN — Princeton University.
M. H. JACOBS — University of Pennsylvania.
FRANK R. LILLIE — University of Chicago.

GEORGE T. MOORE — The Missouri Botanic Garden.
T. H. MORGAN — Columbia University.
W. M. WHEELER — Harvard University.
E. B. WILSON — Columbia University.

JEMtor

C. R. MOORE — The University of Chicago.

VOLUME LV.

WOODS HOLE, MASS.

JULY TO DECEMBER, 1928

LANCASTER PRESS, INC.
LANCASTER, PA.

Contents of Volume LV

No. i. JULY, 1928.
Thirtieth Annual Report of the Marine Biological Laboratory. I

No. 2. AUGUST, 1928.
KING, ROBERT I. The Contractile Vacuole in Paramecium

trichium 59

MAIN, HOLLAND J. Observations of the Feeding Mechanism

of a Ctenophore, Mnemiopsis leidyi 69

AMBERSON, WILLIAM R. The Influence of Oxygen Tension

upon the Respiration of Unicellular Organisms 79 «

BOYD, MARJORIE. A Comparison of the Oxygen Consumption

of Unfertilized and Fertilized Eggs of Fundulus heteroditus 92 *
CALKINS, GARY N., AND BOWLING, RACHEL. Studies on

Dallasia frontata Stokes 101

PARPART, ARTHUR K. The Bacteriological Sterilization of

Paramecium 113*

HUESTIS, R. R. The Effect of Maternal Age and of Temper-
ature Change in Secondary Non-Disjunction 121

MELVIN, ROY. Oxygen Consumption of Insect Eggs 135 •

No. 3. SEPTEMBER, 1928.

HILL, SAMUEL E. The Influence of Molds on the Growth of
Luminous Bacteria in Relation to the Hydrogen Ion
Concentration, Together with the Development of a
Satisfactory Culture Method 143

IVAROL, JOHN J. The Sex Ratio in Peromyscus 151

PAYNE, NELLIE M. Cold Hardiness in the Japanese Beetle,

Popillia japonica Newman 163

NELSON, THURLOW C. Pelagic Dissoconchs of the Common
Mussel, Mytilus edulis, with Observations on the Behavior
of the Larv > of Allied Genera 180

TURNER, C. L. Studies on the Secondary Sexual Characters
of Crayfishes. — VI. A Female of Cambarus immunis
with Oviducts Attached to Openings of Sperm Ducts. ... 193

^(LIBRARY

.'51570

IV CONTENTS OF VOLUME LY.

TURNER, C. L. Studies on the Secondary Sexual Characters
of Crayfishes. — VII. Regeneration of Aberrant Secon-
dary Sexual Characters 197

SAYLES, LEONARD P. Regeneration of Lumbriculus in

Various Ringer Fluids 202

ALPATOV, W. W. Variation of Hooks on the Hind Wing of

the Honey Bee (Apis mellifera L.} 209

No. 4. OCTOBER, 1928.

HARMAN, MARY T., AND ROOT, FRANK P. The Development

of the Spermatozoon in Cavia cobaya 235

TURNER, C. L. Studies on the Secondary Sex Characters of
Crayfishes, VIII. Modified Third Abdominal Ap-
pendages in Males of Cambarus virilis 255

GRAVE, B. H. Natural History of Shipworm, Teredo

navalis, at Woods Hole, Massachusetts 260

NEWMAN, H. H. Studies of Human Twins, I. Methods of

Diagnosing Monozygotic and Dizygotic Twins 283

NEWMAN, H. H. Studies of Human Twins, II. Asym-
t metry Reversal, of Mirror Imaging in Identical Twins. . 298

No. 5. NOVEMBER, 1928.

HUMPHREY, R. R. Sex Differentiation in Gonads Developed
from Transplants of the Intermediate Mesoderm of
Amblystoma 317

MOORE, CARL R. On the properties of the Gonads as Con-
trollers of Somatic and Psychical Characteristics, XI. . . . 339

JUST, E. E. Initiation of Development in Arbacia, VI. The
Effect of Slowly Evaporating Sea-Water and its Signifi-
cance for the Theory of Auto-Parthenogenesis 358

CHAMBERS, ROBERT. Intracellular Hydrion Concentration
Studies, I. The Relation of the Environment to the pH
of Protoplasm and of Its Inclusion Bodies 369

REZNIKOFF, PAUL, AND POLLACK, HERBERT. Intracellular
Hydrion Concentration Studies, II. The Effect of In-
jection of Acids and Salts on the Cytoplasmic pH of Amoeba
dubia 377

POLLACK, HERBERT. Intracellular Hydrion Concentration
Studies, III. The Buffer Action of the Cytoplasm of
Amoeba dubia and Its JJse in Measuring the pH 383

CONTENTS OF VOLUME LY.

GREGORY, LOUISE H. The Effects of Changes in Medium
during Different Periods in the Life History of Uroleptus
mobilis and Other Protozoa 386

No. 6. DECEMBER, 1928.

BODINE, JOSEPH HALL. Insect Metabolism 395 •

LLOYD, FRANCIS E., AND BEATTIE, J. The Pulsatory

Rhythm of the Contractile Vesicle in Paramecium 404

THRELKELD, W. L., AND HALL, S. R. Observations on Hydra
and Pelmatohydra Under Determined Hydrogen Ion
Concentration 4J9

MAN WELL, REGINALD D. The Occurrence of Nuclear Vari-
ations in Pleurotricha lanceolata (Stein) 433

QUIGLEY, J. P. Observations on the Life History and Physio-
logical Condition of the Pacific Dog Fish (Squalus sucklii) 439

FARLOWE, VIVIAN. Algce of Ponds as Determined by an

Examination of the Intestinal Contents of Tadpoles 443

PAGE, IRVINE H. Further Observations on the Chemical
Composition of Woods Hole Sea Water — The Chlorine
Content and Salt Analysis 449 •

KAPP, ELEANOR M. The Precipitation of Calcium and

Magnesium from Sea Water by Sodium Hydroxide 453

HARVEY, E. NEWTON, HARVEY, ETHEL B., AND LOOMIS,
ALFRED L. Further Observations on the Effect of High
Frequency Sound Waves on Living Matter 459

Vol. LV

July 1928

No. i

BIOLOGICAL BULLETIN

THE MARINE BIOLOGICAL LABORATORY.

.1.

II.
III.

IV.

V.

VI.

THIRTIETH REPORT FOR THE YEAR 1927—
FORTIETH YEAR.

TRUSTEES AND EXECUTIVE COMMITTEE (AS OF AUGUST

9> 1927) *

LIBRARY COMMITTEE 3

ACT OF INCORPORATION 3

BY-LAWS OF THE CORPORATION 4

REPORT OF THE TREASURER 5

REPORT OF THE LIBRARIAN 1 1

REPORT OF THE DIRECTOR 17

Statement 17

Addenda :

1 . The Staff, 1927 27

2. Investigators and Students, 1927 30

3. Tabular View of Attendance 41

4. Subscribing and Cooperating Institutions, 1927 42

5. Evening Lectures, 1927 43

6. Members of the Corporation 44

I. TRUSTEES.

EX OFFICIO.

FRANK R. LILLIE, President of the Corporation, The University of

Chicago.

MERKEL H. JACOBS, Director, University of Pennsylvania.
LAWRASON RIGGS, JR., Treasurer, 25 Broad Street, New York City.
L. L. WOODRUFF, Clerk of the Corporation, and Secretary of the Board

of Trustees pro tan, Yale University.

EMERITUS.

CORNELIA M. CLAPP, Mount Holyoke College.
OILMAN A. DREW, Eagle Lake, Florida.

TO SERVE UNTIL IQ3I.

H. C. BUMPUS, Brown University.
W. C. CURTIS, University of Missouri.
1 i

2 MARINE BIOLOGICAL LABORATORY.

B. M. DUGGAR, University of Wisconsin.

GEORGE T. MOORE, Missouri Botanical Garden, St. Louis.
W. J. V. OSTERHOUT, Member of the Rockefeller Institute for Med-
ical Research.

J. R. SCHRAMM, University of Pennsylvania.
WILLIAM M. WHEELER, Bussey Institution, Harvard University.
LORANDE L. WOODRUFF, Yale University.

TO SERVE UNTIL I93O.

E. G. CONKLIN, Princeton University.
OTTO C. GLASER, Amherst College.
Ross G. HARRISON, Yale University.

H. S. JENNINGS, John Hopkins University.

F. P. KNOWLTON, Syracuse University.

M. M. METCALF, Johns Hopkins University.
WILLIAM PATTEN, Dartmouth College.
W. B. SCOTT, Princeton University.

TO SERVE UNTIL 1929.

C. R. CRANE, New York City.

I. F. LEWIS, University of Virginia.

R. S. LILLIE, The University of Chicago.

E. P. LYON, University of Minnesota.

C. E. McCLUNG, University of Pennsylvania.
T. H. MORGAN, Columbia University.

D. H. TENNENT, Bryn Mawr College.

E. B. WILSON, Columbia University.

TO SERVE UNTIL 1928.

H. H. DONALDSON, Wistar Institute of Anatomy and Biology.

W. E. GARREY, Vanderbilt University Medical School.

CASWELL GRAVE, Washington University.

M. J. GREENMAN, Wistar Institute of Anatomy and Biology.

R. A. HARPER, Columbia University.

A. P. MATHEWS, The University of Cincinnati.

G. H. PARKER, Harvard University.

C. R. STOCKARD, Cornell University Medical College.

EXECUTIVE COMMITTEE OF THE BOARD OF TRUSTEES.

FRANK R. LILLIE, Ex. Off. Chairman.
MERKEL H. JACOBS, Ex. Off.
LAWRASON RIGGS, JR., Ex. Off.
OTTO C. GLASER, to serve until 1928.
CASWELL GRAVE, to serve until 1928.
E. G. CONKLIN, to serve until 1929.
C. R. STOCKARD, to serve until 1929.

ACT OF INCORPORATION.

THE LIBRARY COMMITTEE.

C. E. McCLUNG, Chairman.
ROBERT A. BUDINGTON.
B. M. DUGGAR.
E. E. JUST.
FRANK R. LII.LIE.
M. M. METCALF.
ALFRED C. REDFIELD.
A. H. STURTEVANT.
L. L. WOODRUFF.

II. ACT OF INCORPORATION.

No. 3170

COMMONWEALTH OF MASSACHUSETTS.

Be It Known, That whereas Alpheus Hyatt. William San ford Ste-
vens, William T. Sedgwick, Edward G. Gardiner, Susan Minns.
Charles Sedgwick Minot, Samuel Wells, William G. Farlow, Anna D.
Phillips and B. H. Van Vleck have associated themselves with the
intention of forming a Corporation under the name of the Marine
Biological Laboratory, for the purpose of establishing and maintaining
a laboratory or station for scientific study and investigation, and a
school for instruction in biology and natural history, and have complied
with the provisions of the statutes of this Commonwealth in such case
made and provided, as appears from the certificate of the President,
Treasurer, and Trustees of said Corporation, duly approved by the
Commissioner of Corporations, and recorded in this office;

Now, therefore, I, HENRY B. PIERCE, Secretary of the Common-
wealth of Massachusetts, do hereby certify that said A. Hyatt, W. S.
Stevens, W. T. Sedgwick, E. G. Gardiner, S. Minns, C. S. Minot, S.
Wells, W. G. Farlow, A. D. Phillips, and B. H. Van Vleck, their asso-
ciates and successors, are legally organized and established as, and are
hereby made, an existing Corporation, under the name of the MARINE
BIOLOGICAL LABORATORY, with the powers, rights, and privi-
leges, and subject to the limitations, duties, and restrictions, which by
law appertain thereto.

Witness my official signature hereunto subscribed, and the seal of
the Commonwealth of Massachusetts hereunto affixed, this twentieth
day of March, in the year of our Lord One Thousand, Eight Hundred
and Eighty-Eight.
[SEAL] HENRY B. PIERCE,

Secretary of the Commonwealth.

4 MARINE BIOLOGICAL LABORATORY.

III. BY-LAWS OF THE CORPORATION OF
THE MARINE BIOLOGICAL LABORATORY.

I. The annual meeting of the members shall be held on the second
Tuesday in August, at the Laboratory, in Woods Hole, Mass., at
12 o'clock noon, in each year, and at such meeting the members shall
choose by ballot a Treasurer and a Clerk, who shall be, ex officio,
members of the Board of Trustees, and Trustees as hereinafter pro-
vided. At the annual meeting to be held in 1897, not more than
twenty-four Trustees shall be chosen, who shall be divided into four
classes, to serve one, two, three, and four years, respectively, and
thereafter not more than eight Trustees shall be chosen annually for
the term of four years. These officers shall hold their respective
offices until others are chosen and qualified in their stead. The Presi-
dent of the Corporation, the Director and the Associate Director of
the Laboratory, shall also be Trustees, ex officio.

II. Special meetings of the members may be called by the Trustees to
be held in Boston or in Woods Hole at such time and place as may be
designated.

III. The Clerk shall give notice of meetings of the members by pub-
lication in some daily newspaper published in Boston at least fifteen
days before such meeting, and in case of a special meeting the notice
shall state the purpose for which it is called.

IV. Twenty-five members shall constitute a quorum at any meeting.

V. The Trustees shall have the control and management of the af-
fairs of the Corporation ; they shall present a report of its condition at
every annual meeting; they shall elect one of their number President of
the Corporation who shall also be Chairman of the Board of Trustees ;
they shall appoint a Director of the Laboratory; and they may choose
such other officers and agents as they may think best ; they may fix the
compensation and define the duties of all the officers and agents ; and
may remove them, or any of them, except those chosen by the members,
at any time; they may fill vacancies occurring in any manner in their
own number or in any of the offices. They shall from time to time
elect members to the Corporation upon such terms and conditions as
they may think best.

VI. Meetings of the Trustees shall be called by the President, or by
any two Trustees, and the Secretary shall give notice thereof by written
or printed notice sent to each Trustee by mail, postpaid. Seven
Trustees shall constitute a quorum for the transaction of business.
The Board of Trustees shall have power to choose an Executive Com-

TIFK KKPORT OF THE TREASURER. 5

mittee from their own number, and to delegate to such Committee such
of their own powers as they may deem expedient.

VII. The accounts of the Treasurer shall be audited annually by a
certified public accountant.

VIII. The consent of every Trustee shall be necessary to dissolution
of the Marine Biological Laboratory. In case of dissolution, the prop-
erty shall be disposed of in such manner and upon such terms as shall
be determined by the affirmative vote of two-thirds of the Board of
Trustees.

IX. These By-laws may be altered at any meeting of the Trustees,
provided that the notice of such meeting shall state that an alteration
of the By-laws will be acted upon.

X. Any member in good standing may vote at any meeting, either in
person or by proxy duly executed.

IV. THE REPORT OF THE TREASURER.

To THE TRUSTEES OF THE MARINE BIOLOGICAL LABORATORY.

Gentlemen: As Treasurer of the Marine Biological Laboratory,
I herewith submit my report for the year 1927.

The books have been audited by Messrs. Seamens, Stetson &
Tuttle. A copy of their report is on file at the laboratory and is
open to inspection by any member of the Corporation.

There were no changes in the investments in the Endowment
Fund and that Fund at the close of the year consisted of securi-
ties of the book value of $906,337.50 and cash of $112. The in-
come from the Endowment Fund for the year was $47,583 and
the fee of the Trust Company as Trustee was $787.50, leaving a
net income from the Endowment Fund of $46,795.50. The full
list of the securities will be found in the Auditors' report.

At the end of the year the Lucretia Crocker Fund consisted of
securities of the book value of $3,590.59 and cash of $1,093.17.

During the year a fund of Two Thousand Dollars ($2,000)
invested in a note secured by Chicago real estate was presented to
the Laboratory to found the IDA H. HYDE SCHOLARSHIP and the
fund remained invested in this security at the end of the year.

The Retirement Fund consisted of Seven Thousand Dollars
($7,000) invested in participations in bonds secured by mortgages
on New York City real estate and $147.76 in cash.

6 MARINE BIOLOGICAL LABORATORY.

The land, buildings, library and equipment of the Laboratory
including the new apartment house and dormitory but excluding
the Gansett and Devils Lane property represents an investment of
$1,545,616.36, and after deducting $136,181.18 for depreciation,
a book value of $1,409,435.18.

During the year the following donations were received :

From General Education Board for improving the facilities of the

Library $15,000

From Dr. Frank R. Lillie for grading and planting 1,500

During the year Two Thousand Dollars ($2,000) was paid off
on account of the Danchakoff mortgage, and the indebtedness of
the Laboratory at the end of the year consisted of $6,542.83 in
accounts payable, and $42,500 in mortgages on its real estate.

The expenditures closely approximated the estimates for the
year and including an item of depreciation of almost $29,000
exceeded the income for the year by $91.06. Against this item of
depreciation the sum of almost $20,000 was expended out of
current funds upon permanent improvements and equipment.

Since January I, 1916, the Laboratory has adopted the policy
of charging income and crediting reserve for depreciation each
year with 2 per cent, of the book value of the buildings and 5
per cent, of the book value of equipment and library. This de-
preciation at the end of the year 1927 amounted to $136,181.18.
It is interesting to note, however, that against this the Laboratory
has spent from current cash approximately $141,000 in perma-
nent improvements, thus more than meeting the depreciation
charge by improvements paid for out of income.

Following is the balance sheet at the end of the year and the
condensed statement of income and outgo for the year, also the
Surplus Account. The figures are those reported by the Auditors,
arranged in the case of Exhibit B so as to conform to the system
followed in previous reports.

THE REPORT OF THE TREASURER.

EXHIBIT A.

MARINE BIOLOGICAL LABORATORY BALANCE SHEET,
DECEMBER 31, 1927.

Assets.
Endowment Fund Assets :

Securities in Hands of Trustee — Schedule I. $ 906,337.50
Investment Cash in Hands of Trustees . 112.00

$ 906,449.50
Lucretia Crocker Fund Assets,

Securities — Schedule II $ 3,590.59

Cash — Schedule II 1,093.17 4,683.76

Ida H. Hyde Fund Assets,

Securities $ 2,000.00 $ 913,133.26

Plant Assets :

Land — Schedule III $i 13,603.05

Buildings — Schedule III 966,279.78

Equipment — Schedule III 126,197.40

Library — Schedule III 90,682.87 $1,296,763.10

Less Reserve for Depreciation 136,181.18

$1,160,581.92

Cost of New Dormitory and Apartment
House Buildings to December 31, 1927—

Schedule IV $ 248,853.26

Cash in Dormitory Building Fund 3,590.31 $1,413,025.49

Current Assets :
Cash,

In New York Bank $ 2,335.99

In Hands of Trustee 2,200.00

In Falmouth Bank 1,558.19

Petty Cash 500.00 $ 6,594.18

Accounts — Receivable 20,440.55

Inventories,

Supply Department $ 30,802.33

Biological Bulletin 6,237.30 37,039.63

Investments,

Devil's Lane Property $ 33,395.51

Gansett Property 1,769.35

Stock in General Biological

Supply House, Inc 12,700.00

Retirement Fund Assets ... 7,147.76 55,012.62

Prepaid Insurance ... 4-3/8.33 123,465.31

$2,449,624.06

\u~l L

', -*-«••*•

8 MARINE BIOLOGICAL LABORATORY.

Liabilities.
Endowment Funds :

Friendship Fund, Inc $ 405,000.00

John D. Rockefeller, Jr 400,000.00

Carnegie Corporation 100,000.00

Gain on sale of Securities 1,449.50

906,449.50

Lucretia Crocker Fund 4,683.76

Ida H. Hyde Fund 2,000.00 $ 913,133.26

Plant Funds :

Rockefeller Foundation 500,000.00

Friendship Fund Gift of 1925 221,608.61

General Education Board for Buildings .... 250,000.00
General Education Board for

Books $25,000.00

Less Unexpended in Current

Cash 1,300.50 23,699.50

Other Investments in Plants from Gifts

and from Current Funds 399,525.38

$1,394,83349

Mortgages on Drew and Danchakoff Estates 17,500.00
Suspense — Interest on Building Fund Cash ; 692.00 1,413,025.49

Current Liabilities and Surplus :

Mortgage Note on Devil's Lane Property.. 25,000.00

Accounts — Payable 6,542.83

Items in Suspense (net) 222.47

$ 31,765-30
Current Surplus— Exhibit C 91,700.00 123,465.31

$2,449,624.06

EXHIBIT B.

MARINE BIOLOGICAL LABORATORY, INCOME AND EXPENSE,
FOR THE YEAR ENDED DECEMBER 31, 1<)2"J .

Total. Net.

Expense. Income. Expense. Income.

Income, Endowment Fund $ 47,583.00 $47,583.00

Donations (See Current
Surplus)

Instruction 7,829.30 10,640.00 2,810.70

Research 3,598.86 14,525.00 10,926.14

BIOLOGICAL BULLETIN and
Membership Dues 7,117.54 7,659-51 54J-97

THE REPORT OF THE TREASURER.

Supply Department, Sched-
ule IV 52,174.46 59,820.90 7,646.44

Mess, Schedule V 33,08542 36,180.30 3,«94-88

Dormitories, Schedule VI 25,870.11 12,865.29 13,004.82

Interest and depreciation
charged to above three de-
partments, See Schedules
IV, V and VI 29,719.11 29,719.11

Dividends on Stock, Gen-
eral Biological Supply
House, Inc 2,540.00 2,540.00

Rent of Danchakoff Cot-
tages 449-99 750.00 300.01

Rent of Microscopes 3?o.oo 370.00

Rent of Garage, Railway,

etc 356.91 356.91

Rent of Newman Cottage 164.08 150.00 14.08

Interest on Bank Balances. . I37-2O I37-2O

Medical Fees 114.00 114.00

Sundry Items 15-31 15-31

Maintenance of Plant :

New Laboratory Expense 15,044.43 15,044.43

Maintenance of Buildings

and Grounds 13,296.19 13,296.19

Chemical and Special Ap-
paratus Department . . 9,088.83 9,088.83
Library Department Ex-
penses 7,958.73 7-958.73

Carpenter Department Ex-
penses 1,123.04 1,123.04

Truck Expenses 1,203.52 1,203.52

Sundry Expenses 814.22 814.22

Bar Neck Property Ex-
penses 405.00 405.00

Evening Lectures 159-51 I59-5I

Workmen's Compensation

Insurance 627.43 627.43

General Expenses :

Administration Expenses. 12,335.00 12,335.00

Interest on Loans 1, 168.00 1, 168.00

Endowment Fund Trustee 787.50 787.50

Bad Debts 230.20 230.20

Contribution for Research
Space, Naples Zoolog-
ical Station 250.00 250.00

Reserve for Depreciation.. 28,736.23 28,736.23

Excess of Expense over In-
come carried to Current
Surplus— Exhibit C 91.06 91.06

$193.798.48 $193.798.48 $106,246.73 $106,246.73

IO MARINE BIOLOGICAL LABORATORY.

I

EXHIBIT C.

MARINE BIOLOGICAL LABORATORY, CURRENT SURPLUS ACCOUNT,
YEAR ENDED DECEMBER 31, 1927.

Balance, January i, 1927 $ 83,503.64

Add:

Donations Received,

From General Education Board for purchase of Books

for Library 15,000.00

From Dr. Frank R. Lillie for Grading, Planting, etc.
on Laboratory Grounds and around Drew House,

Apartment House, and Whitman House 1,500.00

Income of Retirement Fund Assets 222.82

Reserve for Depreciation charged to Plant Fund 28,736.23

$128,962.69
Deduct :

Payments from Current Funds during Year
Plant Assets as shown in Schedules III
and Ill-a,

Cost of completing Sea Wall $ 435.50

Buildings 7,303.90

Equipment 4,312.23

Library Books, etc 7,623.94

New Dormitory and Apartment House .... 251.57

$19,927.14

Payments from above Donations charged to
Plant Assets General Education Board,

Purchase of Books 13,744.48

Dr. Frank R. Lillie, Grading, etc 1,500.00

Payment on Danchakoff Mortgage 2,000.00

Balance of Income and Expense Account — Exhibit B 91.06 37,262.68

Balance, December 31, 1927 — Exhibit A 91,700.01

Respectfully submitted,

LAWRASON RIGGS, JR.,

Treasurer.

REPORT OF THE LIBRARIAN. II

V. THE REPORT OF THE LIBRARIAN,
DECEMBER 31, 1927.

The expenditures of the Library during 1927 totalled $26,-
039.67; segregated under the following headings: books, $527.20;
serials, $3,325.57; binding of current serials, etc., $997.25; sup-
plies, $541.88; express, $253.53; salaries, $5,650.00; miscellaneous
salaries, $1,550.50; and General Education Board Fund, appropri-
ated for back sets, $14,093.74. The total appropriation was $27,-
400.00, apportioned as follows: books, $500.00; serials, $3,000;
binding, $1,000; supplies, $500.00; express, $200.00; salaries,
$5,650.00; miscellaneous salaries, $1,550.00; and General Educa-
tion Board appropriation, $15,000. It will be noted that the great-
est part of the unused total occurred under the fund for back
sets. The sum of $906.26 was carried on into 1928 and was in
fact expended before January I5th, so that the expenditures all
along the line, show a slight over-running of the appropriation.
The most interesting item in this respect is that for serials. A
special point of this condition- was made by the Librarian in the
1926 Annual Report and in the informal report given to the cor-
poration last August. Either an increase in appropriations will
have to be made for this purpose, or some of the current serials
must be dropped. For 1928, an increase was granted by the Ex-
ecutive Committee and this will have to be further increased for
1929. It will be noticed that of the total $26,939.67, $19,739.17
was expended on material and acquisitions, and $7,200.50 on sal-
aries to carry on the work of the Library.

The Library now contains 22,762 bound volumes, most of these
coming under the category of serials and books, and in about the
proportion of 8 to I. Of these, 4,154 were acquired this year,
657 by binding current serials ; 674 by binding back sets ; and the
others by new back sets and books. Besides these volumes, the
reprints number 43,000. Only 5,000 of these were catalogued
and filed during this year.

The Library receives 764 serials currently and in addition to
these periodicals, subscribes to 36 books and monographs that are
being issued serially, 800 serial publications in all. Of these 764
periodicals, 136 were new this year; and of the 36 books and

12 MARINE BIOLOGICAL LABORATORY.

monographs, 5 are new. The new periodicals for 1927 are not
all by paid subscription, but 87 were acquired by exchange with
the BIOLOGICAL BULLETIN (67 copies of the BIOLOGICAL BULLE-
TIN being sent out) ; and 19 by gift, while only 30 new paid sub-
scriptions have been added. The total number of paid subscrip-
tions including the serially issued books is now 270 ; and the total
number of exchange, 274 ; and the gifts, 256. We have pur-
chased 180 new books, and have received from publishers, 112,
and from authors, 24, and from other sources, 144; 460 in all.
These gifts were as follows:

From the publisher, P. Blakiston's Son & Co., Craigie, E.
Home : An Introduction to the Finer Anatomy of the Central
Nervous System Based upon That of the Albino Rat; Evans, C.
Lovatt: Recent Advances in Physiology; Gould, George M. :
Medical Directory; Hawk, Philip B. : Practical Physiological
Chemistry; Lewis, F. T. and Bremer, J. L. : A Text-book of His-
tology arranged upon an Embryological Basis ; Meyers, Milton
K., Editor : Lang's German- English Medical Dictionary ; Pryde,
John: Recent Advances in Biochemistry; Stitt, E. R. : Practical
Bacteriology ; Youngken, Heber W. : Pharmacognosy.

Gebruder Borntraeger : Diirken, Bernhard : Allgemeine Ab-
stammungslehrc ; Herter, Konrad : Tastsinn, Stromungssinn und
Temperatursinn der Tiere und die diesen Sinnen sugeordneten
Rcaktionen ; von Buddenbrock, W. : Grundriss dcr vergleichenden
Physiologic.

Chicago University Press: Newman, H. H. : Evolution, Gen-
etics, and Eugenics; Newman, H. H., et al : Nature of the World
and of Man.

Columbia University Press : Chandler Chemical Laboratories :
Contemporary developments in Chemistry.

Detroit Digestive Ferments Co. : Manual of Dehydrated Cul-
ture Media and Reagents.

E. P. Dutton Co.: Einstein, Albert: The Theory of Brownian
Movement; Freundlich, Herbert: New Conceptions in Colloidal
Chemistry; Nernst, W. : The New Heat Theorem: its Founda-
tions in Theory and Experiment', Ostwald, Wolfgang: Practical
Colloid Chemistry; Stock, Alfred: Structure of Atoms.

Friederichsen & Co. : Michaelson, W., Editor : Beitrage sur

REPORT OF THE LIBRARIAN. 13

Kenntnis dcr Land-und Siisswasserfauna Deutsch-Sudwestafri-
kas; Michaelsen, W., Editor: Beitrdgc zur Kcnntiiis dcr M ceres-
fauna irestafrikas.

Ginn & Co. : Miller. Dayton C. : Laboratory Physics.

Harcourt, Brace £ Co. : von Uexkiill, J. : Theoretical Biology.

Hokuryukwan £ Co. Ltd. : Hirase, S. et al : Figuraro dc Jap-
ana] Bostoj.

Alfred A. Knopf, Inc. : Pearl, Raymond : Biology of Popula-
tion Growth ; Perrier, Edmond : The Earth before History, Man's
Origin and the Origin of Life; Wheless, Joseph: Is it God's
Word?

Lea and Febiger : Berkeley, W. N. : The Principles and Prac-
tice of Endocrine Medicine ; DuBois, E. F. : Basal Metabolism
in Health and Disease ; Wiggers, Carl J. : Modern Aspects of the
Circulation in Health and Disease.

Lewis, H. K. £ Co.: Boes, P. K. : X-ray Apparatus; its Ar-
rangement and Use.

J. P. Lippincott Co.: Addison, W. H. F. : Piersol's Normal
Histology; Craig, C. F. : Manual of the Parasitic Protozoa of
Man; Meyer, H. H. and Gottlieb, R. : Experimental Pharma-
cology.

Longmans, Green & Co. : MacLeod, John J. R. : Carbohydrate
Metabolism and Insulin.

McGraw-Hill Book Co.: Allen, E. S.: Six-place Tables; Ban-
croft, Wilder D. : Applied Colloid Chemistry; Fernald, H. T. :
Applied Entomology — an Introductory Text-book of Insects in
their Relations to Man; Pearse, A. S. : Animal Ecology; Rogers,
Charles G. : Textbook of Comparatiz'e Physiology ; Shull, Charles :
Heredity.

The Macmillan Co.: Adams, L. A.: Necturus; A Dissection
Guide; Baitsell, Geo. A.: Manual of Biological Forms; Bernard,
Claude: Introduction to the Study of Experimental Medicine;
Billroth, Theodor: Medical Sciences in the German Universities;
Brinkley, Stuart R. and Kelsey, E. B. : Laboratory Manual ar-
ranged to accompany "Principles of General Chemistry"; Cahn,
Alvin R. : The spiny dogfish ; A Laboratory Guide ; Creaser,
C. W. : The Skate; A Laboratory Manual; Frazer, James G. :
The IVorsIiip of Nature, vol. I ; Jeffrey, Edw. C. : Coal and Civi-

14 MARINE BIOLOGICAL LABORATORY.

lication; Kerr, J. Graham: Evolution; Needham, Joseph: Science,
Religion, and Reality; Newman, Horatio H. : The Gist of Evo-
lution; Smuts, J. C. : Holism and Evolution; Woodruff, L. LL. :
Foundations of Biology.

Open Court Publishing Co. : Brodetsky, S. : First Course in
Nomography; Friess, Horace Leland : Schlciermacher's Solilo-
quies; Leathern, J. G. : The Mathematical Theory of Limits;
Piaggio, H. T. H. : Elementary Treatise on Differential Equations
and their Application ; Silberstein, L. : Protective Vector Algebra.

Oxford University Press: de Beer, G. R. : An Introduction
to Experimental Embryology ; Dobson, G. M. B., Griffith, I. O.
and Harrison, D. N. : Photographic Photometry; Goodrich, Ed-
win S. : Living Organisms, an Account of their Origin and Evo-
lution; Heresy, George and Panetti, Fritz: Manual of Radio-
activity; Hinshelwood, C. N. : Kinetics of Chemical Change in
Gaseous Systems; Smith, G. Eliot: The Evolution of Man; Col-
well, H. C. : Introduction with Study of Roentgen Rays and Ra-
dium; Cooper, Eugenia R. A.: The Histology of the More Im-
portant Human Endocrine Organs at Various Ages ; Dakin, W.
J. : The Elements of General Zoology ; Dodds, E. C. and Dick-
ens, F. : The Chemical and Physiological Properties of the In-
ternal Secretions.

Presses Universitaires de France : Problemes Biologiques, 4-6.

Princeton University Press : Conklin, E. G. : A Synopsis of the
General Morphology of Animals; More, Louis T. : The Dogma
of Evolution ; Morgan, T. H. : Evolution and Genetics.

W. B. Saunders Co. : Arey, Leslie Brainerd : Developmental
Anatomy; Castle, W. E. et al : Our Present Kno^vlcdge of He-
redity; Cecil, R. L. : A Text-book of Medicine by American au-
thors; Borland, W. A. N. : American illustrated Medical Dic-
tionary; Falk, I. S.: Principles of Vital Statistics; Friedenwalt,
J. : Diet in Health and Disease ; Herrick, C. J. : Neurology ; Her-
rick, C. Judson: An Introduction to Neurology; Kolmer, John
A. : A Practical Text-book of Infection, Immunity and Biological
Therapy ; Stollmann, Torald : A Manual of Pharmacology ; Sten-
gel, Alfred and Fox, Herbert : Text of Pathology ; Stevens, A. A. :
The Practice of Medicine ; Stiles, Percy : Human Physiology ;
Todd, J. C. : Clinical Diagnosis by Laboratory Methods; Wells,
H. Gideon: Chemical Pathology.

REPORT OF THE LIBRARIAN. 15

D. Van Nostrand Co.: Baker, A. L. : Thick Lens Optics;
Brownell, B. : The New Universe ; Cathcart, W. L. and Chaffee,
J. I.: The Elements of Graphic Statics; Howe, Harrison E. :
Chemistry in the World's Work ; Lee, W'illit, T. : Stories in Stone ;
Ireland, L. T. : The Mystery of Mind.

Yale University Press: Lewis, Gilbert N. : The Anatomy of
Science; Millikan, R. A.: Evolution in Science and Religion.

From the authors : Bailey's Text-book of Histology, revised
by O. S. Strong and Adolph Elwyn; Conklin, Edwin G. : A
Synopsis of the General Morphology of Animals; Curtis, Win-
terton C. : Textbook of General Zoology (2 copies) ; Harvey, E.
Newton : Laboratory Directions in General Physiology ; Herrick,
C. Judson : Brains of Rats and Men ; Johnson, Chas W. : The
Insect Fauna (Biological Survey of the Mt. Desert Region) ;
Mathews, A. P. : The Nature of Matter, Gravitation and Light ;
McKeough, Rev. Michael J. : The Meaning of the Rationes Sem-
inales in St. Augustine; Means, James H. : Dyspnoea; Patten,
Bradley M. : The Embryology of the Pig; Pratt, H. S.: A Lab-
oratory Course in General Zoology ; Warbasse, James P. : Surgical
Treatment; Wieman, H. L. : General Zoology; Woodruff, L. L. :
Foundations of Biology ; Workman, F. B. and Workman, W. H. :
Algerian Memories; The Call of the Snowy His par; Ice-bound
Heiglits of flic Mustagh; Illustrations of Ice JTilds of Eastern
Karakoram ; In the Ice World of Himalaya ; Peaks and Glaciers
of Nun Kun; Sketches Awheel in Fin de Siecle Iberia; Through
Town and Jungle; Two Summers in the Ice-Wilds of Eastern
Karahoram.

An especially notable gift this year is that of the Library on
Coelenterates of Professor Charles Wesley Hargitt, presented
to the Library of the Marine Biological Laboratory by his son.
Professor George T. Hargitt. The volumes and pamphlets num-
ber about 500 in all. They will not be catalogued and shelved
as a special library, but each will be marked by a special book-
plate, and in addition, the collection as a whole will be marked
by a small brass tablet with an inscription showing that it is in-
corporated with the other books of the Library. This commem-
oration tablet, in size about 7"xQ", will be placed on the walls
in the stack-room of the reading-room floor.

16 MARINE BIOLOGICAL LABORATORY.

A special acknowledgment should be made of a gift of reprints,
books and pamphlets sent by Dr. Ida H. Hyde who had already
presented to this Library in 1917-18, the main collection of her
reprints. The library wishes to acknowledge another gift from
Mrs. Edward G. Gardiner, of reprints from Dr. Gardiner's li-
brary. Dr. Louis Murbach has presented a number of his books
and pamphlets that will be of use in the Library here. And very
especially, the Library makes acknowledgment of a gift of books
and other interesting pamphlets which were sent to us by Dr.
Elizabeth H. Dunn when she closed her shop in Woods Hole.

The sum granted by the General Education Board for back
sets was $5000 more this year than in 1926, and the number of
sets completed for the library was, therefore, correspondingly
larger. 84 back sets were completed and 30 partially completed.
The most interesting of these are :

Periodicals : Jahrbiichcr fiir ivissenschaftliche Botanik ; Journal
of the Franklin Institute ; Philosophical Transactions of the Royal
Society of London ; Proceedings of the Zoological Society of
London; Transactions of the Zoological Society of London; Zcit-
schrift fiir iwissenschaftliche Zoologie ; Zoologica.

Serially issued monographs : Bijdragcn tot de Dierkunde K.
soologisch genootschap Natura Artis Magistra te Amsterdam ;
Oppel, Albert : Lehrbuch dcr vcrgleichendcn Mikroskopischen
Ana-tomie der Wirbeltiere ; Oppenheimer, C. : Handbuch dcr Bi-
ochemie des Mcnschen und der Ticrc ; Chun, Carl : Valdivia Ex-
pedition.

The completion of back sets of serials has not been confined
to those purchased. We owe grateful acknowledgment for 7 sets
completed by gift and 19 partially completed. The back sets of
the Biological Bulletin have also been used to secure missing sets
of serials; 19 having been completed by this method, and 21 par-
tially filled in. In this connection, there should be mentioned
also the use made of the duplicate material which was listed and
arranged alphabetically in 1925-26. Several important gaps have
been filled through exchange for these duplicates. It is greatly
to be regretted that lack of time, both on our part, and the part
of other libraries, hinders a freer advertisement of duplicate ma-
terial.

REPORT OF THE DIRECTOR. l"J

The financial statement and the acquisitions of the year are
simple to enumerate. The uses made of the Library during the
year cannot so easily be appraised. This is especially true be-
ginning with this year when new books and the serial publica-
tions were for the first time restricted in place of use to the Li-
brary itself. 1,846 loans were charged out, however, during the
year. The out-of-town loans were greater than in any previous
year, although but 1 1 . Also more volumes were borrowed from
other libraries, 37 in all, one more than in 1926 and five more
than in 1925.

VI. THE REPORT OF THE DIRECTOR.

To THE TRUSTEES OF THE MARINE BIOLOGICAL LABORATORY.

Gentlemen: I beg to submit herewith a report of the fortieth
session of the Marine Biological Laboratory for the year 1927.

i. Attendance. — In the Tabular View of Attendance on page
41 a departure has been made from the previous customs of list-
ing the investigators under the three headings of Zoology, Physi-
ology and Botany. This change has appeared advisable, first,
because the broadening of the activities of the Laboratory in re-
cent years has brought to it an increasing number of persons
whose work does not, strictly speaking, fall into any of these
categories, and, second, because a considerable number of inves-
tigators in filling out their registration blanks have indicated that
in their own opinion their work belongs equally to Zoology and
Physiology or to Botany and Physiology. For this reason in-
vestigators have been classified merely as "independent" or "un-
der instruction." Following the custom inaugurated in 1926 a
separate class has been provided for research assistants, whose
number during the past few years has been rapidly increasing.

An examination of the figures for the attendance during the
years 1923-7, inclusive, shows that the number of students in
the courses has remained nearly stationary owing to the strict
limitation of the sizes of our classes. The attendance of investi-
gators, on the other hand, being subject to no such restriction
has shown a remarkable growth. For 1927 the increase over the
preceding record-breaking year of 1926 was approximately thir-
2

l8 MARINE BIOLOGICAL LABORATORY.

teen per cent. This increase is especially remarkable in view
of the fact that for a considerable time during the previous sum-
mer every room in both the brick and the wooden buildings was
occupied. The accommodation of over 40 additional investigators
under these circumstances was made possible, first, by the con-
version into laboratories of several rooms formerly used for other
purposes, second, by an extensive sharing of the larger rooms by
two or more workers, and, third, by a considerable lengthening
of the season of greatest activity. To encourage the attendance
of investigators during the less crowded parts of the summer, the
Mess in 1927 was opened about two weeks earlier and closed about
five days later than in previous years. This policy has proved to
be so successful that it is planned to continue it.

A consideration of the unprecedented increase in the number
of investigators during the past two years, which considerably
exceeds that for the previous thirteen, raises the question of the
maximum capacity of the present laboratory buildings. It may
be said in this connection that for the months of July and August
the limit has already been almost, though not quite, reached. A
further sharing of rooms during this period will make possible
the accommodation of a small additional number of investigators.
However, chief reliance in the future must be placed upon a
further extension of the working season. With the Laboratory
at present occupied to its full capacity for only two months of
the year it is evident that it is still far from having reached the
condition of its greatest usefulness.

The possibilities for extending the season in both directions
are strikingly shown by the following tabulation of the numbers
of investigators and research assistants in attendance on selected
clays throughout the summer of 1927 :

April 20 None

" 30 i

May 10 3

" 20 6

" 30 7

June 10 50

" 20 114

" 30 212

REPORT OF THE DIRECTOR. K)

July 10 247

" 20 247

" 30 245

August 10 234

" 20 208

30 168

September 10 no

20 50

30 12

October 10 8

20 2

30 2

November 10 None

2. The Nciu Dormitory and Apartment House. — In the report
of the Directors for 1926 mention was made of the generous gift
of $250,000 by the General Education Board for the purpose of
erecting a Dormitory and an Apartment House. Both of these
buildings were completed, except for certain minor details, early
in June, 1927. The first family moved into the Apartment House
on June I and the first persons into the Dormitory a few days
later. From that time until early in September the buildings
were occupied to almost their full capacity and were not entirely
vacant until the first of November.

The new buildings fill admirably the long-felt need of the Lab-
oratory for suitable accommodations for investigators with fam-
ilies. Together they provide six large apartments, fully equipped
for housekeeping, each consisting of a living-room, two bed-rooms,
screened porch, kitchenette, and bath ; two smaller apartments of
similar character but with only one bed-room and without a porch ;
nine suites of two rooms with bath ; sixteen double rooms which
can be combined in various ways into suites ; nineteen other double
rooms and eighteen single rooms. Each of the rooms which is
not part of a suite containing a bath is provided with hot and
cold running water. For the use of those persons who do not
occupy the furnished apartments there is provided in each of the
buildings a large and comfortable social room and in the basement
facilities for laundry work and simple cooking. An especially
valuable feature of the Apartment House is that it can be heated,

2O

MARINE BIOLOGICAL LABORATORY.

3
O

<U

rt
a

REPORT OF THE DIRECTOR.

21

o

o
Q

H

22 MARINE BIOLOGICAL LABORATORY.

thus making it possible for the working season of the Laboratory
to be extended into the colder months of the year.

According to the original plans, the Apartment House was
to be used for investigators and the Dormitory for students, but
the experience of the first year of operation has shown that the
demand from the investigators alone will be more than sufficient to
fill both buildings, and, on the whole, it seems fairer to give pref-
erence in the assignments to this class of workers, both because
their stay at the Laboratory is longer and because they have
greater difficulty on account of their families in finding suitable
accommodations in the village. The usefulness of the new build-
ings to investigators with families is indicated by the fact that
during the first year of their operation they accommodated a total
of 38 children.

The following is a classification of the occupants of the two
buildings for 1927:

Dormitory Apartment House Total

Investigators 52 28 So

Members of familes of inves-
tigators 47 41 88

Students 7 7

Total 106 69 175

The cost of the buildings, in spite of various unforseen diffi-
culties which arose during their construction has to the time of
this report come well within the appropriation of $250,000,
though a certain amount of interior painting still remains to be
done. This very favorable financial showing would have been im-
possible, however, without the generous assistance of Dr. Frank
R. Lillie in connection with the grading of the grounds and the
planting of shrubbery which have added greatly to the appearance
of the buildings. The following figures taken with slight sim-
plification from the Auditors' Report for 1927 show the status
of the building fund on December 31, 1927:

REPORT OF THE DIRECTOR. 23

ANALYSIS OF COST OF DORMITORY BUILDING
AND APARTMENT HOUSE

Dormitory Apartment House

General Contractor $ 90,016.70 $ 72,086.88

Plumbing 16,451.00 13,170.00

Electrical Work 4,657.98 5,378-73

Heating 649.00 5,800.00

All other Building Expenses... 14,806.75 13,185.18

Total Building Expenses . .$126,581.43 $109,620.79

Furnishings 7,266.26 6,067.33

Total $133,847.69 $115,688.12

Summary

Total New Dormitory $126,581.43

Apartment House 109,620.79 $236,202.22

New Dormitory Equipment .... 7,266.26

Apartment House Equipment . . 6,067.33 13-333-59

Total Cost to Dec. 31, 1927 $249,535.81

Accounted for as follows :

Gifts of General Education

Board $250,000.00

Less Cash on Hand 2,898.31

$247,101.69

Gift of Dr. Frank R. Lillie

for Grading, etc 1,500.00

Payments from Current
Funds 25I-57

Unpaid Balance of Archi-
tects' Commission 682.55 $249,535.81

3. Other Buildings. — Among the additions to the permanent
equipment of the Laboratory during 1927, one which would have
received more notice than it has in a year not marked by such
extensive building operations is the new carpenter shop and boat
house, situated beside the Eel Pond near the Marine railway.

24 MARINE BIOLOGICAL LABORATORY.

The erection of this building involved the previous preparation
of the site by extensive rilling with material from the excavations
for the Dormitory and the construction of a substantial retaining
wall. The building itself measures 66 by 44 feet and is two
stories in height. The first story includes a carpenter shop 33
by 44 feet and a boat shop of the same size ; the latter is on the
side of the building adjacent to the Marine railway and is pro-
vided with doors of sufficient size to admit large boats. The
second story consists of a single large room, used at present
chiefly for storage purposes but suitable for meeting a variety
of needs which may arise in the future. The cost of the building
with the retaining wall was approximately $7,500.

While the new Dormitory and Apartment House were under
construction the Dexter House, which for some years has ac-
commodated a large number of our students and younger in-
vestigators was removed. Many of its former occupants were
cared for during the past summer in the new buildings and an
additional number in the Drew House, whose capacity has been
considerably increased by a more economical use of space, and
which is now used exclusively as a men's dormitory.

4. The Report of the Treasurer shows an increase in the total
assets of the Laboratory from $2,281,219.79 in 1926 to $2,449,-
624.06 in 1927, the largest single item in this increase being ac-
counted for by the completion during the past year of the new
dormitory and apartment house buildings. The income for 1927
was $193,707.42 and that shown in the Auditors' Report for 1926
was $187,979.11 — an apparent increase of approximately $6000.
The actual increase, however, was in reality nearly $16,000, since
the gift of $10,600 from the General Education Board in 1926
was listed under income and the corresponding gift of $15,000
in 1927 under current surplus. By the same system of book-
keeping as that adopted in 1927 the income for the previous year
would be $177,979.11 instead of the figure mentioned above.
The expenses for the two years were $180,182.80 in 1926 and
$193,798.48 in 1927, giving apparent deficits of $2,203.69 and
$91.06, respectively. These deficits are at present on paper only
and are due to the depreciation charged against the plant which
the Treasurer discusses in his report. It may be mentioned in

REPORT OF THE DIRECTOR. 25

this connection that no charge for depreciation was made for 1927
in the case of the new Dormitory and Apartment House build-
ings, since some work still remained to be done upon them and
the construction account had not therefore been closed into the
general plant account. In future years, however, an additional
annual depreciation charge of $5,000 on these buildings will, un-
less additional sources of income are found, still further increase
the unfavorable balance of the past two years. In this connection
it is well to remember that depreciation in the case of the Lab-
oratory buildings occurs during twelve months of the year while
most of the income derived from them is at present limited to
three months. This is an additional reason for making every
effort to increase the length of the active season of the Laboratory.

5. The Report of the Librarian shows a very substantial in-
crease in our library facilities during the past year. The greater
part of this increase has been made possible by a second install-
ment of $15,000 from the gift of $50,000 appropriated by the
General Education Board in 1925. This sum was used chiefly
for the purchase of back sets of serials and the Librarian was
fortunate in being able to fill some of the most serious of the
existing gaps in these sets. Another noteworthy addition to the
library during 1927 was the valuable collection on Coelenterates
of the late Professor Charles W. Hargitt, generously presented to
the Laboratory by his son, Professor George T. Hargitt. The
total number of bound volumes on our shelves at present is over
22,000 and of reprints over 43,000. Of the former, approxi-
mately 4,000 and of the latter approximately 5.000 were acquired
during the past year. The number of current serials regularly
received by the Library is now nearly 800.

6. The Loeb and Gardiner Memorial Tablets. — On August 4
there was unveiled a tablet to Jacques Loeb, whose work, carried
on at Woods Hole over a period of more than twenty years, has
been one of the outstanding contributions to science of the Ma-
rine Biological Laboratory. Addresses were delivered in this oc-
casion by Professor Frank R. Lillie, Doctor Simon Flexner and
Professor Hardolph Wasteneys.

The tablet bears the following inscription :

26 MARINE BIOLOGICAL LABORATORY.

JACQUES LOEB

1859—1924

BRAIN PHYSIOLOGY

TROPISMS, REGENERATION

ANTAGONISTIC SALT ACTION

ARTIFICIAL PARTHENOGENESIS

DURATION OF LIFE
COLLOIDAL BEHAVIOR

A similar tablet was unveiled on September 10 to Edward
Gardiner Gardiner, one of the founders of the Laboratory and
until his death one of its most loyal supporters. Addresses were
delivered by Professor E. G. Conklin and Professor Frank R.
Lillie.

The tablet is inscribed as follows :

EDWARD GARDINER GARDINER

ZOOLOGIST
INCORPORATOR OF THE MARINE BIOLOGICAL

LABORATORY
AND MEMBER OF THE ORIGINAL BOARD OF

TRUSTEES
FOR MANY YEARS CLERK OF THE CORPORATION AND

SECRETARY OF THE BOARD OF TRUSTEES
A MAN WHOSE FINE SENSE OF HONOR AND LOYALTY

SHONE IN HIS DEEDS
A SCIENTIST TO WHOM THE SUCCESS OF THE

INSTITUTION

WAS THE FULFILLMENT OF HIS LIFE INTEREST

1 854 --1907

/. Changes in Personnel. — At the close of the courses in 10,27
the resignation of Dr. Ivey F. Lewis, who for twenty years has
been in charge of the Botany Course, was received and accepted
with regret. The valuable services to the Laboratory of Dr.
Lewis were recognized in the following resolution of the Board
of Trustees :

Voted: That the Board of Trustees learns with regret of the res-
ignation of Doctory Ivey F. Lewis from the headship of the course
in Botany at the Marine Biological Laboratory, and expresses to Doc-
tor Lewis its keen appreciation of his highly efficient and constructive
conduct of the work during his long service of twenty years.

REPORT OF THE DIRECTOR. 2/

Dr. Lewis' successor is Dr. W. R. Taylor, Professor of Bot-
any at the University of Pennsylvania, who both by his scien-
tific attainments and by his long experience with the work of the
course in excellently fitted for the position.

8. The Board of Trustees. — At the annual meeting of the
Board of Trustees held August 9, 1927, Professor B. M. Duggar
of the University of Wisconsin was elected to fill a vacancy in
the Class of 1931 of the Board.

There are appended as parts of this report :

i. The Staff, 1927.

3. Investigators and Students, 1927.

3. A Tabular View of Attendance, 1923-1927.

4. Subscribing and Cooperating Institutions, 1927.

5. Evening Lectures, 1927.

6. Members of the Corporation, August, 1927.

i. THE STAFF, 1927.

MERKEL H. JACOBS, Director, Professor of General Physiology

University of Pennsylvania.
Associate Director : — — .

I. INVESTIGATION.

GARY N. CALKINS, Professor of Protozoology, Columbia University.

E. G. CONKLIN, Professor of Zoology, Princeton University.

CASWELL GRAVE, Professor of Zoology, Washington University.

H. S. JENNINGS, Professor of Zoology, Johns Hopkins University.

FRANK R. LILLIE, Professor of Embryology, The University of Chi-
cago.

C. E. McCLUNG, Professor of Zoology, University of Pennsylvania.

S. O. MAST, Professor of Zoology, Johns Hopkins University.

T. H. MORGAN, Professor of Experimental Zoology, Columbia Uni-
versity.

G. H. PARKER, Professor of Zoology, Harvard University.

E. B. WILSON, Professor of Zoology, Columbia University.

LORANDE L. WOODRUFF, Professor of Protozoology, Yale University.

28 MARINE BIOLOGICAL LABORATORY.

II. INSTRUCTION.

J. A. DAWSON, Instructor in Zoology, Harvard University.
RUDOLF BENNITT, Instructor in Biology, Tufts College.
E. C. COLE, Assistant Professor of Zoology, Williams College.
T. H. BISSONNETTE, Professor of Biology, Trinity College.
MADELEINE P. GRANT, Assistant Professor of Zoology, Mount Holy-

oke College.
E. A. MARTIN, Assistant Professor of Zoology, College of the City

of New York.

A. E. SEVERINGHAUS, Instructor in Zoology, Columbia University.
DONNELL B. YOUNG, Associate Professor of Biology, University of

Arizona.

PROTOZOOLOGY.

I. INVESTIGATION.

(See Zoology.)

II. INSTRUCTION.

LORANDE Loss WOODRUFF, Professor of Protozoology, Yale Univer-
sity.

GARY N. CALKINS, Professor of Protozoology, Columbia University.
(Absent in 1927.)

MARY STUART MACDOUGALL, Professor of Zoology, Agnes Scott Col-
lege.

W. B. UNGER, Assistant Professor of Zoology, Dartmouth College.

MARY STUART MACDOUGALL, Professor of Zoology, Agnes Scott Col-
lege.

EMBRYOLOGY.

I. INVESTIGATION.

(Sec Zoology.)

II. INSTRUCTION.

HUBERT B. GOODRICH, Professor of Biology, Wesleyan University.

BENJAMIN H. GRAVE, Professor of Biology, Wabash College.

CHARLES PACKARD, Associate in the Institute of Cancer Research, Co-
lumbia University.

HAROLD H. PLOUGH, Professor of Biology, Amherst College.

CHARLES G. ROGERS, Professor of Comparative Physiology, Oberlin
College.

REPORT OF THE DIRECTOR. 2Q

PHYSIOLOGY.

I. INVESTIGATION.

HAROLD C. BRADLEY, Professor of Physiological Chemistry, Univer-
sity of Wisconsin.

WALTER E. GARREY, Professor of Physiology, Vanclerbilt University
Medical School.

RALPH S. LILLIE, Professor of General Physiology, The University
of Chicago.

ALBERT P. MATHEWS, Professor of Biochemistry, The University of
Cincinnati.

II. INSTRUCTION.

MERKEL H. JACOBS, Professor of General Physiology, University of
Pennsylvania.

WALLACE O. FENN, Professor of Physiology, University of Rochester.

LEONOR MICHAELIS, Professor in the University of Berlin and Resi-
dent Lecturer in the Johns Hopkins Medical School.

H. K. HARTLINE, Department of Physiology, Johns Hopkins Univer-
sity.

CHARLOTTE HAYWOOD, Department of Physiology, University of
Pennsylvania.

BOTANY.

I. INVESTIGATION.

B. M. DUGGAR, Professor of Plant Physiology, Washington University.

C. E. ALLEN, Professor of Botany, University of Wisconsin.

S. C. BROOKS, Department of Public Health, Washington, D. C.
WM. J. ROBBINS, Department of Botany, University of Missouri.
J. R. SCHRAMM, Editor-in-Chief; Biological Abstracts, University of
Pennsylvania.

II. INSTRUCTION.

IVEY F. LEWIS, Professor of Biology, University of Virginia.
WILLIAM RANDOLPH TAYLOR, Assistant Professor of Botany, Uni-
versity of Pennsylvania.
JAMES P. POOLE, Professor of Evolution, Dartmouth College.

LIBRARY.

PRISCILLA B. MONTGOMERY, (Mrs. Thomas H. Montgomery, Jr.), Li-
brarian.

KATHERINE UNDERWOOD, Assistant Librarian.
DEBORAH LAWRENCE, Secretary.

3O MARINE BIOLOGICAL LABORATORY.

CHEMICAL SUPPLIES.

OLIVER S. STRONG, Associate Professor of Neurology, Columbia Uni-
versity, Chemist.

APPARATUS ROOM.

SAMUEL E. POND, Assistant Professor of Physiology, Medical School,
University of Pennsylvania, Custodian of Apparatus.

SUPPLY DEPARTMENT.

GEORGE M. GRAY, Curator. A. W. LEATHERS, Head of Ship-
Assistant Curator : . ping Department.

JOHN J. VEEDER, Captain. A. M. HILTON, Collector.

E. M. LEWIS, Engineer. J. MC!NNIS, Collector.

F. M. MACNAUGHT, Business Manager.

HERBERT A. HILTON, Superintendent of Buildings and Grounds.

THOMAS LARKIN, Superintendent of Mechanical Department.

RAYMOND E. PHIPP, Mechanician.

WILLIAM HEMENWAY, Carpenter.

ARNOLD H. Bisco, Storekeeper and Head Janitor.

2. INVESTIGATORS AND STUDENTS, 1927.

Independent Investigators.

ADDISON, W. H. F., Professor of Normal Histology and Embryology, University
of Pennsylvania.

ALLEE, W. C., Associate Professor of Zoology, University of Chicago.

ALLEN, EZRA, Research Associate, Carnegie Institution of Washington.

AMBERSON, WILLIAM R., Assistant Professor of Physiology, University of Penn-
sylvania.

ARMSTRONG, PHILIP B., Instructor in Anatomy, Cornell University Medical College.

AUSTIN, MARY L., Lecturer in Zoology, Barnard College.

BAITSELL, GEORGE A., Associate Professor of Biology, Yale University.

BAKER, LILLIAN E., Assistant in Department of Experimental Surgery, Rockefeller
Institute for Medical Research.

BELLING, JOHN, Investigator, Carnegie Institution of Washington.

BENNITT, RUDOLF, Instructor in Zoology, Tufts College.

BERRILL, N. J., Assistant of Zoology Department, University College, London,
England.

BIGELOW, ROBERT P., Professor of Zoology and Parasitology, Massachusetts
Institute of Technology.

BISSONNETTE, THOMAS HUME, Professor of Biology, Trinity College.

REPORT OF THE DIRECTOR. 3!

BOWEN, ROBERT H., Assistant Professor of Zoology, Columbia University.

BRADLEY, HAROLD C., Professor of Physiological Chemistry, University of Wis-
consin.

BREITENBENBECHER, JOSEPH K., Lecturer in Zoology, McGill University.

BRIDGES, CALVIN B., Research Assistant, Carnegie Institution of Washington.

BRONFENBRENNER, JACQUES J., Associate Member, Rockefeller Institute.

BRONK, DETLEV W., Assistant Professor of Physiology and Biophysics, Swarthmore
College.

BROOKS, MATILDA MOLDENHAUER, Associate Biologist, Hygienic Laboratory,
Washington.

BROOKS, SUMNER C., Professor of Physiology and Biochemistry, Rutgers Uni-
versity.

BUDINGTON, ROBERT A., Professor of Zoology, Oberlin College.

CAMPBELL, CLARENCE JAMES, Assistant Professor of Physiology, Syracuse Uni-
versity.

CAROTHERS^ ELEANOR E., Lecturer in Zoology, University of Pennsylvania.

CATTELL, MCKEEN, Instructor in Physiology, Cornell University Medical College.

CATTELL, WARE, Research Fellow in Biology, Memorial Hospital.

CHAMBERS, ROBERT, Professor of Microscopic Anatomy, Cornell University
Medical College.

CHEN, C. C., Professor of Biology, Shanghai College.

CHIDESTER, FLOYD EARLE, Professor of Zoology, West Virginia University.

CHRISTIE, JESSE R., Associate Nematologist. U. S. Department of Agriculture,
Bureau of Plant Industry.

CLARK, ELEANOR LINTON, Private Investigator, University of Pennsylvania.

CLARK, ELIOT ROUND, Professor of Anatomy, University of Pennsylvania, Medical
Department.

CLOWES, G. H. A., Director of Lilly Research Laboratory, Eli Lilly & Co.

COBB, NATHAN A., Technologist and Nematologist, U. S. Department of Agri-
culture, Washington, D. C.

COHEN, BARNETT, Chemist, Hygienic Laboratory, Washington, D. C.

COHN, EDWIN J., Assistant Professor of Physical Chemistry, Harvard Medical
School.

COLE, ELBERT C., Assistant Professor of Biology, Williams College.

COLE, KENNETH, National Research Fellow, Harvard University.

CONKLIN, EDWIN G., Professor of Biology, Princeton University.

COPELAND, MANTON, Professor of Biology, Bowdoin College.

CORDIER, DR. ROBERT, Assistant Professor in Histology, University of Brussels,
Brussels, Belgium.

COVELL, WALTER P., Associate, Rockefeller Institute.

COWDRY, E. V., Associate Member, Rockefeller Institute.

COWLES, R. P., Associate Professor of Zoology, Johns Hopkins University.

CRABB, EDWARD D., Instructor in Zoology, University of Pennsylvania.

CRAMPTON, HENRY E., Professor of Zoology, Barnard College, Columbia Uni-
versity.

CROCKER, WILLIAM, Managing Director, Boyce Thompson Institute for Plant
Research.

CURTIS, WINTERTON C., Professor of Zoology, University of Missouri.

DARBY, HUGH HACKLAND, Instructor, New York University.

DAVVSON, JAMES A., Instructor in Zoology, Harvard University.

32 MARINE BIOLOGICAL LABORATORY.

BELLINGER, S. C., Professor of Zoology, University of Arkansas.

DISALVO, MRS. BEATRIX, Assistant Teacher, Biology Department, George Wash-
ington High School, New York.

DOLLEY, WILLIAM L., JR., Professor of Biology, University of Buffalo.

DONALDSON, HENRY H., Professor of Neurology, The Wistar Institute of Anatomy
and Biology.

DREW, KATHLEEN MARY, Lecturer in Botany, The Victoria University of Man-
chester, England.

DUGGAR, B. M., Professor of Plant Physiology, Missouri Botanical Garden and
Washington University.

DURRANT, EDWIN POE, Assistant Professor of Physiology, Ohio State University.

EDWARDS, DAYTON J., Associate Professor of Physiology, Cornell University
Medical College.

EMMART, EMILY WALCOTT, Associate Professor in Biology, Western Maryland
College.

ESAKI, SHIRO, Department of Zoology, University of Chicago.

FARR, CLIFFORD H., Associate Professor, Washington University.

FENN, WALLACE O., Professor of Physiology, Rochester University. Medical School.

FRY, HENRY J., Assistant Professor, Washington Square College.

GARREY, W. E., Professor of Physiology, Vanderbilt University Medical School.

GATES, FREDERICK L., Associate Member, Rockefeller Institute for Medical
Research.

GLASER, OTTO, Professor of Biology, Amherst College.

GLASER, RUDOLF W., Associate Member, Rockefeller Institute for Medical Re-
search.

GOLDFORB, A. J., Professor of Biology, College of the City of New York.

GOODRICH, H. B., Professor of Biology, Wesleyan University.

GORDON, ISABELLA, n Balloch Road, Keith, Banffshire, Scotland.

GRAHAM, JOHN Y., Professor of Biology, University of Alabama.

GRANT, MADELEINE P., Assistant Professor, Mount Holyoke College.

GRAVE, BENJAMIN H., Professor of Zoology, Wabash College.

GRAVE, CASWELL, Professor of Zoology, Washington University.

GRUENBERG, BENJAMIN O., Managing Director, American Association for Medical
Progress.

HAGUE, FLORENCE, Assistant Professor of Biology, Sweet Briar College.

HALL, RICHARD P., Assistant Professor, New York University.

HANCE, ROBERT T., Associate, Rockefeller Institute for Medical Research.

HANN, HARRY W., Instructor in Embryology, University of Illinois.

HARTLINE, H. KEFFER, Johns Hopkins Medical School.

HARVEY, ETHEL BROWNE, Princeton, New Jersey.

HARVEY, E. NEWTON, Professor of Physiology, Princeton University.

HAYWOOD, CHARLOTTE, Graduate Student, University of Pennsylvania.

HECHT, SELIG, Associate Professor of Biophysics, Columbia University.

HEILBRUNN, L. V., Assistant Professor of Zoology, University of Michigan.

HEYROTH, FRANCIS F., Research Fellow, Harvard University Medical School.

HIBBARD, HOPE, Preparateur, The Sorbonne, Paris, France.

HILLER, S., Assistant, Biological Laboratory, Cracow, Poland.

HOADLEY, LEIGH, Assistant Professor, Brown University.

HOSKINS, MARGARET MORRIS, Assistant Professor of Microscopic Anatomy, New
York University Dental College.

Kl I'ORT OF THE DIRECTOR. 33

HOSKINS, R. G., Research Associate in Physiology, Harvard University Medical
School.

HOWE, H. E., Editor, American Chemical Society.

HOWE, THOMAS D., Instructor in Biology, James Millikin University.

HOWLAND, RUTH B., Assistant Professor of Biology, New York University.

HUETTNER, ALFRED F., Assistant Professor of Zoology, Columbia University.

HLIGGINS, JOHN R., Assistant Instructor, University of Pennsylvania.

HUGHES, THOMAS P., Assistant, Rockefeller Institute for Medical Research.

INMAN, ONDESS L., Professor of Biology, Antioch College.

IRWIN, MARIAN, Associate in General Physiology, Rockefeller Institute.

JACOBS, MERKEL H., Professor of General Physiology, University of Pennsylvania.

JENNINGS, H. S., Professor and Director of the Zoological Laboratory, John Hopkins
University.

JOHLIN, J. M., Associate Professor of Biochemistry, Vanderbilt Medical School.

JUST, E. E., Professor of Zoology, Howard University.

KAUFMANN, BERWIND P., Professor of Biology, Southwestern, Memphis, Tennessee.

KEEFE, Rev. ANSELM M., Professor of Biology, St. Norbert's College.

KINDRED, JAMES E., Associate Professor of Histology and Embryology, University
of Virginia.

KLEINER, ISRAEL S., Professor and Head of the Department of Chemistry, New
York Homoeopathic Medical College.

KNOWLTON, FRANK P., Professor of Physiology, College of Medicine, Syracuse
University.

KUNITZ, MOSES, Associate, Rockefeller Institute for Medical Research.

LANCEFIELD, D. E., Assistant Professor in the Zoology Department, Columbia
University.

LANCEFIELD, REBECCA C., Assistant in Bacteriology, Rockefeller Institute.

LANDIS, EUGENE M., University of Pennsylvania.

LEE, MILTON O., Research Associate, Harvard Medical School.

LEWIS, IVEY F., Professor of Biology, University of Virginia.

LILLIE, FRANK R., Chairman of the Depaitment of Zoology, University of Chicago.

LILLIE, RALPH S., Professor of General Physiology, University of Chicago.

LOEB, LEO, Professor of Pathology, Washington University Medical School.

LOWTHER, FLORENCE DEL., Assistant Professor of Zoology, Barnard College.

LUCAS, CATHERINE L. T., Travelling Fellow, London University.

LUCKE, BALDWIN, Assistant Professor of Pathology, University of Pennsylvania.

LYNCH, RUTH S., Instructor, The Johns Hopkins University.

McCLENDON, J. F., Professor of Physiological Chemistry, University of Minnesota.

McCLUNG, CLARENCE E., Director of Zoological Laboratory, University of Penn-
sylvania.

McCuTCHEON, MORTON, Assistant Professor of Pathology, University of Penn-
sylvania.

MACDOUGALL, MARY STUART, Professor of Zoology, Agnes Scott College.

MANWELL, R. D., Rockefeller Special Fellow, School of Hygiene and Public Health.

MARTIN, EARL A., Assistant Professor, College of the City of New York.

MAST, S. O., Professor of Zoology, Johns Hopkins University.

MATHEWS, ALBERT P., Professor of Biochemistry, University of Cincinnati.

MAYOR, JAMES W., Professor of Biology, Union College.

MAY, DR. R. M., Research Fellow, American Field Service Fellowship.

METCALF, MAYNARD M., Research Associate in Zoology, Johns Hopkins University.

34 MARINE BIOLOGICAL LABORATORY.

METZ, CHARLES W., Member Staff, Department of Genetics, Carnegie Institution
of Washington.

MICHAELIS, LEONOR, Resident Lecturer in Medical Research, Johns Hopkins
University School of Medicine.

MITCHELL, PHILIP H., Professor of Physiology, Brown University.

MORGAN, LILIAN V., 409 West 117 Street, New York City, N. Y.

MORGAN, T. H., Professor of Experimental Zoology, Columbia University.

MORGULIS, SERGIUS, Professor of Biochemist:y, University of Nebraska.

MORRILL, CHARLES V., Associate Professor of Anatomy, Cornell University
Medical College.

NEWMAN, H. H., Professor of Zoology, University of Chicago.

NOBLE, G. KINGSLEY, Curator, American Museum of Natural History, New York.

NONIDEZ, JOSE F., Associate in Anatomy, Cornell University Medical College.

PACKARD, CHARLES, Associate, Institute of Cancer Research, Columbia University.

PARKER, GEORGE HOWARD, Professor of Zoology, Harvard University.

PARMENTER, CHARLES L., Assistant Professor of Zoology, University of Pennsyl-
vania.

PATTERSON, WILLIAM MORRISON, University Club, 5th Avenue and 54th Street,
New York City.

PEARSE, A. S., Professor of Zoology, Duke University.

PERLZWEIG, W. A., Associate in Medicine, Johns Hopkins University Medical
School.

PHELPS, LILLIAN A., Instructor in Zoology, Cornell University.

PINNEY, MARY EDITH, Professor of Zoology, Milwaukee-Downer College.

PLOUGH, HAROLD H., Professor of Biology, Amherst College.

POND, SAMUEL E., Assistant Professor of Physiology, School of Medicine, Uni-
versity of Pennsylvania.

POOLE, JAMES PLUMMER, Professor of Evolution, Dartmouth College.

RAND, HERBERT W., Associate Professor of Zoology, Harvard University.

REDFIELD, A. C., Assistant Professor of Physiology, Harvard Medical School.

REDFIELD, HELEN, National Research Fellow in Zoology, Columbia University.

DE RENYI, GEORGE ST., Assistant Professor in Department of Anatomy, Uni-
versity of Pennsylvania.

REZNIKOFF, PAUL, Associate in Anatomy, Cornell University Medical College.

RICE, KENNETH S., Brown University.

RICHARDS, A., Professor of Zoology, University of Oklahoma.

RICHARDS, MILDRED HOGE, University of Oklahoma.

RINGOEN, ADOLPH R., Assistant Professor of Zoology, University of Minnesota.

ROBBINS, WILLIAM J., Professor of Botany, University of Missouri.

ROGERS, CHARLES G., Professor of Comparative Physiology, Oberlin College.

ROMER, ALFRED S., Associate Professor of Vertebrate Paleontology, University of
Chicago.

SANDISON, JAMES CALVIN, Instructor in Anatomy, University of Pennsylvania.

SAYLES, LEONARD PERKINS, Assistant Professor of Biology, Norwich University.

SCHAEFFER, ASA ARTHUR, Professor of Zoology, University of Kansas.

SCHMITT, FRANCIS O., National Research Fellow, Washington University.

SCHRADER, FRANZ, Associate Professor, Bryn Mawr College.

SCHRADER, SALLY HUGHES, Instructor, Bryn Mawr College.

SCHULTZ, JACK, National Research Fellow, Columbia University.

SCOTT, MIRIAM J., University of Pennsylvania.

REPORT OF THE DIRECTOR. 35

SEVERINGHAUS, AURA E., Instructor in Anatomy, Columbia University.
SHAFTESBURY, ARCHIE D., Associate Professor of Zoology, The North Carolina

College for Women.

SIMPSON, GEORGE E., Assistant Professor, University of Pennsylvania.
SMITH, SEPTIMA CECILIA, Fellow in Medical Zoology, Johns Hopkins School of

Hygiene.

SMITH, WILBUR A., Assistant, University of Pennsylvania.
SRIBYATTA, DR. L., Instructor in Physiology, Chulalongkora University, Medical

School, Bangkok, Siam.
STARK, MARY B., Professor of Embryology and Histology, New York Homoeopathic

Medical College.

STEGGERDA, F. R., Teaching Fellow, University of Minnesota.
STIER, T. J. B., Graduate Student, Harvard University.

STOCKARD, CHARLES R., Professor of Anatomy, Cornell University Medical College.
STOKEY, ALMA G., Professor of Botany, Mount Holyoke College.
STRONG, OLIVER S., Professor of Neurology and Neuro-Histology, Columbia

University.

STUNKARD, HORACE W., Professor of Biology, New York University.
STURTEVANT, A. H., Member of Staff, Carnegie Institution.
SVVETT, F. H., Associate Professor of Anatomy, Vanderbilt School of Medicine.
TAYLOR, W. RANDOLPH, Professor of Botany, University of Pennsylvania.
TENNENT, DAVID H., Professor of Biology, Bryn Mawr College.
TERAO, ARATA, Professor of Zoology, Imperial Fisheries Institute, Tokyo, Japan.
TRACY, HENRY C., Professor, University of Kansas.

TURNER, ABBY HOWE, Professor of Physiology, Mount Holyoke College.
UHLENHUTH, EDUARD, Associate Professor in Anatomy, University of Maryland

Medical College.

UNGER, W. BYERS, Assistant Professor of Zoology, Dartmouth College.
VISSCHER, J. PAUL, Associate Professor of Biology, Western Reserve University.
WARREN, HOWARD C., Professor of Psychology, Princeton University.
WEECH, ALEXANDER ASHLEY, Instructor in Research Medicine, Johns Hopkins

University.

WENRICH, D. H., Assistant Professor of Zoology, University of Pennsylvania.
WHITAKER, DOUGLAS M., Assistant in Zoology and Graduate Student, Stanford

University.

WILSON, J. WALTER, Assistant Professor of Biology, Brown University.
WOLF, ERNST, University of Heidelberg, Germany.
WOLF, E. ALFRED, Instructor in Zoology and Comparative Physiology, University

of Pittsburgh.

WOODRUFF, LORANDE Loss, Professor of Protozoology, Yale University.
WOODWARD, ALVALYN E., Associate Professor, University of Maine.
WRIGHT, SEWALL, Associate Professor, University of Chicago.
WYMAN, JEFFRIES, JR., Instructor and Tutor in Biology, Harvard University.
YAGI, NOBUMASA, Assistant Professor of Entomology, Kyoto Imperial University,

Kyoto, Japan.
YOUNG, DONNELL BROOKS, Professor of Biology and Head of Biology Department,

University of Arizona.

Beginning Investigators.
1927.

ALLEN, ELEANOR, Graduate Student, Brown University.

36 MARINE BIOLOGICAL LABORATORY.

ARNOLD, CONSTANCE W., Demonstrator, Brown University.

EARTH, L. G., Graduate Assistant in Zoology, University of Michigan.

BASKERVILL, MARGARET, Adjunct Professor, University of Texas, Medical School.

BLUMENTHAL, REUBEN, Graduate Student, University of Pennsylvania.

CANAVAN, WILLIAM P., Instructor in Zoology, University of Pennsylvania.

CARPENTER, ESTHER, Assistant in Zoology Department, University of Wisconsin.

CARVER, GAIL L., Professor of Biology, Mercer University.

CHEER, SHEO-NAN, Fellow of the Rockefeller Foundation, Peking Union Medical
College, Peking, China.

CLARK, L. B., Graduate Student, Johns Hopkins University.

CRAWFORD, WILEY W., Fellow, University of Missouri.

ELFTMAN, HERBERT, Assistant in Zoology, Columbia University.

FISH, H. D., Student Investigator, Columbia University.

FREEMAN, LEO BOYES, University of Pennsylvania.

GOODKIND, ROBERT, Student, Harvard University Medical School.

GRISWOLD, SYLVIA M., Instructor of Botany and Bacteriology, Pennsylvania
College for Women.

GRUNDFEST, HARRY, University Fellow, Columbia University.

HADLEY, CHARLES E., Harvard University.

HOP, ANNE, Student, Radcliffe College.

HOFKESBRING, ROBERTA, Instructor in Physiology, Tulane University.

HOLMES, GLADYS E., Graduate Assistant, Brown University.

JOHNSON, Percy L., Graduate Assistant, Johns Hopkins University.

KAPP, ELEANOR M., Assistant in Biology, New York University.

KLEIN, H., University of Pennsylvania.

KOEHRING, VERA, Fellow, University of Pennsylvania.

KROPP, BENJAMIN, Graduate Student, Harvard University.

LIGHT, V. EARL, Student Technician, Johns Hopkins University.

LUCAS, ALFRED M., Instructor, Washington University.

LUCAS, EMILIO R., Instructor, University of Kansas.

Lu, HWEI-LING, Graduate Student of Zoology, Columbia University.

McCARDLE, Ross CLAYTON, University of Michigan.

MACNAB, ALLEYNE, Technician in Department of Experimental Surgery, Rocke-
feller Institute for Medical Research.

MATTHEWS, SAMUEL A., Student, Harvard University.

MITCHELL, WILLIAM HINCKLEY, JR., Thayer Fellow, Harvard University.

MONTGOMERY, HUGH, Student, Harvard University Medical School.

MORRISON, MARY ELINOR, University of Pennsylvania.

MOSES, MILDRED S., Research Assistant, Carnegie Institution of Washington.

NELSON, OLIN E., Instructor in Zoology, University of Pennsylvania.

NOMURA, SHICHIROKU, Assistant Professor of Zoology, Tohoku Imperial University.

PIERCE, MADELENE E., Graduate Student, Radcliffe College.

POLLACK, HERBERT, Cornell Medical College.

RITTER, RAYMOND A., Assistant in Zoology, University of Missouri.

RUNYON, ERNEST H., Instructor, Washington University.

SEARS, MARY, Research Student, Radcliffe College.

SHLAER, SIMON, Student, Columbia University.

SHOUP, CHARLES S., Assistant in Instruction, Princeton University.

SICHEL, FERDINAND J. M., Student, McGill University, Montreal, Canada.

SMITH, GEORGE HUME, Instructor, University of Illinois.

REPORT OF THE DIRECTOR. 37

SONNEBORN, TRACY MORTON, Graduate Student, Johns Hopkins University.

STEEN, EDWIN B., Instructor in Zoology, Wabash College.

STEWART, DOROTHY R., Instructor in Biology, Lake Erie College.

SUMWALT, MARGARET, Instructor, University of Pennsylvania.

TAFT, CHARLES H., JR., Student, Columbia University.

TAYLOR, MRS. JEAN GRANT, 3454 N. 23d Street, Philadelphia, Pennsylvania.

TITLEBAUM, Albert, Assistant in Zoology, Columbia University.

WILLEY, CHARLES H., Instructor in Biology, New York University.

YOUNG, R. A., Assistant Professor of Zoology, Howard University.

RESEARCH ASSISTANTS— 1927

ARZBERGER, E. G., Pathologist, Bureau of Plant Industry, U. S. Department of

Agriculture.

BARTHOLOMEW, THOMAS HAYWARD, Columbia University.
BARTHOLOMEW, WILLIAM WEST, Columbia University.
DOWNING, R. C., Student, Wabash College.

FIELD, MADELEINE E., Assistant in Physiology, Mount Holyoke College.
GENTHER, IDA T., Graduate Assistant in Zoology, University of Wisconsin.
GREENE, EUNICE CHASE, Medical School, Syracuse University.
HANSEN, IRA B., Assistant in Zoology, Wesleyan University.
HARROP, GEORGE A., JR., Associate Professor of Medicine, Johns Hopkins Medical

School.
HIDALGO, FRANCISCO, Technical Assistant, Rockefeller Institute for Medical

Research.

HOLMES, W. C., Rockefeller Institute for Medical Research.
HOSKINS, FRANCES, Research Assistant, Columbia University.
JOHNSON, ROSVVELL HILL, Columbia University.
KALTREIDER, NOLAN L., Swarthmore College.

KELTCH, Anna K., Research Assistant, Lilly Research Laboratory, Indianapolis.
LORBERBLATT, ISAAC, Chemist, Harriman Research Laboratory, New York.
MCNAMARA, HELEN, Technician, Rockefeller Institute.
MORGAN, EDITH, 409 West nyth Street, New York City.
PARPART, ETHEL ROBERTA, Assistant, Amherst College.
REYNOLDS, SAMUEL R. M., Assistant in Physiology and Zoology, Swarthmore

College.
RIOCH, DAVID MCKENZIE, Instructor in Medicine, University of Rochester,

Medical School.

SANDERS, GERTRUDE B., Swarthmore College.
SCHAUFFLER, WILLIAM GRAY, Private Practitioner of Medicine, Princeton, New

Jersey.

ULLIAN, SILKA STOCKER, Research Assistant, Carnegie Institution of Washington.
WALDEN, EDA B., Research Assistant, Lilly Research Laboratory.
WALLACE, EDITH M., Artist and Research Assistant, Carnegie Institution of

Washington.

WATERMAN, HARRIET C., Research Assistant, Carnegie Institution of Washington.
WEARE, J. H., Research Assistant, Harvard University Medical School.

38 MARINE BIOLOGICAL LABORATORY.

STUDENTS

Botany.

BAKER, CAROLYN, Bellair Drive, Dobbs Ferry, New York.

BOWERS, W. B., Student, Harvard University.

DUNBAR, FRANCIS F., Student, Harvard University.

FORT, IRENE, University of Pennsylvania.

HOPPAUGH, KATHERINE W., 1176 East South Temple Street, Salt Lake City. Utah.

HUSTED, DON L., Student, Oberlin College.

JEWETT, FRANCIS L., 273 Woodland Road, Ravinia, Illinois.

KEITH, BERNICE, Hastings, Nebraska.

MACFARLANE, CONSTANCE, 87 Upper Prince Street, Charlottetown, Prince Edward

Island, Canada.

McCLiNTOCK, BARBARA, Instructor in Botany, Cornell University.
NAYLOR, ERNST, Instructor in Botany, University of Missouri.
PATRICK, RUTH M., Student, Coker College.
PINSDORF, KATE, Smith College.
PYLE, THERESA P., Smith College.

WELLS, EVELYN CLARE, Teaching Fellow, University of Tennessee.
ZIMMERMANN, RUTH HELEN, Teacher, Brockton High School.

Embryology.

BAILY, JOSHUA L., JR., Institute for Biological Research, Baltimore, Md.

BAILEY, PERCY L., JR., Graduate Student, Brown University.

BEYER, KATHE M., 93 Benefit Street, Providence, Rhode Island.

BOSWORTH, EDWARD B., Assistant, Yale University.

BOUGHTON, ESTHER MARIE, Box 339, Poughkeepsie, New York.

CHASE, AURIN M., JR., Assistant, Amherst College.

CHEN, NELSON S., University of Pennsylvania.

CRANE, NORMAN F., Bowdoin College.

CURTIS, MARY ELIZABETH, Assistant in Biology, Wilson College.

DALTON, ALBERT JOSEPH, Wesleyan University.

DAVIDSON, MARGARET H., North Carolina College for Women.

DEICHMANN, ELIZABETH, Radcliffe College.

FLETCHER, LYDIA M., Brown University.

GRIZZLE, LUCILE A., University of Southern California.

HAMILTON, SALLY, Elmira College.

H.ARDESTY, MARY, Teaching Fellow in Biology, Newcomb College.

HARLAND, MARGARET, North Carolina College for Women.

HERSKOWITZ, ISIDOR A., Columbia University.

HIRAIWA, YOSHI KUNI, University of Chicago.

HOLLINSHEAD, WILLIAM HENRY, Instructor, Vanderbilt University.

JANSEN, JAN BIRGES, Royal Fredericks University, Oslo, Norway.

LICHTMAN, FRIEDA, Student, New York University.

LUCE, WILBUR M., University of Illinois.

McGouN, RALPH C., JR., Assistant, Amherst College.

MILLER, RUTH A., 13 Poplar Avenue, Woodlawn, Wheeling, West Virginia.

NABRIT, SAMUEL M., Instructor in Zoology, Morehouse College.

PARSONS, ELIZABETH H., Graduate Student, Oberlin College.

ROWELL, LYMAN S., Instructor, University of Vermont.

REPORT OF THE DIRECTOR. 39

TRACY, BARBARA, Connecticut College.

WATERMAN, ALLYN JAY, Graduate Assistant, Western Reserve University.

WEN, I., Medical College, Peking, China.

WOODARD, THOMAS M., JR., Instructor, Vanderbilt University.

Physiology.

BAHRS, ALICE M., Assistant in Physiology, University of California.
BARRON, E. G., 802 North Washington Street, Baltimore, Maryland.
BORQUIST, MAY, Research Fellow, Cornell Medical College.
CLARKE, ROBERT W., New York University.
DEBRUE, GEORGES H., Louvain University.

DOWNEY, HAROLD R., Student, Johns Hopkins Medical School.
FRANK, RICHARD L., Student, Cornell Medical College.
HENDERSON, JEAN T., Lecturer, McGill University, Montreal, Canada.
HOWLAND, ESTHER, 107 East 64th Street, New York City, N. Y.
LIGHT, FREDERICK W., JR., Student, Johns Hopkins Medical School.
MILLER, EVELYN H., Graduate Student, Stanford University, California.
NEWTON, ISABEL M., Assistant in Physiology, Mount Holyoke College.
OLCOTT, CHARLES T., Instructor in Pathology, Cornell Medical College.
PANKRATZ, DAVID S., Instructor, University of Kansas.
PARPART, ARTHUR R., Instructor, Amherst College.
STEELE, CHARLES W., University of Missouri.
TEWINKEL, HELEN, Assistant in Zoology, Oberlin College.

TURNER, EDNA M., Assistant in Biology, Washington Square College, N. Y. U.
UHLENHUTH, EDUARD, Associate Professor, University of Maryland, Medical
School.

Protozoology.

ADAMS, THEODORE G., College of the City of New York.

ALEXANDER, ELEANOR G., Graduate Student, Columbia University.

DEBONE, FRANCES M., Student Assistant in Anatomy, University of Pittsburgh.

DETTMER, CLARA ROWENA, Columbia University.

GOODLOW, SARA, Goucher College.

HETHERINGTON, WILLIAM A., Assistant in Zoology, Columbia University.

HOLLIDAY, GAIL H., Teacher of Biology, Wheeling High School.

HUBBARD, CATHERINE E., Cromwell, Connecticut.

KINNEY, ELIZABETH T., Graduate-assistant, University of Pittsburgh.

MORRIS, HELEN S., Graduate Student, Columbia University.

NELSON, GEORGE E., 3038 Hull Avenue, Bronx, New York City.

RICHTER, MARION C. R., Columbia University.

ROBERTSON, GEORGE, Instructor, Dartmouth College.

SHIELDS, LAWRENCE M., Instructor, Phillips Academy.

VAN RHYN, ELSIE A., Instructor in Biology, University of Porto Rico.

Wu, CHAO-FA, Assistant, University of Wisconsin.

ZIMMER, DOROTHY K., Columbia University.

Zoology.

ABELL, RICHARD G., Instructor in Biology, Hampton Institute.
ANDREWS, AVA LEE, North Carolina College for Women.
APGAR, GRACE M., University of Pennsylvania.

AQ MARINE BIOLOGICAL LABORATORY.

BALLARD, WILLIAM W., Student, Dartmouth College.

BEEBE, MARY ELIZABETH, Oberlin College.

BILSTAD, NELLIE MAE, Assistant, University of Wisconsin.

BLOUNT, RAYMOND F., Instructor, University of Arizona.

BOND, EVELYN, University of Pennsylvania.

BRADLEY, MARY A., Wabash, Indiana.

BROWN, DUGALD E. S., Instructor in Biology, New York University.

BUTLER, ELIZABETH, 257 Newbury Street, Boston, Massachusetts.

CLINE, ELSIE, Teacher, Baltimore Public Schools.

CLOUDMAN, ARTHUR M., Instructor, University of Vermont.

COLDWATER, KENNETH B., University of Missouri.

DRUMTRA, ELIZABETH, 5 Curran Avenue, Binghamton, New York.

ELFTMAN, HERBERT O., Assistant in Zoology, Columbia University.

ELLIS, MARJORIE F., Dalhousie University, Halifax, Nova Scotia.

FERRIS, FRANCES R., Assistant in Zoology, Washington University.

FRAME, ELIZABETH G., Dalhousie University, Halifax, Nova Scotia.

FURTOS, NORMA C., Graduate Assistant, Western Reserve University.

GREGG, WARD I., Harvard University, Cambridge, Mass.

GREGORY, PAUL W., Harvard University.

HALL, EDMUND K., Assistant, Yale University.

HAMPEL, CHESTER W., Wesleyan University.

HARE, LAURA, DePauw University.

HOPKINS, SEWELL H., William and Mary College.

HUSTED, CLARA M., Graduate Student, University of Rochester.

JOHNSON, PORTEOUS E., Amherst College.

KERRIGAN, ALICE M., Teacher of Biology, Teachers College of the City of Boston.

LANE, ELINOR M., Assistant in Biology, Goucher College.

LEONARD, SAMUEL L., Rutgers University.

LOVELL, HARVEY B., Harvard University.

MCCLURE, GEORGE Y., Student, Dartmouth College.

MCCLURE, KATHERINE L., Instructor in Biology, Morningside College, Sioux

City, Iowa.

MACCOY, CLINTON V., Harvard College.
MclNERNEY, KATHRYN M., Tufts College.
McNuTT, DOROTHEA, Illinois Wesleyan University.
MARTIN, STEVEN J., University of Wisconsin.

MARTINOVITCH, PETAR N., Graduate Student, Syracuse University.
MILLIKEN, ELEANOR, Wellesley College.
MOLINA, ANA M., University of Porto Rico.
NEWCOMER, A. VIRGINIA, Goucher College.
PFEIFER, KATHERINE M., Washington University.
PICKETT, NATHAN W., Wabash College.
PREFONTAINE, GEORGE H., Assistant in Biology, University of Montreal, Montreal,

Canada.

RECK, VIRGINIA D., Assistant in Biology, Yale University.
SCHMIDT, LORRIMER M., Treble Cove Road, North Billerica, Massachusetts.
SHINAR, CATHERINE, Hunter College.
SHOREY, DOROTHY E., Radcliffe College.
SMALL, VIRGINIA, Butler University.
SMELSER, GEORGE K., Studer';, Earlham College.

REPORT OF THE DIRECTOR. 4!

SNELL, GEORGE D., Graduate Student, Harvard University.

STABLER, ROBERT M., Swarthmore College.

STEHR, WILLIAM C., Assistant in Zoology, University of Minnesota.

SUN, TSON P., Wusih, Kiangsu, China.

TURNBULL, VIRGINIA E., Teacher, Dorchester High School for Girls.

WILDE, FRANCES, Student, Radcliffe College.

3. TABULAR VIEW OF ATTENDANCE.

1923 1924 1925 1926 1927

INVESTIGATORS — Total 176 194 207 252 294

Independent 126 124 135 156 209

Under Instruction 50 70 72 84 57

Research Assistants 12 28

STUDENTS — Total 146 134 132 141 141

Zoology 59 50 54 56 57

Protozoology 16 17 17 19 17

Embryology 31 29 29 28 32

Physiology 22 18 19 18 19

Botany 18 20 13 20 16

TOTAL ATTENDANCE 322 328 339 393 435

Less persons registered as both

Students and Investigators 8 I

385 434

INSTITUTIONS REPRESENTED — Total .. 107 no 112 119 in

By investigators 62 69 74 84 89

By students 73 68 65 60 63

SCHOOLS AND ACADEMIES REPRESENTED

By investigators I I

By students 4 4 4 4

FOREIGN INSTITUTIONS REPRESENTED

By investigators 17 15

By students 3 8

MARINE BIOLOGICAL LABORATORY

4. SUBSCRIBING AND COOPERATING
INSTITUTIONS, 1927.

Amherst College

Antioch College

Barnard College

Bowdoin College

Brown University

Bryn Mawr College

Butler College

C. R. B. Educational Foundation

Carnegie Institution, Cold Spring
Harbor

Carnegie Institution of Washing-
ton

Columbia University

Commonwealth Fund

Connecticut College

Cornell University

Cornell University Medical Col-
lege

Dalhousie University

Dartmouth College

De Pauw University

Duke University

Elmira College

T. W. Evans Dental Museum and
School of Dentistry

General Educational Board

Goucher College

Harvard University

Harvard University Medical

School

Howard University

Hunter College

Industrial & Engineering Chem-
istry, of the American Chem-
ical Society

International Education Board

Johns Hopkins University

Johns Hopkins University Med-
ical School

Knox College

Eli Lilly & Co.

Massachusetts Institute of Tech-
nology

McGill University

Morningside College

Mount Holyoke College

National Research Council

New York Homeopathic Medical
College

New York University

New York University Dental
School

North Carolina College for
Women

Norwich University

Oberlin College

Princeton University

Radcliffe College

Rockefeller Foundation

Rockefeller Institute for Medical
Research

Rutgers University

Smith College

Sophie Newcomb College

Southwestern

Swarthmore College

Tufts College

Union College

United States Dept. of Agricul-
ture

University of Alabama

University of Arkansas

University of Chicago

University of Illinois

University of Kansas

REPORT OF THE DIRECTOR.

43

University of

School

University of
University of
University of

School

University of
University of
University of

ical School
University of
University of
University of
University of

Maryland Medical

Michigan
Minnesota
Minnesota Medical

Missouri
Pennsylvania
Pennsylvania Med-

Pittsburgh
Rochester
Vermont
Virginia

University of Wisconsin
Vanderbilt University Medical

School

Vassar College
Wabash College
Washington University
Washington University Medical

School

Wellesley College
Wesleyan University
Western Reserve University
Wistar Institute of Anatomy and

Biology
Yale University

SCHOLARSHIP TABLES.

Ida H. Hyde Scholarship of the University of Kansas.

Lucretia Crocker Scholarships for Teachers in Boston.

Scholarship of $100.00 supported by a friend of the Laboratory since

1898.
The Edwin S. Linton Memorial Scholarship of Washington and

Jefferson College.

5. EVENING LECTURES, 1927.

Tuesday, June 28,

PROFESSOR S. O. MAST "Structure, Locomotion and Stim-
ulation in Ameba Proteus."
Friday, July I,

PROFESSOR L. V. HEILBRUNN "The Colloid Chemistry of Proto-
plasm."
Tuesday, July 5,

PROFESSOR ALEXANDER FORBES. . ."Implications of the All-or-none

Principle in the Physiology of
the Central Nervous System."
Friday, July 8,

PROFESSOR B. M. DUGGAR "The Physiological Behavior of

Certain Virus Agencies."
Tuesday, July 12,

PROFESSOR E. G. CONKLIN "Localization Phenomena in Em-
bryology."

44 MARINE BIOLOGICAL LABORATORY.

Friday, July 15,

DR. E. M. LANDIS "The Permeability of the Capil-
lary Wall."
Tuesday, July 19,

DR. JACQUES BRONFENBRENNER. . "The Nature of the Bacterio-

phage Phenomenon."
Friday, July 22,

PROFESSOR FRANK R. LILLIE "The Gene and the Ontogenetic

Process."
Tuesday, July 26,

PROFESSOR R. M. YERKES "The Psycho-biology of the Go-
rilla."
Friday, July 29,

DR. HENRY B. BIGELOW "Oceanographic Problems and

Opportunities."
Tuesday, August 2,

PROFESSOR J. C. DRUMMOND "The Chemistry of Vitamines."

Wednesday, August 3, (Special
Lecture)

PROFESSOR H. D. FISH "Life along the Trails and

Streams of British Guiana."
Friday, August 5,

DR. WILLIAM MANSFIELD CLARK. "A Restricted but New Approach

to Oxidation-reduction in the
Living Cell."

6. MEMBERS OF THE CORPORATION.

LIFE MEMBERS.

ALLIS, MR. E. P., JR., Palais Carnoles, Menton, France.
ANDREWS, MRS. GWENDOLEN FOULKE, Baltimore, Maryland.
BILLINGS, MR. R. C., 66 Franklin Street, Boston, Mass.
CLARKE, PROF. S. F., Williamstown, Mass.
CONKLIN, PROF. EDWIN G., Princeton University, Princeton,

New Jersey.

COOLIDGE, MR. C. A., Ames Building, Boston, Mass.
CRANE, MR. C. R., New York City.
EVANS, MRS. GLENDOWER, 12 Otis Place, Boston, Mass.
FAY, Miss S. B., 88 Mt. Vernon Street, Boston, Mass.

REPORT OF THE DIRECTOR. 45

FOOT, Miss KATHERINE, Care of Morgan Harjes Cie, Paris,

France.

GARDINER, MRS. E. G., Woods Hole, Mass.
HARRISON, EX-PROVOST C. C., University of Pennsylvania,

Philadelphia, Pa.

JACKSON, Miss M. C., 88 Marlboro St., Boston, Mass.
JACKSON, MR. CHAS. C., 24 Congress St., Boston, Mass.
KIDDER, MR. NATHANIEL T., Milton, Mass.
KING, MR. CHAS. A.
LEE, MRS. FREDERIC S., 279 Madison Ave., New York City,

N. Y.

LOWELL, MR. A. LAWRENCE, 17 Quincy St., Cambridge, Mass.
MARRS, MRS. LAURA NORCROSS, 9 Commonwealth Ave., Boston,

Mass.

MASON, Miss E. F., i Walnut St., Boston, Mass.
MASON, Miss IDA M., I Walnut St., Boston, Mass.
MEANS, DR. JAMES HOWARD, 15 Chestnut St., Boston, Mass.
MERRIMAN, MRS. DANIEL, 73 Bay State Road, Boston, Mass.
MINNS, Miss SUSAN, 14 Louisburg Square, Boston, Mass.
MORGAN, MR. J. PIERPONT, JR., Wall and Broad Sts., New

York City, N. Y.
MORGAN, PROF. T. H., Columbia University, New York City,

N. Y.

MORGAN, MRS. T. H., New York City, N. Y.
No YES, Miss EVA J.
OSBORN, PROF. HENRY F., American Museum of Natural

History, New York City, N. Y.

PHILLIPS, MRS. JOHN C., Windy Knob, Wenham, Mass.
PORTER, DR. H. C., University of Pennsylvania, Philadelphia, Pa.
PULSIFER, MR. W. H., Newton Center, Mass.
SEARS, DR. HENRY F., 86 Beacon St., Boston, Mass.
SHEDD, MR. E. A.
THORNDIKE, DR. EDWARD L., Teachers College, Columbia

University, New York City, N. Y.

TRELEASE, PROF. \VILLIAM, University of Illinois, UYbana, 111.
WARE, Miss MARY L., 41 Brimmer St., Boston, Mass.
WILLIAMS, MRS. ANNA P., 505 Beacon St., Boston, Mass.
WILSON, DR. E. B., Columbia University, New York City, N. Y.

46 MARINE BIOLOGICAL LABORATORY.

CORPORATION MEMBERSHIP LIST,
AUGUST 1927.

ADAMS, DR. A. ELIZABETH, Mount Holyoke College, South

Hadley, Mass.
ADDISON, DR. W. H. F., University of Pennsylvania Medical

School, Philadelphia, Pennsylvania.
ADOLPH, DR. EDWARD F., University of Rochester, School of

Medicine and Dentistry, Rochester, N. Y.
AGERSBORG, DR. H. P. K., James Millikin University, Decatur,

Illinois.

ALLEE, DR. W. C., University of Chicago, Chicago, Illinois.
ALLEN, PROF. CHAS. E., University of Wisconsin, Madison,

Wisconsin.
ALLEN, PROF. EZRA, 1003 South 46th Street, Philadelphia,

Pennsylvania.

ALLYN, DR. HARRIET M., Vassar College, Poughkeepsie, N. Y.
AMBERSON, DR. WILLIAM B., University of Pennsylvania,

Philadelphia, Pa.
ANDERSON, DR. E. G., University of Michigan, Ann Arbor,

Michigan.

ATTERBURY, MRS. RUTH R., Great Neck, Long Island, New York.
BAITSELL, DR. GEORGE A., Yale University, New Haven,

Connecticut.
BAKER, DR. E. H., 5312 Hyde Park Boulevard, Hyde Park

Station, Chicago, 111.
BALDWIN, DR. F. M., University of Southern California, Los

Angeles, Calif.
BASCOM, DR. K. F., Medical School of Virginia, Richmond,

Virginia.

BECKWITH, DR. CORA J., Vassar College, Poughkeepsie, N. Y.
BEHRE, DR. ELINOR H., Louisiana State University, Baton

Rouge, La.
BENNITT, DR. RUDOLF, University of Missouri, Columbia,

Missouri.
BIGELOW, PROF. R. P., Massachusetts Institute of Technology,

Cambridge, Mass.
BINFORD, PROF. RAYMOND, Guilford College, Guilford College,

N. C.

REPORT OF THE DIRECTOR. 47

BISSONNETTE, DR. T. H., Trinity College, Hartford, Connecticut.
BODINE, DR. J. H., University of Pennsylvania, Philadelphia, Pa.
BORING, DR. ALICE M., Yenching College, Peking, China.
BOWEN, DR. ROBERT H., Columbia University, New York City.
Box, Miss CORA M., University of Cincinnati, Cincinnati, Ohio.
BRADLEY, PROF. HAROLD C., University of Wisconsin, Madison,

Wisconsin.

BRAILEY, Miss MIRIAM E., 800 Broadway, Baltimore, Maryland.
BRIDGES, DR. CALVIN B., Columbia University, New York City.
BROOKS, DR. S. C., University of California, Berkeley, California.
BUCKINGHAM, Miss EDITH N., Sudbury, Massachusetts.
BUDINGTON, PROF. R. A., Oberlin College, Oberlin, Ohio.
BULLINGTON, DR. W. E., Randolph-Macon College, Ashland,

Virginia.

BUMPUS, PROF. H. C., Duxbury, Massachusetts.
BYRNES, DR. ESTHER F., 1803 North Camac Street, Philadelphia,

Pa.

CALKINS, PROF. GARY N., Columbia University, New York City.
CALVERT, PROF. PHILIP P., University of Pennsylvania, Phila-
delphia, Pa.

CARLSON, PROF. A. J., University of Chicago, Chicago, Illinois.
CAROTHERS, DR. ELEANOR E., University of Pennsylvania,

Philadelphia, Pa.
CARROLL, PROF. MITCHEL, Franklin and Marshall College,

Lancaster, Pa.

CARVER, PROF. GAIL L., 613 Orange Street, Macon, Georgia.
CASTEEL, DR. D. B., University of Texas, Austin, Texas.
CATTELL, PROF. J. McKEEN, Garrison-on-Hudson, New York.
CATTELL, DR. McKEEN, Cornell University Medical College,

New York City.

CATTELL, MR. WARE, Garrison-on-Hudson, New York.
CHAMBERS, DR. ROBERT, Cornell University Medical College,

New York City.
CHARLTON, DR. HARRY H., University of Missouri, Columbia,

Missouri.
CHIDESTER, PROF. F. E., West Virginia University, Morgantown,

W. Va.

CHILD, PROF. C. M., University of Chicago, Chicago, Illinois.
CLAPP, PROF. CORNELIA M., Montague, Massachusetts.

48 MA KIM. lilOLOGlCAL LABORATORY.

CLARK, PROF. E. R., University of Pennsylvania, Philadelphia,
Pa.

CLELAND, PROF. RALPH E., Goucher College, Baltimore, Mary-
land.

CLOWES, PROF. G. H. A., Eli Lilly & Co., Indianapolis, Indiana.

COE, PROF. W. R., Yale University, New Haven, Connecticut.

COHN, DR. EDWJX J., 19 Ash St., Cambridge, Mass.

COKER, DR. R. E., University of North Carolina, Chapel Hill,
North Carolina.

COLE, DR. LEON J., College of Agriculture, Madison, Wisconsin.

COLLETT, DR. MARY E., Western Reserve University, Cleveland,
Ohio.

COLLEY, MRS. MARY W., 1712 Madison St., Madison, Wisconsin.

COLTON, PROF. H. S., Box 127, Flagstaff, Arizona.

CONNOLLY, DR. C. J., Catholic University, Washington, D. C.

COPELAND, PROF. MANTON, Bowdoin College, Brunswick, Maine.

COWDRY, DR. E. V., Rockefeller Institute, New York City.

CRAMPTON, PROF. H. E., Barnard College, Columbia University,
New York City.

CRANE, MRS. C. R., Woods Hole, Mass.

CURTIS, DR. MAYNIE R., Crocker Laboratory, Columbia Uni-
versity, New York City.

CURTIS, PROF. Wr. C., University of Missouri, Columbia, Mo.

DANCHAKOFF, DR. VERA, Timiriaseff Research Institute, Mos-
cow, Russia.

DAVIS, DR. DONALD W., College of William and Mary, Williams-
burg, Va.

DAVIS, DR. ALICE R., 19 Ash St., Cambridge, Mass.

DAWSON, DR. J. A., Harvard University, Cambridge, Mass.

DEDERER, DR. PAULINE H., Connecticut College, New London,
Connecticut.

DELLINGER, DR. S. C., University of Arkansas, Fayetteville, Ark.

DETLEFSEN, DR. J. A., Swarthmore, Pennsylvania.

DEXTER, DR. J. S., University of Porto Rico, Rio Piedras,
Porto Rico.

DODDS, PROF. G. S., Medical School, University of West Virginia,
Morgantown, WTest Virginia.

DOLLEY, PROF. WILLIAM L., University of Buffalo, Buffalo,
N.Y.

REPORT OF THE DIRECTOR. 49

DONALDSON, PROF. H. H., Wistar Institute of Anatomy and
Biology, Philadelphia, Pennsylvania.

DONALDSON, DR. JOHN C., University of Pittsburgh, School of
Medicine, Pittsburgh, Pa.

DREW, PROF. GILMAN A., Eagle Lake, Florida.

DUGGAR, DR. BENJAMIN M., University of Wisconsin, Madison,
Wisconsin.

DUNGAY, DR. NEIL S., Carleton College, Northfield, Minn.

DUNN, DR. ELIZABETH H., 105 North 5th Ave., La Grange,
111.

EDWARDS, DR. D. J., Cornell University Medical College, New
York City.

ELLIS, DR. F. W., Monson, Mass.

FARNUM, DR. LOUISE W., 43 Hillhouse Ave., New Haven, Conn.

FARR, DR. C. H., Washington University, St. Louis, Mo.

FENN, DR. W. O., Rochester University School of Medicine,
Rochester, N. Y.

FIELD, Miss HAZEL E., Occidental College, Los Angeles, Cali-
fornia.

FORBES, DR. ALEXANDER, Harvard University Medical School,
Boston, Mass.

FRY, DR. HENRY J., Washington Square College, New York City.

GAGE, PROF. S. H., Cornell University, Ithaca, N. Y.

GARREY, PROF. W. E., Vanderbilt University Medical School,
Nashville, Tennessee.

GATES, DR. F. L., Rockefeller Institute, New York City.

GATES, PROF. R. RUGGLES, University of London, London,
England.

GEISER, DR. S. W., Southern Methodist University, Dallas,
Texas.

GLASER, PROF. O. C., Amherst College, Amherst, Mass.

GLASER, PROF. R. W., Rockefeller Institute for Medical Re-
search, Princeton, N. J.

GOLDFORB, PROF. A. J., College of the City of New York,

New York City.
GOODRICH, PROF. H. B., Wesleyan University, Middletown,

Conn.

GRAHAM, DR. J. Y., University of Alabama, University, Alabama.
GRAVE, PROF. B. H., Wabash College, Crawfordsville, Indiana.

5Q MARINE BIOLOGICAL LABORATORY.

GRAVE, PROF. CASWELL, Washington University, St. Louis,

Missouri.
GREENMAN, PROF. M. J., Wistar Institute of Anatomy and

Biology, Philadelphia, Pennsylvania.

GREGORY, DR. LOUISE H., Barnard College, Columbia Uni-
versity, New York City.
GU*THRIE, DR. MARY J., University of Missouri, Columbia,

Missouri.

GUYER, PROF. M. F., University of Wisconsin, Madison, Wis-
consin.
HAGUE, DR. FLORENCE, Sweet Briar College, Sweet Briar,

Virginia.

HALSEY, DR. J. T., Tulane University, New Orleans, Louisiana.
HANCE, DR. ROBERT T., University of Pittsburgh, Pittsburgh,

Pennsylvania.
HARGITT, PROF. GEORGE T., Syracuse University, Syracuse,

New York.
HARMAN, DR. MARY T., Kansas State Agricultural College,

Manhattan, Kansas.

HARPER, PROF. R. A., Columbia University, New York City.
HARRISON, PROF. Ross G., Yale University, New Haven,

Connecticut.

HARVEY, PROF. E. N., Princeton University, Princeton, N. J.
HARVEY, MRS. E. N., Princeton, New Jersey.
HAYDEN, DR. MARGARET A., Wellesley College, Wellesley,

Mass.
HAZEN, DR. T. E., Barnard College, Columbia University, New

York City.

HEATH, PROF. HAROLD, Pacific Grave, California.
HECHT, DR. SELIG, Columbia University, New York City.
HEGNER, PROF. R. W., Johns Hopkins University, Baltimore,

Maryland.
HEILBRUNN, DR. L. V., University of Michigan, Ann Arbor,

Michigan.
HESS, PROF. WALTER N.,. DePauw University, Greencastle,

Indiana.
HINRICKS, DR. MARIE A., University of Chicago, Chicago,

Illinois.
HISAW, DR. F. L., University of Wisconsin, Madison, Wisconsin.

REPORT OF THE DIRECTOR. 5!

HOADLEY, DR. LEIGH, Harvard University, Cambridge, Massa-
chusetts.

HOGUE, DR. MARY J., 503 North High Street, West Chester;
Pennsylvania.

HOLMES, PROF. S. J., University of California, Berkeley, Cali-
fornia.

HOOKER, PROF. DAVENPORT, University of Pittsburgh, Pitts-
burgh, Pa.

HOPKINS, DR. HOYT S., New York University College of Den-
tistry, New York City.

HOSKINS, MRS. ELMER R., New York University, College of
Dentistry, New York City.

HOWE, DR. H. E., 2702— 36th Street, N. W., Washington, D. C.

HOYT, DR. WILLIAM D., Washington and Lee University,
Lexington, Virginia.

HUMPHREY, DR. R. R., University of Buffalo School of Medicine,
Buffalo, N. Y.

HYMAN, DR. LIBBIE H., University of Chicago, Chicago, Illinois.

INMAN, PROF. ONDESS L., Antioch College, Yellow Springs, Ohio.

IRWIN, DR. MARIAN, Rockefeller Institute, New York City.

JACKSON, PROF. C. M., University of Minnesota, Minneapolis,
Minnesota.

JACOBS, DR. MERKEL H., University of Pennsylvania, Phila-
delphia, Pennsylvania.

JENNINGS, PROF. H. S., Johns Hopkins University, Baltimore,
Md.

JEWETT, PROF. J. R., Harvard University, Cambridge, Massa-
chusetts.

JOHNSON, PROF. GEORGE E., State Agricultural College, Man-
hattan, Kansas.

JONES, PROF. LYNDS, Oberlin College, Oberlin, Ohio.

JORDAN, PROF. H. E., University of Virginia, Charlottesville,
Virginia.

JUST, PROF. E. E., Howard University, Washington, D. C.

KEEFE, REV. ANSELM M., St. Norbert's College, West Depere,
Wisconsin.

KENNEDY, DR. HARRIS, Readville, Massachusetts.

KINDRED, DR. J. E., University of Virginia, Charlottesville,
Virginia.

52 MARINE BIOLOGICAL LABORATORY.

KING, DR. HELEN D., Wistar Institute of Anatomy and Biology,
Philadelphia, Pa.

KING, DR. ROBERT L., University of Pennsylvania, Philadelphia,
Pennsylvania.

KINGSBURY, PROF. B. F., Cornell University, Ithaca, New York.

KINGSLEY, PROF. J. S., 2500 Cedar Street, Berkeley, California.

KIRKHAM, DR. W. B., Springfield College, Springfield, Mass.

KNAPKE, REV. BEDE, St. Bernard's College, St. Bernard, Ala-
bama.

KNOWER, PROF. H. McE., University of Alabama, University,
Ala.

KNOWLTON, PROF. F. P., Syracuse University, Syracuse, New
York.

KOSTIR, DR. W. J., Ohio State University, Columbus, Ohio.

KRIBS, DR. HERBERT, Ewing Christian College, Al'ahabed,
North India.

KUYK, DR. MARGARET P., Westbrook Ave., Richmond, Va.

LANCEFIELD, DR. D. E., Columbia University, New York City.

LANGE, DR. MATHILDE M., Wheaton College, Norton, Mass.

LEE, PROF. F. S., College of Physicians and Surgeons, New
York City.

LEWIS, PROF. 1. F., University of Virginia, Charlottes ville, Va.

LEWIS, PROF. W. H., Johns Hopkins University, Baltimore, Md.

LILLIE, PROF. FRANK R., University of Chicago, Chicago,
Illinois.

LILLIE, PROF. R. S., University of Chicago, Chicago, Illinois.

LINTON, PROF. EDWIN, University of Pennsylvania, Philadelphia,
Pennsylvania.

LOEB, PROF. LEO, Washington University Medical School, St.
Louis, Mo.

LOEB, MRS. LEO, 6803 Kingsburg Boulevard, St. Louis, Missouri.

LOWTHER, MRS. FLORENCE DEL., Barnard College, Columbia
University, New York City.

LUCRE, PROF. BALDWIN, University of Pennsylvania, Phila-
delphia, Pa.

LUND, DR. E. J., University of Texas, Austin, Texas.

LUSCOMBE, MR. W. O., Woods Hole, Mass.

LYNCH, Miss CLARA J., Rockefeller Institute, New York City.

LYON, PROF. E. P., University of Minnesota, Minneapolis, Minn.

RKPORT OF THE DIRECTOR. 53

MACCALLUM, DR. G. A., 925 St. Paul St., Baltimore, Md.

MACDOUGALL, DR. MARY S., Agnes Scott College, Decatur,
Georgia.

McCLUNG, PROF. C. E., University of Pennsylvania. Phila-
delphia, Pa.

McGEE, DR. ANITA NEWCOMB, 2707 Adams Mill Road, Wash-
ington, D. C.

MCGREGOR, DR. J. H., Columbia University, New York City.

MclNDOO, DR. N. E., Bureau of Entomology, Washington, D. C.

McMuRRiCH, PROF. J. P., University of Toronto, Toronto,
Canada.

McNAiR, DR. G. T., 1909 Ohio St., Lawrence, Kansas.

MACKLIN, DR. CHARLES C., School of Medicine, University of
\Vestern Ontario, London, Canada.

MALONE, PROF. E. F., University of Cincinnati, Cincinnati, Ohio.

MARTIN, MR. E. A., College of the City of New York, New
York City.

MAST, PROF. S. O., Johns Hopkins University, Baltimore, Md.

MATHEWS, PROF. A. P., University of Cincinnati, Cincinnati,
Ohio.

MATSUI, PROF. K., Imperial College of Agriculture and Den-
drology, Morioka, Japan.

MAVOR, PROF. JAMES W., Union College, Schenectady, New
York.

MEDES, DR. GRACE, University of Minnesota, Minneapolis,
Minn.

MEIGS, DR. E. B., Dairy Division Experiment Station, Beltsville,
Md.

MEIGS, MRS. E. B., 1736 M St., N. W., Washington, D. C.

METCALF, PROF. M. M., Johns Hopkins University, Baltimore,
Md.

METZ, PROF. CHARLES W., Carnegie Institution of Washington,
Cold Spring Harbor, Long Island.

MINER, DR. ROY W., American Museum of Natural History,
New York City.

MITCHELL, DR. PHILIP H., Brown University, Providence, R. I.

MOORE, DR. CARL R., University of Chicago, Chicago, Illinois.

MOORE, PROF. GEORGE T., Missouri Botanical Garden, St.
Louis, Mo.

-^ MARINE BIOLOGICAL LABORATORY.

MOORE, PROF. J. PERCY, University of Pennsylvania, Phila-
delphia, Pa.

MORRILL, PROF. A. D., Hamilton College, Clinton, N. Y.

MORRILL, PROF. C. V., Cornell University Medical College,
New York City.

MULLER, DR. H. J., University of Texas, Austin, Texas.

NABOURS, DR. R. K., Kansas State Agricultural College, Man-
hattan, Kansas.

NACHTRIEB, PROF. HENRY F., 2448 Cedar St., Berkeley, Cali-
fornia.

NEAL, PROF. H. V., Tufts College, Tufts College, Mass.

NEWMAN, PROF. H. H., University of Chicago, Chicago, Illinois.

NICHOLS, DR. M. LOUISE, Powelton Apartments, Philadelphia,
Pa.

NONIDEZ, DR. JOSE F., Cornell University Medical College,
New York City.

OKKELBERG, DR. PETER, University of Michigan, Ann Arbor,
Michigan.

OSBURN, PROF. R. C., Ohio State University, Columbus, Ohio.

OSTERHOUT, PROF. W. J. V., Rockefeller Institute, New York
City.

PACKARD, DR. CHARLES, Columbia University, Institute of
Cancer Research, 1145 Amsteidam Avenue, New York City.

PAGE, DR. IRVINE H., Presbyterian Hospital, New York City.

PAPANICOLAOU, DR. GEORGE N., Cornell University Medical
College, New York City.

PAPPENHEIMER, DR. A. M., Columbia University, New York
City.

PARKER, PROF. G. H., Harvard University, Cambridge, Mass.

PATON, PROF. STEWART, Princeton University, Princeton, N. J.

PATTEN, PROF. WILLIAM, Dartmouth College, Hanover, N. H.

PATTERSON, PROF. J. T., University of Texas, Austin, Texas.

PAYNE, PROF. F., University of Indiana, Bloomington, Ind.

PEARL, PROF. RAYMOND, Johns Hopkins University, Baltimore,
Md.

PEARSE, PROF. A. S., Duke University, Durham, North Carolina.

PEEBLES, PROF. FLORENCE, 1095 Marengo Avenue, Pasadena,
California.

PHILLIPS, DR. E. F., Cornell University, Ithaca, New York.

REPORT OF THE DIRECTOR. 55

PHILLIPS, DR. RUTH L., Western College, Oxford, Ohio.

PIKE, PROF. FRANK H., 437 West 59th St., New York City.

PINNEY, DR. MARY E., Milwaukee-Downer College, Milwaukee,
Wisconsin.

PLOUGH, PROF. HAROLD H., Amherst College, Amherst, Mass.

POND, DR. SAMUEL E., University of Pennsylvania, School of
Medicine, Philadelphia, Pennsylvania.

PRATT, DR. FREDERICK H., Boston University School of Medi-
cine, Boston, Mass.

RANKIN, PROF. W. M., Princeton University, Princeton, N. J.

RAPPORT, DR. ANNA YATES, Bryn Mawr College, Bryn Mawr,
Pennsylvania.

REDFIELD, DR. ALFRED C., Haivard University Medical School,
Boston, Mass.

REESE, PROF. ALBERT M., West Virginia University, Morgan-
town, W. Va.

REINKE, DR. E. E., Vanderbilt University, Nashville, Tennessee,

REZNIKOFF, DR. PAUL, Cornell University Medical College.
New York City.

RHODES, PROF. ROBERT C., Emory University, Atlanta, Georgia.

RICE, PROF. EDWARD L., Ohio \Vesleyan University, Delaware,
Ohio.

RICHARDS, PROF. A., University of Oklahoma, Norman, Okla-
homa.

RIGGS, MR. LAWRASON, JR., 25 Broad St., New York City.

ROBERTSON, PROF. W. R. B., Columbia, Missouri.

ROGERS, PROF. CHARLES G., Oberlin College, Oberlin, Ohio.

ROMER, DR. ALFRED S., University of Chicago, Chicago, Illinois.

RUDISCH, DR. J., Fifth Avenue Bank, 44th St. and Fifth Ave.,
New York City.

SAMPSON, DR. MYRA M., Smith College, Northampton, Mass.

SANDS, Miss ADELAIDE G., 348 N. Main St., Port Chester,
New York.

SCHRADER, DR. FRANZ, Bryn Mawr College, Bryn Mawr, Pa.

SCHRAMM, PROF. J. R., University of Pennsylvania, Philadelphia,
Pa.

SCOTT, DR. ERNEST L., Columbia University, New York City.

SCOTT, PROF. G. G., College of the City of New York, New
York City.

56 MARINE BIOLOGICAL LABORATORY.

SCOTT, PROF. JOHN W., University of Wyoming, Laramie,

Wyoming.

SCOTT, PROF. WILLIAM B., 7 Cleveland Lane, Princeton, N. J.
SHULL, PROF. A. FRANKLIN, University of Michigan, Ann Arbor,

Michigan.

SHUMWAY, DR. WALDO, University of Illinois, Urbana, Illinois.
SIVICKIS, DR. P. B., University of the Philippines, Manila, P. I
SMITH, DR. BERTRAM G., 119 Grove St., Tarrytown, N. Y.
SNOW, DR. LAETITIA M., Wellesley College, Wellesley, Mass.
SNYDER, PROF. CHARLES D., Johns Hopkins University Medical

School, Baltimore, Md.
SOLLMAN, DR. TORALD, Western Reserve University, Cleveland,

Ohio.
SPEIDEL, DR. CARL C., University of Virginia, Charlottesville,

Va.

SPENCER, PROF. H. J., 24 West loth St., New York City.
STARK, DR. MARY B., N. Y. Homeopathic Medical College and

Flower Hospital, New York City.
STOCKARD, PROF. C. R., Cornell University Medical College,

New York City.
STOKEY, DR. ALMA G., Mount Holyoke College, South Hadley,

Mass.
STREETER, PROF. GEORGE L., Carnegie Institution, Baltimore,

Maryland.

STRONG, PROF. O. S., Columbia University, New York City.
STUNKARD, PROF. HORACE W7., New York University, University

Heights, New York.
STURTEVANT, DR. ALFRED H., Columbia University, New York

City.

TASHIRO, DR. SHIRO, Medical College, University of Cincinnati,
Cincinnati, Ohio.

TAYLOR, Miss KATHERINE A., Cascade, Washington Co.,
Maryland.

TAYLOR, DR. WILLIAM R., University of Pennsylvania, Phila-
delphia, Pa.

TENNENT, PROF. D. H., Bryn Mawr College, Bryn Mawr, Pa.

THARALDSEN, PROF. C. E., 618 Church St., Evanston, Illinois.

THATCHER, MR. LLOYD E., University of Mississippi, University,
Mississippi.

REPORT OF THE DIRECTOR.

57

TIXKHAM, Miss FLORENCE L., 71 Ingersoll Grove, Springfield,

Mass.
TOMPKINS, Miss ELIZABETH M., 134 Linden Ave., Brooklyn,

N. Y.
TRACY, PROF. HENRY C., University of Kansas, Lawrence,

Kansas.

TREADWELL, PROF. A. L., Vassar College, Poughkeepsie, N. Y.
TURNER, PROF. C. L., Beloit College, Beloit, Wisconsin.
LTHLEMEYER, Miss BERTHA, Washington LTniversity, St. Louis,

Mo.
UHLENHUTH, DR. EDWARD, University of Maryland, School of

Medicine, Baltimore, Maryland.
UNGER, DR. W. BYERS, Dartmouth College, Hanover, New

Hampshire.

VAN DER HEYDE, DR. H. C., Galeria, Corse, France.
VISSCHER, DR. J. PAUL, Western Reserve University, Cleveland,

Ohio.
WAITE, PROF. F. C., Western Reserve University Medical School,

Cleveland, Ohio.
WALLACE, DR. LOUISE B., Constantinople Woman's College,

Constantinople, Turkey.

WAXX, PROF. FRANK B., Utah Agricultural College, Logan, Utah.
\VARD, PROF. HENRY B., University of Illinois, Urbana, Illinois.
WARDWELL, DR. E. H., Chappaqua, New York.
WTARREN, PROF. HOWARD C., Princeton University, Princeton,

New Jersey.

WARREN, DR. HERBERT S., White Plains, New York.
WENRICH, DR. D. H., University of Pennsylvania, Philadelphia,

Pa.
WHEDON, DR. A. D., North Dakota Agricultural College, Fargo,

No. Dakota.

WHEELER, PROF. W. M., Bussey Institution, Forest Hills, Mass.
WHERRY, DR. W. B., Cincinnati Hospital, Cincinnati, Ohio.
WHITE, DR. E. GRACE, Wilson College, Chambersburg, Pa.
WHITESIDE, DR. BEATRICE, Detroit College of Medicine and

Surgery, Detroit, Michigan.

WHITING, DR. PHINEAS W., Bussey Institute, Forest Hills, Mass.
WHITNEY, DR. DAVID D., LTniversity of Nebraska, Lincoln, Neb.
WIEMAN, PROF. H. L., University of Cincinnati, Cincinnati, Ohio.

f<^*

LjlLIBRARY

,<*V^!A

58 MARINE BIOLOGICAL LABORATORY.

•

\YILLIER, DR. B. H., University of Chicago, Chicago, Illinois.
WILSON, PROF. H. V., University of North Carolina, Chapel

Hill, N. C.

WILSON, DR. J. W., Brown University, Providence, R. I.
WOGLOM, PROF. WILLIAM H., Columbia University, New York

City.

WOODRUFF, PROF. L. L., Yale University, New Haven, Conn.
WOODWARD, DR. ALVALYN E., 5507 New Medical Building,

Ann Arbor, Michigan.

YOUNG, DR. B. P., Cornell University, Ithaca, New York.
YOUNG, DR. D. B., University of Maine, Orono, Maine.
ZELENY, DR. CHARLES, University of Illinois, Urbana, Illinois.

Vol.LV.

August, 1928.

No. 2.

BIOLOGICAL BULLETIN

THE CONTRACTILE VACUOLE IN PARAMECIUM

TRICHIUM.

ROBERT L. KING,
ZOOLOGICAL LABORATORY, UNIVERSITY OF PENNSYLVANIA.

INTRODUCTION.

The contractile vacuolar apparatus of Ciliata seems to be more
highly specialized than that of other Protozoa. As to the form
of the vacuole itself, we may recognize for convenience two gen-
eral types among the Ciliata: (a) Vesicle-fed vacuoles, in which
the contracting vacuole is surrounded by a series of small vacuoles
(called vesicles in this paper) which seem to coalesce and form
a new contracting vacuole after systole, new smaller vesicles ap-
pearing around the contracting vacuole after systole, and growing
larger during diastole as described by Taylor ('23) in Euplotes;
(b) Canal-fed vacuoles, in which the vacuole is fed by canals,
which vary in number from one in Spirostomum and Stcntor
(Roux, '01) to thirty in Ophryoglena (Biitschli, '87-' '89). A
condition which seems to be intermediate has been described by
Schewiakoff ('89) for Prorodon tercs. Here the vacuole is fed
by four radial rows of vesicles which decrease in size distally from
the contractile vacuole.

The contractile vacuole usually communicates with the exterior
by a tubule through a pore. This excretory tubule leading from
the vacuole to the pore may be very short as in Nassula and Fron-
tonia relatively long as in Urocentrutn, or very long as in Lemba-
dion (Schewiakoff, '89) where the vacuole is located near the
central part of the body and the excretory tube leads to the pore
which is about one quarter of the body length posterior to the
vacuole.

5 59

6o ROBERT L. KING.

The purpose of this paper is to describe the contractile vacuolar
apparatus of Paramecium trichium which has been described by
Stokes ('88) as consisting of two vacuoles "close together, con-
tracting quickly, the one beginning to again form almost before
the completion of the other's systole " and by Wenrich ('26) who
was unable to reach a satisfactory conclusion as to " whether the
two main vacuoles empty alternately into the same outlet tube or
whether there is an auriculo-ventricular relationship between
them."

Without the use of methods introduced by Bresslau ('21) as
modified in this paper, the structures to be described could not
have been studied. I am also particularly indebted to Doctor D.
H. Wenrich who first pointed out the need for further study of
the contractile vacuoles of Paramecium trichium.

MATERIALS AND METHODS.

The Paramecium trichium used in this series of observations
were obtained from two sources in the vicinity of Swarthmore,
Pa. : from the east branch of Crum Lynne Creek where they were
found among the jelly of toads eggs and from the west branch of
Stony Creek, the source of Wenrich's ('26) Swarthmore race.
The material was first studied in mixed cultures but later pure
lines were established. For purposes of comparison, Colpidium
colpoda from the same sources was studied.

Observations on the contractile vacuolar apparatus of Parame-
cium trichium were first made on material prepared according to
the methods described by Bresslau ('21) and later those described
by Coles ('27). These methods consist essentially of drying the
protozoa in concentrated solutions of stains, which are relatively
low in toxicity. The dye is precipitated in and on surface struc-
tures such as the depressions from which cilia arise, the cilia
themselves, the mouth and the cytopharynx. The routine used is
essentially as follows : A small drop of concentrated culture of
the protozoa to be examined is placed upon a perfectly clean glass
slip and mixed with a similarly sized drop of the stain. The two
are then mixed and spread evenly over the slip with a needle or a
glass spreader. The slides are then allowed to dry in the air and
are examined under oil or mounted in damar.

CONTRACTILE VACUOLE IN PARAMECIUM. 6 1

Bresslau ('21) has used the following staining solutions: (a)
Three parts saturated aqueous solution China Blue to one part
saturated aqueous solution Cyanosin (a mixture previously used
in bacteriological technique) and (&) one cc. 10 per cent, aqueous
solution Opal Blue to 4-6 drops 6.5 per cent, aqueous solution
Phloxinrhodamin.

Coles ('27) has used numerous stains for Protozoa in the same
manner, obtaining his best results with a saturated aqueous solu-
tion Nigrosin.

The stains recommended by Bresslau are very toxic to Para-
mecium trichium which are killed quickly and usually burst open
before they become dry. However, with solutions of (a) 10 per
cent. China Blue (Coleman Bell), (b) 10 per cent. Nigrosin
(Coleman Bell or Griibler), (c) Mixture of equal parts of above,
and (rf) 10 per cent. Opal Blue (Coleman Bell), beautiful prepara-
tions have been made. The above methods have been placed in
order of their apparent usefulness for the structures studied in
Paramecium trichium.

Observations made on such dried preparations have been veri-
fied on material fixed with Bouin's or Schaudinn's and stained
with hsemalum and on living animals in hanging drops of the
culture medium or of the culture medium mixed with China Blue,
Opal Blue or Nigrosin. Paramecium trichium will survive over
two hours in such hanging drops containing stains and remain
apparently typical during that time.

OBSERVATIONS.

There are two contractile vacuoles in Paramecium trichium.
one located in the anterior and one in the posterior part of the
animal (Wenrich, '26). When dried in the staining solutions
described above the coloring matter collects in the contractile
vacuolar apparatus sometimes penetrating into the vacuole itself.
The contractile vacuole is seen to be connected with the exterior
by a long convoluted tubule (Figs. 2, 3, 4 and 9) which terminates
in a small pore (Figs, i, 2 and 4) located between the longitudinal
rows of cilia on the dorsal surface of the body opposite the mouth.
The pore of the anterior vacuole and that of the posterior open
to the exterior between the same or adjacent rows of longitudinal

62 ROBERT L. KING.

rows of cilia (Figs. 8, 15 and 14). The vacuoles themselves do
not seem to be fixed in position in living animals but move about
more or less in the endoplasm; this apparently is made possible
by the presence of the convoluted tubule. The inner end of the
tubule terminates in a cup-like valve (Figs. 7 and 9) with which
the vacuole is in contact when undergoing systole. This cup-like
valve has been observed both in prepared slides and in living
animals; while it seems always to be in contact with the vacuole
it may be exterior or interior, anterior or posterior in relation to
the vacuole. In fact the whole structure seems to be continually
in active movement. The proximal portion of the tubule shows
the effect of increased pressure at the beginning of systole, tend-
ing to straighten out as does a coiled hose when water is first
turned into it under pressure.

In living animals each of the contractile vacuoles appears to
be made up of two vacuoles which contract alternately. However,
if they be carefully compared with a typical vesicle-fed vacuole
such as that of Colpidium colpoda, it is found that the two are
fundamentally alike. Observations made with a stop-watch show
that the time relations of various parts of the cycle of systole and
diastole in the two species are quite different. The table gives a
set of typical observations on Paramecium trichiuni and Colpidiuin
colpoda. These observations were made upon a single individual
of each species on the same slide. For the interval between con-
tractions seven groups of three contractions each were timed in
Colpidiiim colpoda; ten groups of three contractions each in Para-
mecium trichiuni (anterior vacuole). A number of separate ob-
servations of the time from beginning to end of systole were
made on the same individuals.

TABLE I.

TIME RELATIONS IN SECONDS OF CONTRACTILE VACUOLES OF Paramecium
trichiuni AND Colpidiiim colpoda AT 25° C.

Time between Completion Time
of Two Systoles. for Systole.

Paramecium tricliium 3.1 2.0

Colpidium colpoda 7.4 0.4

In Colpidiiim colpoda the vacuole empties rapidly (about 0.4
seconds) ; the small feeding vesicles gradually enlarge, coalesce

CONTRACTILE VACUOLE IN PARAMECIUM. 63

into one vacuole which finally reaches its maximum size about
seven seconds after the last contraction, in the meantime there
have appeared more tiny feeding vesicles around its periphery.
The vacuole then contracts rapidly and the cycle is repeated.

In Paramecium trichium the vacuole (Fig. 120) empties slowly
(about 2 seconds). As it contracts the feeding vesicles grow
(Fig. 126) and by the time the vacuole has half completed systole
(Fig. I2c) the feeding vacuoles coalesce with an apparent passing
of membranes over their surface marking their fusion [called by
Taylor ('23) "vestiges of the adjacent walls" in Euplotes].
When systole is completed (Fig. i2cQ the new contractile vacuole
has reached its maximum size ; as it comes in contact with the cup-
like valve a membrane (as described by Wenrich, '26) appears
to pass over its surface, and it begins to contract (Fig. I2c).

The two processes are fundamentally alike; in Paramecium
trichium the systole of the old contractile vacuole and the diastole
of the new occur almost simultaneously and are of approximately
the same duration while in Colpidium colpoda they occur consecu-
tively with systole very brief and diastole protracted.

SUMMARY.

1. There is an anterior and a posterior contractile vacuolar
apparatus in Paramecium trichium.

2. Each is permanent and consists of feeding vesicles, contrac-
tile vacuole, excretory tube and pore.

3. The contractile vacuoles of Paramecium trichium are vesicle-
fed, differing here from those of other well known species in the
same genus which are canal-fed.

4. Diastole of the new vacuole is practically complete before the
prolonged systole of the old is over. This gives the appearance of
two vacuoles contracting alternately.

5. The excretory tube is long and convoluted with its flattened
cup-like end in contact with the contractile vacuole.

6. The excretory tube opens as an excretory pore on the surface
of the body opposite the mouth.

7. The pore of the anterior and that of the posterior apparatus
are located between the same or adjacent longitudinal rows of
cilia.

64

ROBERT L. KING.

8. The long duration of systole may be correlated with the
presence of the convoluted excretory tube.

9. The excretory tube and pore were first demonstrated by the
use of Bresslau's relief staining method.

LITERATURE CITED.

Bresslau, E.

1922 Die Gelatinierbarkeit des Protoplasmas als Grundlage ernes

Verfahrens zur Schnellanfertigung gefarbter Dauerprapar-
ate von Infusorien. Arch. f. Protistenk., 43: 467-480.
Butschli, A.

1887-89 Infusoria. Bronn's Thier-reichs, I, III Abt.
Coles, A. C.

1927 Relief Staining of Bacteria Protozoa, Infusoria. Watson

Microscope Record, No. 10: 23-25.
Roux, J.

1901 Faune infusorienne des eaux stagnantes des environs de Gen-
eve. Geneve: Henry Kiindig.
Schewiakoff, W.

1889 Beitrage zur Kenntniss der holotrichen Ciliaten. Bibliotheca

zoologica, Heft 5.
Stokes, A. C.

1888 Fresh-water Infusoria. Jour. Trenton Nat. Hist. Soc., i:

7I-344-
Taylor, C. V.

1923 The Contractile Vacuole in Euplotes. Jour. Exp. Zool., 37:

259-289.
Wenrich, D. H.

1926 The Structure and Division of Paramecium trichium. Jour.
Morph. and Physiol., 43: 81-103.

66 ROBERT L. KING.

PLATE I.
Explanation of Figures.

Figs. i-n. Contractile vacuolar apparatus of Paramccium trichuun,
dried in 10 per cent. China Blue unless otherwise stated X 1000.

1. Posterior apparatus of animal photographed for Fig. 13.

2. Anterior apparatus of animal photographed for Fig. 14.

3. Posterior apparatus.

4. Posterior apparatus.

5. Anterior apparatus.

6. Posterior apparatus of same animal as Fig. 5.

7. Posterior apparatus of animal photographed for Fig. 16.

8. Entire animal showing cytopharynx, anterior and posterior vacuolar
apparatus. Same as photograph Fig. 15.

9. Anterior apparatus.

10. Anterior apparatus (10 per cent, nigrosin).

u. Two tubules and pores from anterior end of same animal.

Fig. 12. Diagrams of various stages of contracting vacuole. a. Begin-
ning of systole, b. Systole half-completed, feeding vesicles grow larger.
c. Fusion of feeding vesicles, d. Systole complete. Cup-like inner end of
tubule, e. Fusion of new contractile vacuole with cup-like inner end of
tubule preparatory to systole.

BIOLOGICAL BULLETIN, VOL. LV.

PLATE I.

* i

II

12 I f a

ROBERT L. KING.

68 ROBERT L. KING.

PLATE II.
Photographs X 750,

BIOLOGICAL BULLETIN, VOL. LV

PLATE II.

ROBERT L. KING.

OBSERVATIONS OF THE FEEDING MECHANISM OF'
A CTENOPHORE, MNEMIOPSIS LIIIDYI.1

ROLLAND J. MAIN.

Observations of the feeding habits of ctenophores are scattered
through the literature dealing with these organisms (Bigelow,
'15; Lebour, '22-23; Mayer, '12; Nelson, '25), but as yet no
detailed study of the feeding mechanism has appeared.

The ciliation of a hydromedusa has been studied ( J. F. Gemmill,
'19), but this compares in no way with the complex food catching
apparatus of a ctenophore such as Mnemiopsis Icidyi. The mor-
phological work done upon this ctenophore is incomplete, for
neither Agassiz (1849), Fewkes (1881), nor Mayer ('12), men-
tion the presence of its remarkable mechanism for the capture of
food.

Mnemiopsis Icidyi through its habit of devouring the free-
swimming larvse of the oyster and of other molluscs becomes of
such economic importance that it is of interest to determine by
what means these organisms are captured and carried into the
stomodseum, and how the undigested residues are discharged.2

MATERIALS AND METHODS.

The specimens of Mnemiopsis Icidyi were obtained in the
northern half of Barnegat Bay, N. J., a shallow estuary, in water
of a specific gravity approximating i.oio, with temperatures close
to 20° C, during August and the first part of September, 1926.
The animals were caught in a net, placed in jars without injury
and within ten minutes after capture they were being examined
under the binocular.

Living plankton was used to determine the feeding mechanism,
and it is felt that to this the success of the experiment is pri-

1 From the Zoological Laboratory of Rutgers University, Publication
No. n, New Jersey Oyster Investigation Laboratory.

- The writer is indebted to Dr. Thurlow C. Nelson of Rutgers University
for aid and advice during this investigation and for reading the manuscript.

69

JO ROLLAXD J. MAIN.

marily due, since it is doubtful for reasons given below whether
any other material could have been successfully used. The plank-
ton was secured by pouring sea water through a 200 mesh plankton
net, and concentrating the organisms in a small amount of water.
A Mncmiopsis was placed in a watch crystal under the binocular,
a little of the plankton culture was added, and the reactions of the
ctenophore noted.

STRUCTURE AND OPERATION OF THE FOOD CATCHING

MECHANISM.

To understand the mechanism of the food catching apparatus,
it is first necessary to have a clear idea of the gross anatomy of
Mncmiopsis, Fig. I. Although considerable work has been done

FIG. I. Adult Mncmiopsis Icidyi from Barnegat Bay. Photographed im-
mediately after fixation in 10 per cent, hydrochloric acid. The oral lobes
have contracted to approximately 2/3 the length characteristic of the
living animal. Photographed by T. C. Nelson.

on the morphology of the animal, all the writers have apparently
disregarded the presence of a definite ridge, an extension of the

FEEDING MECHANISM OF M XKM IOPSIS. yi

lips of the mouth, which the writer has named the " labial ridge."

There are four furrows formed by the juncture of the oral lobes
with the body. In each furrow along the inner side of the labial
ridge is a line of tentacles. Through the base of this labial ridge
runs a branch of the paragastric canal, which finally unites with
the auricular canal. On the opposite side of this ridge is the
ciliated channel for conveying food to the mouth, Figs. 2 and 3.

FIG. 2. Adult Miiciniopsis Icidyi. Part of the right lobe and the tip of
the right auricle have been omitted. It is difficult to represent the turning
of the labial ridge. The lips are in the plane of the paper. As the lip be-
comes the labial ridge, it turns so that it lies in a plane at right angles to
the paper, i. The tentacular bulb. 2. The tentacular ridge, with ten-
tacles. 3. The labial ridge, along which runs the tentacular ridge. 4. Lip.
5. Auricular groove. 6. Cilia of auricle.

To this channel, or trough, will be applied the term " labial trough."
It is formed by the labial ridge on one side, and the oral lobe on
the other. It runs along the ridge to the point where the ridge
becomes the lip, and here the trough runs directly into the corner
of the mouth, Fig. 4. The labial ridge is separated from the cilia
of the auricles by the auricular groove in which the cilia of the
auricle beat, and at the bottom of which lie the tentacles stretched
out in the current.

3 The writer calls attention to some apparent discrepancies in earlier work
on Mnciniopsis Icidyi. Fewkes pictures an adult of this species which differs
widely from the type obtained from Barnegat Bay. The latter, save for
the contraction of the oral lobes, is well illustrated in Figure I. Fewkes'
figure shows the surface of the animal covered with discoidal warts which
are claimed by Mayer to be present in M. inccradyi and in .!/. f/ardcni but
absent in M. Icidyi. Fewkes' figure differs also in the shape of the body.

72 ROLLAND J. MAIN.

Near the mouth the line of tentacles curves away from the
labial ridge up to the tentacular bulb. The tentacles are placed
irregularly along this line, usually in groups, some animals having

8

FIG. 3. A. Cross section of auricular groove. The cilia of the auricle
(i) beat up and down as indicated by the arrow and dotted line. The
other two arrows show the direction of the current produced by the cilia.
I. Cilia of auricle. 2. Auricular canal. 3. Auricular groove. 4. Tentacle.
5. Tentacular ridge. 6. Labial ridge. 7. Labial trough. 8. Branch of para-
gastric canal. 9. Oral lobe. B. View of auricular groove from above.
The oral lobe has been laid back. Parts correspond to Fig. ^A. Three ten-
tacles are here shown putting food in the labial trough, where it will be
drawn off and conveyed to the mouth.

many more tentacles than others. This may be due to the fact
that they have been broken off in securing food, for often food
may be seen entering the stomoclaeum with portions of tentacles at-
tached.

FIG. 4. Oral view of adult Mncmiopsis leidyi. This shows how the lips
are continued into the labial ridge and how the trough runs into the mouth.
I. Lip. 2. Mouth. 3. Labial ridge. 4. Labial trough.

When a particle of food is caught in the current produced
by the cilia in the grooves it is whirled about until it finally
touches a tentacle. This entangles it, often with the aid of
several other tentacles. These tentacles then contract, and

FEEDING MECHANISM OF MNEMIOPSIS. 73

apparently are drawn over the labial ridge into the labial
trough, presumably by cilia, Fig. ^B. Here they stretch out in
the direction of the mouth, the food is drawn off, and passes down
toward the mouth. The tentacles then relax, and resume their
normal position. Often several pieces of food are beaten about
for some time in the groove. Dirt in the groove is gradually en-
tangled in mucus into a long thread which slowly passes out at the
aboral end of the groove. If much dirt be present, the whole
animal pulsates, contracting the groove and forcing out all material
present. The tentacles were never seen placing any foreign
material into the labial trough, unless a little happened to be caught
up with the food. Possibly it is for these reasons that Muciniop-
sis Icidyi is not found in muddy waters, since it will not seize
food if much dirt be present. Carmine introduced directly into
the labial trough is drawn along but for a short distance, and then
is passed out over the labial ridge. For this reason the use of the
natural plankton food organisms in studying the mechanism is
imperative.

It is here that we must search for the explanation of why
Mncniiof>sis Icidyi lives so largely upon bivalve larvae, in spite of
the great preponderance of other plankton in the water (Nelson,
'25). The writer has observed that often the ctenophore is un-
able to hold an active copepod. Possibly the stronger swimmers
are able to escape the ciliary currents, whereas the young oyster
shuts its shell on contact and is therefore an easy prey. Poly-
chaste larvae were found in Mncnnopsis at this time, although never
more than one or two per animal. This is contrary to Nelson's
('25) belief that it would be almost impossible for this ctenophore
to ingest such a prey.

Food captured by the tentacles about the mouth was passed
down directly over the lips into the mouth, often aided by a con-
traction of the lips, bringing them near the tentacular bulb. After
the food has passed into the stomodaeum, it usually proceeds slowly
to the center, between the two paragastric canals, close to the con-
voluted tubules which probably secrete the digestive juices. It
may, however, lodge below this point, Fig. 5. Sometimes it is
caught in the swifter current at the very edge of the stomodaeum,
and is whirled up to the beating cilia at the aboral end. Here it

74

ROLLAND J. MAIN.

is usually turned back, for these cilia seem to act partly as filters.
At times, however, a particle may be squeezed through and enter
the funnel to pass around in the food canals.

FIG. 5. The stomodfeum of Mnemiopsis Icidyi. In order to avoid con-
fusion, the paths taken by ingested food are shown on the right side only.
The larger arrows are the more usual paths. The smaller arrows on the
extreme right denote a swifter current, in which the food sometimes
travels. On the left half only, are shown the paths taken by the excreted
materials. I. Mouth. 2. Paragastric canal. 3. Digestive glands? 4. Cilia.
5. Funnel.

The undigested material in the stomodaeum is passed down as
indicated, and ejected through the mouth. These paths are not
definite, for incoming food will pass a certain spot, and immediately
afterwards outgoing wastes will cross the same spot going in the
opposite direction. Those particles which have passed through
into the food canals may reenter the stomodaeum and pass out
through the mouth, or they may follow the usual procedure for
material in the canals, and be voided at the anus.

Just before defecation occurs, particles may be seen gathering
about in the funnel and in the axial funnel canal. Then one of
the branches of this canal elongates above the surface and the
particles are forced out through the pore. The current in all the
food canals now seems to be in the direction of the funnel. By

FEEDING MECHANISM OF MNEMIOPSIS.

75

this time the cilia of the aboral end of the stomodseum have ceased
beating, and the whole upper end of the stomodseum presents a
contracted appearance, Fig. 6. After the particles of waste have

FIG. 6. Aboral portion of stomodaeum, and axial funnel canal of Mnciiri-
opsis leidyi. A. Before defecation. I. Paragastric canal. 2. Cilia. 3.
Food canals. 4. Funnel. 5. Axial funnel canal. 6. Sense organ. 7. Ex-
cretory pore. B. During defecation, arrows showing direction of waste.
Note shrunken appearance of stomodseum.

all passed out the cilia begin beating again, and the branch of the
funnel canal slowly retracts. Although several successive defeca-
tions of specimens have been observed, only one branch was used,
and in no specimen was the use of both branches observed.

THE EARLY DEVELOPMENT OF THE FOOD CATCHING

MECHANISM.

Since the complex food catching apparatus is present only in
the adult Mnemiopsis, the question of its ontogeny naturally
arises. The young were plentiful at the time of this study, and
various stages were examined.

The smallest specimens obtained were in the Cydippidse-stage,
approximately 2 mm. high and 2 mm. broad. Fig. 7. These pos-
sess two long branching tentacles with no trace of the tentacular
ridge, labial ridge, or labial trough. They feed by capturing the
food with the tentacles, retracting them, and drawing them down
over the lip and into the stomodseum, where the food is drawn
off. Another contraction, and the tentacles emerge, to again float
up above the animal.

j6 HOLLAND J. MAIN.

The next step in development was found in a 6 mm. specimen,
Fig. 8. This stage has still the two compound tentacles.

FIG. 7. Young Mnemiopsis Icidyi, 2 mm. high. i. Branching tentacle,
partially contracted. 2. Paragastric canals, only unbranched terminations

shown. 3. Mouth.

\

The 8 mm. specimens are much further advanced, Fig. 9. The
auricles are now forming, and the tentacular ridge has appeared
as a slight fold or line as shown, but it is not connected to the
tentacular bulb, and possesses no tentacles. It was observed that
tentacles never appeared along the tentacular ridge until it had
joined the tentacular bulb.

FK;. 8. Young Mnemiopsis Icidyi, 6 mm. high. i. Tentacular bulb.
(Tentacle omitted, being same as in Fig. 7.) 2. Juncture of paragastric
and auricular canals. 3. Mouth.

It is now easy to see how the adult structures are completed.
As the junction of the paragastric and auricular canals moves up-
ward forming the auricular groove, the tentacular ridge and labial
ridge grow with it. The large branched tentacle disappears, and
small tentacles appear along the tentacular ridge.

FEEDING MECHANISM OF MXEMIOPSIS.

77

This remarkable food catching apparatus of Mnemiopsis, in
which the conveying system seems to foreshadow that of the
bivalves, is certainly a great advance over that of the Scyphozoa.

FIG. 9. Young Mnemiopsis Icidyi, 8 mm. high. i. Branched tentacle en-
tirely retracted, but same as in Fig. 7. 2. Tentacular ridge. 3. Paragastric
canal, termination shown with branches. 4. Mouth. 5. Beginning of labial
ridge. 6. Developing auricles.

Of its efficiency there can be no doubt, for compare Bigelow's
('15) statement that the plankton was greatly diminished in a
swarm of ctenophores. Nelson ('25) also brings forth evidence
of a correlation between the abundance of Mnemiopsis lcld\i
and the intensity of shipworm infestation and oyster sets. More-
over, the fact that the ctenophores are usually found in such vast
and dense swarms, argues well for their ability to obtain food.
Possibly it is due to this efficient apparatus that we find in many
species of ctenophores the small compact bodies and absence of
long trailing tentacles.

SUMMARY.

The mode of feeding was studied in young tentacled forms and
in the adult Mnemiopsis Icidyi. The young capture food with
their branched tentacles, and deposit it in the mouth. The adults
entangle the food with the small tentacles along the tentacular ridge,
and deposit it in the labial trough, whence it is carried to the mouth.

78 HOLLAND J. MAIN.

Food enters the stomodaeum and after digestion is cast out of
the mouth, or it may enter the food canals and pass out of the
anus.

BIBLIOGRAPHY.
Agassiz, A.

'65 North American Acalephae. 111. Cat. Mus. Comp. Zool., No.

II. Harvard.
Bigelow, H. B.

'15 Exploration of the Coast Waters between Nova Scotia and
Chesapeake Bay, July and August, 1913, by the U. S-. Fisheries
Schooner, Grampus, Oceanography and Plankton. Bull. Mu-
seum of Comp. Zool. Cambridge, Vol. LIX., No. 4.
Fewkes, J. W.

'81 Studies of the Jelly-fishes of Narraganset Bay. Bull. Museum
Comp. Zool. Harvard, Vol. IX. On the Acalephae of the
East Coast of New England. Ibid., Vol. VIII.
Gemmill, J. F.
'19 The Ciliation of the Leptomedusan Mclicertidium octocostatum.

Proc. Zool. Soc., 1919.
Kincaid, T.

'J5 Oyster Culture in Washington. Trans. Second Ann. Meet-
ing Pacific Fisheries, San Francisco, p. 4.
Labour, M. V.
'22 The Food of Plankton Organisms. Journ. Mar. Biol. Assn.

Plymouth, N. S., Vol. XII., No. 4, p. 644.
'23 Ibid., Vol. XIII., No. I p. 70.
Mayer A. G.

'12 Ctenophores of the Atlantic Coast of North America, Publ. No.

162. Carnegie Inst. of Washington.
Nelson, T. C.
'23 On the Occurence and Food Habits of Certain Ctenophores.

Anat. Rec., Vol. 26, No. 5, p. 381.

'25 On the Occurrence and Food Habits of Ctenophores in New
Jersey Inland Coastal Waters. BIOL. BULL., Vol. XLVIIL, No. 2.

THE INFLUENCE OF OXYGEN TENSION UPON THE
RESPIRATION OF UNICELLULAR ORGANISMS.

WILLIAM R. AMBERSON.

(From the Department of Physiology, School of Medicine, University of
Pennsylvania, and the Marine Biological Laboratory, Woods Hole, Mass.)

Our knowledge of the influence of oxygen tension upon the
oxygen consumption of unicellular organisms is quite incomplete.
The literature contains many studies of the influence of oxygen
tension changes upon growth and activity of such forms, but
relatively few direct measurements of oxygen consumption have
been made. In some studies in which the consumption has been
measured the problem has been complicated by changes in the
number of respiring cells during the course of the experiment.
This would appear to be true of such observations as those of
Stephenson and Whetham (1924) who have found that the oxygen
intake of B. coli is much greater in pure oxygen than in air, and
of Novy and Soule (1925) who report that the tubercle bacillus
grows best in an atmosphere containing 40-50 per cent, oxygen,
the growth and the oxygen consumption falling off progressively
above and below this value. It is not possible to infer that a
change in division rate indicates a change in the oxygen intake of
the individual bacterium. The influence of the oxygen tension
may be more indirect, possibly through the formation of such
growth-promoting substances as Burrows (1924) has described,
whose production is increased by an increased oxygen supply.

In other studies of bacterial respiration in which there has
probably been no significant change in the number of respiring
cells, Piitter (1924) and E. N. Harvey (1926) have secured
evidence that the respiratory rate is not influenced by changes in
the oxygen tension. In unicellular animal organisms the weight
of the somewhat meagre evidence so far secured indicates that
oxygen consumption is independent of oxygen tension over a wide
range. Lund (1918) found this to be true for I\innnccinm.
Henze (1910) and Warburg (1908) found a similar situation in

79

So WILLIAM R. AMBERSON.

sea-urchin eggs, in which there was little change in oxygen intake
when the oxygen tension varied from double that in air to one-
fourth of the same value.

In all of the studies in this last group in which oxygen has
actually been measured, the Winkler method has been employed.
It is well known that this method, while very satisfactory for the
determination of dissolved oxygen in pure water or in salt solu-
tions, becomes untrustworthy when organic material is present in
the fluids tested. Heilbrunn (1915) and others have objected to
the use of the method in the study of heavy suspensions of pro-
tozoa and marine eggs. The presence of iron, found by Warburg
(1914) to be contained in sea-urchin eggs in considerable amounts,
is known to introduce large errors in the titration. (See Alster-
berg, 1926.)

I became interested in this problem after making the observa-
tion (1924) that the oxygen consumption of a number of marine
invertebrates is directly proportional to the oxygen tension in the
sea water, over a considerable part of the normal physiological
range. This observation has led me to a reexamination of the
problem in other forms. The present communication deals with
some results obtained on unicellular materials in an attempt to
confirm the conclusions of previous workers by methods not open
to the criticisms which can be leveled against the Winkler technique.
This confirmation has been secured. The data are submitted in
support of the older observations, and as giving a more complete
account of the oxygen tension relationships in the Arbacia egg
than has previously been published.

On the technical side an attempt has been made to apply stand-
ard methods of gas analysis to the study of the problem. Novy
and his collaborators have previously successfully used such meth-
ods in their study of bacterial respiration. I find that the oxygen
consumption of unicellular animal organisms can be similarly
followed by such methods, with an accuracy at least as good as
that possible in human and mammalian metabolic studies. The
carbon dioxide production is more difficult to determine because of
the high solubility of the gas in the liquid phase, and the possi-
bility of its chemical fixation. No great reliance can therefore be
placed upon the carbon dioxide values given below, or upon the

INFLUENCE OF OXYGEN TENSION UPON RESPIRATION. 8l

respiratory quotients calculated. The large variations in the value
of the quotient is sufficient to indicate the magnitude of the errors
which must be present in the determination of carbon dioxide.
My main concern has been to study the oxygen consumption.

EXPERIMENTS WITH Panuncciitin.

A group of experiments was first carried out with Parameciuin,
in an attempt to develop a satisfactory technique. For several
reasons the data obtained are not as complete or accurate as the
values secured later on Arbacia eggs. The results are, however,
fairly consistent and give a satisfactory confirmation of Lund's
report on this organism.

A thick suspension of the protozoa was prepared by centrifug-
ing several liters of fluid from a number of cultures. The or-
ganisms were then washed through several changes of tap water,
being concentrated with the centrifuge after each washing. The
suspension in its final form was practically free from bacteria.
The cultures were never entirely pure, but P. caudatum always
constituted at least 95 per cent, of the protozoa present. The
presence of other unicellular organisms, either animal or plant,
cannot appreciably have modified the results.

A preliminary obstacle was encountered when it was observed
that it is exceedingly difficult to secure two samples of such a
suspension which will contain the same number of animals. This
difficulty arises from the high mobility of the organisms which are
negatively geotropic, and tend to rise to the surface even while the
sample is being drawn. After many unsuccessful attempts to
secure two identical samples, the procedure was abandoned. In
its stead it was found possible to carry out two consecutive meas-
urements of respiration upon the same suspension, the first at
atmospheric pressure, the second at some lower or higher pressure.
Under the conditions of the experiments division was absent, yet
the measurements were completed before starvation intervened.

20 cc. of the suspension finally obtained were introduced into a
cylindrical glass vessel, of about the size and form of a Haldane
gas collecting tube. This tube was fitted with three-way stopcocks
at both ends. The volume was 80.85 cc. After the introduction
of the suspension the volume of gas in the tube was, therefore,

82 WILLIAM R. AMBERSON.

60.85 cc. Air delivered by a pump under a small pressure was
now bubbled through the suspension for five minutes. This air
was taken by the pump from a large room in the basement of the
medical building; its oxygen content was slightly lower, and its
carbon dioxide content slightly higher, than in outside air. The
actual percentages were determined by later analysis. At the end
of the equilibration period the tube, completely filled with the room
air, and with the suspension, in gaseous equilibrium with this air,
was closed off, leaving the contained gas completely saturated with
water, at atmospheric pressure, and at approximately 25° C., the
temperature of the room. The tube was then placed horizontally
within a water bath at a temperature of 25° C. ± .2°. From
time to time the tube was gently rocked by hand to keep the sus-
pension approximately in gaseous equilibrium with the air above
it. At the end of three hours the tube was removed and the sus-
pension vigorously shaken into complete equilibrium with the
gaseous phase. A sample of the contained gas was now withdrawn
into a Bailey collector, and set aside for later analysis.

As quickly as possible the same suspension was again equi-
librated with room air. The tube was then partially exhausted by
a water pump, the residual pressure being measured by a mer-
cury manometer connected with one inlet. Upon the attainment
of the desired low pressure the stopcocks were closed, and the
tube placed again within the water bath. At the conclusion of a
second three hour period the gas in the tube was brought to at-
mospheric pressure and a sample collected. At the end of this
second period the organisms were alive and active.

The gas samples were now analyzed by the use of a Haldane-
Henderson gas analyser. Whenever possible duplicate or tripli-
cate determinations were made, and the results averaged. As-
suming the gaseous solubilities to be those given by the standard
tables for pure water at this temperature, the total oxygen and
carbon dioxide present at the beginning and at the end, in both
air and water, were now calculated, the usual corrections for ba-
rometer, water vapor, etc., being applied.

The results obtained in fovirteen experiments are given in Table
I. It is seen that the oxygen intake is practically constant from
200 to 50 mm. Hg partial pressure of oxygen. Below 50 mm. the

INFLUENCE OF OXYGEN TENSION UPON RESPIRATION.

values are somewhat reduced, but down to n mm. the intake is
still at least 80 per cent, of that at atmospheric pressure. Since,
in these experiments, an oxygen gradient must have been present
from air to water, the actual tensions in the water were somewhat
lower than those given in the table, which represent the tensions in
the air. The ability of these organisms to utilize oxygen at low
tensions therefore becomes even more evident.

TABLE I.
RESPIRATION OF Paramcchim AT DIFFERENT OXYGEN TENSIONS.

Respiration in

Respiration in

Ratio be-

Oxygen

First Period.

Second Period.

tween C>2

Ex-

Pres-

Consumption

peri-

sure in

in Second

ment.

Second
Period.

O2
Cons.

CO2

Prod.

R. Q.

02
Cons.

CO2

Prod.

R. Q.

Period and
that in First

Period.

mm. Hg.

*c.c.

c.c.

c.c.

c.c.

I

208-192

1.030

• 703

.683

1.027

• 753

• 733

•997

2

211-195

1.107

•663

• 598

1.167

.762

•653

1.054

3

IS4-I39

.969

• 565

.583

1.025

.640

•625

1.058

4

IS4-I35

1-345

.849

.632

1.390

•903

.649

1-033

5

122-109

•933

.763

.817

1.029

1.016

.986

1.103

6

92-68

2.088

1.490

.714

2. OO2

1.446

.722

•952

7

Qi-74

.216

i. 086

.893

1-245

1.204

.967

1.024

8

70-48

•654

1.302

.787

1.612

1.448

.898

•975

9

70-60

.698

•390

•559

.724

•458

•633

1-037

10

70-57

•131

.676

.598

•973

•553

• 568

.860

ii

60-42

•045

1.028

.686

1.440

1.008

.699

• 875

12

49-28

•592

1.093

.686

1.546

l.«5

.721

.971

13

28-n

1.146

.766

.668

•977

•638

.652

-853

14

28-11

1.642

I-I34

.691

1.290

1.038

.804

.786

Average R. Q. .685 .736

* Volume measured at 760 mm. Hg and O° C.

The average of the respiratory quotients obtained in twenty-
eight determinations comes out to be .710. Considering the wide
range of the individual values it is hardly possible to attach any
great significance to this figure, although it may be taken to suggest
the presence of a fat metabolism under the conditions of the ex-
periment, when the normal food supply is absent.

These preliminary experiments indicated that the method is ap-
plicable to such problems, but certain difficulties were encountered
which made it advisable to complete the study on another material.

84

WILLIAM R. AMBERSON.

These consisted in (i) the impossibility of controlling the activity
of the organisms, (2) the manipulation of gases at pressures very
much below atmospheric, which prevented the exploration of very
low oxygen tensions, and (3) the lack of complete gaseous equi-
librium between air and water during the course of the experiment.
The study was therefore continued with a modified method at
Woods Hole on fertilized Arbacia eggs, which have no independent
motility during the first hours of their development.

EXPERIMENTS WITH FERTILIZED Arbacia EGGS.

In these experiments it has been found possible to secure two
suspensions of eggs containing equal numbers of cells, whose
respiratory exchanges check well with each other when the two
are studied simultaneously under identical conditions. The eggs
were freed from ovarian debris and body fluid and washed through
several changes of sea water. A heavy suspension of cells was
secured by permitting the eggs to sediment in a large beaker and
then pouring off the greater part of the supernatant sea water.
These were then fertilized. About ten minutes after fertiliza-
tion two 60 cc. samples of this suspension were taken up by pipette
and introduced into two tubes similar to that used for Paranic-
chtin but of a somewhat larger volume.

The lower oxygen tensions were secured by mixing oxygen and
nitrogen, or air and nitrogen, in the desired proportions. Eight
liters of such a gaseous mixture were collected in a large bottle,
over water. One tube (B) was then brought into equilibrium
with this mixture, the gas being bubbled through the suspension
for at least five minutes. For the same period the second tube (A}
was equilibrated with outside air. In every case a sample of gas
was collected from the low pressure tube toward the end of the
equilibration, and its later analysis accepted as giving the value
of the initial oxygen and carbon dioxide percentages. The air
which had passed through tube A was analyzed in several experi-
ments and this value accepted for the rest as giving the initial oxy-
gen and carbon dioxide percentages in the high pressure tube. It
showed, after passing through the egg suspension, a slight diminu-
tion in oxygen and a slight increase in carbon dioxide.

At the conclusion of the equilibration the two tubes were closed

INFLUENCE OF OXYGEN TENSION UPON RESPIRATION. 85

in such a manner that the contained gas was left at atmospheric
pressure and at approximately 20° C. They were then placed
side by side within a water bath, and rotated continually throughout
the experiment, turning at the rate of about thirty times a minute.
Under these conditions the eggs were always evenly distributed
throughout the suspension, and kept in constant motion, the water
was always nearly in equilibrium with the gas, and cleavage pro-
ceeded in a perfectly normal manner.

Running sea water was used in the water bath. Its tempera-
ture varied slightly from day to day. The lowest temperature re-
corded in any experiment was 18.2° C., the highest 20.2° C. The
experiments continued in most cases for two hours ; in a few cases
for three hours. The first division occurs about one hour after
fertilization at this temperature ; subsequent divisions follow about
every thirty minutes. At the end of the two-hour experiments
the eggs were in the four and eight cell stage; at the end of the
three-hour experiments they were in the sixteen and thirty-two
cell stage. The material is not, therefore, unicellular throughout
the whole experiment. The individual cells, however, in all of
these early stages are all at the surface of the dividing egg in
intimate relation with the oxygen supply in the water ; there seems
every reason to believe that the relationship under investigation
will not be materially modified by this increase in number of cells,
unaccompanied by any change in the mass of respiring tissue. We
have reason to believe from the work of Gray (1925), that
cleavage itself does not affect the rate of oxygen consumption, and
that, after the first sharp rise following fertilization the consump-
tion is practically constant during the first three hours of develop-
ment. The unfertilized egg has so low a gaseous exchange that it
has not proven practicable to follow its respiration by the present
method.

At the end of the experiment samples of gas were secured from
both tubes and analysed. The oxygen and carbon dioxide in the
gas and in the sea water were then calculated for the beginning
and for the end of the experiment. For this calculation the ab-
sorption coefficients for oxygen and carbon dioxide in sea water
given in Tabulae Biologicae (Vol. 4, pp. 571-578) were used. The
results of a typical experiment are as follows :

86 WILLIAM R. AMBERSON.

Tube A. Tube B.

Oxygen tensions during experiment . 155 to 142 mm. Hg. 61 to 49 mm. Hg.

Volume of tube 106.15 c.c. 105.39 c.c.

Volume of suspension 60 c.c. 60 c.c.

Gas Analysis at beginning:

O0 20.87% 8.22%

CO., 05% .02%

N0_~ 79-08% 91-76%

Gas analysis at end (corrected for volume change) :

O, 19.16% 6.59%

c60 62% .58%

N, " 79.08% 91-76%

Oxygen in air and water :

At beginning 9-96i c.c. 3.863. c.c.

At end 9.147 c.c. 3-095 c.c.

Oxygen Consumption 814 c.c. .768 c.c.

Carbon dioxide in air and water :

At beginning 045 c.c. .018 c.c.

At end 55§ c.c. .517 c.c.

Carbon dioxide production 513 c.c. .499 c.c.

Volumes corrected to dry values at O° C. and 760 mm. Hg.

Oxygen consumption .741 c.c. .699 c.c.

Carbon dioxide production 467 c.c. .454 c.c.

Respiratory quotient .630 .649

Oxygen consumption at low pressure = 94.4% of that at atmospheric pres-
sure.

Carbon dioxide production atJow pressure = 97.3% of that at atmospheric
pressure.

The results obtained in twenty experiments, carried out after
the preliminary tests, are given in Table 2, and shown graphically
in Fig. i. The oxygen consumption is seen to be practically con-
stant from an oxygen pressure of 228 mm. Hg. down to about
20 mm. Hg. Between 80 and 20 mm. there is a definite downward
trend in the values, but at 20 mm. the consumption is still about
90 per cent, of that at atmospheric pressures. Below this point
the consumption falls off sharply.

In Fig. i the experimental values are shown as rectangles. The
height of this rectangle corresponds to i per cent, on the oxygen
consumption scale; the length indicates the oxygen tension range
in tube B during the course of the experiment. Each rectangle
shows that over this range the oxygen consumption of the egg
suspension in tube B was the indicated percentage of the con-
sumption in tube A, run at atmospheric pressure. The absolute

INFLUENCE OF OXYGEN TENSION UPON RESPIRATION.

TABLE II.
RESPIRATION OF FERTILIZED Arbacla EGGS AT DIFFERENT OXYGEN TENSIONS.

Respiration in

Respiration in

Ratio be-

Ex-
peri-

Oxygen
Pressure in

Tube A.

Tube B.

tween Ot
Consumption
in Tube B

ment.

Tube B.

02

CO2

O2

C02

and that in

Cons.

Prod.

R. Q.

Cons.

Prod.

R.Q.

Tube A.

mm. Hg.

*c.c.

c.c.

c.c.

c.c.

I

228.8-220.0

•423

•394

•931

•433

.294

.679

1.024

2

155.2-147.2

•473

•430

.909

.470

•386

.821

•994

3

152.2-144.6

•443

•329

•742

•447

•350

•783

1.009

4

142.0-135-7

•309

•257

•832

-315

• 237

• 753

1.019

5

123.2-112.4

•524

•436

•832

• 533

.326

.611

1.017

6

116.8-104.3

.691

•443

.641

• 733

.448

.611

1.061

7

85.5- 76.0

.650

.496

•763

.621

•403

.648

•955

8

70.6- 61.7

-572

.417

.729

.520

.412

.792

-909

9

66.6- 55.4

-665

.528

•794

•653

•493

•754

.982

10

61.2— 49.0

.741

.467

•630

.699

•454

.649

-944

ii

44.6- 38.3

-390

•334

-856

•370

-339

•915

-949

12

36.8- 24.2

.818

.611

.768

-735

• 535

.727

•899

13

30.0- 24.2

.406

.279

.688

•345

.217

.628

.850

14

23.9- 14.9

• 592

•524

• 885

15

23-7- 8.7

.856

• 593

.697

.862

• 558

.648

1.007

16

17.3- 10.1

.674

•444

•658

.419

.360

• 859

.622

17

ii-S- 6-3

.636

• 543

• 854

•367

•427

1.160

• 577

18

7-9- 3-3

-565

.667

1.181

.268

.269

1.004

•457

19

7.1- .8

.746

• 527

.706

•309

.488

1.582

.414

20

4-3- 1-7

.665

•445

.669

•151

.222

1-473

.227

Average R. Q. .783 Average R. Q. (1-16) .725

* Volume measured at 760 mm. Hg. and O° C.

100;?

120

340

^r

I TJKSIOJ. 1QL Hg. /£

FIG. I. Oxygen consumption of fertilized Arbacia eggs at different Lj I j_ | B F
oxygen tensions. The range of tensions within which the division rate \,
is affected is also graphically shown.

88

\VILLIAM R. AMBERSON.

values vary considerably from experiment to experiment, but the
graph of these percentages assumes a fairly regular and consistent
form.

Correlated with the diminished oxygen intake at very low oxy-
gen tensions retardation in development was observed in experi-
ments 17-20. In all other experiments the eggs in the low pres-
sure tube had developed as far as had those at atmospheric pres-
sure. In every case 95-100 per cent, of the eggs developed. In
experiment i/, continuing for two hours, a slight retardation in
division rate was evident. Counts on 100 eggs from each suspen-
sion gave the following values :

i-cell.

2-cell.

4-cell.

8-cell.

Tube ,4 (HighO2)

4

6

^7

-1-1

Tube B (Low Oz)

c;

14

74

7

In experiment 18 (2 hours) a more marked effect was observed.
Counts on 100 eggs gave the following values:

i -cell.

2-cell.

4-cell.

8-cell.

i6-cell.

Tube A (High O>) . . .

2

o

C2

4O

6

Tube B (Low O)

"?O

7

o

o

In experiment 19 (3 hours) the eggs at atmospheric pressure were
in the sixteen and thirty-two cell stage. In tube B about 80 per
cent, had reached the four-cell stage, but none were found in later
stages. In experiment 20 (2 hours) the eggs at atmospheric
pressure were in the four and eight-cell stage. In tube B a care-
ful search failed to reveal any cleavage whatsoever. It has long
been known that in the complete absence of oxygen cleavage in
these eggs is prevented. (See E. B. Harvey, 1926.) My own
observations would suggest that a certain minimal concentration
of oxygen is necessary for division, but the matter has not received
a thorough study. The range of oxygen tensions within which
development is either retarded or prevented is indicated graphically
in Fig. i. The values, taken from four experiments, are to be
considered as approximations only. Taken in conjunction with
the curve of oxygen consumption they show the great ability of

INFLUENCE OF OXYGEN TENSION UPON RESPIRATION. 89

these eggs to carry out a normal development clown to very low
oxygen tensions.

It is of interest to note that in all four of these experiments in
which retardation or inhibition of development occurred the res-
piratory quotient rose ahove unity; in experiments 19 and 20 the
quotient reached the high values of 1.58 and 1.47. These figures
suggest the presence of anaerobic respiratory processes at these
low oxygen tensions. It is not possible to be certain concerning
the matter, since, under these conditions of oxygen lack, acid
metabolites may collect in the suspension and liberate carbon
dioxide from the carbonates of the sea water.

In none of these experiments has the tension of carbon dioxide
risen to such a point that it can have materially affected develop-
mental rate. Haywood (1927) has shown that, in high concen-
tration, carbon dioxide behaves as a narcotic and completely pre-
vents cleavage when its tension rises above 230 mm. Hg. Below
this value cleavage occurs at a rate slower than normal. The
threshold tension for this carbon dioxide effect to appear was not
determined, but it seems evident that at very much lower con-
centrations the retardation of development must become negli-
gible. The highest carbon dioxide value observed in the present
study was at the end of experiment 18, when the partial pressure
reached 7 mm. Hg in tube B. The retardation of development
observed at low oxygen tensions must therefore be caused by
oxygen lack rather than by a narcotic effect of the carbon dioxide
produced. Haywood also reports experiments on the influence of
low oxygen tension upon developmental rate which agree with
my own findings in showing practically no influence down to quite
low values.

In most experiments carried out below an oxygen tension of
50 mm. Hg there was observed, at the end of the experiment, a
liberation of pigment in the suspension in the low pressure tube
which became more and more marked as the oxygen tension was
lowered. This liberation of pigment apparently arose from the
cytolysis of a certain number of cells. The actual percentage of
eggs thus destroyed was not determined, but must have been small,
since at the end of the experiment the volume of the eggs after
sedimentation was not appreciably diminished. The downward

GO WILLIAM R. AMBERSON.

trend in the oxygen consumption values below 80 mm. Hg may
be in part due to this destruction of a small number of the eggs,
although we know, from the work of Warburg (1914) that
respiratory exchanges may continue for some hours even in com-
pletely fragmented sea-urchin eggs, at a level not far below that
fovind when the cells are intact.

The ability of both protozoa and sea-urchin eggs to carry on a
normal respiratory exchange down to very low oxygen tensions
points very definitely to the normal presence, within the cells, of
a considerable oxygen tension. Oxygen is present in such amount
that it does not limit the metabolism, whose rate is determined
by other than oxidative reactions.

SUMMARY.

By standard methods of gas analysis the respiratory exchanges
of Paramecium and of fertilized Arbacia eggs have been studied.
The respiratory rate in both materials is found to be practically
constant over a wide range of oxygen tensions, thus confirming
older work done by other methods.

In the fertilized Arbacia egg the oxygen consumption is prac-
tically constant between 228 and 20 mm. Hg partial pressure of
oxygen. Between 80 and 20 mm. Hg there appears to be a slight
diminution in oxygen intake, but at 20 mm. Hg the consumption
is still about 90 per cent, of that at atmospheric pressure. Below
20 mm. Hg the consumption is sharply reduced.

The cleavage of Arbacia eggs proceeds at a normal rate down
to very low oxygen tensions. No retardation in development has
been observed above n mm. Hg. Below this value the rate be-
comes slower and cleavage ceases entirely below 4 mm. Hg.

BIBLIOGRAPHY.
Alsterberg, G.

'26 Die Winklersche Bestimmungsmethode fur in Wasser gelosten

elementaren Sauerstoff. Biochem. Zeits., 170, 30-75.
Amberson, W. R., Mayerson, H. S., and Scott, W. J.
'24 The Influence of Oxygen Tension upon Metabolic Rate in In-
vertebrates. Journ. Gen. Physiol., 7: 171-176.
Burrows, M. T.

'24 Relation of Oxygen to the Growth of Tissue Cells. Amer.
Jour. Physiol., 68: no.

INFLUENCE OF OXYGEN TENSION UPON RESPIRATION. 91

Gray, J.

'25 The Mechanism of Cell Division. Oxygen Consumption during

Cleavage. Proc. Camb. Phil. Soc., i: 225-236.
Harvey, E. B.
'27 The Effe,gt of Lack of Oxygen on Sea Urchin Eggs. BIOL.

BULL., ^53: 147-160.
Harvey, E. N.
'25 The Total Luminous Efficiency of Luminous Bacteria. Jour.

Gen. Physiol., 8: 89-108.
Haywood, C.

'27 Carbon Dioxide as a Narcotic Agent. BIOL. BULL., 53:450-464.
Heilbrunn, L. V.

'15 The Measurement of Oxidation in the Sea-Urchin Egg. Sci-
ence, N. S., 42: 615-616.
Henze, M.
'10 Uber den Einfluss des Sauerstoffdrucks auf den Gaswechsel

einiger Meerestiere. Biochem. Zeits., 26: 255-278.
Lund, E. J.

'18 Relation of Oxygen Concentration and the Rate of Intracel-
lular Oxidation in Paramecium Cau.datitni. • BIOL. BULL., 45:

351-364-
Novy, F. G., and Soule, M. H.

'25 Respiration of the Tubercle Bacillus. Jour. Inf. Dis., 36: 168-

232.
Putter, A.

'24 Die Atmung der Planktonbakterien. Arch. ges. Physiol., 204,

94-126.
Stephenson, M., and Whetham, M.

'24 The Effect of Oxygen Supply on the Metabolism of Bacillus

Coli Communis. Biochem. Jour., 18: 498-506.
Warburg, O.
'08 Beobactungen fiber die Oxydationsprozesse in Seeigelei. Zeits.

Physiol. Chem., 57: 1-16.
'14 Zellstruktur und Oxydationsgeschwindigkeit nach Versuchen

am Seeigelei. Arch. ges. Physiol., 158: 189-208.

'14 liber die Rolle des Eisens in der Atmung des Seeigeleis. Zeit.
Physiol. Chem., 92: 231-256.

A COMPARISON OF THE OXYGEN CONSUMPTION OF

UNFERTILIZED AND FERTILIZED EGGS OF

FUNDULUS HETEROCLITUS.

MARJORIE BOYD.
(From the Marine Biological Laboratory, Woods Hole.)

Since Warburg (i) in 1908 measured the oxygen consump-
tion of Arbacia eggs and observed the marked increase follow-
ing fertilization, the oxidation processes in marine eggs and em-
bryos have been carefully investigated. The rate at which the
oxygen is removed from the surrounding air or sea water has
been correlated with the stages in development. Thus Shearer (2)
measured the oxygen consumption during fertilization of Echino-
derm eggs, and found a decided increase upon the addition of
the sperm ; " more oxygen is taken up in the first minute of the
process than at any subsequent interval of the same time." In
another article Shearer (3) states that, in the first hour of devel-
opment, the fertilized egg consumes six to seven times as much
oxygen as the unfertilized egg. In the star fish egg, however, ac-
cording to Loeb and Wastenys (4) there is no increase in the
oxidation rate after fertilization.

The rate of oxygen consumption is also correlated closely with
heat production. Rogers and Cole (5) in their work on Arbacia
eggs have shown how the heat production varies before, during,
and after fertilization ; according to them " the rate of heat pro-
duction at the instant of fertilization is ten to twelve times that
of the unfertilized egg."

The literature upon this subject reports work done almost ex-
clusively upon invertebrate eggs. Apparently no previous study
of the influence of fertilization upon respiratory rate has been
made on any vertebrate egg. Scott and Kellicott (6) and Hyman
(/), who have measured the oxygen consumption of Fimdnlus
embryos at various stages of development, made no observations
on the respiration during the first two hours after fertilization,
and secured no information as to the influence of fertilization

92

OXYGEN CONSUMPTION OF EGGS OF FUNDULUS.

93

itself. The present study represents an attempt to secure such
information. It has been possible to show by several methods
that fertilization markedly increases the oxygen consumption of
the eggs of Fiuidiilns hctcroclitus. The time relations of this
increase are of some interest.

i. WINKLER METHOD FOR DETERMINATION OF DISSOLVED

OXYGEN.

The Winkler method as applied to this problem was employed
in the manner described by Amberson, Mayerson and Scott (8).
600 eggs were placed in 500 cc. of sea water in each of two small
Erlenmeyer flasks. Samples for analysis were withdrawn through
siphons. The water surface was covered with paraffin oil to
minimize the diffusion of new oxygen from the air into the water.
The sea water was analyzed for dissolved oxygen previous to
experimentation ; the initial sample was withdrawn after twenty
to forty minutes. In order to secure successive determinations of
the dissolved oxygen during an extended time, it was necessary to
adopt a micro-Winkler method as suggested by Lund (9). For
these analyses small vials of 6.5 cc. capacity were used. Fig. i
represents the graph resulting from plotting the data shown in
Table I. below. The values for the amount of oxygen consumed
during a given period are obtained by subtracting the amount
remaining in the sea water at the end of that period from the
amount originally present in the sea water used for the experiment.

TABLE I.

Time.

Sea Water Originally Contains 5.1 cc. Oxygen per Liter.

Unfertilized Eggs.

Fertilized Eggs.

02

Remaining.

02
Consumed.

O2
Remaining.

Ot

Consumed.

20 min

5-0
4.9

4.6
3-8
3-i

O.I

0.2

0.5

1-3

2.O

4-7
4-5
3-7
3-4

2.8

1.8

0.4
0.6
1.4
i-7
2-3
3-3

45 min

7 hrs

10 hrs and 35 min

25 hrs. and 25 min
31 hrs. and 25 min

94

MARJORIE BOYD.

<u .S

.> S

en Q.

« rr

qj •<

o <y
o

o ^

rt ~

o £

-4— '

rt -c

— o

w ir.

O u

^H a;

S 'rt

oj >

00 i*

CJ

p rt .S

o 55 ""

-*-t

, Wl

CU <U r-
*> *""* • ^-

X *-• rt
o be

.2 -S ^

2 ^1

"o 13 ^

S'l^
|'i-=

^ r- JS

aj C 5
— M —

SH

'd - . ~
X "^ s-<
2 <u o

rO > •*-"

-" T— i OJ

<u o *rr
^ in

.— 2

>- ^ ^

^ «4- ^

•^ ° ."

• — -4->

^ 5 S

o P

g S 3

o « S

Si 8

TJ
<U

r oj
5 H M

nj O ^

X!
O

X
O

m

^ O

C/5

•S *"

5 O
tn u

OJ M-c

« .

O ^~

rt rt

O
fe

CT, f t

OXYGEN CONSUMPTION OF EGGS OF FUNDULUS.

95

It is to be noticed that the fertilized eggs had at the time of each
determination consumed more oxygen than the unfertilized eggs.
Data from similar experiments, as shown in Table II., show that
the rate of oxygen consumption is most rapid during the first two
hours following fertilization.

TABLE II.

Time.

Sea Water Originally Contains 5.2 cc. Oxygen per Liter.

Unfertilized Eggs.

Fertilized Eggs.

Oz Remaining.

Oz Consumed.

Oz Remaining.

Oz Consumed.

20 min. . . .
i hr. . . .

5-i
5-0
4.9
4.8

O.I
0.2

0.3
0.4

4-7
4-4
4.0

3-9

0.4
0.8

1.2

1-3

2 hrs.

4 hrs.

2. MICRO-RESPIROMETER METHOD FOR DETERMINATION OF

OXYGEN.

The type of micro-respirometer that was used for the study of
oxygen consumption by the Fnudulus eggs is one that has been
devised by W. O. Fenn for similar studies of Arbacia eggs. A
small glass bottle with a ground glass neck is fitted with a ground
glass stopper which is connected with a horizontal fine-bore man-
ometer provided with a centimeter scale. In the center of the
bottom of the bottle is a small compartment into which 0.5 cc. of
15 per cent. NaOH is introduced; the eggs to be studied are
placed in the space surrounding the compartment. The NaOH
serves to absorb the CO2 produced by the eggs. A small drop
of kerosene is introduced into the manometer and its movement
across the tube from the outer to the inner end indicates both the
amount of oxygen consumed and the rate of the process.

Five cc. of sea water, containing fifty Fund id us eggs, were
pipetted into the micro-respirometer. Two micro-respirometers
were used so that experiments on unfertilized and fertilized eggs
could be carried out at the same time under identical conditions.
The constants of each apparatus were found by calibration of the
respective manometers. The micro-respirometers were immersed
in a bath of running sea water; the temperature for all of the

LIBRARY:;
~.-

96

MARJORIE BOYD.

aiwnsMoo lo jo ixnawv -moo.

X

W

OXYGEN CONSUMPTION OF EGGS OF FUNDULUS. 97

experiments proved to be 20.2 ± .6° C. The readings of the
meniscus of the kerosene drop were taken at intervals of fifteen
minutes. Over twenty series were run, hut in only the last five
experiments were the temperature variations observed with a Beck-
man differential thermometer.

Figure 2 shows typical curves for the results obtained. In
Experiment I. the number of cubic millimeters of oxygen con-
sumed by the fertilized eggs is a little less than twice the corre-
sponding amount in the case of the fertilized eggs in Experiment
II. Nevertheless both curves show the same marked increase in
oxygen consumption 45 minutes after fertilization. This in-
creased oxygen utilization reaches a maximum during the period
from 60 to 90 minutes following fertilization. From this time
on, the amount of oxygen consumed per unit time falls so that the
rate of utilization approximates that for the unfertilized eggs.
It would appear, therefore, that the oxygen requirements of the
unfertilized Fundulus eggs are increased by fertilization. The
time relations of this increase are of theoretical interest ; they are
indicative of some oxidation process occurring within the egg for
which an increased oxygen intake is a necessity. To follow the
development in relation to the time, two control sets of 50 Fundulus
eggs were placed in sea water and the stages of development were
traced by microscopic inspection. It was found that the increased
rate of oxidation occurs at a time before and during the appearance
of the groove in the surface of the blastodisc which initiates the
first cleavage. The subsequent cleavages evidently do not require
such a marked rate of oxygen intake. A single run with twenty
9-day old Fundulus embryos revealed a later rise in the oxygen
consumption which can be correlated with the marked rise in the
rate which Scott and Kellicott found to occur at the time circula-
tion is established.

To show still further the peculiarity of the time relations, the
average amount of oxygen consumed per 10 minutes was calcu-
lated from the data of Experiment I. and is shown in Fig. 3. The
difference in the rates of the fertilized and unfertilized eggs is
markedly contrasted.

A few experiments were carried out by a third method and
gave results that checked qualitatively with the two mentioned

98

MARJORIE BOYD.

'Ot.

-

OJ
N

O <L>

11

o c
" o

o ^

<u o

•*-* bjO

^ X

OJ O

*^J <-t-H

^ o

OJ 15

U. _r;

c H

E «
c o

rt

OJ

T3

<u

>,

_. -

en C t«

^ S -

« I 8

. •"' 03

o ^ o

£ 2 'tJ

-S 03

g ^

OXYGEN CONSUMPTION OF EGGS OF FUNDULUSi 99

above. This method involved the analysis of air above 25 cc. of
water in a 100 cc. Haldane gas collector. At the beginning of the
experiment 200 Fundulus eggs were pipetted into each of two
collectors with the sea water, and the water was equilibrated with
atmospheric air which filled the vessel. The collectors were then
sealed, immersed, and rotated in a bath of running sea water for
two hours. More oxygen was found to have been lost from the
sample of air taken from the collector containing the fertilized
eggs than from that containing the unfertilized. This method
proved to be only approximate as the rotation caused the eggs to
stick together in a clump and 'normal development did not take
place. The data secured gave a qualitative confirmation of the
more accurate results obtained by the two other methods.

SUMMARY.

By three methods it has been shown that the oxygen consump-
tion of the eggs of Fundulus heteroclitus is greatly increased after
fertilization. This increased rate of oxygen consumption is at
its maximum from 60 to 90 minutes after the addition of the
sperm, in a period immediately preceding the first cleavage. The
oxygen consumption then falls to a level practically identical with
that of the unfertilized eggs.

The writer wishes to express her appreciation to Dr. \Y. R.
Amberson and Dr. W. O. Fenn for their suggestions, and to Mr.
J. O. Pinkston for his assistance in the oxygen determinations.

BIBLIOGRAPHY.

1. Warburg, O.

'08 Beobactungen iiber die Oxydationsprozesse in Seeigelei. Zeits.
Physiol. Chem., 57: 1-16.

2. Shearer, C.

'22 On the Oxidation Processes of the Echinoderm Egg During
Fertilisation. Proc. Roy. Soc., B, 93: 213-229.

3. Shearer, C.

'22 On the Heat Production and Oxidation Processes of the Echi-
noderm Egg during Fertilisation and Early Development.
Proc. Roy. Soc., B, 93: 410-425.

4. Loeb, J., and Wasteneys, H.

'12 Die Oxydationsvorgange in befruchteten und unbefruchteten
Seesternei. Arch. f. Entwick-Mechanik., 35: 555-557-

IOO MARJORIE BOYD.

5. Rogers, C. G., and Cole, K. S.

'25 Heat Production by the Eggs of Arbacia punctulata during Fer-
tilization and Early Cleavage. BIOL. BULL., 49: 338-353.

6. Scott, C. G., and Kellicott, W. E.

'17 The Consumption of Oxygen during the Development of Fim-

dulus heteroclitusi. Anat. Record, n: 531-533.
/. Hyman, L. H.

'21 The Metabolic Gradients of Vertebrate Embryos. I. Teleost
Embryos. BIOL. BULL., 40: 32-72.

8. Amberson, W. R., Mayerson, H. S., and Scott, W. J.

'24 The Influence of Oxygen Tension upon Metabolic Rate in In-
vertebrates. Jour. Gen. Physiol., 7: 171-176.

9. Lund, E. J.

'21 A Micro-Winkler Method for the Quantitative Determination
of Dissolved Oxygen. Proc. Soc. Exp. Biol. & Med., 19: 63-
64-

STUDIES ON DALLASIA FRONTATA STOKES.
I. POLYMORPHISM.

GARY N. CALKINS AND RACHEL BOWLING.

Dallasia frontata is a common fresh-water ciliate classified in
the family Chiliferidse, sub-order Trichostomina, order Holo-
trichida. It was originally described in 1886 by Stokes and char-
acterized by him in 1888 as follows : " Body elongate-obvate, sub-
cylindrical, transparent, longitudinally striate, and finely
reticulated, five times as long as broad, the lower or ventral sur-
face convex, the dorsal slightly concave, tapering posteriorly to
a somewhat retractile tail-like prolongation forming about one-
fifth of the entire body ; anterior extremity narrowed, obtusely
pointed ; oral aperture narrow, ovate, obliquely placed on the
ventral or convex surface at some distance from the anterior ex-
tremity, enclosing two small vibratile membranes ; contractile
vacuole single, spherical, near the center of the dorsal or concave
border ; nucleus presumably represented by large, ovate, sub-cen-
tral clear space. Length of body 1/180 of an inch. Habitat.—
Still water, with Myriophyllum" (1888, p. 17 1).

This characterization is not adequate to describe the many form
changes which this remarkable organism passes through in its life
history, changes which we are not yet ready to interpret as to cause
or full significance, but which we will describe in the following
pages.

The organism may be found without much difficulty in the water
of Van Cortlandt pond in the environs of New York. Unlike Uro-
leptits inobilis it cannot be regarded as a rare form and is probably
widely distributed in fresh-water ponds throughout the country.
Many individuals were encountered in the autumn of 1927 and
individuals were isolated in different types of media in isolation
culture dishes usually employed for this work. Initial experi-
ments with culture media including pond water, hay infusion, and
the combination of hay and flour soon showed that the latter, as

101

IO2 GARY N. CALKINS AND RACHEL BOWLING.

in the case of Uroleptus mobilis, was the most favorable. This
medium, made up in the same way as for Uroleptus mobilis during
tight years of culture, has been consistently used for some of our
material since October 6, 1927. Later, media made up with rice
and with cracked wheat were tried and some of our material is
now successfully running on the latter. In this, as in the hay-
flour medium, individuals in the period of maximum vitality di-
vide from four to six times in twenty-four hours.

The material of the isolation cultures is run in " series " and
" lines." A series is made up of all the progeny of a single indi-
vidual isolated as an ex-con jugant; lines, usually five in number,
are isolation cultures made from the first five individuals formed
by division of the ex-conjugant. Individuals from each line are
picked up with a capillary pipette and transferred daily to another
culture dish of fresh medium. After such isolations are made the
unused individuals of a series are transferred to a Syracuse dish
containing about 10 cc. of fresh medium. Such reserve material
is allowed to multiply with no change of the medium for from six
to ten days. It constitutes a " conjugation test " such as proved
successful with Uroleptus mobilis. In this way abundance of
material is available for study. With Dallasia after a few weeks,
epidemics of conjugation occurred in the Syracuse dishes and pedi-
greed series were started.

Material for preparations has been fixed in osmic fumes, Flem-
ming's, Hermann's and Schaudinn's fluids. The latter, made up
as a saturated solution in 95 per cent, alcohol is most satisfactory
for general staining. Iron haematoxylin is good for general
topography but inner cellular structures are obscured by the dense
cortical zone of deeply staining granules. This however, may
be avoided by prolonged treatment with turpentine. Auerbach's
combination of methyl green and acid fuchsine (without orange
G) is excellent for cortical structures and for the mouth parts.
Vital stains are useful for demonstrating some structures par-
ticularly the capsules about the " couples."

The derived organization of Dallasia is so delicately adjusted to
its environment that small changes in the latter cause remarkable
changes in form. This leads to polymorphism which, more than
with any other free-living protozoon known to us, is character-

STUDIES ON DALLAS1A FROXTATA STOKES.

103

8

104

GARY N. CALKINS AND RACHEL BOWLING.

istic of this ciliate. Certain well-marked types of organization
follow in the same order. To these we have applied purely col-
loquial names which have no resounding classical roots indeed,
but which enable us to distinguish clearly between the forms indi-
cated by them. These forms are (i) tails; (2) boats; (3) couples
(gametes) ; and (4) pairs, and they will be described in this order.

i. Tails. — This term is an abbreviation for " tail-bearing forms "
such as indicated by the original description of Stokes. They are
relatively large (105^ to 140/^X22^ to 36^) and, owing to the
remarkable mouth have a curious resemblance to a microscopic
shark a resemblance to which Stokes called attention. The an-
terior end is rounded and in most cases this is the broadest part
of the organism which tapers gradually to the posterior end where
it narrows into a well-marked tail (Fig. I., 4 and Fig. III., i).
The tail is quite variable in length and shape. Sometimes it is
long, resembling the handle of a skillet; again it is reduced until
it is little more than the sharply-pointed posterior end of the cell.
In other cases the tail disappears entirely. These forms are
lairly sluggish, richly stored with food, and are usually attached
to the substratum by the tip of the tail where they swing about
in circles with the tips of the tails as centers. The cilia are long
and closely set in longitudinal rows of which there are about
forty.

Another type of tailed form is much longer and somewhat more
slender but unlike the fat form is very active and rarely becomes
attached.

So far as the visible structures are concerned the most com-
plex part of the organism is the mouth. (Fig. II., i). This is
relatively large and lies in the anterior quarter of the cell. The
external aperture varies in shape from an elongated slit to a cir-
cular opening. It leads into a spacious buccal pouch (B. p.) which
extends inwards and diagonally from a region slightly anterior to
the mouth, to the gullet which is posterior to the mouth. The
entire apparatus is about 27/4, long and 15^ wide, thus taking up
about one-fifth of the organism. On the floor of the buccal pouch
is a long tongue (T}, triangular in cross section, which runs al-
most the full length of the pouch (Fig. II.). On one side of this
and at the anterior end is a broad endoral membrane which fre-

STUDIES ON DALLASIA FRONTATA STOKES.

105

quently protrudes from the mouth (Fig. II., E.m.}. At the re-
gion of the gullet and on the opposite side of the tongue is a
long, narrow undulating membrane, the adoral membrane (Fig. II.
A.m.}. From the base of this membrane to the gullet is a long
ladder-like structure recalling the " railroad track " of Chlamy-
dodon (A.c). There is evidence of a complicated neuro-motor
system which will be described in a later paper on the cytology
of these forms.

The macronucleus, like the organism, is polymorphic. In many
individuals it appears to be emarginate, frayed out and of a de-

E.m

A. m.

B

FIG. II. Mouth and buccal pouch of Dallasia frontata.

A. Total preparation of tailed form ; mouth and buccal pouch only.

B. Transverse section of tailed form.

A.c., ladder; A.m., adoral membrane; £./>., buccal pouch; E.m., endoral
membrane ; M.t mouth opening ; T., tongue and supporting bars.

cidedly unhealthy appearance. It is often splinter-like and irregu-
lar, but at the approach of division it becomes more condensed and
homogeneous and ellipsoidal in form. It divides without mitosis
and in the characteristic manner of most macronuclei.

The micronucleus is usually single, spherical, and homogeneous,
and is closely applied to the macronucleus. It divides by mitosis.

The contractile vacuole is a single vesicle, in the middle of the
ventral surface; feeding canals are absent but a ring of feeding
vesicles, clearly visible after contraction of the vacuole, are present.

The cytoplasm is filled with great vacuoles which are frequently
so abundant as to distort the organism. They are gastric vacuoles
for the most part but are frequently merely fluid-filled vesicles.

IO6 GARY N. CALKINS AND RACHEL BOWLING.

Granules of large size and great number are present in all stages
of these tailed forms and are a nuisance in preparations stained
with iron haematoxylin. The majority of them stain well with
the vital dyes neutral red, brilliant cresyl blue and methylene blue.
All in all these tailed forms are remarkably variable in size and
shape. They appear to be highly sensitive to environmental
stimuli readily becoming amorphous and variously distorted. If
the medium is too rich such distortions are more numerous. For
some unaccountable reason, possibly because of incomplete reor-
ganization after division, minute dwarf forms with grotesquely
large mouths are frequently seen (Fig. III., 4). Such types are
prone to change into distinctly amoeboid forms with protoplasmic
processes which cannot be distinguished from pseudopodia

(Fig. III., 4a).

2. Boats. — In form and character boats are quite different from
the tailed forms. They are considerably smaller (68/x to 83^)
and are derived from the tailed forms by gradual absorption of
the attenuated caudal extremity. Both anterior and posterior ends
become rounded and the organism becomes navicular and sym-
metrical (Fig. I., 8; Fig. III., 10). The environmental condi-
tions under which the transition occurs have not yet been fully
determined but the period required for it varies according to the
age of the series. It is a striking phenomenon to see a rich stock
culture in fresh medium yield nothing but boats twenty-four hours
later. Such boats are not attached but shoot about the culture dish
with amazing speed. After another 24 hours the majority of
them have divided four times, giving rise to sixteen minute cells
which separate off in pairs to form the couples. After the first
division of the boats the daughter cells (gamonts number i) are
more quiet than the original boat and have a tendency to rest on
the bottom where the second division takes place. The daughter
cells of this second division (gamonts number 2) still have the
ability to move but their movement is more or less spasmodic and
irregular and their daughter cells (gametocytes) derived from
a third division, are now quiescent (Figs. I., 10, and III., 13).
These forms, however, are rarely found on the bottom but, to-
gether with the couples, are suspended in the medium.

The early stages of the boats show mouth parts only slightly

STUDIES ON DALLASIA FRONTATA STOKES.

107

FIG. III. Dallasia frontata Stokes. Life cycle. Camera lucida sketches
from preparations. Same magnification throughout.

1. Vegetative individual from isolation culture.

2. Boat-shaped individual before tail is formed which may originate at any

time from the anterior end of dividing tailed form.

3. Early stage of division of tailed form.

4. Degeneration type of tailed form which may give rise to an amoeboid

form 40.

5. 6. Later stages of division of tailed form.

7. Conjugation.

8, 9. Ex-conjugants which reorganize into tailed forms.

10. Typical boat-shaped form characteristic of mass cultures after two to

three days.

11. First binary division of boat.

12. Second division of boat.

13. Third division of boat giving eight products.

14. 15. Fourth division of boat and fusion of psedogamous gametes in cap-

sules to form eight zygotes.

16. Development of zygote into young individual.

17. Intermediate young individuals from couples.

IO8 GARY N. CALKINS AND RACHEL BOWLING.

different from 'those of the tailed forms but these parts are slowly
absorbed and in the second generation of the boats they have en-
tirely disappeared. The macronucleus of the boat form is more
condensed, more definite in form, and stains more readily than in
the tailed form, and this intensity of staining is retained
throughout all of the later stages. The protoplasm likewise is
denser and has lost its included vacuoles but there are still many
granules which are partitioned out at each division with apparently
no increase in their total number. The contractile vacuole varies
considerably in position, sometimes on the ventral surface, some-
times on the dorsal and frequently nearer one pole than the other.
In the third and fourth divisions of the boats the onset of di-
vision is always indicated by the presence of two vacuoles sym-
metrically placed in the cell.

The first two divisions are fairly slow, requiring several hours
but the last two divisions follow one another in quick succession.

Boat-shaped forms may appear at any stage and appear to be a
palingenetic phase of the organism. Thus in ordinary division
of the tailed form the anterior half is navicular until the tail is
regenerated. Also just as pathological tailed forms turn into
amoeboid cells so the boat-like individuals may undergo a similar
pathological change. In some cultures the entire population ap-
parently becomes thus transformed into amoebae.

3. Couples. — The first two divisions of a " boat" lead to small
individuals (2 I/A to 25^) of broadly ellipsoidal form and with
relatively large nuclei (Fig. III., 12). The daughter cells formed
by the second division still have the power to move and usually
become widely separated. Each divides into two and these two
quickly give rise to four. Chains of four cells are characteristic
and as there is a tendency at this period for the boats and their
products to agglomerate, great masses of these chains are fre-
quently found in the Syracuse dishes. The four cells of a chain
soon become associated as two pairs and these are the " couples "
(functionally gametes) of our terminology. These pairs measure
from 22/x, to 26/A, each individual, from n/* to 13/x. About each
couple is a delicate capsular membrane resembling a fertilization
membrane, but there is as yet no fertilization, hence the resemb-
lance is closer to a sporocyst membrane of two gregarines in
pseudo-conjugation (Fig. I., 13; Fig. III., 14).

STUDIES ON DALLASIA FRONTATA STOKES. 109

We have repeatedly watched the process of couple formation in
the living cells and the further changes which take place within
the capsule. An instructive picture is obtained by use of neutral
red which stains some of the endoplasmic granules and these fur-
nish points of orientation. The two cells of a couple fuse to form
a zygote (Figs. I., 13, and III., 15). The nuclei also fuse. We
have watched this fusion in living couples under an immersion
lens and have noted a center in each gamete where brownian move-
ment of granules is evident. Stained preparations show that
these centers are nuclei. After fusion of the cell bodies these
centers approach and melt into one immediately after which there
is a more violent brownian movement of the granules.

These activities show that the boats are gamonts which give
rise to gametocytes and the latter to gametes of which there are
sixteen from each gamont. Fertilization is strictly paedogamous
and nothing like it has been described for any type of ciliate.
The nearest approach to it is Brumpt's account of encystment and
fusion in Balantidium coli, but here two gamonts come together,
no gametes are formed and the two individuals, as hologametes,
fuse within a membrane analogous to the sporocyst membrane of
gregarines.

This period of copulation is a critical one in the history of
DaUasia. Up to the present time we have not succeeded in rear-
ing a single zygote in isolation culture. Many young forms are
found in the Syracuse dishes in which an epidemic of copulations
has occurred (Fig. I., 16) ; some of these are not yet provided
with mouths and their development into mouth-bearing forms
has been repeatedly observed (Fig. I., 15). The origin of these
young forms from the stage of the encapsulated zygote has also
been observed but we have not yet succeeded in providing a suit-
able environment for their continued life in isolation culture. In
many cases, but not in all, the zygote apparently encysts within
the capsule (Fig. III., I5(/) and such cysts are liberated by the
dissolution of the capsular membrane. The further fate of these
cysts is unknown.

4. Pairs or Conjugants. — We have cultivated Dallasia in isola-
tion cultures for four months and now have eight series of dif-
ferent ages under observation each series derived from an indi-

HO GARY X. CALKINS AND RACHEL BOWLING.

viclual ex-con jugant. Paedogamous copulation, described above,
occurs in conjugation tests made within a week of the first divis-
ion of an ex-conjugant and epidemics of such unions still occur
at intervals in our oldest series. They occur less frequently and
in much milder form when the individuals of a series are mature
for conjugation.

Conjugation epidemics are rare. Tests have been made daily by
placing the reserve individuals left over after the usual isolations
are made, in a Syracuse dish with about 10 cc. of fresh medium.
These dishes are set aside in a moist chamber and left for at least
one week and usually without the addition of fresh medium.
They are examined daily and the observations recorded. In the
early life of a series boats usually appear within two or three days
and the boats usually give rise to couples. If, however, such
boats are transferred to fresh medium they change again into tails.
Sometimes fully 100 per cent, of the original tailed forms change
into boats and these into couples and zygotes but as a series grows
older there is an increasing percentage of tailed forms which do
not become transformed into boats and an increasingly diminish-
ing number of couples. In Syracuse dishes with material from
older series there is thus a predominance of tailed forms at all
stages. These are somewhat smaller (Fig. I., 6) than are the
individuals of the isolation cultures and they show the same type
of agglomeration as does Uroleptus mob His in similar conjugation
tests. As with Uroleptus such agglomerations are usually although
not invariably, followed by conjugation of the individuals.

The first epidemic of conjugations occurred after thirty-five
days of culture of a wild individual and gave us material for
Series 2 and 2a of our pedigreed races. In one of these (2a) a
mild epidemic occurred in the 96th generation or 25 days after the
first division of the original ex-conjugant, and Series 3 and 4
were derived from it. A second epidemic occurred in the i6oth
generation or 47 days after the first division of the ex-conjugant
and from this epidemic Series 5 and 6 were started. Three other
epidemics have appeared in Series 3 and 4 and have furnished ma-
terial for Series 7 and 8.

The conjugating individuals are relatively small (77/x. to iO2ju.)
and are always tailed forms. Union occurs as in Uroleptus or

STUDIES ON DALLASIA FRONTATA STOKES. Ill

Spathidhiui, etc., at the anterior ends and, again as in Uroleptus
the mouth parts are not involved. The mouths, however, are
greatly reduced and apparently are absorbed, new ones being
formed by the ex-conjugants. The period of actual fusion varies
from twelve to twenty-four hours and the period of reorganiza-
tion of the ex-conjugant varies from one to four days. The cyto-
logical details have not yet been worked out but meiotic divisions,
interchange and fusion of nuclei appear to follow the customary
history.

The average division rate for the initial lo-day period is high
and is higher in most cases than the division rate for the same
calendar period of the parent series. As with rrolcptus, however,
this is not invariable as the following table shows :

Series 20 division rate ist. 10 days, 38.6. Parent ser es same per od 19.2

Series 3 36.8.

Series 4 40.8.

Series 5 29.6.

Series 6 " " " " 33.4.

40.2
40.2
24.4
25.2

It is too early to draw any conclusions as regards vitality before
and after conjugation, this subject will be discussed in a later
study.

DISCUSSION.

So far as we are aware Dallasia frontata presents a unique phe-.
nomenon hitherto undescribed for the ciliated protozoa. This is
the interpolation of a paedogamous fertilization stage in the other-
wise ordinary cycle from ex-conjugant to conjugant. Two dis-
tinct and entirely different fertilization phenomena in the same
life cycle certainly furnish food for reflection, particularly as
regards the significance of fertilization in general. The nearest
parallel case that we know is Cryptochilum echini, as described by
Russo. The high death rate, in cultures, after copulation may be
significant. It may mean that the culture medium is not suitable
for this stage of the organism or it may mean that the encapsulated
stage is taken into some other organism where part of the life his-
tory is spent as a parasite or as a commensal. Further study of
the organisms in culture with experiments to test the effect of dif-
ferent media, which are now under way, may throw more light on
this problem.

The novelty of Dallasia does not lie in the copulation of micro-

H2 GARY N. CALKINS AND RACHEL BOWLING.

gametes; this phenomenon is known in the Opalinidae. Nor does
it lie in the union of paedogamous gametes as this phenomenon
is well established in the case of Actinophrys sol and in the case
of Actinosphaerium eichhornii. There is certainly no novelty in
the phenomenon of conjugation of Dattasia for in this it agrees
with the vast majority of ciliates. The novelty lies in the combi-
nation of fertilization by copulation and fertilization by conjuga-
tion.

It is well known through isolation culture work with infusoria
that a reorganization process without union of individuals occurs
and has the same effect on vitality as does conjugation, it is a
process of parthenogenesis termed endomixis by Woodruff and
Erdmann (1914) ; and in some form or other it occurs in prac-
tically every ciliate that has been studied. It takes place prior to
and during the early phases of encystment in the Hypotrichida,
without encystment in various species of Paramecium. In U ro-
le ptus mobilis endomixis with encystment is a characteristic phe-
nomenon of the early stages of the life cycle (Calkins, 1926) ; it
becomes infrequent with maturity of the protoplasm and is absent
altogether in the later stages. In Dallasia front at a the incidence
of couple formation in the early stages of the life cycle, the for-
mation of capsules, together with the absence of any evidence up
to the present, of the ordinary forms of endomixis, lead us to the
conclusion that we have here a very unusual, perhaps primitive,
type of endomixis. If this conclusion is correct the further hypoth-
esis is permissible that endomixis as ordinarily observed in ciliates
is a reminiscence of ancestral gamete-brood formation.

REFERENCES.
Calkins, G. N.

'26 The Biology of the Protozoa.
Russo, A.

'26 Gli exconiuganti, derivati dalla ia coniugatione accessoria fra Gameti
impuri in " Cryptochilum echini," producona Gametogeni puri e
Gameti puri, che rinnovanno il ciclo principale. Rend. Accad. dei
Lincei., Ser. 6, Vol. 3.
Stokes, A. C.

'88 A preliminary Contribution toward a History of the Fresh-water
Infusoria of the United States. Jour. Trenton Natural History
Society, Vol. i, No. 3, Jan., 1888.
Woodruff, L. L. and Erdmann, R.

'14 A Normal Periodic Reorganization Process without Cell-fusion in
Paramecium. Jour. Exp. Zool., XVII.

THE BACTERIOLOGICAL STERILIZATION OF

PARAMECIUM.

ARTHUR K. PARPART.

(From the Biological Laboratory, Amherst College, Amherst, Mass.)

I.

If for one reason or another it is necessary to control the bac-
terial content of the medium in which Paramecium is living, the
first step is a reliable method for the bacteriological sterilization
of the animals.

Hargitt and Fray ('17) devised a method which they believed
accomplished this end. Their procedure, in brief, consisted in
transferring a single animal, by means of sterile pipettes, through
five successive washings of sterile water contained in sterile de-
pression slides, the latter being enclosed in Petri plates. There is
no evidence in their paper as to how many animals were treated
in this way to determine the efficiency of the method. According
to Philipps ('22) the technique of Hargitt and Fray " is undoubt-
edly reliable." However, she used a procedure in her experiments
which " made it necessary to wash each animal seven times instead
of five."

II.

In his first attempts to sterilize Paramccia the present writer
increased the number of washings to ten. The animals were ob-
tained from a pedigreed culture of Paramecium caudal urn, grow-
ing on a 0.7 per cent, infusion of pure timothy hay in tap water.
This same solution was used for washing. The solution was ster-
ilized in an autoclave at 12 to 15 pounds pressure for 45 minutes.1

As a first step eight animals were washed ten times, with the
purpose of determining, first, the diminution in the number of

1 The difficulty, experienced by Hargitt and Fray, of getting Pannnccia
to live on media sterilized in an autoclave under high pressure, has never
been encountered in these experiments though pure lines of Paramccia have
been carried for a number of months on hay infusions and beef extracts
treated in this way.

ARTHUR K. PARPART.

bacteria that occurred during the ist, 3d, 5th, and 7th washes;
second, the number of animals sterile in the loth wash. To ac-
complish the first purpose, the ist, 3d, 5th, and 7th wash fluids
were plated, these plates incubated at 37.5 degrees C. for 72 hours
and examined. For the second purpose the loth wash fluid to-
gether with the animal was broth cultured, and the cultures treated
in the same way as the plates. The results are recorded in Table I.

TABLE I.

REDUCTION IN NUMBER OF BACTERIA IN WASHES, i, 3, 5, AND 7.
EFFICACY OF 10 SUCCESSIVE WASHINGS.

Animal

Number of Colonies on Plates of wash Fluid No.

Broth Culture of
Wash No.

No.

10 + Animal.

i

3

5

7

i

9,000

I

0

o

Infected

2

6,000

3

o

o

Infected

3

10,000

o

o

o

Sterile

4

9,000

2

o

o

Infected

5

13,000

0

o

0

Sterile

6

9,000

O

o

0

Infected

7

16,000

2

o

o

Infected

8

16,000

O

o

o

Infected

The diminution of the number of bacteria in successive washes
up through the 5th, as brought out in the above table, concurs
very well with the results obtained by Hargitt and Fray. The
discrepancy between the number of bacteria present in their first
wash and the infection I found may be accounted for by the fact
that they transferred only a small portion of these washes to agar
plates, while I transferred the entire amount.

Superficially the fact that no animals contaminated the 5th wash
might be taken to indicate their sterility. However, 80 per cent,
do contaminate the loth wash. These, of course, could not have
been sterile at the time of the 5th washing.

To further test this particular point 18 animals were washed
10 times, and the loth wash fluid together with the animal broth
cultured. The latter was incubated at 37.5 degrees C. for 72 hours
and examined. The results are tabulated in Table II.

Of the 26 animals included in these tables only 5 were sterile
in the loth wash.

BACTERIOLOGICAL STERILIZATION OF I'AKA M ECIUM .

TABLE II.
EFFICACY OF 10 SUCCESSIVE WASHINGS.

Total Number of
Animals Tested.

Broth Cultures of the lotli Wash -Fluids + the animals.

Sterile.

Infected.

18

3

16

III.

The Hargitt and Fray sterilization method differs from the
above method in that they employed a sterile tap water solution
for the washing and passed the animals through only 5 wash
fluids. Conceivably sterile tap water might be a better sterilizing
agent. At any rate, it speeds up animals put into it and leads to
rapid reversals which possibly enable the Paramccia to throw off
more readily the bacteria lodged between their cilia.

Accordingly, 30 animals were washed 5 times in sterile tap
water. Broth cultures of the 3d wash fluid and the 5th together
with the animal were incubated at 37.5 degrees C. for 72 hours.
The results are tabulated in Table III.

TABLE III.
EFFICACY OF 5 SUCCESSIVE WASHINGS.

Total Number
of
Animals Tested.

Broth Cultures of the
3d Wash Fluids.

Broth Cultures of the sth \Yasii
Fluids + the Animals.

Sterile.

Infected.

Sterile.

Infected.

30

28

2

3

27

Only one conclusion is possible. In the majority of cases 5 and
even 10 washings in sterile media cannot be relied .upon to sterili/e
a Paramccinui.

IV.

As washing is the only practical method for ridding Parameda
of bacteria, the following technique was devised.

i. The washings were performed in depression slides, each
slide being enclosed in a Petri plate. Those plates in which
Paramccia were cultured had a thin glass slide under the de-

Il6 ARTHUR K. PARPART.

pression slide, so that the sterile water poured into the plates to
make them serve as moist chambers could not get into the de-
pression.

2. The pipettes for transferring the animals through successive
washes were made from soft glass tubing having an inner diam-
eter of 2 mm. and a wall of I mm. thickness, by drawing this out
to capillary fineness. Of the 60 pipettes made in this manner,
10 were chosen at random and the inner diameter measured at
the tip of the capillary portion. The average inner diameter was
213 micra; none varied more than 30 micra from this average.
The large end was plugged with cotton, and each pipette plugged
into a separate test tube.

3. The sterilization of the pipettes and of the depression slides
in Petri plates was carried out in a dry oven at between 160 and
170 degrees C. for 45 minutes.

4. The actual washing of the Paramecia was performed under
a hood which was placed at one end of a large table, and con-
sisted of a wooden frame (3 ft. by 15 in. by n in.), with a glass
top and cloth sides. The front cloth, which served as entrance,
was loose at the bottom. Toward one end there was a binocular
microscope with sufficient focal length so that its oculars extended
through and above the top. Cloth, with slits in it for the oculars,
was glued to the edges of the glass surrounding the oculars.

In handling the animals the transfer pipettes were attached to
a rubber tube plugged with cotton and operated by means of
mouth suction while the operator was looking through the oculars.

5. All of the various types of culture and wash media used
were put into separate, one-liter flasks fitted with glass siphon
tubes. Rubber tubing with a glass pipette at one end led off from
each siphon tube. After the flasks, tubing and pipettes (the lat-
ter plugged into small test tubes) had been sterilized in an auto-
clave at 12 to 15 pounds pressure for 45 minutes and the corks
surrounding the siphon tubes sealed with paraffin, they were ar-
ranged outside of the hood and the rubber tubing and pipette led
through the back into the hood. The pipettes were suspended
at the back of the hood in such a way that their tips, after the
test tubes had been removed, did not touch anything. By use of
carefully adjusted pinch clamps the size of the drops flowing from

BACTERIOLOGICAL STERILIZATION OF PARAMECIUM. 117

these pipettes was regulated, and hence the volume of media
could be determined. The fact that none of the culture media
thus treated became infected, although the pipettes were exposed
continually for a number of months to the air of the hood, is very
good evidence of the efficiency of the hood.

6. The wash fluid was prepared by placing 250 ing. of Liebig's
beef into 200 cc. of sterile tap water. This solution was bacter-
ized from the cultures of the pedigreed series of P. caudatnin
being cultured on 0.25 per cent, beef extract, incubated at 37.5
degrees for 2 days and diluted up to I liter with tap water. It
was placed in one of the liter flasks, sterilized and arranged for
use.

7. The actual steps in the washing of an animal were :

(a) Three piles of 5 Petri plates each were placed under the
hood, and 6 drops (about ^ cc.) of wash fluid was put into each
of the enclosed depression slides. The lowermost Petri plates
served as moist chambers for the 5th wash and hence contained
slides under the depression slides.

(b) The 15 pipettes necessary for the transfers were placed
under the hood.

(c) The culture containing the Paramecia was placed on the
microscope stand and a single individual transferred to the upper-
most slide in each* stack of plates. Each animal was transferred
successively to the depression slide in the Petri plate immediately
beneath. By working in rotation from stack to stack, the ani-
mals remained in each wash about one minute.

(d) When all three animals were in the 5th wash, from 3 to
4 cc. of sterile distilled water was added to the lowermost plates.
This prevented excess evaporation from the depression slide while
the 5th wash fluid and the animals were being incubated for 5
hours at 25 degrees C.

(0) At the end of 5 hours, each animal was again transferred
through 4 washes.

(/) From the last of these, the Qth, the animal was trans-
ferred to the desired culture media. The Petri plate of this, the
loth wash, was converted into a moist chamber as above (d).

ARTHUR K. PARPART.

V.

The data demonstrating the efficiency of this method are sum-
marized in Table IV. In this summary are included the data on
those animals which were placed on some type of sterile medium
after the 9th wash. The data were obtained by broth culturing
the 5th wash after the animal had been in it 5 hours, and the loth
wash together with the animal after the latter had died in it. The
death, in some cases, came only after a number of days, during
which time the loth wash and the animal were incubated at 25
degrees C. and examined every 24 hours until the death of the
animal. The broth cultures were incubated at 37.5 degrees C.
for 72 hours before being examined.

TABLE IV.

SUCCESS OF 10 WASHINGS; THE ANIMAL REMAINING IN THE sth WASH

FIVE HOURS.

Total Number
of
Animals Tested.

Broth Cultures of the
5th Wash Fluids.

Broth Cultures of the loth Wash
Fluids + the Animals.

Sterile.

Infected.

Sterile.

Infected.

50

17

33

50

o

The number of animals tested and the fact that all were sterile
in the loth wash shows conclusively that the method adopted will
rid Paramccia of bacteria.

The length of time that an animal is allowed to remain in the
5th wash fluid is a significant factor in accomplishing the steri-
lization. It raises the question as to why many animals shed
bacteria into the 5th wash fluid, but a few do not.

VI.

Tables III. and IV. appear to demonstrate that Paramecium
caudatuui defecates bacterial spores. According to Table III.,
over 93 per cent, of the Paramccia were no longer shedding
bacteria into the 3d wash fluid, yet over 90 per cent, of these
later proved infected. Table IV. shows that 66 per cent, of the
animals left in the 5th wash for 5 hours shed bacteria, yet when
these animals were washed four times more they proved sterile.

BACTERIOLOGICAL STERILIZATION OF PARAMECIUM.

119

If the majority of Paramecia can be passed through a 3d wash
fluid without shedding any more bacteria, and then later do so, it
seems highly improbable that the bacteria are on the outside of
the animals.

To test this point the following experiments were performed.
Seventeen bacteriologically sterile Paramecia were left for 24
hours in a pure culture of Bacillus prodigiosus in beef extract,
while 21 sterile animals were left for the same length of time in
a pure culture of Bacillus subtilis in beef extract. At the end of
this time the animals were washed, individually, ten times ; the
time of the 5th wash fluid being varied. Those fed on B. pro-
digiosus were allowed to remain in the 5th wash for ^2 hour.
Those fed on B. subtilis were left in the 5th wash from 2 to 5
hours. About 24 hours after the animals had been put into the
loth wash, this together with the animal was transferred to a
broth culture, incubated at 37.5 degrees C. for 72 hours and ex-
amined. The results are tabulated in Table V.

TABLE V.
EXCRETION OF SPORES BY Paramecia FED ON B. subtilis.

Brotli Cultures of the loth

Total Number

Bacterial Culture

Time Each

Wash Fluids + the Animals.

of

on Which Animal

Animal Spent

Animals Tested.

Was Placed.

in the 5th Wash.

Sterile.

Infected.

17

B. prodigiosus

}/2 hour

17

O

4

B, subtilis

i hour

i

3

7

B. subtilis

2 hours

I

6

4

B. subtilis

3 hours

I

3

6

B. subtilis

5 hours

6

o

Bacillus prodigiosus has never been known to produce endo-
spores, while Bacillus subtilis produces endospores very readily.

In every case the Paramecia that had been cultured in B. pro-
digiosus were able to throw off all bacteria, although the animals
remained in the 5th wash fluid only l/2 hour. Eighty per cent, of
the Paramecia cultured on B. subtilis and then left in the 5th
wash fluid up to 3 hours were not sterile when washed five times
more.

Since both of these species of bacteria have the same type of

9

I2O ARTHUR K. PARPART.

flagellae, it seems very unlikely that B. sub tills could remain
lodged between the cilia of the Paramecia for a longer period of
time than B. prodigiosus.

The above data seem to clearly indicate that Paramecium def-
ecates solid material, in this case bacterial spores.

The efficiency of the sterilization technique adopted is further
attested by the data in Table V. Those animals washed after
having been cultured in B. subtilis, and left in the 5th wash for
5 hours, were all sterile in the loth wash.

The author wishes to thank Professor Otto C. Glaser for aid
rendered in the preparation of this manuscript.

LITERATURE CITED.

Hargitt, G. T., and Fray, W. W.
'17 Paramecium in Pure Cultures of Bacteria. Jour. Exp. Zool., Vol.

22, pp. 421-454.
Phillips, R. L.

'22 The Growth of Paramecium in Infusions of Known Bacterial Con-
tent. Jour. Exp. Zool., Vol. 36, pp. 135-183.

THE EFFECT OF MATERNAL AGE AND OF TEMPERA-
TURE CHANGE IN SECONDARY NON-
DISJUNCTION.

R. R. HUESTIS,
UNIVERSITY OF OREGON.

INTRODUCTION.

The following figures present the results of the raising of XXY
Drosophila females in an effort to ascertain, particularly, how in-
creasing maternal age and how differences in the temperature, at
which mother flies are maintained, affect the percentage of excep-
tions to sex linkage.

The first part of the paper deals with the effect of maternal age
and here, because of some lack of conformity in the results ob-
tained at different times, the problem has not been solved. How-
ever, I feel that the figures should be published for they have been
accumulating for three years and I am not, at present, continuing
the investigation. The second part of the paper deals with the
effect of differences in temperature. Here the different tests
which were made check in a fairly satisfactory manner. I am
also able to give some figures, in the last part of the paper, which
have a bearing upon the genetic variation in exception-producing
ability.

EFFECT OF MATERNAL AGE.

The first experiment was carried out during the time I was a
graduate student in Genetics at the University of California, the
use of successive subcultures being a routine method of rearing
flies there and fermented banana the food medium. I undertook,
under the direction of the genetic staff, a rather complete repeti-
tion of Bridges' 1916 experiments and obtained comparable re-
sults in most particulars. However, in a group of females con-
tinuously outcrossed and presumably producing a " normal "
percentage of exceptions, I obtained 3423 (5.90 per cent.) of

121

122

R. R. HUESTIS.

these in 58037 flies ; this being somewhat higher than the 4.3 per
cent, reported by Bridges.

An inspection of my material, made after most of the data were
in, showed that a given female's later subcultures produced, dur-

16

14

12

10

30

FIG. i. Curves of smoothed percentages of exceptional young obtained
from groups of females kept at different temperatures. Abscissas ap-
proximate the age in days of mother females when eggs were ripened.
Ordinates are percentages of exceptional young. Curves end with the
group number. Under each curve is the temperature in degrees C. Broken
lines denote inbred groups. Thick-lined curves are based on more flies.

ing most of her lifetime, almost one per cent, more exceptions
per week than her earlier subcultures. The total production of
offspring and of exceptions to sex linkage (exc.) of 109 white-
eyed females mated with red-eyed males, and changed each week
to a new subculture, is given in the totals column of Table I. The
curve of these percentages is that of group i, Fig. i.

In order to eliminate the possibility of the weighting of the later
subcultures by reason of the longer life, or the greater production
of exceptions in certain subcultures of a few high-producing fe-
males, I excluded, in the figures presented in Table I., the data
from all females which produced over 10 per cent, of exceptions
or which had any sub-culture failure prior to the one which ended
the females production. When the material is divided into groups
of females which produced young for the same length of time,

EFFECT OF AGE IN SECONDARY NON-DISJUNCTION. , 123

DED INTO CULT

Q
M

£

z

w

K
*

M

a,
o

Z ^
w o

dS

£ O
Z 7

M -f-
3^

CO <

8 §•

M W

§ 5

H

CU r5
W §

U M

X >

W .^

w

ta
O

H

u

^ a

Q W

I s

g §
g £

o
z

o

rt

C.

ca

aJ

•a

b t-

o

M

tf a>

U ^

M y

B 3

H

B
H

CO

u

-r ^

-r M.

X

M Tt O

N "

CO

.y

5

5

o

q q
M d\od

O 06 t^

r- oo

M CO t^-

U

t—

"

N

co

O

00

to

ts

._!

r-»

co

o

s

M

t^ O

t^ oo

T^ o

00 0\

u

O oo t^

00 •O O

\O oo r»

-* t> o

M

o

M

X)

O

r?

^

o

o

N

(N

i-H

OO

^

H

ro

r^

W

fc

H

M

ro

t^ oo

0) M

to r^

3\ "O

0 •*

o

u

O O O

T o r-

oo 10 10

ro Tf O

^f ^D ^D

3

a

H

M
M

O

10

rO

0

t/3

CO

M

00

co

o\

<N

M

O\

O

O

o»

£

hH

M

M

-1

•0

00 O

O\ M

O\ M

oo ^>

ro M

0) 0

o

M LO LO

O O O

OO O O

o -rf 10

roo >o

OO O 1O

w

ro

1-H

o

M

vO

"*

oo

^

^

M

o

N

^.

*o

^_j

(--^

-t

oo

^o

0)

M

IO

0

oo

fa

0)

cs

M

M

l>

ro

O »O

(N >O

^ N

O\ t-~

LO »-l

Ox O

o
X

a

O 1O

oo

M IO >O
O)

)-H

\O ^ IO

o\

M IT) ^

Ov •* 10
ro

SO

00

_cu

o

0

o\

M

ro

o

Ov

O
r-

ro

oo

0

<*

E

M

M

N

M

M

CN

0

M

O O\

•^ t~

r^ M

10 cq

O\ oo

IN •*

o

ro ro

oo 10

ro *^ (*O

00 ^" ^"

oo ro ^~

ro ro ^"

ro ^t ^t

o •*• ^r

W

ro

*^

o

oo

O

^

o

00

ro

00

s

Tf

PO

rO
O
O

O
00

Ov

M

oo
0

(N

^O

U.

H

CM

IN

M

H

M

i-H

.

^O

• -

~

vO

• »^

. . ^

o^

C^

o^

~

. O^

. s^

•d

•a

•a

—

; -3

* ; -o

cu

o

<u

K

. s

• —

cj
X

0

-

6

£
^

^ 0

2 2
;>C«

'^ 0

Qj O

^1

^ o
£ o

s

^ I

H> w

d

||

- x o

3 S 2

124

R. R. HUESTIS.

the same sort of " curve of age " is obtained. The production of
groups of females, so arranged, being also presented in Table I.

All the females used, in obtaining the above figures, descended
from the one original white-eyed female which started the XXY
line, so the X chromosomes were presumably constant throughout.
Male flies, however, were taken from a number of stocks of flies
trapped around Berkeley, or present in the laboratory, so the Y
introduced into each generation was not constant. The flies were
kept in a large cabinet incubator at 25 degrees C. and counted
daily. It is perhaps worth recording that the mother female, of
each group of subcultures, was removed from the incubator in
the subculture tray during the counting period. This was long
enough, at times, to cool the vials down to room temperature.

Before leaving an account of this part of the work I should
like to report briefly upon a test of the constitution of 60 reg-
ular daughters of XXY females. Twenty-seven of these daugh-
ters, when mated with Bar males, produced no exceptions.
Thirty-three produced exceptions, but of this number 8 daughters
produced just one or two exceptions in large counts of flies, the
percentage being, in these cases, consistently in the neighborhood
of one third of one per cent. ; the total production being 10 ex-
ceptions (95 i $ ) in 3,148 flies. If these 8 regular daughters
were XX females which produced primary exceptions these latter
are not predominantly male as in Safir's results (1920).

LATER EXPERIMENTS.

After some lapse of time during which I was engaged in breed-
ing Peromyscus I returned to the problem of non-disjunction with
the idea of checking my results prior to publication, and also
of trying out the effect of temperature differences upon the
percentage of exceptional offspring. I obtained a stock of flies,
through the courtesy of Dr. R. E. Clausen, and after inspecting
the progeny of white-eyed females, mated with normal males,
picked up an exception-producing strain. The flies used in these
latter experiments consisted of a number of white-eyed females
from this stock, a number of white-apricot compound females,
obtained from a mating of white and apricot, and finally of a

EFFECT OF AGE IN SECONDARY NON-DISJUNCTION. 125

number of white and apricot females obtained by equational non-
disjunction from the white-apricot stock, females of which pro-
duced XXY daughters pure both for white and for apricot. I
could not observe that these allelomorphs, white and apricot, dif-
fered from one another in the capacity for exceptional production,
in comparable experiments.

The culture methods in this latter part of the work were modi-
fied somewhat. Yeast-seeded banana agar was used for food.
Females which were producing young were left continuously in
the incubator except during the interval when they were changed
to a new culture. Subcultures were made up every six days, at
20 or 21 degrees C., and every three days (in group 6, Table
III., every four days) at temperatures higher than this. Two
Freis electric incubators and one electrically controlled cabinet
incubator was used and the temperature checked daily. These
machines will fluctuate in temperature, within a degree up or
down, but since the' routine involved the growth of flies, at each
different temperature, over a considerable period of time such
fluctuations should cancel out.

Table II. summarizes and Fig. I depicts graphically the results
obtained when groups of females were kept continuously at cer-
tain temperatures each female being transferred to new subcul-
ture vials as long as she remained fertile. Except in groups 6
and 7, in which inbreeding was the rule, male parents came from
several cultures. In group 6 one inbred stock of wild males was
used, in group 7 exceptional brothers.

A comparison of Tables I. and II. and of the curves in Fig. I
brings to light some very obvious differences in the characteris-
tics of the females in comparable groups. In group I the average
fertile lifetime, in round numbers, was 28 days. In groups 4
and 5, kept at comparable temperatures, the average fertile life-
time was 16 days. It appears improbable that this difference is
an environmental one for banana agar is generally conceded to be
a better food medium than fermented banana.

The percentage of exceptional progeny produced by the fe-
males in groups 4 and 5 (3.14 and 3.88 per cent.) is considerably
below the 5.49 per cent, of exceptions obtained in group I, and /c
this is not due to the difference in longevity for the group
percentage is higher at any comparable age.

126

R. R. HUESTIS.

Id

\
U
fe

X

to

O

o,
o

O

S
o

Qj

t,

Q

Z

NH

H

n
O

o

Z

? w

J^ C^

in H

H

3

(/)

PU

.

Z

o

s

,w

h— 1

H

H

TABLE

OF EXCEP

IFFERENT

w

Q

o

H

H

•<!

Z

Id

CJ

H

PL,

<A
W

W

0,

Q

Z

B!

w

n

Q

Z

<

o
o

O

CA

Subcultures

"3

o
H

0
x
W

Ov

i— (

4}
N
o

N M
t^

•*

M

-H

r—

00

r^ IN

O

«t

0\

M
•\\

5?

(M ro

IO

cs

•H

00

oo

•sf ro

ro

IN

O

<t

•H

M
IO

M O
10
CM

oo

IO

-H

ro

M
M M

ro M

ro

6
2

r^
10
•*

Tf

o
r^

M
't
HH

\r.
ON

ro

06"

10

CM

O
O

r^
to

00

ro

-f
r^

Ov
CM

u
il

6

X

W

O\ M

MOW

<3- ro

O M IN

M

\O OO
ro IN ro

M

6
Z

ro

M

M

t^

rq
r^

CM

O
IO

SG

e>

u

X

W

ro O\

MOO

•* OA

O\ M M

10 r»
M ro CM'

^"

M O\

•* 10 ro
•<t

6
£

•*

>^5

n

<X>
VO
O

ro
O\

M

O
LO

00

fc,

o
X

W

o f)

O\ M M

ro ro

1O CM <N

M

00 N

oo CM ro

CM

ro "P

O\ ro •*

M

6
Z

ro

<D
LO

IN

0\

q

00

o\
o\

o\

t^

u">

tq

0

X

W

1O IO

M M M

M

\o t-»

O M IN
"t

M M

10 ro ro
ro

r^- u^

oo ^t •*
ro

r- °. "?

M vQ 10

o ON

10 d\ -f

IN M M

6

O
•*

t^

O

n
10

M

M

ro

M
M

10

O
oo

r^
r^

IN

r^

cs

M

Q

d

X

H

to -sf

IO i-i i-l

^-(

O N

oo ro ro

LO

T "*

ro ro ro
Tt

M CM

Tj- IT) 4
•*

10 10

O\ ^ 10
CM

ro ro

t^ M CM

ro M M

6
Z

Tl-

oo
o

O
ro
ON

M

0
IO

M

M

ro

^0
00

o\

ro
vO

00
CM

ro

U

u
X

U

<N 00
M M M
tH

O M

M ro ro
0

M

CM ro
M ro ro
ro

r^ O
ro CM ro

CM

CM t^

IO v6 1O

Tl-

ro oo

•* M 6

\O M M

6
£

U1

O\

iri

00

r^
cs"

•*
»o

O\

•^f
•<t

oo

0
CM

r^

0

-o

1O

CQ

6

X

W

ro O

O\ ro N

t-H

q\ M
ro r^ ro

00

oo ro

t- ci ro

ro

a M

oo CM ro

CM

1 9

O O r^

O

ro Is-

odd

t>» M M

6
£

M

r-

10

O
oo

00
M

M
0

ro

M

o

•&

vO

t^-

o\

O

•*

;^

•<

0

X

a

00 1O
>O H N

OO 00
O CS CM

o\

M \q
oo •* ro
ro

M O

w »j- ro

CM

M 1O

O 06 r^-

O

M

O 10

o\ d d

MUM

M

6
Z

O
ro
ro

oo
oo

M

ro

>o

ro
0

0

M

0

o\

ro

CM

M

ro

CM

M

a

<u
H

0

M fe§!%S

M S^O^
g

Cfl

o

rOK§K$

pq «JxiJ^

C/3

VeS^
en

VfcSSS

B

en

%^£?

fi

°°,^^

Cfl

t-i

0

N

ro

4

UO

O

t~-

EFFECT OF AGE IN SECONDARY NON-DISJUNCTION.

127

The curve of age, so obvious in group i, is not exactly repro-
duced in any of the later groups although a curve of age, as may
be noted in Fig. I, is usually apparent instead of a randum
fluctuation from a model percentage. This curve of age differs
at different temperatures, in the figures presented, but the possi-
bility that genetic differences in the females used may be respon-
sible for this cannot be excluded.

A noticeable thing in Table II. is the increase in exception
production at higher temperatures. Based on the totals in each
group the increase was slow, for a given temperature increment,
up to 25 degrees, then more rapid between 25 and 28 degrees and
most rapid between 28 and 30 degrees. For these temperatures
the curve of increased exception production is comparable to the
curve of increased crossing-over of second chromosome genes in
Plough's results.

EFFECT OF TEMPERATURE CHANGE DURING THE LIFETIME OF

XXY FEMALES.

In order to make tests of this kind females were hatched, and
their first eggs ripened at a high temperature and then kept for
the remainder of their lifetime at a temperature ten degrees lower.
In addition, females were hatched at a relatively high tempera-
ture and, after a period at a temperature eight degrees lower,
they were returned to the high temperature again. A difference
of eight degrees was substituted for ten because too many females
died when transferred from a 20 to a 30 degree incubator.

It was found by Plough and Metz that females about twelve
hours old have approximately 140 oocytes in their ovaries. The
practice in the experiments here recorded being to empty cultures
every 24 hours it can be assumed that the average age at mating,
of mother females, was not greater than twelve hours, for younger-
appearing females were always chosen for mothers since they
are most likely to be virgins. It takes such females six to eight
days (this period must differ at different temperatures) to lay
the eggs ripened at the temperature at which they were hatched,
the lag of layed after ripened eggs continuing all through a fe-
male's lifetime.

12

10

8

30

6

N

o-

20

20

20

8

6

28

R

28

20

8

28

28

28

T

20

0 10 20 30

FIG. 2. Curves of three tests of temperature change. Abscissas are
days of life of mother females, ordinates the percentages of exceptional
young. Figures at intervals, under the curves, are temperatures at which
the exception-producing eggs in mother females presumably ripened, inter-
mediate temperature plateaus being left blank.

EFFECT OF AGE IN SECONDARY NON-DISJUNCTION.

129

Table III. and curve N, Fig. 2, presents the results obtained
when white-apricot compound females, born at 30 degrees C.,
were mated with normal males and kept at 20 degrees C. Sub-
cultures are six days apart.

TABLE III.

THE NUMBER OF FLIES AND THE NUMBER AND PERCENTAGES OF EXCEPTIONS

PRODUCED BY XXY FEMALES BORN AT 30° C. AND

THEN MATED AND KEPT AT 20° C.

Temperatures are those at which the eggs, producing the exceptions, are pre-
sumed to have ripened.

Temperature.

30

30-20

20

20

20

C K It

A

B

C

D

E, F

Flies

Exc.

Flies.

Exc.

Flies.

Exc.

Flies.

Exc.

Flies.

Exc.

1,751

197

2,005

146

I.93I

40

1,023

28

482

5

C7

II. 2

7-3

2.1

2.7

I.O

It would be expected that the eggs ripened at 30 degrees would
be laid in the A and in the first part of the B subcultures, being
followed by eggs ripened at 20 degrees, which would be laid in
the latter half of the B and in subsequent subcultures. If the
percentage of exceptions varies with the temperature at which
the eggs are ripened then the A subculture should contain the
highest, B an intermediate and all subsequent subcultures a low
percentage of exceptions. It may be seen that this expectation is
realized completely.

Table IV. and curve R, Fig. 2, presents the results obtained
when white-apricot females were born at 28 degrees C., mated
with normal males and held for two subcultures (12 days) at 20
degrees C., then returned to 28 degrees C. again.

The expectation of a high A, intermediate B, low C, intermedi-
ate D and of high subcultures subsequent to D, is only partly
met. The poor fit in the A and B subcultures may very well be
due to the chance selection of very young mother females. The
final cultures are higher than expected.

130

R. R. HUESTIS.

TABLE IV.

THE NUMBER OF YOUNG AND THE NUMBER AND PERCENTAGES OF EXCEPTIONS
PRODUCED BY XXY FEMALES BORN AT 28° C. MATED AND .
HELD FOR Two SUBCULTURES (12 DAYS) AT 20° C.

AND THEN RETURNED TO 28° C.

Temperatures are those at which eggs are presumed to have ripened.

Temperature.

28

28-20

2O

20-28

28

.

A

B

C

D

E, F

Flies.

Exc.

Flies.

Exc.

Flies.

Exc.

Flies.

Exc.

Flies.

Exc.

1,084

42

1,146

23

834

IS

562

18

434

37

'.

3-9

2.0

1.8

3-2

8.5

Table V. and curve T, Fig. 2, presents the results obtained when
apricot females were born, mated with one inbred stock of wild-
type males and held for one subculture (4 days) at 28 degrees C.,
then kept in a subculture (6 days) at 20 degrees C. and finally
returned to 28 degrees C. for all subsequent four-day subcultures.

TABLE V.

THE NUMBER AND PERCENTAGE OF EXCEPTIONAL YOUNG PRODUCED BY XXY

FEMALES BORN, MATED, AND KEPT 4 DAYS AT 28° C., THEN

KEPT Six DAYS AT 20° C. AND THEN RETURNED TO 28° C.

Temperatures are those at which eggs are presumed to have matured.

Temperature.

28

28

20

20 28

28

Subcult.

A

B

C

D

E. F

Flies.

Exc.

Flies.

Exc.

Flies.

Exc.

Flies.

Exc.

Flies.

Exc.

2,342

175

3.825

297

2,108

67

1,033

49

485

45

'

7-5

7-8

3-2

4-7

9-3

This was the most adequate test of the effect of temperature
for a fair number of flies was raised, the males used as parents
were all of one type, and other females (group 6, Table III.)
were held continuously at 28 degrees. The C subculture should

EFFECT OF AGE IN SECONDARY NON-DISJUNCTION. 131

be the low one but should not go down to 2 per cent, for the 6
days at 20 degrees would hardly complete the oviposition of all
the eggs brought into the culture which were ripened at 28 de-
grees. The final cultures are again too high and the same cause
may be operative here as in Table IV. I suggest an over-reaction
to heat after a transfer from a low temperature.

When it is considered that the curve produced by the percent-
age of exceptional young in different subcultures, which received
eggs ripened at different temperatures, is superimposed upon a curve
of age which has been shown to be somewhat variable, it may be
concluded that the above figures are as close to the requirements
for a demonstration of the effect of temperature as could be ex-
pected. I think the statement that the percentage of exceptional
progeny, produced by XXY females, varies in being higher at
high temperatures and lower at low temperatures between 20 and
30 degrees C. is thus quite warranted by the results.

GENETIC DIFFERENCES IN EXCEPTION PRODUCTION.

By breeding to reintroduce the maternal Y chromosome,
Bridges obtained a " high " line of exception producing females.
He suggested that this high production was due to increased
heterosynapsis resulting from the peculiar constitution of the
introduced Y. Bonnier, using Bridges' high line, showed that the
introduction of a new Y is not followed by a return to the " low "
percentage of exceptions, and presented evidences to show that a
high percentage of exceptions is a matter of the constitution (gene
basis) of the two X's.

The following data suggest that neither of these theories will
account for all the inherent differences in exception producing
ability. I mated 24 matroclinous daughters of a white-eyed XXY
female (9 No. 116), mated with one male, which had herself
produced 14.5 per cent, of exceptions in 447 young. These
daughters (they are included in Table I. above) produced an aver-
age of 6.2 per cent, of exceptions in 11,692 young, the percentage
of individual cultures being from 2.9 per cent, in one culture of
820 young to 10 per cent, in another culture of 600 young. This
difference between 2.9 and 10 per cent., is five times its probable

132

R. R. HUESTIS.

error 1 although these daughters all had, theoretically at least,
equivalent sex chromosomes. Here an apparent segregation of
exception-producing ability is manifest.

It can readily be demonstrated in another way that this varia-
bility is not just a matter of random sampling, for if the percent-
age of exceptions produced during the first part of a female's life-
time is compared with that produced during a later part of her
lifetime, positive correlation may be observed. I made such cal-
culations within groups of females which had produced a se-
quence of complete subcultures at 23, 26 and 30 degrees (groups
3, i, and 5 already tabulated) and obtained the following corre-
lation coefficients : -|- .56 ± .12, + .39 ± .07 and -f- .70 ± .11. This
suggests that the ability to produce a certain percentage of excep-
tions, tends, like any other measurable character, to stay within
the limits proscribed by genetic constitution of the individual in
question. That being the case one would expect parent-offspring
correlation, in exception-producing ability, of about the same
magnitude as that which has been found for other characters. I
had one series in which I could test this point, for in group I
there were 60 daughters which had produced 300 or more young
which had mothers with an equal productivity. The mother-
daughter correlation in exception production was here -f- .37 ± .07.

I selected a few generations of flies for increased exception-
production by choosing the daughters of my higher producing fe-
males and mating these daughters with their exceptional brothers.
In the first selected generation I obtained 6.5 per cent, of excep-
tions in 4,493 flies, in the second generation 8.4 per cent, of ex-
ceptions in 4,414 flies, and in the third generation 14.6 per cent,
of exceptions in 1,772 flies. Exceptional females from average
exception-producing mothers which had been similarly mated
with their exceptional brothers produced 5.6 per cent, of ex-
ceptions in 5,674 flies, so inbreeding alone did not increase the
per cent, of exceptions. An inspection of the pedigrees showed
the selected high line to altogether come from one high-producing

1 When the probable errors are obtained by the formula ±\l— -, where

/> = the percentage, q = ioo-p and Af = the number of observations upon
which p is based.

EFFECT OF AGE IN SECONDARY NON-DISJUNCTION. 133

female. After one generation produced by outcrossing to non-
related stock, two females from this line still produced 14.1 per
cent, of exceptions in 546 flies.

I also repeated Bridges' scheme of mating to reintroduce the
maternal Y chromosome from a high-producing female by mating
her daughters with their exceptional uncles. In two generations
of this line I obtained 15.5 per cent, of exceptions in 2,072
flies, a figure comparable, since sister females were used in the
two cases, to the third and one later generation of the high se-
lection tabulated above. Either of these methods of inbreeding
would tend to concentrate factors favorable to exception produc-
tion in a given line of flies and these need not be intrinsic char-
acteristics of the sex chromosomes.

CONCLUSIONS.

The data presented above together with those of previous in-
vestigations show that the percentage of exceptions to sex link-
age may be affected by a number of variables which, in order of
importance, are : the genetic constitution of the female, the tem-
perature at which eggs are ripened, and maternal age.

With regard to genetic constitution, XX females produce well
under i per cent, of primary exceptions. XXY females may
produce from i to over 20 per cent. Our knowledge of the causes
of this latter variability does not appear to be complete. Bon-
nier's outcrossing experiments appear to remove the probability
that the constitution of the Y chromosome is responsible for
Bridges' high eosin line and although Bonnier was not able to
exclude the possibility that autosomal genes were implicated, his
experiments pointed to an exclusively X chromosomal effect in
the production of different percentages of exceptions. It is the
rule to have a constant pair of X chromosomes in all lines ot
secondary non-disjunction, except as the X's may interchange
material with each other or with the Y, and yet genetic variability
is still present. The only inference that is possible is that the
percentage of exceptions may be affected directly by interaction
of the sex chromosomes themselves and indirectly by autosomal

134

R. R. HUESTIS.

genes.2 This is what has been found to be true in cross-over per-
centages.

The direct effect of temperature upon the percentage of ex-
ceptions appeared in all the tests I made. The results of tem-
perature differences upon protoplasm in modifying physiological
activity are so well known that some temperature effect would be
an a priori expectation in secondary non-disjunction. Although
my data suggest that the temperature effect increased as 30 de-
grees C. was approached, I cannot exclude the possibility that this
was due to genetic differences in the groups of females kept at
these different temperatures.

Maternal age appears to affect the percentage of exceptional
young to some degree but apparently by interacting with other
variables for different age curves were found in different groups
and at different temperatures.

The inference that autosomal genes, temperature and ma-
ternal age all affect the allocation of the sex chromosomes, in
XXY females, to gamete-forming cells, follows the conclusions
reached above.

LITERATURE CITED.

Bonnier, Gert.

'23 Studies in High and Low Non-disjunction in Drosophila mclano-

gaster. Hereditas, IV., 81-110.
Bridges, C. B.
'16 Non-disjunction as Proof of the Chromosome Theory of Heredity.

Genetics, I., 1-52, 107-163.
Plough, H. H.
'17 The Effect of Temperature in Crossing Over in Drosophila. Jour.

Exp. Zool., 24, 147-208.
Safir, S. R.

'20 Genetic and Cytological Examination of the Phenomena of Primary
Non-disjunction in Drosophila melanogastcr. Genetics, 5, 459-487.

1 This would explain why a high-producing female may or may not es-
tablish a high exception-producing line.

OXYGEN CONSUMPTION OF INSECT EGGS.1

ROY MELVIN,
IOWA STATE COLLEGE.

Although insect eggs present unique material for studies in
metabolism the literature contains comparatively few reports in
this interesting field of insect physiology. Bodine (i) has pre-
sented data from a study of eggs of Orthoptera which show the
velocity of development to increase in direct proportion to in-
crease in temperature within the normal limits of development.
This is in accordance with other findings on the effects of tem-
perature on poikilothermos species. This author shows that it
is possible to calculate the time of hatching of eggs if previous
temperature history is known. Such knowledge of insect pests
may lend itself to practical application.

Fink (2) has conducted studies which lead him to conclude
that the formative period in the development of eggs of certain
insects is dependent upon whether they are deposited upon foliage
or in the soil. Data to be presented in this paper tend to dis-
prove the above explanation. For further references to literature
in this field see the papers of Bodine and Fink cited above.

Thanks are due Doctor Erma Smith, Professor of Physiology,
and other members of the Zoology and Entomology staff at Iowa
State College, for many helpful suggestions and encouragement
throughout the course of this work.

PURPOSE.

The purpose of this paper is to present briefly a preliminary
report of a study of the oxygen consumption during embryonic
development of certain insects.

1 Contribution from the Department of Zoology and Entomology, Iowa
State College, Ames, Iowa.

10 135

136

ROY MELVIN.

METHODS AND MATERIAL.

Bodine's modification of Krogh's manometer was used for
determining the oxygen intake. Constant temperatures were
maintained by use of a Freas electric water bath. The data pre-
sented were determined from the eggs of the following insects:
Squash bug. Anas a tristis De G. ; Luna moth, Tropcca luna L. ;
Cecropia moth, Samia cccropia L. ; and Smartweed borer,
Pyransta dinslici Hein.

The O2 consumption is expressed in millograms of O2 per gram
live weight (exclusive of shell) per hour. Apparently previous
workers have not taken into consideration the weight of the shell.
If the weight of the shell be deducted, as it evidently should be,
the O2 consumption curve will be raised from 10 to 30 per cent.

The per cent, of shell at the beginning of incubation for sev-
eral species of insects was found to be as follows: A. tristis, 29.2;
S. cccropia, 22; T. luna, 23.3; and P. ainslici, 31.

Assuming the weight of the egg shell to remain constant
throughout the incubation period, the percentage of shell varies
directly with changes in the weight of the egg. It is thus evident
that changes in the weight of the egg will alter the type of curve
representing O2 consumption. For this reason the weight was
determined just prior to each gas determination and calculations
made accordingly.

Determinations were made on egg masses as soon as they were
deposited and every 12 or 24 hours thereafter, depending on the
length of the incubation period, until hatched. Calculations were
made according to the formula of Krogh (3).

TEMPERATURE.

The effects of temperature upon biological processes are too
numerous and too well known to warrant detailed discussion.
Numerous investigators have studied the effect of temperature
upon the length of the incubation period, but few reports have
been found dealing with the effect of temperature upon the rate
of metabolism as determined by the oxygen consumption. With
this in mind experiments were undertaken to determine the ef -
fects of temperature upon embryonic development of insects.

OXYGEN CONSUMPTION OF INSECT EGGS.

137

«

ta

O

t/}

H

z

u
S

-

o

(d

w

Q

u

o
S

Q

0.

S
5

(/)

o
U

z

o

w
oi

H
<

ft!
W
Hi

W

H
u,

O
(/)

H

^
En
t

W

£

O

H
•<

Q

ta

O

>H

ft!

<

W2

O

O

S
t/3

p.

o

HH

M

OO

M

H

N

oo

0
0

O

GO

00

N

o\

-t

t^-

a

HH

ro

N

Q

£L

O

!N

0

O

3
O

r-

•O
00

0

o

O

q

K

N

N

a

^

o

o

N

[M

0

oo

o

0

M

5

>

3

(S

o
o

0

"0

0

o

'o

to

^o

U">

H

Oi

S

M

CtJ
Ui

M

Ui

8.

.0

O

00

ro

(N

•a

OJ

«

O

S

6

I4H

O

ro

o
o

oo

0

0

i

*9

°?

"?

't

o!

bo

o\

O

^

^

N

10

IN

H

M

«

ro

IO

O

M

"^j-

T

oo

ro

U

•*

rO

ro

"o

M

y>

10

N

N

6

3
o
H

Masses.

M
M

I-H

»— (

-

-

c

'•3

"a;

CO

"a!

0}

J3
w

"5
~

u

3

o

"S

O

o

O
"o

_>

>

>

V

3

_3

3

tn

3

e
c
ri

W

TJ

HH

"o
W

"o

d

6

00

OO

ro

ro

H

138

ROY MELVIN.

ij

s

.5

H

o
ffi

w

to
O

()

3

CJ

X

W

H

=
O

3

w

W

D
W
I

1/1
Z

O

u
o

to
o

K
c/i

a

(-H

oo

h-l

tN

O

OO

tN

o
o

O
0

8

t*

oo

00

00

tN

tN

O

^

Q

o

O

oo

Q

o

-o

q

0

O

o

tN

tN

0

tN

CN

O

o

in

0
tN

M

tN

o
q

oo

>0

O

o
o

*t

q\

o

o

tN

tN

0

0

M

O

o\

O

tN

o

n

oo

10

Tf

oo

M

o

O

O

O

00

M

o

IO

•

M

00

3

s

O

o

ro

ro

ro

M

l*H

O

CO

bo

DO

w

O
O
vO

10

H

O

0

6

13

ii

CO

0

H

M %

W rt

oo

M
M

ON

't

a

a

tu

CO

C

•H

1

«_•

O

tu

E

rt

Z

•2

•»

.Co

^

•S

'•2

o

Q

CO

.Co

1

s

•«*

t3

OXYGEN CONSUMPTION OF INSECT EGGS.

139

The same eggs were used to make the determinations at both
temperatures. After sufficient acclimatization, two to five hours,
depending on the size of the animal chamber, the manometer was
closed and at the end of two hours the reading was made. In like
manner the reading for the next temperature was made and the
eggs returned to the incubator at 28 degrees until the next day.

25
2.0
1.5
1.0
0.5
0.0

02MG M

TEMP. C

15

20

25

30

35

FIG. I. Showing the effects of temperature on the O> consumption of
T. lima eggs. (A) last day of incubation; (B) first day of incubation.

The results of these experiments are shown graphically in Fig. I.

It is observed from Fig. i that the effects of temperature on
O2 consumption are very slight during the first day of incu-
bation and very pronounced during the last day. In order to
determine the effect of temperature on O, consumption during
the entire incubation period daily determinations were made on
two series of S. cecropia eggs at 28 and 34 degrees C. respectively.
These results are summarized in Table I. and shown graphically in
Fig. 2.

For comparative purposes the eggs of four species of insects
were run at 28 degrees C. and the rate of O2 consumption thus
determined is shown graphically in Fig. 3 and summarized in
Table II.

140

ROY MELVIN.

FORMATIVE PERIOD.

Upon examination of Figs. 2 and 3 we note that during the
early part of the incubation period temperature has very little
stimulation on the O2 consumption but as the incubation period

i;

QJ

7

/ a

B/

0
DAYS1 23 + 5G781

FIG. 2. Shows the effects of temperature on O2 consumption during the
entire incubation period of S. cecropia. (A} exclusive of shell 34° C.,

(a) same as above but including shell; (B) exclusive of shell 28° C.,

(b) same as above, but including shell.

progresses its effects become pronounced. This is in accord with
and substantiates the existing theory which states that during
early embryonic development there is a formative period during
zvhich metabolic activity is comparatively low and only influenced
slightly by environmental changes. Mention has been made,
above, of the explanation offered by Fink for the variation in the
length of this formative period among different species of in-
sects. In the case of S. cecropia and T. luna, both species laying
eggs on foliage, the formative period is somewhat lengthened.
This is contrary to Fink's explanation. Data presented in this
paper shows the length of the incubation period to be a greater

OXYGEN CONSUMPTION OF INSECT EGGS.

141

factor in determining the length of the formative period 'than
the type of place where the eggs chance to be deposited. Fig. 3

cu 3

0.

cu

0
OAY51 23 + 5Z1Z1

FIG. 3. O:; consumption of eggs of (A) P. ainslici, (B) A. tristis, (C)
T. lima, and (D) S. cccropia.

bears out this explanation. A study of Fink's curves will show
that they too substantiate the explanation here offered.

SUMMARY.

From a preliminary study of the factors accompanying and
influencing metabolism as determined by the O2 consumption
during embryonic development made on four species of insects
the following conclusions are drawn :

I. The weight of the egg shell is an important factor and
should be taken into consideration.

II. The effects of temperature are not as pronounced during
the formative period as during the period of late incubation.

III. The explanation offered for the variation in the length of
the formative period is the length of the incubation period and
not the place where the eggs chance to be laid as has been sug-
gested.

MELVIN.

LITERATURE CITED.

1. Bodine, J. H.

'25 Effect of Temperature on Rate of Embryonic Development of
Certain Orthoptera. Jour. Exp. Zool., 42: 91-109.

2. Fink, D. E.

'25 Metabolism during Embryonic and Metamorphic Development of
Insects. Jour. General Physiol., 7: 527-543.

3. Krogh, A.

'15 Microrespirometrie. In Ahderhalden, E., Handbuch der Biochem-
ischen Arbeitsmethoden, Berlin, 8: 519-528.

Vol. LV. September, IQ28 No. 3.

BIOLOGICAL BULLETIN

THE INFLUENCE OF MOLDS ON THE GROWTH OF

LUMINOUS BACTERIA IN RELATION TO THE

HYDROGEN ION CONCENTRATION, TO -

.GETHER WITH THE DEVELOPMENT

( )F A SATISFACTORY CULTURE

METHOD.

SAMUEL E. HILL,

PHYSIOLOGICAL LABORATORY,
PRINCETON UNIVERSITY.

At the Marine Biological Laboratory, Woods Hole, during
August and September, 1927, it was observed that cultures of
luminous bacteria (Bacillus Fischcri, Rcijerinck, Migula) tended
to deteriorate rapidly, the deterioration being progressive, so that
finally subcultures were made daily, the luminescence becoming
steadily less, and the culture was finally lost. A fresh culture was
obtained by plating out luminous material of the same stock,
recovered from an old Petri dish culture contaminated with mold.
This culture grew vigorously for a time, and then degenerated.
Since the bacteria in contact with the mold continued to grow and
glow for some time, new cultures were isolated when necessary.

These bacteria were being used for physiological experimental
material, and it was considered necessary to learn the reason for
the deterioration of the cultures and devise a cultural method
by which bacteria of the same strain could be maintained in
vigorous condition throughout any given series of experiments.
Luminous bacteria live normally in sea water, which is maintained
constantly in a fairly definite alkaline pH range. They are con-
sidered to grow best on culture media of about the same pi I
value as the sea water. All of these cultures were grown on the
same medium, supposedly of the proper pH, and it was sug-
10 143

!44 SAMUEL E. HILL.

gc.stt-d bv Professor Harvey that the trouble might be due to
insufficient alkali reserve, the acid produced by the bacteria
rapidly lowering the pH of the medium to a value unfavorable to
their growth. The influence of the mold in causing continued
light and growth might be due to alkali production. This expla-
nation was favored by the result of pouring a solution of M/2
NaCl to which Clark's phosphate buffer, pH 8.0, had been added,
over the surface of several Petri dish cultures which had ceased
to glow. One, in which the light had been out only a few hours,
again began to glow, and the luminesceice lasted for over eight
hours. Others, in which the light had been extinct for longer
periods, were not revived.

Friedberger and Doepner (1907) had studied the influence of
various molds on the light intensity of cultures of luminous bac-
teria. They grew molds in bouillon, filtered the bouillon, and used
this material in making up culture media. They found a greater
intensity of light in cultures grown on these media than on con-
trols prepared with ordinary bouillon. The one difference which
they could establish between ordinary bouillon and bouillon in
which mold had been grown was an increased alkalinity in the
latter. Their figures show that 10 cc. normal bouillon neutral-
ized .4 cc. n/io NaOH to phenolphthalein, while 10 cc. of their
" mold bouillon " neutralized .2 to .4 cc. n/io H.,SO4 to phenol-
phthalein. They arrived at the conclusion that the greater inten-
sity of the light of cultures grown on " mold bouillon " was due
in part to the increased alkalinity, and in part to " other proper-
ties " of the mold.

Molisch (1912) had shown that in general the intensity of
light of cultures of luminous bacteria depended on the rate of
growth. It is the opinion of the writer, for reasons given below,
that the only cause for the increase in intensity of light and
length of life of cultures of luminous bacteria grown in contact
with mold is that of alkali production by the mold, which thus
acts as an alkali reserve.

A series of experiments using solutions of M/2 NaCl plus
Clark's phthalate, phosphate, and borate buffers, found to be non-
toxic, showed that these bacteria glowed brightly in the pH
range 5.7 to 8.7, the luminescence lasting for over an hour.

INFLUENCE OF MOLDS ON BACTERIA. 145

(Observations were not made after more than an hour had
elapsed.) Below pi I 5.2 the light lasted only a few seconds,
above pH 9.0 for three minutes or less. The pH range in which
growth can be expected lies then between 5.7 and 8.7, pH values
outside this range being productive of rapid injury.

The culture medium in use was a peptone, beef-extract, glycer-
ine agar, made up in sea water, the pH being adjusted to 8.2 with
NaOH. As these bacteria live normally in an environment con-
taining NaCl in about one half molecular concentration, favor-
able conditions are provided for the use of buffer mixtures.
Molisch (1912) had shown that a number of salts other than
NaCl might be used in culture media for luminus bacteria. A
culture medium was made up in which one fifth mol of sec-
ondary potassium phosphate in 500 cc. distilled water was sub-
stituted for one half of the sea water. After sterilization the pH
was adjusted with NaOH to 8.2. Separate lots of the same batch
were colored with the Clark and Lubs selection of indicator dyes,
covering the pH range from 1.2 to 9.8. Cultures were started on
slants prepared from these media, six tubes of each being inoculated
with luminous bacteria and three of each six being inoculated
also with a common mold at one end of the slant. (The mold
used was kindly identified for me by Dr. Charles Thorn, as
Penicillinin sp.. in the same section with P. commune (Thorn)
and P. soHtitin (Westling).) These were all allowed to develop
somewhat below room temperature for two weeks.

Some of the indicators used were accumulated by the bacteria.
These are being studied further to determine whether they pene-
trated the cell, or were merely adsorbed on the surface. They
were of little value for this study, since not enough dye was left
in the medium to indicate its pH value. However, in the case of
several of these, the pH was indicated roughly by the color of the
dead bacteria, which was not markedly different from the medium.
With brom cresol green (yellow at 3.8, blue at 5.8), the dead
bacteria near the mold were a more intense blue than elsewhere,
and the acid range of the indicator had not been reached any-
where in the slant.

On the chlor-phenol red slants (yellow at 5.2, red at 6.8) the
color of the medium indicated that the pH had been reduced to

146

SAMUEL E. HILL.

5.4 : : .2. The pH of the medium near the mold was well above
the alkaline range of the indicator.

On the cresol red slants (yellow at 7.2, red at 8.8) the color
of the medium indicated pH below the range of the indicator ex-
cept near the mold, where a pH of 8.6 =p .2 was indicated. The
results with meta cresol purple were about the same. With thymol
blue, the color of the indicator was masked by the color of the
medium at the critical value, and it was of no value.

On the same date six cultures were started on medium of the
same batch without addition of indicator. At the end of two weeks
all were alive and glowing brightly. These cultures decreased
slowly in brilliance during the next month, but were still glow-
ing faintly at the end of six weeks, and viable transfers were
made at the end of the seventh week. The final death of these
cultures appeared to be caused by the drying up of the medium.

As a further check on the alkali influence, several cc. of
M/NaOH was introduced at the bottom of each of six slants of
unbuffered medium colored with brom thymol blue, and an equal
number without indicator. Streaks made on these slants devel-
oped rapidly on the upper half of the slant, away from the alkali,
and grew well, the cultures on the uncolored medium lasting for
several weeks (average of six, 22.2 days), until the alkali was
exhausted. On one of these, more alkali was added and a fresn
inoculation made, the growth lasting this time for less than a week.
It was observed that no growth took place below the line which
marked the limit of diffusion of strongly alkaline NaOH. This
limit was well marked on the uncolored medium by the precipi-
tation of magnesium hydroxide.

The most characteristic activity of luminous bacteria seems
to be that of acid production. They are killed in a few days in
their own acid if some method of neutralization or removal is not
employed. In their natural environment the excess acid would
simply diffuse into the surrounding sea water, but within the
limits of the test tube this cannot occur. The base used in the
culture medium was NaOH, which in contact with carbon diox-
ide becomes NaHCO3, and since NaHCO, in the concentration
used (.01 M ) is not particularly acid when saturated with carbon
dioxide, it is not likely that the acid limiting their growth is car-

IM'l.rF.NCE OF MOLDS ON BACTERIA. 147

bon dioxide. That it is a non-volatile acid is shown by the fol-
lowing experiment :

A constant stream of sterile air was drawn in series through
three bottles of slightly buffered culture medium colored with
cresol red. The first of these was the control, without bacteria.
The other two were inoculated with luminous bacteria. At the
end of 24 hours the control was red, as at the start, and un-
contaminated as shown by the absence of turbidity, and this con-
dition lasted until the close of the experiment. The two inocu-
lated bottles at the end of 24 hours were down to about pH 7.4.
Enough NaOH was added to the third bottle to restore the orig-
inal pH of approximately 8.2. and this was repeated every two
hours until the close of the experiment. At the end of 36 hours,
the PH in bottle No. 2 was down to about 5.5 (determined by
withdrawing some of the material and testing with other indi-
cators) and the light was extinct. In bottle No. 3, in which
pH 8.2 was maintained, the bacteria continued to glow for an-
other 24 hours, when the light failed, due presumably to failure
of the food supply. Carbon dioxide and any other acids volatile
at room temperature (if any were formed) would have been
swept out by the stream of air, leaving behind the non-volatile
acid. This is probably lactic acid.

Other culture media were tried in which calcium and barium
carbonates were employed as buffers, and also higher concen-
trations of K2HPO4 and sea water, and lower concentrations.
Luminous bacteria can tolerate a considerable range of salt con-
centration. It was found that on phosphate buffered media
where the total salt concentration was greater than in sea water, but
not in excess of molar concentration, growth was slower than on
media of the proper concentration, and the tendency to diffuse
growth was absent. The resultant crowding gave the streaks a
fictitious brilliance for a few days, after which the light intensity
decreased to a low value. These cultures were viable for fairly
long periods of time, average 21 days, but not as long as cultures
on media of the proper salt concentration. When media of lower
total salt concentration (as about *4 molar) were used, there was
an initial rapid growth, accompanied by flowing over the surface
of the medium, and a rapid decay, so that such cultures were

148

SAMUEL E. HILL.

viable for only a few days, 'the average of six cultures being five
days. Since a heavy precipitate of calcium and magnesium phos-
phates was formed when the phosphate buffer was added to sea
water, media were prepared containing various concentrations of
NaCl, from .25 M to .75 M, as substitutes for sea water, but
these were unsatisfactory, the best of them lasting for only
14 days.

On medium buffered with calcium carbonate, growth was vig-
orous, but the life of the cultures was less than with the best of the
phosphate buffer mixtures. The average length of life of eleven
cultures without indicator was 17.8 days. Curiously enough, the
death of these cultures was due to excess alkalinity. The initial
growth was rapid, but on the third or fourth day there was a
decrease in brilliance of light and a slowing down of growth,
caused by the rapid diffusion of the acid through the agar, using
up the small amount of calcium hydroxide in solution. This was
followed by an increased brilliance and renewed growth as the
pH rose again, due to the solution of more calcium hydroxide
(produced by hydrolysis from the calcium carbonate), and its
diffusion through the medium. The calcium salt of the acid pro-
duced by the bacteria is much more soluble than calcium car-
bonate, and is evidently hydrolyzed in solution, for the medium
becomes steadily more alkaline until the alkaline range of the
available indicators is passed. Since the bacteria are soon killed
by alkali above pH 9.0, the limiting value is passed, and lumines-
cence ceases. This can happen only when the calcium carbonate is
in excess. When the pH of the medium was adjusted with calcium
carbonate, and the excess carbonate filtered off, initial growth
was rapid, but the decline following it continued until the death
of the culture occurred on the sixth day (average of six cul-
tures), caused by acidity as shown by the use of a suitable
indicator.

On the medium prepared with barium carbonate from which
the excess carbonate was filtered off, the initial fair growth was
followed by a rapid decline, the average length of life of 14 such
cultures being 6.5 days. When an excess of barium carbonate
was present, the initial growth was fair, and slowly decreased,
the cultures growing steadily more alkaline, the average length

IM'I.rKNCE OF MOLDS OX BACTERIA. 149

of life of 14 cultures being 17.5 days. Although theoretically
about the same pH value should be produced 1>y barium and cal-
cium carbonates, in practice the medium prepared with barium
carbonate was always the more alkaline, and was too alkaline for
good growth of the bacteria. The vigorous growth obtained on
calcium carbonate was never obtained on media prepared with
barium carbonate.

The medium prepared with calcium carbonate has the advan-
tage that no adjustment of pH is required, the hydrolysis of
the carbonate giving approximately the right value. It is by far
the best buffer substance to use, both for slants and for Petri
dish cultures. The medium should contain 20 grams " Bacto "
nutrient agar, 10 cc. glycerine, and 5 grams calcium carbonate
per liter, made up in sea water. If a transparent medium is de-
sired, the phosphate buffer mixture with the same amount of
nutrient substance, made up in sea water and filtered, may be
used. The optimum pH value for this medium, probably about
8.6, may be secured by titrating the hot medium by the drop
method until a good red is secured with cresol red, and a barely
perceptible color with thymol blue. When one fifth mol of
buffer is added to sea water, the average life of cultures emitting
strong light is 18 days. After this time, very little light is
emitted, but viable transfers may be made for several weeks.

Of the indicators used, several appeared to be slightly toxic
to the bacteria, but the evidence on this point is inconclusive.

SUMMARY.

The influence of molds on the length of life of cultures of
luminous bacteria may be simulated by the use of buffer mix-
tures, or by supplying fresh alkali continually. The maximum
alkalinity produced in these experiments by the influence of
Penicilliitin sp. was pH 8.6 q= .2. Degeneration of cultures of
luminous bacteria may be caused by growth on media insufiiciently
alkaline, or so slightly buffered that it soon becomes acid. Dif-
fuse growth and spreading over the surface of the slant is caused
by too low salt concentration. Long life of cultures may be se-
cured by growing on media sufficiently alkaline, and sufficiently
buffered to resist rapid change by the acid production of the bac-
teria, which are killed by their own acid at about pH 5.6.

, -0 SAM CEL E. HILL.

BIBLIOGRAPHY.

Clark, W. M.

'27 The Determination of Hydrogen Ions. Baltimore.
Fischer, B.

'88 Ueber einen neuen lichtentwickelnden Bacillus, Centralbl. f. Bakt.,

etc., 3, Nos. 4 and 5.
Friedberger, E., and Doepner, H.

'07 Ueber den Einfluss von Schimmelpilzen auf die Lichtintensitat in
Leticbtbacterien-culturen, etc., Centralbl. f. Bakt., etc., ist Abt.,

43, i.
Migula.

'00 System der Bakterien Zweiter Band, Jena.
Molisch, Hans.

'12 Leuchtende Pflanzen, eine physiologiscbe Studie, Zweite Auflage,
Jena, 1912.

THE SEX RATIO IX PEROMYSCUS.
JOHN J. KAROI..

The data herein presented are based on the records of breeding
experiments with Pcrontyscits, conducted by Dr. F. B. Sumner.
In an earlier paper l the sex ratio in Pcroinyscns was discussed at
considerable length and data covering the years 1915—1921 in-
clusive were presented. The present report is based on the rec-
ords of births from 1922-1926 inclusive. The material is made
up partly of the various mutant strains of the niaiiicitlatus series,
variously hybridized and partly of the three subspecies of Pero-
inyscus polionotus, P. p. polionotus, P. p. leucocephalus, and P. p.
albifrons, both pure and hybrid. No attempt will be made to give
comprehensive interpretation of the findings but reference may be
made to the paper cited above for more detailed discussion.

I take this opportunity of acknowledging my indebtedness and
sincere thanks to Dr. F. B. Sumner who suggested the subject
and under whose general guidance the work was carried out.

The influences which might affect the sex ratio in Peromyscut
were considered in this treatment of the data to be (i) season,
(2) size of litter, (3) race, (4) hybridization.

The total number of broods recorded in the records from
1922 to 1926 is 760, comprising 2,522 young, or an average of
3.32 mice per brood. According to sex these were distributed as
follows :

Males 1,316

Females 1,1 14

Sex undetermined (dead or escaped) 61

.The sex ratio (number of males per hundred females) for those
of known sex is 114.93 ± 3- 19-~ lt is interesting to note here
that the sex ratio for the data from 1915 to [922 was 97.37 ± 1.93.

1 Sumner, McDaniel and Huestis, BIOL. BULL., No. 2, 1922.

D

-The probable error here employed is ±67.45 (i + R) <J—, in whicn
R = sex ratio.

JOHN J. KAROL.

Since the number of individuals considered here is about half
as great as that in the previous paper on Peromyscus, we shall
present the data of this later period only for what they may be
worth. At the suggestion of Dr. Sumner it was considered per-
missable to combine these additional data with the earlier records
and thus, in a sense, bring some of the results on the sex ratio of
Pcromyscus up to date.

The total number of broods in the combined data from 1915
to 1926 is 2,321, comprising 7,547 young, or an average of 3.25
mice per brood. According to sex these were distributed as fol-
lows :

Males 3-597

Females 3,49^

Sex undetermined (dead or escaped) 458

The sex ratio for the combined data is thus 103.01 ± 1.64.

SEASON.

The following table gives the sex ratio for each month of the
year and also the number of individuals upon which this ratio is
based. The table contains the total data for the years 1922-1926.

k

January ( 162) 123.61 + 13.29

February (154) 120.59 i !3-21

March (360) 106.43 + 7.64

April (290) H4-39 i 8.94

May (390) 1 10.50 + 7.65

June (277) 123.77 + 10.12

July (184) 111.90+11.30

August (220) 143.68 + 13.50

September ( 140) 140.35 + 19.42

October (143) 92.96 + 10.67

November (109) 76.67 + 10.15

December (68) 120.00 + 19.30

As it is obvious from the graph that the differences between the
consecutive months are of little significance we may combine our
monthly birth records into four seasons of three months each.
In both the earlier data alone and in the combined data we may
distinguish two high periods and two low periods annually. The
sex ratios for these four periods applied to the later data are as
follows :

THE SEX RATIO IN PEROMYSCUS.

153

1 I ) February-April 1 1 1.86 + 5.43

(2) May-July 1 14.99 ± 5-36

(3) August-October 126.55 ± 7-77

(4) November-January 105.56 + 7.92

The greatest difference between two of these ratios is that be-
tween the third and fourth periods This difference is
20.99 — 1 1-09.

145

140

135

130

125

120

115

110

105

100

95

90

B5

80

75

137

553

JAN FEE MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

FIG. I. The sex ratio of Pcrotiiyscits for each month of the year com-
puted for the data from 1922 to 1926. The figures along the graphs denote
the number of individuals born during each month of the year.

Grouping the same data according to the seasons of the year, as
employed by King and some others, we get the following sex
ratios :

Spring 1 10.12 + 4.90

(March-May)
Summer 126.28 + 6.71

(June-August)
Autumn 102.13 + 3.08

( September-November )
Winter 121.76 + 8.53

(December-February)

Here the greatest difference is between summer and autumn,
being in this case 24.15 ± 7.38. Inasmuch as our figures are
small we make no attempt to attach any particular significance

JOHN J. KAROL.

to these values but we may say in passing that they are of the
same order of magnitude as the findings of King 1 in the Norway
rat. In both we find a maximum in summer followed by a mini-
mum in autumn.

Combining the earlier data (1915-1921) with these additional
data we get the following monthly sex ratios :

January (395) 103.53 + 7.41

February (469) 99-54 ± 6.17

March (1,129) 106.26 + 4.30

April (660) 1 13-65 + 6.20

May (967) 101.13 + 4-61

June (707) 100.59 ± 5-29

July (592) 98.11 + 5.74

August (818) 113.46 + 5.46

September (617) 108.45 + 6.03

October (564) 103.3? ± 6.02

November (367) 78.12 + 5.76

December (279) 96.38 + 7.93

145

140

135

130

125

120

115

110

105

100

95

90

85

80

75

6Z6

346

342

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

FIG. 2. The sex ratio of Pcromyscus for each month of the year com-
puted for the combined material from 1915 to 1926. Numbers along graphs
indicate numbers born in each month.

• Here, as in the earlier data alone, we find two annual maxima,
one occurring in March and April; the other from August to Oc-

1 Arc hir fiir Entwickungsmeckanik, 1927, 61.

'I UK SEX RATIO IN PEROMYSCUS. 155

tober. In the graph we have the appearance of a fairly well
marked biennial rhythm.

Now grouping the combined data according to 3-month periods
we find the following sex ratios :

1 i ) February-April 106.96 + 3.07

(2) May-July 100.19 + 2.96

(3) August-October 107.87 + 3.36

(4) November-January 91.80 + 4.01

The difference between the third and fourth periods is
16.07 — 5-23 ar>d may be considered of probable significance
according to the conventional statistical standard. These figures
still show a rather marked biennial rhythm despite the fact that
the later data showed reversed relations for the February-April
period.

Again, if we regroup the combined data by the ordinarily rec-
ognized seasons the figures become :

Spring 106.22 + .2.78

(March-May)
Summer 104.76 + 3.18

(June-August)
Autumn 97.43 + 3.45

( September-November )
Winter 100.00 + 4.18

( December-February)

Here the greatest difference, between spring and autumn, is
8.79 ± 4.43 and of no probable significance. Likewise the bi-
ennial rhythm, apparent in the case of the later data seems to
have been eliminated by the addition of the earlier data. This,
we may say, is typical of the conflicting results pervading the entire
literature on the sex ratio.

In the previous paper on the sex ratio in Peromyscus it was
stated that the records were " unfortunately not adapted to re-
vealing definite periods of increased or diminished reproductive
activity, since the matings were to a large extent controlled in
accordance with the demands of the breeding experiments."
Since this statement is equally applicable to the later data, we
wish to stress the point that only the number of matings was
controlled and we cannot understand how this could possibly
affect the normal seasonal trend of the sex ratio1

1 Cf. King, 1927.

156

JOHN J. KAROL.
SIZE OF THE BROODS.

The mean size of the 760 broods considered in the later data
i- 3.32. The following table gives the sex ratios for mice be-
longing to broods containing from one to seven individuals re-
spectively. Double broods or broods in which individuals of
unknown sex are known to have died have been excluded.

No. in Brood.

Males.

Females.

Ratio.

i

17

12

141.67 ± 36.07

2

119

103

iiS-53 ± 10.44

3

423

336

125.89 ± 6.25

4

352

3i6

111.39 ± 5.83

5

141

129

109.30 ± 9.02

6

59

3i

190.32 ± 28.36

7

18

17

105.88 ± 24.17

Summarizing the combined data we get the following table for
the sex ratios according to the size of the brood :

Xo. in Brood.

Males.

Females.

Ratio.

i

81

73

110.96 ± 12.09

2

35i

355

98.87 ± 4.96

3

1,047

993

105.44 ± 3-18

4

1,029

983

104.68 ± 3.18

5

405

385

105.19 ± 4.97

6

159

in

143.24 ± 11.96

Considering either the single or combined data we can find no
significant differences in the sex ratios of various sized litters and
we can only conclude that the size of the brood does not seem to
have any well-defined relation with the sex ratio in Pcromyscus.

Separate calculations were made for the litters in which no
deaths were recorded and for the litters in which deaths are
known to have occurred. In the later data we find the sex ratio
for incomplete broods, comprising 43 broods, to be 83.64 ± 11.26.
For the 673 complete broods the sex ratio is 118.09 + 3.38—
the difference between incomplete and complete broods being
3445 — H-7^- \Yhile this difference is large enough to be of
interest we cannot attach any great significance to it inasmuch as
only 43 incomplete broods were considered. In the combined
data we find sex ratios of 91.45 ± 4.76 and 104.65 ± 1.79 based

THE SEX RATIO IN PEROMYSCUS.

157

on 309 and 1,974 broods for the incomplete and complete broods
respectively. Thus we do find a difference between the sex ratios
of complete and incomplete broods but we do not feel justified
in regarding it as significant in view of the meagre record of
identified dead.

145
140
135
130
125
120
115
110
105
100
95

270

790

2012

706

1

FIG. 3. Variations in the mean sex ratio, according to the size of the
broods. Numbers along graphs indicate numbers of individuals.

Combinations of the Sc.vcs in Individual Hroods.

It is interesting to consider the possible tendency of members
of a litter to agree with one another in respect to sex, that is,
whether or not we encounter broods consisting entirely of the
same sex more frequently than would result from chance. In the
following table, using the combined data, we have arranged
broods of each size in groups according to the number of each
sex present. For example, broods of three present four possible
combinations : 3 <5 , 2 £ + i 9 , r <J -j- 2 9,39. The actual num-
ber of complete broods containing a given combination of males

158

JOHN J. KAROL.

W
J
-
<

H

w

fe

tft
Id
J

ta

O

O

I

-

§

u

Vl
!/)

o

o
W

|

S
<

o
u

b
o

§

pa

B

s

0 0 ~

0

0 I- ^

IO

0

O t^ O

LO

_j. cs ro

'b

O

O t^ «

^ IN Tf

^D

H

O

o

O "^O

ro ^ IN
.MM

^00 0

M

M

^

O 1*3 ~O

O

o

^00 O

"

^ M M

'b ^

M

CN

o

o

o

. 00 t—

M ro 00
. oo oo

^ ro •*

HH

^" M

%

O

0

'b M °°

M O Os
, \o ^o

N 10 O

M

•b M "

•b ^

N

ro

O

O

*t> ro H

M o O

N 0 "

M °° ^

"b " "

•b

ro

r;i'

O

PV) O) O\

M oo o

^_ O\ OO
"

" IN (N

-t

o

fV) O fO
^ ro PO

M
^ lO 't

'b

i3 oo 10

''b o M

~ ;

15 •

15 •

15 •

15 •

15 •

3 •

3 •

3 •

3 •

3 •

3 •

4-J 1-1

-*-» u

O Ol
sS--0

C_) dj

cr3 ^2

a HI
cd —

O OJ

O o

cfl .a

o a;

||§

ll§

ill

II §

ll §

1 1 §

ts 2-0

•3 2-a

ts 2-a

t! o _
rt >-c ™

c< .0 ^

C -Q ^

c n ^

C J^ ^

C J2 _lj

13 "-• t)

"P O <U

S • a

o o x

fs ^*^ O

e 0 <"

lo&

IQ "o u
Id |

o ,0 x

1|S

•1° V

B • a
o o x

U 2 W

U 2 U

U 2 W

U 2 W

U 2 U

o 0

-t; 0

|ffl "

'/ —

IN

ro

o

TIIK SKX RATIO IN PEROMYSCrS.

ami females and the " expected " number to the nearest integer
are computed. Since the percentage of males in the combined
data is 50.74, I have computed these last figures by expanding
the binomial (1.015 -)- .985)". In the case of an equality ratio
we should use the ordinary formula for probability, e.g., (i -f- i)n.
Considering the comparatively small number of broods present
in most of the groups we find a rather close agreement between
the actual and the expected figures for all of the broods in which
all members were of the same sex. The actual number of such
homosexual litters, among broods containing from 2 to 6 indi-
viduals inclusive, was 409; while the most probable number on
the assumption of purely random sex-production, was 422. If
we consider fractions (a more exact procedure) this last figure
becomes 420. In the earlier data alone a closer agreement than
this was found, the figures being 276 and 274 for the actual and
expected number of broods respectively. It would appear that
the distribution of the sexes in single broods follows the laws of
chance and there seems to be no tendency for fetuses (or germ
cells) developing in the same parents at the same time to give
rise to organisms of the same sex. We may likewise reiterate the
conclusion of Sumner, McDaniel and Huestis, namely ; ;' the
non-occurrence of polyembrony or true twinning, at least with
sufficient frequency to afreet the results."

RACE.

For the later data we have computed the sex ratio separately
for the "pure" (non-hybrid) polionotus series and in the fol-
lowing table we have listed in addition the sex ratios for some of
the other geographic races (subspecies) as computed by Sumner,
McDaniel and Huestis.

Subspecies.

Males.

Females.

Ratio.

polionolus

1 20

89

i *4.8 } ± I 3. IS

e,ambeli (La Jolla)

77°

840

91.67 dh 3.07

sotioriensts . ....

SSo

373

93.83 ± 4.70

rubidus

ISO

124

120.97 ± 9-91

The difference between " polionotus ': and " gambeli " is
43.16 ± 13.50 and may possibly be regarded as significant. But
11

JOHN J. KAROL.

-f cannot say definitely that these figures imply the existence of
any actual racial differences with regard to the sex ratio.

It is interesting to also observe here that in the subspecific hy-
brids of Pcromyscus pblionotus we find a sex ratio of
114.61 ± 5.79 while the subspecific hybrids considered in the
earlier report (mainly P. maniculatus) give a mean sex ratio of
only 104.76 ± 3.41.

Parkes ' briefly summarizes the data of many workers on
specific variations in the sex ratio in man and other mammals.

HYBRIDIZATION.

In 235 broods comprising 735 individuals of Fx hybrids in
the later series we find a sex ratio of 114.61 ± 5.79. While this
is lower than the ratio for the pure " polionotus " stock (see p.
159), we cannot attach any significance to the latter figures since
they are so small. For the same reason we do not feel justified
in combining the later group with the earlier, in a comparison of
pure and hybrid ratios. We may say, however, that in the earlier
series alone the difference between the ratios for pure and hybrid
stock was found to be 11.49 ±4.1, tne hybrid series giving the
higher ratio. These results are in agreement with the conclu-
sions reached by other workers, e.g., Pearl (1908), King (1911),
and Little (1919), that hybridization "per se " may result in
raising the sex ratio.

THE YEAR.

The sex ratios and the number of individuals upon which they
are based for the year 1922-1926 are as follows :

1922 (296) 106.34 + 8.33

1923 (355) 106.43 ± 7.64

1924 (519) 120.80 + 7.27

1925 (966) II3-93 + 5-03

1926 (386) 125.60 + 8.80

Although it is quite evident that there are no significant differ-
ences here it was thought worth while to present the figures in
view of the fact that the earlier data on Peromyscus (1915-1921)
showed such marked yearly variations. While these results were

] A. S. Parkes, " The Mammalian Sex Ratio," Biol. Review, Vol. II.,
No. i, Nov., 1926.

THE SEX RATIO IN PEROMYSCUS. l6l

inexplicable, they were statistically speaking, the most significant
of all and the likelihood of obtaining one of the differences by
" accident " was less than one in 40,000. It was further proven
that these differences were " not due either to the seasonal dis
trilmtion of births, to the preponderance of hybrid births in one
year as compared with another, or to the operation of any of the
other factors previously considered."

Inasmuch as it is evidently exceedingly difficult to correlate the
annual variation in the sex ratio with any known influences,
accurate data on the subject are generally lacking. Of course it
is not impossible that the most " significant " figures may result
from chance.

SUMMARY.

Data have been presented based upon 2,522 deer mice as re-
corded during the breeding experiments of Dr. Sumner, from
1922 to 1926. Earlier records (1915-1921) were added to the
above and the combined data have also been presented.

The following results seem to be of most importance.

1. The mean size of 760 broods in the later records is 3.32.
For the combined data comprising 2,321 broods the mean size is
3.25 mice per brood.

2. The sex ratio for the later data is 114.93 — 3-I9> while
that for the entire lot is 103.01 ± 1.64.

3. Considering the possibility of a seasonal cycle in the pro-
portion of males and females born, we can only say that we find
in the later data a maximum sex ratio in the August-September
period followed by a minimum during October and November.
In the combined data we find two annual maxima, one occurring
in March and April, the other from August to October, and hence
a fairly well marked biennial rhythm. Grouping the combined
material according to 3-month periods we find in one arrange-
ment that the biennial rhythm is practically eliminated while in
another it is rather well marked. The existence of a seasonal
cycle in the sex ratio of Peromyscus is not definitely proved.

4. The size of the brood in the combined material does not
seem to have any well defined relation with the sex ratio in
Peromyscus.

Although we find a difference between the sex ratio of complete

JOHN J. KAROL.

and incomplete broods we cannot regard it as significant in view
of the meagre records of identified dead.

5. When the number of each possible combination of males and
females, in broods of each size, is compared with the number
expected according to chance, the conformity is found to be, on
the whole, very close. For example, if we compare the actual and
expected totals for all of the broods in which all members were
of the same sex we find 409 as the actual number and 420, the
" expected "' number. Thus there is no preponderant tendency
toward the production of homosexual litters and thus the non-
occurrence of polyembrony or true twinning to any great extent.

6. While the sex ratio for the three subspecies of polionotus is
" significantly ' higher than that for other pure races of Pcro-
myscus we cannot say definitely that these figures imply the ex-
istence of any actual racial differences with regard to the sex ratio
in Peromyscus. The sex ratio of polionotus hybrids is likewise
considerably higher than that of other Peromyscus hybrids which
have been studied. 4

7. No significant yearly variations were found in the sex ratio
of Peromyscus from 1922 to 1926.

COLD HARDINESS IN THE JAPANESE BEETLE,
POPILLIA JAPONICA NEWMAN.

NELLIE M. PAYNE,
NATIONAL RESEARCH FELLOW IN THE BIOLOGICAL SCIENCES.

Cold hardiness, or the ability of an organism to withstand low
temperature may be considered from two points of view, (i)
cold hardiness to the intensity factor or the ability to survive ex-
treme low temperatures, and (2) cold hardiness to the quantity
factor or ability to withstand long periods of low temperature.
By low temperature is meant, temperature below that required for
normal development.

The Japanese beetle, which was introduced into the United
States about 1916, can be secured in large numbers, thus making
intensive study possible. This insect represents a type of ecolog-
ical group, the soil dwelling insects. It passes the winter in the
larval stage; about 97 per cent, in the third instar; about 3 per
cent, in the second. Cold hardiness to both the quantity and in-
tensity factor of low temperature was studied. Both external
and internal factors are involved in cold hardiness. These in-
clude such environmental factors as relative humidity and tem-
perature, and such physiological conditions as nutritional state,
health, blood conductivity and metabolic rate. Most of the work
was done on larvae. Some studies were made on adults and a few
observations were made on cold hardiness in pupae.

METHODS AND APPARATUS.

Respiratory rate and quotient were determined by the modified
Krogh manometer of Bodine and Orr (1925). Conductivity of
blood and body fluids was determined by the ionometer, described
by Gram and Cullen (1923). pH was determined with the type
K potentiometer, using a small vessel capable of testing the pH of
a drop. By this method several readings could be taken on the
same larva. This method was described by Bodine and Fink

163

NELLIE M. PAYNE.

(1925). Occasionally a larva was found that would not bleed
freely enough to give sufficient blood for a reading. Blood was
usually taken from one of the feet. Relative humidity was main-
tained by pulling air over different concentrations of sulfuric acid
by means of a suction pump.

COLD HARDINESS TO THE INTENSITY FACTOR OF Low TEMPERA-
TURE.

In comparison with the oak borers previously studied by the
author, Payne (1926), the Japanese beetles are less cold hardy and
also exhibit less variation to low temperature. In Pennsylvania
the most cold hardy Japanese beetle withstood - - 28° C. : the most
cold hardy oak borer - - 47° C. The most cold hardy Japanese
beetle collected in the field thus far withstood - - 15° C.

Periodicity in cold hardiness to the intensity factor of cold is
not as marked in the Japanese beetle as in the oak borers Synchroa
punctata and Dendroidcs canadcnsis. Comparison of the three
species in question tested at the same dates is shown in Table I.

Conditions other than seasonal which modify the cold hardiness
of the Japanese beetle to the intensity factor of low temperature
are (i) degree of dehydration, (2) disease incidence, (3) nu-
tritional state, and (4) temperature at which the larvae were
kept. Although these larvae are seldom collected in dry places
normally, they are able to withstand a high degree of dehydra-
tion. Larvae dried down to a pulpy condition in which the free
water is reduced to a minimum are cold hardy to both intensity
and quantity factors of low temperature. Severe dehydration is
accompanied by a high death rate. Larvae can be dried down to
one third of their body weight. In the dehydrated condition the
Japanese beetle larvae reach their greatest cold hardiness. Since
eighty per cent, of dehydrated larvae die the effect of dehydration
may be considered highly selective, killing off those larvae unable
to'hold water. Those larvae capable of resisting dehydration are
cold hardy. Relative humidity affects cold hardiness in a de-
cided manner. The results of a series of different experiments
with varying relative humidities is shown in Table

COLD HARDINESS IN THE JAPANESE BEETLE.

( •

O ao •$•

*•**

r-* O *o

u

N N q

N
O

cs o to
CN O ro

H

b

f*5 ro '

H

-H-H-H

W

E

01

u

s S c5

-t

•^ IT) IN

01

* "§ —

(S 1-^. O

H^ y

T T i

>

i i

t-» O t

ro N O

N

O\00 M

»— i

b

t^ O r-i

10
IH

M ^

-H-H-H

OJ

1

O

« i M

>00 Ov

£

i|s

t^. «o 4

M

5 C o

i^ i-) O

1 1 1

HH

u

W

n

H

-^.

OO
O ro 1-1

f

U

O ro O

O\

IO *t HI

.

M O I-

Ov

b

01

-H^ -H

1

+J

O,

i M

10 10

OJ

c/3

oj a-1 g

r- -^ cs
<s od ri

S !D o

1 1

^ u

BJ

OJ

s

y

•2 "w e

o

a

^ Q .M

s.

R 3 §

§35

i J3

vj >3 '**

1 66

NELLIE M. PAYNE.

TABLE II.

1 '.\DucrrviTY OK BLOOD OF JAPANESE BEETLE LARVAE KEPT AT DIFFERENT

TEMPERATURES.

(Conductivity shown in % NaCl equivalent uncorrected for protein.)

o°C.

10° C.

20-22° C.

25° C.

•65

.6

.38

•33

.68

.61

-45

•35

.72

.604

.42

•38

.64

•58

.41

•39

.70

.604

.42

.40

.69

•55

.41

.41

.67

•58

-45

-375

.66

.6

•44

•39

.68

.6

•435

.40

• 71

.46

.42

•45

•39

-43

.41

•445

Starvation at high temperatures, 20° C. or above, is fatal
to the larvae unless the relative humidity is kept high. When
kept at high humidity, larvae are able to withstand comparatively
long periods of starvation. One hundred larvae were kept with-
out food for the month of May, 1927, but under conditions of
100 per cent, relative humidity or saturation. Each larva was
placed in an individual vial and weighed before and after the
starvation period. During the process they lost about one half
of their body weight. None of them survived freezing, the low-
est freezing point was - - 1.7 'C. ; the highest - -.65° C. Larvae
kept at -|- io°C., or below their developmental temperature, lost
one half of their body weight. Starvation conditions were as-
sured by keeping the larvae in sterile white sand kept moist
with distilled water. About one fourth of these larvae survived
freezing. Changes in body weight under different conditions of
starvation and dehydration are shown in Fig. i. The effect of
prolonged exposure to low temperature as well as starvation was
involved in the experiment described above. The effect of differ-
ent temperatures on cold hardiness as measured by blood con-
ductivity is shown in Table "Hrl. Larvae starved for one week at
-f 20° C. increased in cold hardiness. In general early stages
of starvation are marked by (an increase in cold hardiness, later

COLD HARDINESS IN THE JAPANESE BEETLE.

I67

BODY WEI

(0(o

rtu

WEIGHT

2 MONT
PERATUR

POPlLLI/>

I

J

or LJ

$

(T

u h
h-

LJ

<

<0

O(-

1 — — 1

rr5(0

NORMAL

ARVAL WEIGH
TARVATION AT

WEIGHT AFTEI
H'S STARVATION
TEMPERATURE

i

b -i

-JCO

<t;i:

L_ Q

t LJ

cz^

CD <£

LJ cr

j

> >

n

LiJ

Q

IN

FIG. i.

158 NELLIE M. PAYNE.

TABLE III.

EFFECT OF DIFFERENT RELATIVE HUMIDITIES ON CONDUCTIVITY AT

TEMPERATURE OF 22° C.
(Conductivity shown in % NaCl equivalent uncorrected for protein.)

Saturation.

80%.

50%.

•33

46

72

.38

• .45 - •

75

•35

•47

76

•39

465

78

•40

44

73

.41

455

77

•375

.. -46

74

•39

44

76

40

455

77

•375

46

76

•38

45

82

.32

46

77

•445

78

•43

stages by a decrease. The point of decrease in cold hardiness
from starvation comes when the digestive tract clears. In con-
nection with this observation it is interesting to note that freshly
molted larvae are unable to withstand freezing until they have
eaten. Pre-pupae with clear digestive tracts are not cold hardy.

The occurrence of wilt disease in many of the specimens col-
lected in the field offered an opportunity for the study of the
effect of this disease on cold hardiness. Larvae were collected at
the same date and subjected to the same conditions of temperature
and relative humidity, only healthy larvae were studied. No
larva showing typical symptoms of wilt disease or polyhedral-
skrankheit was able to survive freezing. Since thermocouples
used in diseased larvae were difficult to sterilize and might infect
healthy larvae, cold hardiness was studied by measuring blood
conductivity rather than freezing point depression. Conductiv-
ity decreases as the disease progresses. On the first day of
apparent infection, conductivities of blood of diseased larvae were
below that of healthy larvae. To produce such a marked change
on the first day of infection, the causative organism must affect
the blood very profoundly and very rapidly. On the other hand
the change in conductivity may not be as rapid as it appears. The
disease may be present in larvae before it is detected by discol-

COLD HARDINESS IN THE JAPANESE BEETLE.

169

\35

\
\
\

RELATION BETWEEN WILT
DISEASE AND CONDUCTIVITY
FIG. £

OD
IT

-§ 20
o
(O

c

15

o
O

10

Period of Apparent Infection in Days

7

8

FIG. 2.

10

oration or wilting, and may be producing conductivity changes in
the blood before other symptoms can be observed. A graph
showing the relationship between day of apparent infection and
blood conductivity is shown in Fig. 2.. Table IV. shows the re-

-Q NELLIE M. PAYNE.

TABLE IV.

CONDUCTIVITY OF HEALTHY AND DISEASED JAPANESE BEETLE
(Conductivity shown in % NaCl equivalent uncorrected for protein.)
Wilt Disease. Healthy. Blackened by Freezing.

OT

.38

• 6

T7

.4

6

•l/

TO

.41

604

T T

.4 ...

604

58

375

575

(17

43

61

•w

18

.425

63

12

.41

62

06

."?Q

625

.4.

64

-25
.27

•15

suits of conductivity readings made on the blood of diseased
larvae in comparison with healthy ones.

Wilt disease is characterized by a pronounced blackening that
precedes the final softening that occurs just before death. Black-
ening also has been observed when larvae are frozen and thawed
quickly. Blood from larvae blackened after thawing always
showed high conductivity. In these cases discoloration was be-
lieved to be due to changes in cell permeability releasing certain
oxidative enzymes, which on escaping blackened the cells. The
prothoracic segment is the first portion of the larvae to discolor
.after freezing, both in the Japanese beetle and in the oak borers
studied. Changes in permeability could be observed during the
thawing process. Water apparently passes through the body wall
where the chitin is thinnest. This water was frequently reab-
sorbed when the larva? were kept under small bell jars. Larvae
losing water alone were generally able to survive freezing. When
the fluid exhuding from the larva gave tests for amino-acids or
proteins the larvae always died. The exudate remained colorless
for several days unless hydrogen peroxide was added, in which
case it blackened quickly. Larvae which showed the exudate
after thawing were fixed and sectioned, but in these sections no
gross differences from normal tissue could be detected. Broken

COLD HARDINESS IN THE JAPANESE BEETLE. 1JI

cell walls were not in evidence. The direct cause of death from
extreme low temperature has been interpreted as due to an irre-
versible change in permeability rather than to a breaking of the
cell walls.

If larvae capable of surviving low temperature are ground up
and filtered and the filtrate precipitated with lead acetate, there
occurs in the filtrate an enzyme capable of breaking proteins down
to amino acids at low temperatures and of building up proteins
from amino acids at high temperatures. A similar enzyme has
also been found in tussock moth eggs. Reversible reactions with
proteases have been reported by Abderhalden (1914) from auto-
lyzing tissues. Taylor (1909), found that a protein— ;' plastein "
—could be formed from albuminose and proteolytic enzymes.
The reversible reaction of starch to sugar at low temperatures
and sugar to starch at high temperatures is a well known reaction
that takes place in potato storage. The cold hardy mechanism
of these larvae studied may. in part, be due to enzyme action which
transforms large protein molecules into smaller amino-acids. The
larger number of osmotically active units thus formed would lower
the freezing point.

Periodicity to cold hardiness is not as marked in the Japanese
beetle as it is in some of the insects that are exposed to extremes
of low temperature. Larvae of the Japanese beetle live close
enough to the surface of the ground to experience some seasonal
change. During the spring and fall they are in addition subjected
to diurnal temperature change. Cold hardiness in the larvae ap-
pears to be closely related to their environment. These or-
ganisms are somewhat seasonal in their resistance to low tem-
peratures. This periodic cold hardiness is shown in Table I.
Comparison with oak borers and aquatic insects is shown more
fully in a previous article by the author (Payne, 1926). Al-
though the larva stage is the only one which overwinters in this
climate, it was thought that studies on the cold hardiness of the
adults would yield valuable material for the comparison of a stage
exposed to winter conditions and a stage not normally exposed.
Adults captured in summer and frozen without previous condi-
tioning were able to survive ice formation within their tissues
and to survive temperatures as low as - - 20° C. Since it was im-

172

NELLIE M. PAYNE.

possible to obtain enough blood from the adults to make a con-
ductivity reading none were made.

A beginning was made on the study of cold hardiness of the
Japanese beetle pupae. From present observations the age of the
pupae and consequently the degree of hydrolysis they are under-
going determines cold hardiness.

No changes in blood pH were found to be associated with cold
hardiness in healthy larvae. The pH obtained from a series of
blood samples is shown in Table V. In the early stages of wilt

TABLE V.
PH OF JAPANESE BEETLE LARWE BLOOD THIRD INSTAR.

Each reading is an average of 3.

Healthy. With Wilt Disease.

6.5 5-8

6.78 5-7

6.92 5.56

6.94 6.

6.5 : 5-Qi

7.16 5-82

7-i8 5-83

6.66 5-97

6.77 6.1

7-1 5.84

7-17 5-92

7 5-96

6-54 5.98

6.66 5.96

6.82 5.94

7 5-9

6-35 • 5-84

6.51 5-97

7-i 5-95

7-i

disease the pH was lower than in healthy larvae. In the late
stages of the disease the larvae were in such condition that it was
difficult to obtain blood by cutting off the feet.

The respiratory quotient tends to be high in both cold hardy
and non-cold hardy specimens, ranging from .67 to .72. The
respiratory quotient of starving larvae tended to be higher than
well fed larvae regardless of the temperature at which they were
kept. The respiratory rate in larvae in which cold hardiness had

COLD HARDINESS IN THE JAPANESE BEKTI.K.

173

been induced was much lower than in the non-cold hardy indi-
viduals. Associated with the low respiratory rate of hibernating
forms was the slight change in body weight occurring over a
period of several months, as shown in Fig. i.

COLD HARDINESS TO THE QUANTITY FACTOR OF Low

TEMPERATURE.

Both the second and the third instars of the Japanese beetle
larva are cold hardy to the quantity factor of low temperature
except directly after molting or when the digestive tract is clear.
Larvae are markedly adapted to withstand long periods of low tem-
perature. At the present writing there are still ten larvae alive
of one hundred which were placed at -|- 10° C. on December 6,
1925. These larvae have now been kept over two years below
their developmental temperature. Similar lots have been kept
from six to twelve months at -f- 10° C. Graphs showing the
number of larvae surviving plotted against time in months in these
experiments are shown in Fig. 3.

The relationship between survival for long periods at low
temperatures and cold hardiness to the intensity factor of low
temperature is shown in Table VI. The two types of cold hardi-

TABLE VI.

SURVIVAL AFTER FREEZING OF JAPANESE BEETLE LARV.TI.
Kept at constant temperature of + 10° C. for varying periods of time.

Length of Time Kept
at + 10° C.

Number
Frozen.

Number
Survived.

% Survived.

2 weeks

I J.sO

1,426

08.^4

4 weeks. . . .

I 4OO

1,078

77

8 weeks

I ,OOO

64=5

64. S

3 months

5OO

2QO

58

6 months

2OO

48

24

ness appear to be inversely related after a certain point has been
reached. This decrease in cold hardiness to the intensity factor
of low temperature cannot be interpreted as a simple loss in vital-
ity since larvae kept at low temperatures are able to complete
their development when placed at room temperature with no
higher death rate than larva- maintained at room temperature.

174

NELLIE M. PAYNE.

SURVIVAL OF POPILLIA
AT 10° C.

^o

10

45678
TIME IN MONTHS

FIG. 3.

9 10 II \&

Col. I) HARDINESS IN THE JAPANESE BEETI.K. [75

Long periods of dormancy accelerate development when the larva-
kept at low temperatures are raised to developmental tempera-
tures. Blood conductivity at first rises, then falls after two or
more months when larvae are placed at or below -)- 10' C.

The effect of rapid alternation between high and low tempera-
tures on cold hardiness was tried with one hundred third instar
larvae. Temperatures of o° C. and -j- 30° C. were alternated
every twenty- four hours for one month. Neither of these tem-
peratures is fatal. As controls one hundred larva- were kept at
o° C. and one hundred at -j- 30° C. None died at -f- 30° C.
Those alternated between -|- 30° C. and o° C. died more rapidly
than those kept at o° C. Results of these experiments are shown
in Fig. 4. In larva? which had been exposed to wilt disease alter-
nating temperature had no effect ,on length of life. None of
these larva; lived longer than ten days except when they were
kept at or below o° C. Healthy larvae were considered exposed
when they had been bitten by larvae having wilt disease.

The respiratory quotient of larvae cold hardy to the quantity
factor of cold was somewhat variable but not connected to length
of survival at low temperatures. In larvae with clear digestive
tracts it tended to become lower. In larvae kept at -(- 10° C. it
ranged from .69 to .73, or slightly higher than in larva- cold hardy
to the intensity factor of low temperature. In larva- with clear
digestive tracts low respiratory quotients were associated with
lack of cold hardiness.

Low respiratory rate is associated with cold hardiness to the
quantity factor of low temperature. Changes in body weight, as
has been stated before, were very small with larva- kept for IOHL;
periods of time at + 10° C. These changes occurring in differ-
ent states of nutrition and under varying temperature and hu-
midity conditions are shown in Fig. I.

Dehydration of larvae is associated with cold hardiness to the
quantity factor of low temperature as well as to the intensity fac-
tor of low temperature. Dehydration beyond two thirds of the
body weight decreases cold hardiness to the quantity factor of
low temperature. Over dehydrated larva- lived but one day at
20° C. and not more than three days at -f 10° C. or not move
than four days at o° C. Dehydrated larvae have been kept for
12

NELLIE M. PAYNE.

\

SURVIVAL OF POPILLIA

\

\

\

\

\

80

70

60

o:

50 JLJ

T

z

40

10

10

•At 0° C

'At temperatures alternating
between 0°and 30° every ^4 hrs.

30 40

TIME IN DAYS

FIG. 4.

5O 60

COLD HARDINESS IN THE JAPANESE BEETLE. 177

one year at -f- 10° C. The experiment has not been continued
long enough to determine whether or not dehydration increases
the cold hardiness to the quantity factor of low temperature.
Untreated larva are able to live two years or more below their
developmental temperature. Dehydrated larvae show very nearly
the same death rate as undehydrated ones.

LITERATURE.

Since the literature pertaining to cold hardiness has been re-
cently brought together it seems hardly necessary to make a de-
tailed list and discussion of it. Robinson (1927) has discussed
and given experimental data on water binding capacity as a factor
in cold hardiness. Robinson (1926), and the author (1926) have
summarized the literature. Hibernation in regard to both its
ecology and physiology has been recently treated by Fink (1925),
Townsend (1926), and Holmquist (1926).

From a survey of the literature it would appear that no one
factor is an adequate measure of cold hardiness. The develop-
ment of a cold hardy from a non-cold hardy insect is a deep-
seated physiological process which affects blood, and body fluids,
respiratory rate and permeability. Nutritional state and environ-
mental conditions also influence cold hardiness.

ACKNOWLEDGMENTS.

The facilities for this study of cold hardiness were furnished
by the Zoological Laboratory of the University of Pennsylvania.
To Dr. J. H. Bodine, of the Department of Zoology, I owe thanks
for many helpful suggestions, especially as regards apparatus and
methods. To Dr. Henry Fox, of the Japanese Beetle Labora-
tory at Moorestown, New Jersey, I am indebted for most of the
material used and for suggestions that have proved of service
during the progress of the work.

SUMMARY.

i. Cold hardiness, both to the intensity factor and to the quan-
tity factor of low temperature, were studied in the second and
third instars of the Japanese beetle. Brief observations were
made on pupse and adults with regard to cold hardiness.

NELLIE M. PAYNE.

2. Japanese beetle larvae are somewhat periodic in their cold
hardiness to the intensity factor of low temperature, less so than
the oak borers previously studied and more so than the aquatic
insects.

3. Disease incidence, nutritional state, and degree of dehydra-
tion are associated with cold hardiness to the intensity factor of
low temperature.

4. Development of cold hardiness to the quantity factor of low
temperature is associated with loss of cold hardiness to the in-
tensity factor except in extremely dehydrated individuals.

5. Marked permeability changes associated with enzyme action
occur at the vital temperature minimum.

LITERATURE CITED.

1. Abderhalden, Emil.

'14 Versiiche iiber die Synthese von Polypeptiden, Peptonen, und Pro-
teinen mittels Fermenten. Fermentforschung 1 : 47-57.

2. Bodine, Joseph Hall, and David E. Fink.

'25 A Simple Micro-vessel with Electrode for Determining the Hy-
drogen Ion Concentration of Small Amounts of Fluid. Jour.
Gen. Physiol., 7 : 735-740.

3. Bodine, Joseph Hall, and Paul Rudbert Orr.

'25 Respiratory Metabolism. Physiological Studies on Respiratory
Metabolism. BIOL. BULL., 48 : 1-14.

4. Fink, David E.

'25 Physiological Studies on Hibernation in the Potato Beetle, Lcpti-
notarsa decemlineata Say. BIOL. BULL., 49 :38i-4O5.

5. Gram, H. C., and Glenn. E. Cullen.

'23 The Accuracy of the " lonometric " Method and of the Protein
Correction in Measuring Conductivity. Jour. Biol. Chem., 67 :
477-491.

6. Holmquist, A. M.

Studies in Arthropod Hibernation. I. Ecological Survey 01
Hibernating Species from Forest Environments of the Chicago
Region. Ann. Ent. Soc. Amer., 19 : 395-428.

7. Payne, Nellie M.

'26 Freezing and Survival of Insects at Low Temperatures. Quart.
Rev. Biol., 1 : 270-286.

8. Payne, Nellie M.

'27 Measures of Insect Cold Hardiness. BIOL. BULL., 52 : 449-457.

9. Robinson, William.

'26 Low Temperature and Moisture as Factors in the Ecology of the
Rice Weevil, Sitophilus orysa L. and the Granary Weevil,
Sitophilus (/ranarins L. Minn. Agri. Expt. Stat. Tech. Bull.,
41 : 43 p.

COLD HARDINESS IN THE JAPANESE BEETLE. 179

10. Robinson, William.

'27 Water Binding Capacity of Coloids a Definite Factor in the Win-
ter Hardiness of Insects. Jour. Econ. Ent., 20: 80-88.

11. Taylor, Alonzo Englebert.

'09 On the Synthesis of Protamin through Ferment Action. Jour.
Biol. Chem., 5: 381-387.

12. Townsend, M. T.

'26 The Breaking-up of Hibernation in the Codling Moth Larva. Ann.
Ent. Soc. Amer., 19: 429-439.

PELAGIC DISSOCONCHS OF THE COMMON MUSSEL,

MYTILUS EDULIS, WITH OBSERVATIONS ON

THE BEHAVIOR OF THE LARV.E OF

ALLIED GENERA.1

THURLOW C. NELSON.

The larvae of the common black mussel, Mytilus eduJis, are
abundant in plankton samples taken throughout most of the
summer in all regions where this mollusc occurs, Stafford, '12.
Recognition of the larva, as Stafford points out, is rendered easy
owing to its horn yellow color, its relatively small umbones and
its small depth. To these characteristics may be added the dis-
tinctive shape of the shell, being more pointed and of shorter
height at the anterior end. Fig. i. The size of the mature pro-
dissoconch when ready to attach varies considerably as judged
from measurements of the largest larvae obtained from the plank-
ton, and from measurements of the prodissoconch shell of newly
attached dissoconchs. Measurements of -ten of the largest larvae
found in the plankton in Maine waters are as follows, antero-
posterior axis being given first.

360 X 338 n, 368 X 320 P>

360 )( 320 M, 336 X 304 p,

376 >( 344 /*, 360X312^,

35° X 312 /A, 360 X 320 ju,, exclusive of dissoconch rim,

336 X 304 ju, 304 X 280 p, exclusive of dissoconch rim.

The last two larvae, although caught in the plankton, each bore
a narrow rim. of purple dissoconch shell, Jackson, '88. From
these and from other measurements made upon Mytilus larvae it
appears that dissoconch shell may be secreted at any time after
the larvae attain a length between approximately 300 and 360 u.
Stafford, I.e., gives the measurements of two mature prodisso-
conchs as 345 X 310/4 and 400 X 331/1.

1 From the Zoological Laboratory of Rutgers University. Paper No.
14, New Jersey Oyster Investigation Laboratory.

1 80

PELAGIC DISSOCONCHS OF THE COMMON MUSSEL. I X I

4

The observations here reported were made during August,
1924 and August, 1927, in Frenchman Bay, Mt. Desert Island,
Maine.1 A collecting station some 100 meters from the labora-

FIG. i. Stages in the development of the prodissoconch larva of My til us
cdulis: U, umbones ; A, anterior end.

tory point was marked with a buoy. With approximately 12
meters depth at mean low water this station lay in the full sweep
of the tide through Frenchman Bay. Plankton samples of 25
liters were pumped here from various depths using a hose and
oscillating clock pump, the majority of the samples being taken
at the surface and at 7 meters depth. The Mytilits larvie were
collected by passing the water through a No. 18 treble extra
heavy bolting cloth net, adding two or three drops of formalin to
the catch and then drawing off the supernatant water bearing great
quantities of the diatoms Chcetoceros and Rhisosolenia. Table I.
gives the numbers and stages of development of the mussel
larva; taken at the station, together with other data.

1 It is a pleasure to acknowledge my indebtedness to the former Director,
Professor Ulric Dahlgren, for the facilities given me at the Mount Desert
Island Marine Biological Laboratory at Salisbury Cove, and for making
early summer plankton catches for me.

1 82

THURLOW C. NELSON.

TABLE I.

\VATKK CONDITIONS AND NUMBERS OF My til us LAKV.I; AT STATION OFF
LAHOKATORY POINT, FRENCHMAN BAY, MT. DESERT ISLAND IN 1924.

Uatr.

Time.

Tide.

Depth.

Temper-
ature
0 C.

Mytilus Larvae in
25 Liters.

Prodisso-
conch.

Disso-
conch.

Aug. i

—

—

Towing

. — .

— .

i

5-

11:30 A.M.

2/3 flood

2 m.

10.6

Many

i

7 •

3:00 P.M.

High

7 m.

II. 0

262

o

13-3

753

8.

3:45 P.M.

High

7 m.

10.9

336

o

17.1

7i

9-

n :3o A.M.

Low

7 m.

11.9

278

2

o

15-6

5

1 1 .

10:15 A.M.

1/2 ebb

7 m.

13-7

63

o

14-3

6

n .

10:30 A.M.

1/2 ebb

Towing

14-3

Many

8

12.

10:10 A.M.

1/3 ebb

7 m.

ii. i

1,500

o

13-9

4

1.3-

ii :oo A.M.

i/6 ebb

7 m.

11.7

650

o

13-8

6

13-

2:50 P.M.

3/4 ebb

7 m.

ii. 7

213

o

iS-3

20

14-

3:30 P.M.

1/6 flood

7 m.

12.3

240

i

o

12.9

320

IS-

10:45 A.M.

High

7 m.

II.O

390

7

o

12.9

i

16.

11:20 A.M.

5/6 flood

7 m.

II.O

i

2

0

12.9

177

18.

10:50 A.M.

2/3 flood

7 m.

10.8

65

o

iS-9

i

19.

3:10 P.M.

High

7 m.

12.2

152

o

13-2

2

20.

II :30 A.M.

1/3 flood

7 m.

II.9

56

o

12.7

2

21 .

2:50 P.M.

5/6 flood

7 m.

1 1. 6

30

o

12.0

4

22 .

5:00 P.M.

High

7 m.

II.4

24

I

o

14.7

5

23-

3:10 P.M.

2/3 flood

7 m.

II.O

17

0

13.7

7

25-

3:00 P.M.

1/3 flood

7 m.

ii. 7

40

I

o

i3-i

10

27.

3:45 P.M.

1/6 flood

7 m.

12.3

19

o

13-9

7

PELAGIC DISSOCONCHS OF Mytilus.
(a) Buoyancy through Gas Secretion.

In the tow sample taken by Professor Dahlgren August i was
found one Mytilus larvae which bore the distinct rim of purple
shell which marks the commencement of the dissoconch stage.

PELAGIC' DISSOCOXC'IIS OF TIIK COMMON M L'SSFL. 183

Since only one such mussel was found it was believed to have
been accidentally introduced through the townet striking some
object bearing attached mussels. Subsequent collections, how-
ever, revealed numerous dissoconchs up to 941 ^ in length freely
floating about at various depths up to / meters. A 25 liter sample
pumped August 13, from the surface, 20 cm. from a Mytilits-
covered pile, yielded 200 Mytilus larva- from mature prodis-

FIG. 2. Pelagic dissoconch of Mytilus cditlis approximately .8 mm. in
length, bearing a large bubble of secreted gas within the branchial cham-
ber. This specimen came from a depth of 7 meters.

soconchs to advanced dissoconchs over 900 ^ in length. A similar
sample pumped August 15 from a depth of 7 meters at the col-
lecting station, more than 100 meters from the nearest mussel
beds, gave 390 prodissoconch Mytilus and 7 dissoconchs which
ranged in length from 445 to 784 p.

The presence of well-developed dissoconchs floating freely in
the water at once raises the question of the means by which this
is effected in the absence of the swimming organ or velum of the
prodissoconch. When brought to the laboratory for examina-
tion these dissoconchs were found to be identical with others
removed from sea weeds, save for the presence of a large clear
space in the posterior portion of the pallial cavity. Believing
that some change might have occurred in the molluscs even dur-
ing the fifteen minutes to half an hour which elapsed between
their capture and subsequent examination in the laboratory, a

1 84

THURLOW C. NELSON.

binoular was taken in the boat and the young mussels were ex-
amined immediately after their capture. The result is shown in
Fig. 2. A large bubble was found to occupy the posterior part
of the pallial cavity, its buoyancy causing the young bivalve to
bang suspended in the water umbone downward, with the postero-
ventral margin of the valves turned upward. On one occasion
the bubble was seen to form through the coalescing of many
minute bubbles, which, passing slowly down between the gill
filaments, united to form a single large bubble. In several in-
dividuals two or three smaller bubbles were found. Where a
single bubble was present its size was such as to cause a thin-
ning of the mantle on either side and a forward displacement of
the posterior gill filaments, thus accounting for the large clear
space already noted in the posterior pallial cavity of the young
Mytilus dissoconchs first taken.

Failure to observe the bubble in larvae first brought to the
laboratory was due to the fact that as soon as a Mytilus dissoconch
comes in contact with any object the foot is rapidly extruded
from between the valves and brought into contact with the sur-
face. The extrusion of the foot, accompanied as it is by a
separation of the valves and of the applied lobes of the mantle,
results in the immediate escape of the bubble in nearly every
instance.

The composition of the gas in the bubble was not determined
owing to its small size and lack of adequate facilities for a micro -
chemical test. The fact that it forms within the gills would indi-
cate that it is mainly oxygen. The composition of gas secreted
into the swim bladders of fishes renders this still more probable.

The possibility that the bubbles within the branchial chamber
of these young Mytilus might have been air introduced acci-
dentally during passage through the pump or while in the net,
was tested in the following manner. The hose was disconnected
from the pump and allowed to siphon water from a depth of 7
meters into the net held in the bottom of the boat. The stream
entered the net under water and great care was taken not to
agitate the net or to break the water surface. Dissoconchs of
Mytilus collected in this way revealed the same large bubbles as
before. Dissoconchs of Mytilus collected from sea weeds and

PELAGIC DISSOCONCHS OF THE COMMON MUSSEL. 185

violently shaken with a little sea water in a bottle failed to ac-
quire any bubbles of air in the process : thus, with the above
experiment, proving that the bubbles of gas were not accidentally
introduced.

To determine the possible effects of pressure in bringing about
gas secretion pieces of glass tubing 2 cm. long were cut and an
early Mytilus dissoconch obtained from sea weeds was intro-
duced into each. A piece of coarse bolting cloth was tied over
each end of the tubes which were then fastened to a line at one
meter intervals and suspended from a float at the collecting sta-
tion. One string bore seven tubes which hung at depths of from
two to eight meters. A second string was attached to a weight
on the bottom with a float of sufficient size to hold the string
vertically in the water, the lowest tube being at u meters depth
at low water and approximately 15-16 meters at high water.
When removed 48 hours later all of the bivalves were found to
have attached by the byssus to the inside of the tubes or to the
bolting cloth ends. When removed to a dish of sea water they
crawled actively about with the foot. In no case was a bubble
present. Either the stimulus to gas secretion is absent when the
mussels are attached, or the frequent extrusion of the foot which
occurs while the mussel is attached permits the escape of such
gas as rapidly as it is formed.

(b) Attachment to the Surface Film.

If the surface of the water near a mussel bed or near a mus-
sel covered piling be skimmed with a plankton net during the
latter part of the breeding season, numerous dissoconchs will be
found. They are most numerous as the rising tide first sweeps
over the mussels and attached sea weeds. A microscopic exami-
nation of these dissoconchs shows that none contains a bubble,
hence it is obvious that these young mussels must maintain them-
selves at the surface through means other than the gas secretion
employed by larvae at a depth. When placed in a dish of sea
water such larval mussels exhibit great activity, gliding about upon
the long, highly adhesive, ciliated foot as rapidly as a snail. Ob-
servations were made upon these young molluscs using a cham-
ber 0.5 cm. wide made of two microscopic slides, filled with sea
water and viewed horizontally through the binocular.

1 86

THURLOW C. NELSON.

Once in contact with a solid object, such as the wall of the
chamber, a rock, or a fragment of sea weed, the mussels exhibited
a marked negative geotropism and climbed straight upward until
the surface was reached. Here the distal one third to one tenth
of the foot was extended in the surface film. Fig. 3, and with
a quick contraction of the foot, aided apparently by contrac-
tion also of the pedal retractor muscles, the ventral margins of

FIG. 3. Ventral view of My til us dissoconch, 4 mm. long, hanging from
the foot in the surface film, as seen from the side and partly from above.

FIG. 4. Lateral view of 4 mm. Mytilus dissoconch hanging from the
surface film. The siphons are fully extended.

FIG. 5. Lateral view of 3 mm. Mytilus dissoconch hanging from byssus
thread attached to holdfast secreted in the surface film. The foot, which
was fully extended in the surface film while secretion of the holdfast was
effected, has been wholly withdrawn between the valves.

FIG. 6. Lateral view of 3 mm. Mytilus dissoconch holding to the sur-
face film with the aid of the tentacles of the incurrent siphon. The foot
which serves to hold the mussel close to the film until the siphon is in-
serted therein, has been withdrawn between the valves.

PELAGIC DISSOCONCHS OF THE COMMON Ml'SSKL. 187

the mantle were brought into contact with the surface film.
While lying with the entire ventral margin of the body in con-
tact with the surface film the byssus gland in a few seconds se-
creted onto the surface film a small holdfast similar in appearance
to that which is laid down on rock or piling for the attachment of
each byssus thread. A thread 1-2 mm. long serves to support
the young mussel from this float and with foot withdrawn it
may hang suspended indefinitely, Fig. 5. At times it thrashes
about with the foot fully extended as though in search of some
solid surface for attachment. When the foot strikes such an
object the mussel glides quickly upon it, trailing the byssus thread
behind or breaking it off. The "float" is not a buoyant struc-
ture, since when pushed beneath the surface film it rapidly sinks.
It maintains its position in the surface film, supporting mussels
up to 4 mm. in length, solely through surface tension.

A float and connecting thread are not always secreted when the
young mussel reaches the surface. At times it supports itself
solely by the distal end of the foot in the surface film, Fig. 4,
after the manner described for the prodissoconch oyster larva,
Nelson, '243. With the aid of the numerous short cilia covering
the foot the animal glides slowly along the surface film, rocking
the body slowly from side to side and occasionally through a
quick contraction of the proximal portion of the foot, bringing
the entire ventral margin of the shell in contact with the surface
film. This behavior will recall the familiar habit of pond snails
of hanging from the entire foot spread out in the surface film.

A third mode of suspension from the surface consists in ex-
tending the tips of the tentacles of the incurrent siphon into the
film and hanging from these, Fig. 6. This behavior, though sel-
dom observed, serves to support the mussel quite as effectively
as does the foot.

Such floating dissoconchs have never been found further than
approximately 25 meters distance from mussel beds or mussel
covered piling. Their abundance, 5 to 100 per 25 liters of water,
close to such habitats indicates that young mussels frequently
make use of this mode of transportation for covering short dis-
tances.

Examination with the low power binocular of several small

j88 THURLOW C. NELSON.

tide pools close to the laboratory revealed numbers of Mytilus
3-4 mm. long moving actively over the rocks and barnacles while
others were at the surface. With the incoming tide the latter are
carried away and may eventually reach a place of attachment at
a considerable distance.

METAMORPHOSIS IN ALLIED LAMELLIBRANCHS.

The water samples taken in Frenchman Bay contained in ad-
dition to the larvae of Mytilus edulis, great numbers of the young
of the soft clam, Mya arenaria, together with occasional speci-
mens of the larvse of V enericardium, of Anemia and possibly also
of Astarte. During more than ten years study of the oyster
larvse of the New Jersey Coast I have become familiar also with
the larvse of Mytilus recurvus (the southern oyster mussel),
/ 'cnus mcrcenaria, and Teredo navalis. In no instance have I
ever observed gas secretion in any of these forms nor have I
found pelagic dissoconchs. When the time for setting arrives
the mature larvse of all of the above species disappear from the
water within 24 to 36 hours.

Reproduction and dispersal of marine pelecypod molluscs occur
through the medium of pelagic veligers which are free-swimming
for periods ranging from a few days in such incubatory forms as
Ostrea edulis and Teredo bartschi, to approximately three weeks
in Mytilus edulis, Mya arenaria, and in probably most of the
marine bivalve molluscs which reproduce at temperatures below
20° C., Nelson, '28. Owing to the sessile or sedentary habits of
the adult molluscs, the activities of the larvse become of first
importance in the dissemination and preservation of the species.
Through the aid of the velum the larval bivalve, while unable to
make progress against a current, can control its vertical distri-
bution and thus secondarily may determine to a marked degree its
horizontal distribution by tides and currents, Nelson, '22.

The rate of growth and of development during larval life is
determined chiefly by the temperature. The long series of ob-
servations on the life history of the American oyster, Ostrea
•rinjinica Gmelin (J. Nelson, Stafford, Churchill, T. Nelson and
others), indicates that at a given temperature the duration of the
pelagic period is remarkably constant. With an average tern-

PELAGIC DISSOCONCIIS OF THE COMMON MUSSEL. 189

perature of 23-24° C. the period from spawning to the attach-
ment of the spat in New Jersey waters is 13 days. In Richmond
Bay, Canada, J. Nelson, '17, found that at temperatures approxi-
mating 20° C. the minimum time required for oyster larvae to
mature was 17 days. Stafford, '13, considers three weeks to be
the average time required to reach maturity in Canadian waters.

The close of the free-swimming period of pelecypod larva: is
determined apparently by internal developmental factors : when
the time for " setting " arrives the larva: must attach or die.
Since the veligers during their pelagic existence have been dis-
tributed widely by currents it follows that for those which through
chance happen to " fall upon good ground " there will be many
more which through this same chance will " fall by the wayside "
and be destroyed.

Observations of the oyster larva, T. Nelson, '22, '24, show that
approximately 24 hours prior to attachment the young bivalve
becomes positively stereotropic and that it may explore numerous
surfaces with the aid of the foot before it finally attaches. Such
behavior, while of the utmost importance in securing a favorable
spot for attachment, is without avail if no substrata suitable for
attachment are present. Little is known of the factors necessary
to provide a favorable bedding ground for such burrowing spe-
cies as Mya and J^ciius. Although attachment of young Mya
by the byssus to sea weeds or other objects may occur, as shown
by Ryder, '89, and by Kellogg, '99, it is pointed out by fielding,
'12, that survival of both Mya and Venus depends largely upon
the character of the mud and sand forming the surface layers
of the bottom. All who have studied the habits of larval bi-
valves agree that the vicissitudes of larval life and subsequent
attachment form one of the chief barriers to wide dispersal of
the species.1

1 A survey of our present knowledge of the habits and life histories of
both fresh-water and marine pelecypods shows that of all environmental
influences the presence of a suitable substratum is the most important
single factor limiting distribution. The following papers may be cited in
this connection: fresh water mussels, Coker et al., '21; Mya, Belding, '09;
Pcctcn, Belding, '10; Venus, Belding, '12; Ostrca cucitllata, O. angasi,
Roughley. '25; Mytilns, Card him, Sa.ridomus, Siliqua, Paphia, and other
genera of the Pacific Coast, Thompson, '13, and Weymouth, '20; Ostrc<i

190

TIM'RLOW C. NELSON.

Of the known genera of marine pelecypods, Alytilus cdnlis and
Teredo navalis alone are circumpolar in their distribution over
the shores of the northern hemisphere. General adaptability to
changing conditions and the power to resist adverse surround-
ings together with relatively low spawning temperatures, Nelson,
'28, have aided these two forms in attaining their present wide
distribution. Transportation through attachment to vessels or
to other floating wood has likewise aided in their dispersal, being
for Teredo the only means by which any great distance could
be covered. In the case of Mytilus edulis, however, the ability
to bridge the period of metamorphosis while remaining pelagic
must have been an important factor in securing the wide disper-
sal which this mollusc now enjoys; as well as a great aid in
bringing to a suitable place of attachment a fair proportion of
the larvae produced each season. The largest of the pelagic dis-
soconchs found in Frenchman Bay was fully a month old, dur-
ing which time it must have been transported over long distances
by the tide. If during this period it had come in contact at any
time with a solid object attachment could easily have been effected.

SUMMARY.

At the close of the larval or prodissoconch stage the young of
Mytilus edulis which fail to secure immediate attachment may
remain pelagic through the secretion of gas into the mantle
chamber.

Short distances may be covered through the aid of a holdfast
secreted on the surface film or through holding the foot or the
tentacles of the incurrent siphon in the surface film.

The ability to bridge over the critical period of metamorphosis
while remaining pelagic has been an important factor in securing
the present wide distribution of the black mussel.

Absence of a similar adaptation in the larvae of other bivalves
renders them still dependent largely upon chance in securing at
the close of the pelagic period a proper substratum for attach-
ment. This dependence upon the substratum is one of the chief
barriers to the wide dispersal of the species.

I'irginica, Moore, '97, Grave, '01, Stafford, '13, J. Nelson, '17, Churchill, '20,.
T. C. Nelson, '22.

PELAGIC DISSOCONCHS OF THE COMMON MUSS1.I-.

191

CITATIONS.
Belding, D. L.

'09 A Report upon the Mollusk Fisheries of Massachusetts. Boston,

243 pages.
Belding, D. L.
'10 A Report upon the Scallop Fishery of Massachusetts. Boston, 150

pages.

Belding, D. L.
'12 A Report upon the Quahog and Oyster Fisheries of Massachusetts.

Boston, 134 pages.
Churchill, E. P.

'20 The Oyster and Oyster Industry of the Atlantic and (iulf (Toasts.

Rept. U. S. C. F. Appendix 8-51 pages.
Coker, R. E., Shira, A. F., Clark, H. W., and Howard, A. D.

'21 Natural History and Propagation of Fresh-water Mussels. Bull.

U. S. B. F., 37: 77-i8i.
Grave, C.
'10 The Oyster Reefs of North Carolina. J. H. U. Circulars, No. 151.

1-9-
Jackson, R. T.

'88 The development of the Oyster with Remarks on Allied Genera.

Proc. Bost. Soc. Nat. Hist., 23 : 531-556.
Kellogg, J. L.

'99 Observation on the Life History of the Common Clam, Mya arc-

naria. Bull. U. S. F. C., 1899.
Moore, H. F.

'97 Oysters and Methods of Oyster Culture. Rpt. U. S. C. F., 1897,

263-340.
Nelson, J.

'17 Oyster Propagation in Prince Edwards Island. Contr. to Canad.
Biol. Supplement to 6th Ann. Rpt. Dept. Naval Service. Ottawa,
53-76.
Nelson, T. C.

'22 Aids to Successful Oyster Culture, i. Procuring the Scvd. Bull.

351, N. J. Expt. Sta., New Brunswick.

'24a The Attachment of Oyster Larvse. BIOL. BULL., 46: 143-151.
'24b Metamorphosis to the Dissoconch Stage without Attachment in
the Veligers of Ostrea and of Mytihts (Abstract.) Anat. Rec.,
29: 97.
'28 On the Distribution of Critical Temperatures for Spawning and for

Ciliary Activity in Bivalve Molluscs. Science, 67: 220-221.
Roughley, T. C.
'25 The Story of the Oyster. Australian Museum Magazine, 2, Nos.

5, 6, 7

Ryder, J. A.

'89 The Byssus of the Young of the Common Clam, Mya art'iutria L.
Am. Nat., Jan., 1889.

13

[92

THURLOW C. NELSON.

Stafford, J.

'12 On the Recognition of Bivalve Larvae in Plankton Collections.

Contr. to Canadian Biol., 1906-10, 221-242.
Thompson, W. F.
'13 Report on the Shellfish of British Columbia Report of B. C. Comni.

Fisheries, 1913.
Weymouth, F. W.

'20 The Edible Clams, Mussels and Scallops of California. Calif. Fish
and Game Comm. Fish Bulletin Xo. 4.

STUDIES ON THE SECONDARY SEXUAL CHARAC-
TERS OF CRAYFISHES.

VI. A FEMALE OF CAMBARUS IMMI'MS WITH OVI-
DUCTS ATTACHED TO OPENINGS OF
SPERM DUCTS.

C. L. TURNER.
ZOOLOGICAL LABORATORY, NORTHWESTERN UNIVERSITY.

The specimen to be described was taken from Root River near
Racine, Wisconsin, on July 10, 1924.

The presence of a normal annulus ventralis and of the usual
rudimentary appendages upon the first abdominal segments mark

FIG. I. Diagram illustrating external secondary sexual characters. S,
openings of oviducts at base of fifth walking leg. A.V., annulus ventralis.
R.A., rudimentary first abdominal appendages. O, position at base of third
walking leg of oviducal pore in normal female.

it as a female but there are no oviducal pores at the bases of the
third walking legs. On the other hand, a pair of openings ap-
pears at the bases of the fifth walking legs at the position ordi-

194

C. L. TURNER.

narily occupied by the openings of the sperm ducts. No other
male characters are found, however (Fig. i).

Upon dissection the specimen was found to have a well-devel-
oped ovary and oviducts which were attached to the ovary at the
usual point. In normal specimens the oviducts make their way
laterally and ventrally and open through pores located at the bases
of the third legs. In this case, however, the oviducts slope pos-
teriorly as well and finally attach themselves to the openings at
the bases of the fifth walking legs. A closer examination of the
pores themselves shows that they are essentially the openings of
sperm ducts in structure although they are considerably modi-
fied (Fig. 2).

FIG. 2. Diagram illustrating internal relations of ovary, oviduct and
walking legs. Position of oviduct in this specimen shown in solid black.
Position of oviducts in normal female shown by lines. Positions of walk-
ing legs indicated by numerals.

The normal oviducal pore is oval in outline and has slightly
raised and thickened rims, but the structure, as a whole, is not
raised conspicuously above the surface of the shell. The greatest
width of the oviducal pore in a specimen of this size is about one
mm. with the length a little more. The plane of each pore is
tilted a little toward the median line and a little toward the rear.

The normal openings of sperm ducts are much smaller and are
extended by means of membranous projections. They open toward
the median line and there is practically no deflection ventrally or
posteriorly.

SECONDARY SEXUAL CHARACTERS OF CRAYFISHES. 195

The openings of the specimen described here resemble the pores
of sperm ducts in their general position, in the direction in which
they open and in being projected somewhat by membranes. On
the other hand, they are much larger than the openings of sperm
ducts but smaller than normal oviducal pores and in shape re-
semble oviducal pores.

DISCUSSION.

Several questions of interest arise in connection with this case
but most of them can be treated only as speculations, (i) How
does it happen that the oviducal pores are absent in a female ani-
mal? (2) How does it happen that the openings of sperm ducts
are present in a female animal? (3) Is the duct to be interpreted
as an oviduct which has become attached at its distal end to the
opening of a sperm duct or as a vas deferens which has become
attached at its proximal end to an ovary? (4) To what extent
has there been a modification during its development of the pore
of the sperm duct by virtue of its attachment to the oviduct?
(5) To what extent does the oviduct possess the potentiality of
shaping a structure to which it is not ordinarily attached (the
opening of the sperm duct) in the direction of an organ to which
it is usually attached (the oviducal pore) ?

As regards the first and second questions, it is being repeatedly
shown in crayfishes that the contemporary occurrence in an ani-
mal of the ovary or spermary and a fixed set of secondary char-
acters is by no means necessary for normal functioning. Indeed
it has been shown that in some localities the occurrence of a char-
ter supposed to be fixed upon one sex occurs upon the other sex
considerably more than half of the time. The lack of one or both
oviducal pores in otherwise normal females has also been re-
corded. The absence of oviducal pores is not surprising, there-
fore, nor the presence of the openings of sperm ducts. The
presence of the openings of sperm ducts in the absence of ovi-
ducal pores does not mean that these are mutually exclusive struc-
tures, for oviducal pores are sometimes found upon males normal
with respect to the sperm duct openings as well as otherwise
normal.

The tube connecting the pore at the base of the fifth walking
legs and the ovary may be considered an oviduct for the follow-

10,6 C. L. TURNER.

ing reasons. First, it is straight and shows none of the coils of
the vas deferens. Second, it is thin walled and wide while the
vas deferens is rather narrow and dense in texture. Third, it
exhibits the same types and arrangement of tissues as a normal
oviduct.

In Cainbarus the only misplaced oviducal pores discovered have
occurred on the second or the fourth walking legs of females and
then only as supernumerary pores, the normal oviducal pores
being present on the third walking leg as usual. It is assumed,
therefore, that the pores of the fifth walking legs are the open-
ings of sperm ducts and not of oviducts. The extent to which
the pores depart from the features of the normal sperm duct is
taken to represent the extent of the influence of the attachment of
the pores to oviduct instead of sperm duct.

There is no evidence to show that the influence of the oviduct
is required in embryology for the proper shaping of the oviducal
pore. Rather to the contrary the oviducal pores are sometimes
present and perfectly developed in the absence of oviducts. At
the same time the resemblance of the pores of this specimen to
oviducal pores must have been due to the influence of the oviduct
to which they were attached. The case is roughly parallel to the
results of the experiment in which an optic cup in a vertebrate
embryo was transplanted under ectoderm in another part of the
body and under the influence of the optic cup the ectoderm de-
veloped a lens although normally it would not have done so.

The condition described here is not to be confused with the
normal condition found in some South American species belong-
ing to the genus Parastacus. In these there are regularly both
oviducts and sperm ducts but only one set is functional, that one
being appropriate to the sex in which it occurs. The case de-
scribed in this paper cannot be considered as a parallel, for here
only one set of tubes is present and only one set of external pores.

STUDIES ON THE SECONDARY SFXTAL CHARAC-
TERS OF CRAYFISHES.

VII. REGENERATION OF ABERRANT SKCONDAKY

SEXUAL CHARACTERS.

C. L. TURNER.
ZOOLOGICAL LABORATORY, NORTHWESTERN UXIVKRSITY.

It is rather generally conceded that intersexuality in insects has
as a background either genetic disturbances or metabolic inter-
ferences due to parasitism. The evidence for the causes of inter-
sexuality in the Crustacea, however, is by no means as distinct,
due to the fact that it has not been possible to separate the ques-
tion of the development of secondary sex characters from that
of their control by hormones located in the gonads. It has been
rather concisely demonstrated in the insects that secondary sex
characters are independent of spermatic or ovarian hormones,
whereas some of the evidence obtained in the Crustacea seems to
indicate the possibility of such hormone control.

Among some six or seven causes which might be able to ac-
count for intersexuality among the decapods there is the repeated
suggestion that some ambiguous or accidental event in embryol-
ogy has been responsible. When in one locality eighty-six per
cent, of all the females of Cambarus virilis and in another lo-
cality thirty-seven per cent. of*Cambarus propinquus possess well-
defined male characters, it would seem that no accidental event
in embryology could account for this unusual occurrence. Rather,
it would seem necessary to seek for some orderly and fixed in-
fluence. It occurred to the writer that the tendency to form
aberrant structures might be tested for its duration beyond the
embryonic stage by detaching the appendages bearing these struc-
tures and determining whether there would then be developed
normal or aberrant structures. The accumulation of this body of
data and the interpretation of it has been carried out with this
inquiry in mind.

197

198

C. L. TURNER.

RKGKNKKATION OF NORMAL SECONDARY SEX CHARACTERS.

When any appendage is lost in the crayfish, unless the crayfish
is too old, a new appendage will grow in place of the old one and
if the crayfish is young when the accident occurs the regenerated
appendage will come to have almost completely the size and char-
acteristics which the original appendage would have had. In an
i >ld specimen, however, there will not be sufficient time for the
complete regeneration of the appendage before the crayfish dies.

Some of the external secondary structures which have to do
with sex are either modified appendages or structures located
upon the appendages. Such are the first and the second abdom-
inal appendages which are modified for copulation in the male
but are rudimentary in the female, the hooks located upon the
third walking leg of the male, and the openings of the oviducts
and sperm ducts upon the third and the fifth walking legs re-
spectively. When one of these appendages is broken off so as
to include one of the modified structures, the appendage will be-
gin to regenerate, beginning with the first moult. At first it is
juvenile in character and unmodified but eventually it becomes
completely differentiated and contains the modified structure. The
regenerated secondary sex characters never completely resemble
the normal secondary sex characters. In the case of the hooks
which are found upon the third walking legs, the regenerated ones
are blunt and flatter than the original ones, but occurring as they do
in a definite position they are easily recognizable.

OBSERVATIONS ON THE REGENERATION OF ABERRANT SECONDARY

SEX CHARACTERS.

Not more than thirty cases have been recorded in Cambarus of
male-like modifications of the abdominal appendages in females
and the possibility of finding cases in which the aberrant appen-
dages have been injured and regenerated would be very remote.
The occurrence of female structures upon males is also too rare to
give any expectation of finding regenerated aberrant structures.
The copulatory hooks upon the third legs have been selected,
therefore, as the most likely structures for observation because
their occurrence upon females furnishes the most common aberra-
tion.

SECONDARY SEXUAL CHARACTERS HE CRAVEISHES. 199

In Lake Delavan (Wisconsin) eighty-six per cent, of the fe-
males of Cambarus virilis carry the copulatory hooks like those
which occur upon the third legs of the males. It is as fully de-
veloped in the female as it is in the normal male and developes
in ontogeny at the same stage. In the Menomonee River (Wis-
consin) thirty-seven per cent, of the females of Cambarus propin-
quus also bear these hooks. These two localities were chosen as
most likely to produce the desired specimens.

In the course of three summers collecting after several thou-
sands of specimens had been examined, seven males and three
females of Cambarus virilis from Lake Delavan were found, each
of which had lost and regenerated one of the third legs. In each
specimen the third leg had regenerated to a point where it was
possible to determine whether or not a normal copulatory hook was
being formed. Males, ranging from 56 to 88 mm. in length, had
all somewhat imperfectly developed new hooks upon the regener-
ated legs (Fig. i). These normal males with regenerated hooks

FIG. i. Diagram illustrating three basal segments of third walking legs
and copulatory hooks in male crayfish which has lost and regenerated a
part of left leg containing copulatory hook. Regenerated copulatory hook
is short and blunt.

were used as controls with which to compare the females which
had likewise lost their third walking legs and had regenerated
the third legs together with the hooks upon them. Of the fe-
males, two had well developed hooks upon the uninjured third
legs and upon the regenerated third legs the hooks had reformed
as in the males (Fig. 2). In the third female the uninjured third
leg carried no hook and upon the regenerated third leg no hook
had formed.

In the Menomonee River, two specimens of Cambarus propin-
quus were found which could be used for this study. The first
was a male in which the uninjured third leg was entirely normal
and carried the usual hooks. The left third leg had been lost at

2QO C. L. TURNER.

an early stage and had regenerated. Upon it was the blunt type
of hook usually found upon the regenerated third leg. The second
was a female which also carried a hook upon the uninjured third

FIG. 2. Diagram illustrating three basal segments of third walking legs
in an aberrant female which had lost and regenerated a part of the third
walking leg. The copulatory hooks are aberrant and the left one has been
regenerated.

leg and had also developed a blunt hook upon the regenerated
third leg.

EXPERIMENTS.

Young specimens of Cambarus virilis, about thirty-six mm. in
length, were selected for experiment. They were taken from
Lake Delavan on July 17. Eighty-two males in which the copu-
latory hooks on the third legs were visible were divided into two
equal lots. One lot was used as a control and in the other, one
of the third legs was detached in each specimen. Fifty-five fe-
males in which a copulatory hook was visible upon the third leg"
were divided into two lots and similarly operated upon or used as
controls. All were kept in the laboratory under conditions as
nearly natural as possible for ten months and about one fourth of
the specimens survived. Six of these were males in which one
third leg had been removed and upon the regenerated leg there
had developed the copulatory hook. Seven were females from
which one third leg had been removed. All such females had re-
generated the third legs together with the copulatory hooks upon
them. The hooks compared favorably with those regenerated by
the males of approximately the same age.

CONCLUSIONS.

In all the cases cited above, whether observed in nature or ex-
perimentally produced, females bearing aberrant male hooks upon
their third walking legs regenerated hooks whenever an injured
leg had sufficiently developed. Some were one year of age and

SECONDARY SEXUAL CHARACTERS OF CRAYFISHES. 2OI

others were older. It is reasonable to state, therefore, that all the
evidence, though meager, tends to show that whatever influence
was present in the first place to produce this aberrant secondary
sex character was also present and operative in the animal later
during any regeneration period. The permanency of this influ-
ence during the life of an animal would seem to take it out of
the classification of accidental or temporary embryonic agencies.
When it is considered together with the fact that the same aber-
rancy is repeated in this crayfish population (observed for six
years) it seems logical to give the influence a genetic status and
to postulate that there has been a definite change within the germ
cell.

REGENERATION OF LUMBRICULUS IN VARIOUS

RINGER FLUIDS.

LEONARD P. SAYLES,
TUFTS COLLEGE.

INTRODUCTION.

In the course of work with Ringer solution on Planaria, J. W.
Wilson ('26) has noticed that wound closure may be more or
less completely suppressed in an isotonic solution. With the in-
tention of making use of this peculiarity if it held true for Lum-
briculus, I have experimented with various strengths of Ringer
solution on this form. Finding various modifications of the usual
method of wound closure and regeneration, I have made studies
on the effects of various strengths of Ringer fluid on regeneration
in this form. It is my purpose to report these at this time.

As a preliminary, the approximate osmotic pressure of the body
fluids of this worm were determined in order that it might be
possible to know something concerning the relative strengths of
the internal fluids and the external solutions used. Adolph ('25,
p. 332) concludes that we can " probably regard the maximum
survival concentration for freshwater animals as a measure of
the osmotic pressure of their body fluids." The maximum sur-
vival concentration of Ringer solution for Lumbriculus at the end
of 24 hours (the arbitrary time adopted by Adolph, '25, for
Phagocata) was found to be O.I47M. When corrected for ioni-
zation this gives a figure of O.257M as compared with O.2I5M
for Lumbricus, as assumed by Adolph and Adolph ('25). Ap-
parently a Ringer solution of between O.14M and O.I5M concen-
tration is approximately isosmotic with the body fluids of Lum-
briculus.

WOUND CLOSURE IN VARIOUS CONCENTRATIONS OF RINGER

SOLUTION.

The usual method of wound closure in microdrilous annelids
has been quite completely described (von Wagner, 'oo and '06,

202

REGENERATION OF LUMBRICULUS. 2O3

Iwanow, '03 and Krecker, '10, among others), both from the
gross and from the microscopical points of view. Briefly it
takes place in somewhat the following manner: Immediately after
a worm is cut the muscles of the body wall begin to contract.
They continue to draw the cut edges of the hypodermis together
until only a small opening is left. This aperture is then clogged
by a plug of cells, many of which have been torn free by the
cut. The wound is thus completely closed and the body fluids
once more prevented from freely mixing with the solution in
which the worm is cut. At the same time the intestine also con-
tracts somewhat, closing over at the end and withdrawing slightly
from the contracting body wall. This preliminary wound closure
is completed in from 10 to 15 minutes after the cut is made.

The behavior in an isotonic solution is in distinct contrast to
this usual behavior. The following description of what occurred
in one series of observations might well apply to many cases which
have been followed for considerable periods.

An individual is cut in a O.I4M Ringer solution, without anaes-
thetization, at 2 154 P.M. Both pieces move about quite rapidly
at first but in 3 or 4 minutes they have quieted down to ordinary
" crawling " movements, such as are commonly found in unin-
jured individuals. During this time there is a loss of some blood
and a number of cells due to the fact that there is no semblance
of contraction of the body wall. At 3:03 there is evidence of a
protrusion of the gut beyond the end of the body wall. This
protrusion of the gut gradually becomes more pronounced until
a portion, perhaps a segment in length, extends beyond the plane
of the cut at 3:10. At this time the cut end of the gut begins to
show evidence of a rolling back upon itself. This process con-
tinues until at the end of an hour there is a well formed bulb of
everted gut present at the cut end (Fig. i). During this time
there has been a gradual contraction of the body wall until it has
reduced the diameter of the opening resulting from the cut to
about two thirds of its original size. The gut in the course of its
eversion has now come in contact with the body wall so that
there is very little opportunity for interchange of materials be-
tween the body fluids and the external solution. This is the end
of wound closure from the macroscopical point of view.

204

LEONARD P. SAYLES.

As a result of this process of " wound closure " there is present
at the cut surface, at the end of i or 2 hours, a bulb of everted
gut usually of almost as great diameter as that of the body. The
ciliated portion of the gut cells are thus exposed to the outside
solution, in which they continue to beat with apparently the usual
rapidity.

s

^

FIGS, i AND 2. Bulbs of everted gut at the posterior end of pieces re-
generating in o.ogiM Ringer solution. Fig. I after 12 hours; Fig. 2 after
6 days.

In the case of slightly hypotonic solutions the resulting bulb of
gut is usually smaller, in some instances not more than one fourth
the diameter of the worm. The presence of these smaller bulbs
is due, in part at least, to the fact that in these cases the gut does
not protrude as far at first, so that when the eversion occurs
there is only a short piece involved. A contributing factor is the
slight pulling together of the body wall ; this probably tends to
hinder the protrusion of the gut and in addition reduces some-
what the diameter of the portion which does pass through the
aperture.

POSTERIOR REGENERATION IN VARIOUS RINGER SOLUTIONS.

If we assume that the maximum survival concentration at the
end of a twenty-four-hour period is isotonic with the body fluids
of an animal, it is difficult to keep individuals in an isotonic fluid
for a very long period. One experiment may be cited to show
what occurs when individuals are left for a long period in O.I4/M
Ringer, the solution being changed each day at the time of ob-
servation. Fifteen worms were put into such a solution and on
the following day all were alive, with no ill effects apparent. On
the second day, 2 had died and 3 others were clearly not far short
of death. On the third day, 4 more were dead and 4 others were
beginning to disintegrate at the posterior end. On the fifth day,
only 5 were alive and I of these was beginning to disintegrate

REGENERATION OF LUMBRICULUS.

205

<it the posterior end. Because of this high mortality in Ringer
stronger than o.i_|.M, it has been found advisable to observe the
regenerative processes in the latter concentration in which the
majority of the regenerating head ends will live for a considerable
period. Even in this strength many tails die in the course of }
or 4 days.

In the case of heads, regenerating new tails, which have been
cut and left in O.I4M Ringer, many individuals have bulbs very
similar to those found after 2 or 3 hours, even at the end of a
week. For example, of 12 heads put into such a solution, 9 had
bulbs of everted gut, i had a short and very slender bud of re-
generating tissue and 2 had died at the end of a week. This pro-
portion of individuals with bulbs of gut holds true within fairly
narrow limits in all experiments using O.I4M Ringer.

In light of these results obtained with a solution almost iso-
tonic with the body fluids, it is of interest to determine the effect
of slightly greater dilutions of Ringer solution. The number of
bulbs of gut present on individuals '" regenerating " in various
strengths of Ringer solution for / days is given in Table I.

TABLE I.

NUMBER OF INDIVIDUALS WITH BULBS OF GUT ix DIFFERENT CONCENTRA-
TIONS OF RINGER SOLUTION.

10 worms were cut in each concentration.
Figures in parenthesis indicate number which also show new tissue.

Molar cone,
of Ringer.

Days of Regeneration.

i

2

3

4

5

6

7

O.OII ....

o
o
o

3d)

2(2)

3d)

8(2)

7(3)
8d)
10(0)
9(o)
10(0)

O
O
O

id)

o

3(2)

8(4)
7(4)
8(5)
1 0(0)

9(o)
10(0)

o
o

0

id)

o

1(1)

8(7)
7(7)
8(8)

10(1)
9(o)
10(0)

o
o
o

id)

o

id)

8(8)

7(7)
8(8)

10(4)
9(3)
1 0(0)

o
o
o

id)

0

id)

8(8)

7(7)
8(8)
10(10)

9(5)
10(0)

0

o
o

id)

o

id)

8(8)

7(7)
8(8)
10(10)
9(6)
9(0)1

0

o
o

Id)

o

id)

8(8)

7(7)
8(8)
10(10)
9(6)
9(o)

0.023

0.034

o.oj^

O.O^l

o-.o68

O.O7O

0.091

O.IO2

O. I I }

0.125

0.136

1 I dead on the sixth day.

206 LEONARD P. SAYLES.

After one day of regeneration all the bulbs of gut present are
practically in line with the central longitudinal axis of the body
(Fig. i). The anal opening (not a true anal opening, for it is
completely surrounded by gut which forms part of the outside
wall in these cases) is, therefore, terminal. On the second day,
however, those pieces in O.IO2M and weaker solutions have bulbs
tipped at an angle so that the opening is not exactly terminal
(Fig. 2). This change in the position of the bulbs does not occur
as soon in o.ii3M and O.I25M; it does appear, however, usually
in all individuals by the fifth or sixth day. But few individuals
show any evidence of this tipping in O.I4M. In this respect the
individuals cited in Table I. under O.I36M were exceptional,
since none happened to show evidence of new tissue. The under-
lying cause of this slight shifting of the position of the bulb is
apparently the growth of tissue on one side of the wound region
and not on the other. The region in which growth occurs is
found, on microscopical examination, to be always ventral.

There is, then, from the macroscopical point of view, a graded
effect on regeneration in Lumbriculus produced by a series of
various concentrations of Ringer solution. In an approximately
isotonic solution the majority of individuals show little evidence
of the production of new tissue ; except in rare cases those indi-
viduals which do produce sufficient new tissue to make it apparent
do so only on the ventral side. In weaker solutions new tissue
is produced in practicaly all cases. In dilutions but slightly hypo-
tonic (o.oSM to O.I3M) this process is confined to the ventral
region in most instances. In most cases where new tissue is
produced throughout the wound region, no bulb of gut is present.
Occasionally, however, a very small bulb may be found at the
end of a regenerating bud which is i or 2 mm. long. In solu-
tions below o.oSM, on the other hand, practically every individ-
ual produces new tissue throughout the wound region and bulbs
of gut are rare.

SUMMARY AND DISCUSSION.

In Lumbriculus, therefore, it would appear that an important
factor in arousing the cells to regenerative activity is the dilution
of the body fluids with water. When an individual is cut or

REGENERATION OF LUMBRICULUS. 2(>7

broken into two or more pieces, there are open wounds through
which the body fluids may flow out and water may enter. This
entrance of water is perhaps increased by the " writhing " and
" crawling " movements of pieces immediately after they are cut.
These movements would also extend the region of dilution a
slight distance from the wound. The author (Sayles, '27) has
reported that for mesoderm and intestine regenerative activity
involves between 10 and 12 segments from the wound region.
In the case of the hypodermis, which is commonly bathed on the
outside by a hypotonic solution, activation occurs only in a re-
stricted region near the wound. This limited activity of the hypo-
dermis is due perhaps to the fact that the diluted fluids come in
contract with its inner surface underneath the muscles which are
pulled away near the wound. Farther away than that, however,
the diluted body fluids probably never reach the hypodermal cells
through the relatively thick layer of muscle cells.

While the hypotonicity of the water to the body fluid seems to
be an activating factor in Lumbriculus, too general conclusions
cannot be drawn from such results. In other fresh water forms
this factor may be found to be of importance but in marine ani-
mals regeneration can certainly take place in a medium which is
presumably of the same osmotic pressure as their body fluids.
In these animals, however, greater regenerative activity may occur
in slightly diluted rather than in normal sea-water. This has been
reported by Goldfarb ('07, p. 353) in the hydroid, Eudcndrimn,
in which " the maximum number of polyps regenerated does not
occur in normal sea-water but in solutions diluted with about 20
per cent, of tap- water."

REFERENCES.
Adolph, E. F.

'25 Some Physiological Distinctions Between Freshwater and Marine

Organisms. BIOL. BULL., 48.
Adolph, E. F., and P. E.

'25 The Regulation of Body Volume in Fresh-water Organisms. Jour.

Exp. Zool., p. 43.
Goldfarb, A. J.

'07 Factors in the Regeneration of a Compound Hydroid, Eudendrium

Ramosnin. Jour. Exp. Zool., 4.
Iwanow, P.

'03 Die Regeneration von Rumpf- und Kopfsegmenten bei Lumbriculus

variegatus. Zeit. f. wiss. Zool., 75.
14

2O8 LEONARD P. SAYLES.

Krecker, F. H.

'10 Some Phenomena of Regeneration in Limnodrilus and Related Forms.

Zeit. f. wiss. Zool., 95.
Sayles, L. P.

'27 Origin of the Mesoderm and Behavior of the Nucleolus in Regener-
ation in Lumbriculus. BIOL. BULL., 52.
von Wagner, F.
'00 Beitrage zur Kenntnis der Reparationsprozesse bei Lumbriculus

variegatus. I. Teil. Zool. Jahrb. Anat, 13.
von Wagner, F.

'06 Beitrage zur Kenntnis der Regenerationsprozesse bei Lumbriculus

variegatus. II. Teil. Zool. Jahrb. Anat., 22.
Wilson, J. W.

'26 Regeneration of Planaria maculata in Isotonic Ringer's Fluid. Anat.
Rec., 34.

VARIATION OF HOOKS ON THE HIND WING OF THE
HONEY BEE (APIS MELLIFERA L.).1

W. W. ALPATOV,
RESEARCH FELLOW OF THE INTERNATIONAL EDUCATION BOARD.

The present paper represents partial results of a series of in-
vestigations carried on by 4he author since 1924 in the field of
biometry of the honey bee (See Alpatov, i-io). The material
for this work has been collected partly during the author's work
in the Zoological Museum of the Moscow University, and partly
(material on American bees) during the summer of 1927 in the
apicultural laboratory of the Agricultural College, Cornell Uni-
versity. The definite calculations and the preparation of the
manuscript have been completed during the winter 1927-28 in
the Institute for Biological Research. The author is glad to ex-
press his deep indebtedness to Professor Koshewnikov (Moscow),
Professor E. F. Phillips (Cornell University), and Professor Ray-
mond Pearl for their interest and help. The author also appreci-
ated very much the help given by beekeepers of Russia and U. S.A.
in collecting bees from different parts of both countries. Profeessor
E. F. Phillips has also been so kind as to show the author the
manuscript of his unpublished paper.

In spite of the fact that the beekeeper's literature contains a
tremendous number of observations on differences in bee races,
a scientific basis of racial studies in bees is practically absent,
especially in comparison with racial and genetical studies on other
domestic animals. The cause of this lies chiefly in certain pe-
culiarities which characterize the honey bee. Firstly, the bees
being fecundated in air do not allow us to control the mating and
therefore to conduct exact genetical experiments. Secondly, it
is more difficult to study the characteristics of such small animals
as the honey bee than those of domestic mammals and birds.
Only quite recent progress in artificial insemination of the queen

1 From the Institute for Biological Research, Johns Hopkins University.

209

2IQ W. W. ALPATOV.

(Watson, '20) gives us the hope of being able to overcome the
first of these obstacles.

The present author believes that a careful investigation of varia-
tion should be made before any attempts to study the heredity of
the honey bee. In this direction the present paper brings evi-
dence of the importance of a certain characteristic, namely, the
number of the hooks, characterizing different biological groups in
the limits of the species Apis incllifcra L. Thanks to the mod-
ern investigations mostly of Russian scientists (Koschewnikov,
Chochlov, Michailov, Alpatov, Alpatov and Tjunin) two very
important facts in the field of variation of the honey bee have
been discovered.

The first of them is the geographical regularity in the varia-
tion. The changes in tongue length of the worker bee is the
most striking fact in the geographical variation of the honey bee.
We are able to say that for countries with native bee population
each locality is characterized by a definite tongue-length of bees
inhabiting the given locality. Moreover, the change from one
locality to another is regular and gradual. A general rule can
be established ; the more to the south, the longer the tongue length.
Other body characteristics also show some regularities in geo-
graphical variation (Alpatov, 8). The author of the present
paper believes that it is perfectly justifiable to compare the dif-
ferent r> races " of the honey bee with geographical subspecies
of wild animals.

Family variation is the second important fact which every in-
vestigator in the field of variability of social insects has to bear
in mind. It was shown by several investigators, Thomson, Bell
and Pearson (23, 24), Warren (25), Arnddi (12), Z. G.
Palenitschko (20), Alpatov and Tjunin (i) and Alpatov (3, 4,
6, 9, 10), that the variation of single families is smaller than
the variation of the whole population. Therefore, in establishing
racial characteristics we have to collect our material from as many
families as possible.

Turning our attention to the literature devoted to the special
question of hook variation we find only a small number of papers
dealing with that particular subject. Professor Koschewnikov
(19) was the first who introduced the number of hooks in the

HOOKS ON HIND WING OF IIONKY BEE. 211

taxonomy of the honey bee. E. B. Casteel and K. F. Phillips
(14) without using biometrical methods, tried to solve the prob-
lem of comparative variability of drones and worker bees. Kel-
log's (i/) data have also a very restricted value from the point
of view of modern biometry. Wright, Lee and Pearson (27)
then attempted, by recalculating Casteel's and Phillips' data, to
draw some more definite conclusions. The most extensive work
has been done by Bachmetjew (13). The conclusions of this
author found just criticism in Koschewnokov's (18) and Ray-

20

10

/3 H 15 16 17 /g 19 20 21 22 23 24 25 26 27 28 29

FIG. i. Hook variation of 3 colonies of the Bulgarian drones.

mond Pearl's (21) papers and need not be mentioned further.
Fortunately, Bachmetjew published in his paper all his numerous
countings (about 2,500 bees were examined). His data have
been worked out biometrically by the author of the present paper,
and published in Russian (4). Professor Phillips did the same
in the paper which is now in press. In this paper Professor
Phillips turns his attention mostly to the individual variation in

212

W. W. ALPATOV.

"t OO

c/5

0

II M ||

HJ

M

>o - ro M

Crf

O N t-.

<!

6 6

s

M M

^

ro IH

M in

z

-H S-tt o

o\ o

Pi

W

OO • ON IH

O IH rt

b

06 6

b

Q

ro H
w T}-

1

00

^S^ro
ro • r^» ^

b

IS) M (V^

Z

M 00

U

in ro

fr"

_LJ ^^ IJ

Tl O Tl O

C/l

r—

o o

U

ro • N M

M M UO

BULGARIAN

V.)

a

_o
"o
U
D

O

M ^-

-H g-H g

vO « O

»*"

0 '«

"8

O vO
IH \n

u S

IH
OJ
.0

-ii lr> -U ^
Tl in Tl r~

1 — 1
W

ij

" f*

£ o

O ^
iJ CQ

s

3
Z

O • cs M
M cs o

CN M

U S

ro O

H

o

2 &

a

it

" "1"

-U M JJ f
Tl O ~n O

O M

>0 ^ 0 M

8

t^ N oo

IN 00
<N

z

ro M

0

M T}~

M

C

-tiX-ti H

ci

o

OO • M ^

IH M ro

CN t-^.

Q

H

M ?

<

"H ro "H ro

M

M

O\ °^ 00 °°

U

J

. .

M 00

r.

CN

^^

M ro

O

-H "" -W

H

M

vn"

g

\O • ro ^

>

M 1^.

K^

S

V

^

rt

<;

o

fe

"o

o

IH

U

.a
E

HOOKS ON HIND WING OF HONEY BEE.

213

the honey bee, and on that account his conclusions do not parallel
those of the present paper.

The number of bees examined by the author of the present
paper exceed three thousand — a number which has never been
reached by previous investigators.

Table I. shows us the variation of Bulgarian drones belonging
to different colonies. It is evident that the difference between
the averages are in many cases more than five times larger than
their probable errors. Fig. I represents 3 variation curves of the
4th, Qth and loth colonies, proving the conclusion just made.
Table II. shows the same for worker bees. It can be seen that in

TABLE II.

CONSTANTS FOR WORKERS OF 5 COLONIES OF BULGARIAN BEES
(DATA FROM BACHMETJEW).

Number of the Colony.

i

2

3

4

5

M.

21.60 ± .10
1.487
6.88 ± .33
99

21.01 ± .OQ
1.367
6.51 ±-3I

99

21.76 ± .10
1.566
7.20 ± .33
no

21.91 dh .12
1.867
8.52 ± .41
IOO

21. II ± ,IO

1.467

6-95 ± -33
98

<7

COT

Number of cases

the last case the differences are not so pronounced as in the case
of the drones. Fig. 2 compared with Fig. i gives the same im-
pression. If we consider the coefficients of variation we find that
for the drones they vary in the limits 7.52-13.02 per cent.; for
the worker bees 6.88-8.52 per cent. It is obvious that the aver-
age variation of worker bees of the colony is smaller than the
variation of the drones.

Are we justified in concluding that the drones are more vari-
able than the worker bees ? There is a certain weak point in such
conclusions. We are not convinced that the method of collecting
the material was safe enough to provide us with bees really rep-
resenting the progeny of single queens — i.e., members of one
family. The proper way to get such a material would be to put
a sealed brood in an incubator and collect the emerging bees. In
collecting bees directly from the hive there is a danger of pick-
ing up bees belonging to the population of a neighbour hive. It

214

W. W. ALPATOV.

is known that the bees and especially the drones sometimes pene-
trate into neighboring hives. The only way to avoid this diffi-

30

20

10

17 /Z 19 20 21 22 23 24 25 26
FIG. 2. Hook variation of 3 colonies of the Bulgarian worker bees.

culty is to calculate the coefficient of variation for the whole mass
of bees. The results of such processes are shown in Table III.
It can be seen that the variation of the worker bees belonging to

*

TABLE III.

BULGARIAN BEES IN DIFFERENT GROUPINGS.

Queens

All Drones

Workers from

from 10

Drones from

Drones from

i-io Colonies.

5 Colonies.

localities

1-5 Colonies.

5-10 Colonies.

in Bulgaria.

21.39 ± -05

21.49 ± -05

18.46 ± .11

21.82 ± .07

20.98 ± .07

2.352

1.586

1.892

2.157

2.438

11.00 ± -I?

7.38 =fc .16

10.25 ± .42

9.89 ± .42

11.62 ± .25

997

507

139

490

507

HOOKS OX HIND WING OF HONEY P.M..

215

the 5 colonies is lower than the variation of the two groups of
drones each representing members of 5 colonies. The coeffi-
cients of variation calculated from our original material on worker
bees are also in general lower than 8 per cent. Even for 1000
worker bees from Middle Russia taken from 106 colonies the
coefficient of variation is only 8.539 — -I29> as can be seen from
Table X. We believe that the present material permits us the
definite conclusion of a larger variability of drones in respect to
the number of hooks.

Table III. contains also data on variation of hooks in queens.
Firstly, it is evident that the average number of hooks is far lower
than in the drones and workers, which have practically the same
averages. This conclusion is given here in statistical form for
the first time, although G. A. Koschevnikov has already given
a few analogous data. In respect of the coefficient of variation
the queens are nearer to the drones than to the workers. A very
incomplete material collected in Table IV. shows that Middle Rus-
sian, German and American black and yellow queens have also a
much lower average number of hooks than the worker bees of
the corresponding races.

TABLE IV.

NUMBER OF HOOKS OF DRONES AND QUEENS FROM DIFFERENT LOCALITIES.

Drones.

Queens

Mos-
cow.

Kaluga
(M. Russia).

N. Wodolaga
(S. Russia).

Black-
Ontario.

Italians.

Moscow and
Darmstadt.

M . . .

20 72

2O 22 ± 2O

20 83 ± 26

16 2?

1 8 oo

18.67 ± .22

c%

0.76 ± .60

12 oo ± 87

9.06 ± .83

No. of cases

2 ^

AS

48

8

10

27

It is interesting to ,note that among the bees the relations of
castes in respect of variation differ from those found in other
social insects. It was shown (Palenitschenko, 20) that among
wasps, termites and ants, the workers are more variable than the
sexual forms — males and females. The worker caste among
bees is therefore an exceptionally constant and standardized group
of individuals.

Already in an earlier paper (4) some evidence has been brought

2l6

W. W. ALPATOV.

together to show that the bees of southern localities have a greater
average number of hooks than the northern ones. In order to
check that statement on a more solid basis, a special material has
been collected from different parts of European Russia and the
Caucasus. The map in Fig. 3 shows the localities which supplied a

FIG. 3. Map of European Russia and Caucasus. The figures correspond
to localities where the material has been collected.

corresponding material. The plain of European Russia is pop-
ulated by the black variety of Apis mellifera L — A. iiiellifcra.
mcHifcra L. An introduction of foreign blood, mostly of Italian
queens, was according to certain statistical studies a compara-

HOOKS ON HIND WING OF HONEY BEE.

217

tively rare phenomenon and could probably not produce any
significant influence on the whole mass of the bee population of
Russia (the number of hives in Russia according to certain esti-
mations runs over 5,000,000). Tables V. and VI show the fre-

TABLE V.

FREQUENCY DISTRIBUTIONS AND CONSTANTS OF THE NUMBER OF HOOKS OF

BEES FROM MIDDLE RUSSIA.

No. of Hooks.

Localities.

9

10

II

12

13

14

15

1 6

17-18

2O

19

2oa

21

22

20

i

2

i
6
3

20
21
24
13

8
I

4
14
19

22

25
14
I
I

I

I

4

12

14
13

8

2
2
2

I

3
9
8
6

13

8

5
i

i

I

0

I
6

13
16

13
II
I

2
2
8
IO
12

18

4
4

i

I

5
II
18
13
4
7
i
i

I
I
4
9
16

12

IO

5

2

2

3

13
8

25

22

18

4

2

3

I
O

3

0

7
13

10
12

3
I

4

4

IO

ii
5
4

2

i

2
10
10

17

18
17
13
9

2

I

I
2

4
ii

21
26
22
9

3

3

9

10

8
9

10

3

28

27

26

2"C ,

24

2"? . .

22

21

2O

10

18

17 .

16

No. of cases

IOO

IOO

59

55

62

61

61

60

IOO

49

40

99

IOO

52

No. of colonies

vt

M

O

o

o

a>

<^

c>

O

M

co

0

4^

M

O

*o

o

M

to

p

Co
O

fr

M

\0

>b

VD

H-

tO

p

La

O

If

K>
O
VI
Ca

If

to

o
*b

K>

H-

M

\0

i>

~J

H-

to

p

Cn

ON

H-

to
p

i-

co

H-

to
O

t!a

cc

H-

M

to

to

M
i

co

H-

M

O

to

O
to
CO

H-

b
o

to

b

4^

H-

M

La

to

M

to
vi

If

M

to

to

O

<b

CO

If

M

o

Co

o

0

00

to

La

4^

4-

C%

oo
>b

-4
If

•^1

to
VI

If

oo

CN
*.

ft-

'O

A
La

If

ON

vj
vt

H-

oo
On

M

H-

^j
vi
<I

H-

oo

b

4-

H-

co

O
*.

H-

VI

\b
^j

If

~j

bv

4^

ff

\0

|H
M

H-

00

M

4^

H-

cc
b

CO

H-

4^

w

00
La

La

4-

O

l-t

4^

M

C^l

to

4-
-J

La

O

4^

CO

La
«•

La

CO

4^
*.

CO

VO

On

CO

quency distribution for different localities, number of colonies
and corresponding biometrical characteristics. Table X. gives
summarized frequencies for the bees distributed in Middle Rus-
sia at the level of 55° of N. latitude and in South Russia, 50° N.
latitude. The difference between the averages is 8.85 times
larger than the probable error. We may conclude, therefore, with
.a high degree of certainty, that there is an increase in the average

218

W. W. ALPATOV.

TABLE VI.

•

FREQUENCY DISTRIBUTION AND CONSTANTS OF THE NUMBER OF HOOKS OF

BEES FROM SOUTH RUSSIA.

Xo. of Hooks.

Localities.

i

2-3

4

5

6

7

8

26

i
3
4
10

i?

IO
10

5
3
I

I
7

12

18
19

22
15
3

3

i
o

2

6
II
8
6
9
4
i

i

2

5

7

10

ii

12

5
6

!•

i

2

4
6

12

24
21
15

7

3

8

4
13
8
8

4
i

i

2

7

10
20

14

4
3

2?

24

2^

22

21

°O

10

18

17

No of cases

64

IOO

48

60

92

50

60

No. of colonies

o

W

00

h-»
o

O

o

Cn

o*

M

M

Ul
to.

H-

M

o

to

0

*b
-u

H-

W
(0

so

9

bo

00

H-

00

to

M

M
W

H-

M

00

(0

p
•Li
tj\

H-

t— i
>o

to

M

c>

OO

H-

M

oo

to

O

•b

00

H-

M

W

C%

00
Cn

*O

If

Cn

00

'M

<5

H-

4^
O

00
MO

oo

H-

<>

M

vp

'~r\

OJ

H-

c/i

\o

00
to

M
ff

ji

M

oo
<>

U)

H-

Cn

00

o

j>
o

H-

-t*

M

number of hooks in the southern direction even between groups
of bees in comparatively closely situated localities. Turning our
attention to the Caucasus (Fig. 5) we must say that the situa-
tion here is more complicated than in the plain of European
Russia. Zoogeographically, the Caucasus is divided into several
sharply limited provinces, each of them with peculiarities in the
composition and the origin of the organic life. The Caucasus
bees are also not homogenous. The best characterized is the gray
Caucasian mountain bee Apis mellifera caucasica Gorbatschev
and the yellow Transcaucasian so-called Persian bee. This bee
was first recognized as an independent species by Pallas ; although
he did describe the Caucasian bee he has never published his
manuscript. The specimen with the original label is preserved
in the Berlin Zoological Museum and was briefly described by

HOOKS OX HIM) \\TXG OF IIONKY 1!KH.

219

20%-

16 17 IS 19 20 21 22 23 24 25 26 27 2S 29

FIG. 4. The continuous curve represents the variation of hooks of bees
from Middle Russia. The dotted, is based on material from South Russia.
The frequencies are expressed in percents.

Black
Sea

Intermediate

fhrm^

•*••••
••••••
• •• •• •
• •••••
••••••

lUf //7O.

A. mellifera
Caucasica 6.

oooooo
oooooo
oooooo

Yellow trans-
Caucasian bee.

FIG. 5. Map of Caucasus showing the distribution of fifUvn variations
of bees (after A. Gorbatschev).
15

220

\V. W. ALPATOV.

«

M N W M

Os

O

I-

HH (*O
t— 1 11

M ^1-

inscauca

Q.I

ro I^» Tf OO O '-| *H

o

-

oi4i

"O ~O
o jj

e

H

.So

W M O <N "M" 00 Tj-OO -H

o\

(N

ro

O\ ro

S -H

01 10

•rf r*

p-i

— s^^^ OOH

(-1
W

M 00

Ov O

6 4)
01 n

to o

O\ <N

9

^OOOJ^OO-OOOOM

o

1 0
c3

01 in

OO M

6 jj
Oi Tl

to f^

r*- LO

* -H

\bchasia

1—1 00

M O O\ O <N i-D <N IN
M M

o

M

Ov H

01 "H

01 \o

1-4 00

1— 1 M

M

*

O

M

O t-

M M

oi -H

00 00

-'«

M C Oi 'O O\ oO O\ to

rf

os

-t t-

00 M

6 4J

01 I

*t oo
i-i in

06 -H

•«

IH ^f r** O t^» O ^^ '-O ^*

01

ro O

M _l_|
01 ~

M Ol

rn

c
'3

tfi

M O ^ W ON O\ "^1* *-^j

ro

„.

H OS

ro M
el -H

Os ro

3
O

3

1

o'S
(/}'£

c^ O O O ^J" 0) N

$

several

i— i t~*
^O i— '
h-l 11

Os t^

1-1 m

1

cd

U

03

<*O fO P^ ^O O ^" *~t

01

-

O Tt-
O 01

c-i jj

01 I

oo r—

H

h- 1 IH

O
m

00

01 <>

M 00
M ^^~

'ji

»

to <"O CO to ^t" to O O\ 'O HH

M "^J* \O t^* CO

01

t~t \O

^o o

O _L|

^" M
M O)

Caucus

a
14

CS N

M c^ ro M

O

O

many

OS O

6 4J

01 Tl

^ ro

Ov ro

O JJ

y,

i*

P

M M rj- ^ O»

vO

rv.

t^- M

° -H

M 00

ro 01

03

O

Q

HH

03

C

Qj

o

0

*~?

o

^

M-l

'-*-<

O

o

Kp

r-o >OTf(V)Cs M O O\oo *--xO lO1^-

O

o

^

§

CJ

CQ

•/.

<
5!

<;
U
D
i;

O

fc
O

t/l

O

o

ffi

u.
o

04
W

a
g

u
a

H

fa

O

71
H

z

-r.

7)
X

O

o

HOOKS ON HIND WING OF HONEY BEE.

221

W

(J

in

a
a

S3

§

1/1

a,
D
O

Q*

O
H

z

u

Q
z

o
o

K

§

M

S

5

z

K
O

W
S
H

Z

u

Id

n

u

w

to
to

Q

Os

£

q

a *

.y u w

i ri QJ

41

Ctf (ft

O

P CQ 03

M

c .

q

-H \f>

ll

41

O

41^

c "3

o
q

H

t~- II

oo

q

Os

• o

hH 0

j^

41

41 "=>

41 «

ctf

o

M

M II

00 II

KH

•-O

IT)

oo

HH

6 OH

h?

£ CO

hH I~^

0 0

-H O

t Cti

" CO

• o

rrt ^

41

O

41 -

41 «o

41^

2 y

hH

0 II

« II

00 II

t^

cs

q

o

hH

OPS'

C OH

0)

•

O

(-TJ

cs

^o oo

— C

M 00

M U">

-H q

^ '-H

41

4i M

-H S"

41 J?

O r-1

O

— - r*J

r/5

OO ||

CO II

o II

flj ^

ro

M

CO

r-

o <S

hH

do^

O OH

M PH

^H

CS

oo

ro

ly-j

l_l

0

cs O

cd

q

M (O

M IO

HH ^^

• cs

• o

c§

41

41 co

-ti «o

41 «o

41 d

4l2

o

*tj

hH

<N II

0 II

0 II

Os II

ao II

y

OS

oo

MO

q

cs

5j

d

O OH

d <£

d*

M PH

cs

^~

o

^

oo

M

09

O

M 0

cs O

M O

0

O r-

M O

r* 3

• q

M O

• oo

M 0

• oo

• Os

o y

41

0

41 >o

0 II

4lf

II •

0 II

Hif

41 4

Os II

41 06

oo II

x^ — '

*o

cs

h- 1

0

q>

FH CO

U

d

cs

d(*

^ti

M04

dsO^

0 OH

o

CO

o

cs

*^"

o

00

M

o

O CO

H 0

M 00

M

hH

0 0

M CS

• oo

• O

* 10

' CO

• o

•

0 '

41

41 «o

41 M

41 "'

41 4

41 4

41 «

41 M

o

o

co

t- ll

t- II

>o II

CO II

CO II

oo II

U*> ||

O

M

cs

0

0

rj*

hH

O OH

6ti

O OH

004

d A

O OH

M &

(S

TH_-

t^

i^

Os

M

Os

r-^.

O

o

O "1

O Os

O

M CO

^f cs

O O

O Os

M r**

• oo

• cs

• CO

' t^-

H CS

• o

' CS

Q

o '

41

41 °o'

41 M

41 «

41 -o

' t-L

41 o\

41 4

"H M

o

O

cs II

Os II

« u

CO II

"" II

« II

0 II

t- II

O

q

cs

o

F-

oo

CO

o

d

Of*

O PH

O OH

»H Pn

O OH

d 04

M PH

cs

q

0

q

0

q

oo

q

O

CO

M

oo

q

o
o

q

41

41

-H

41

41

41

41

41

41

o

00

hH

M

CO

M

hH

o

ro

t-

o

O

Os

CO

t^

IT)

o

*O

d

,_J

d

d

M

H

M

H

ON

cs

cs

cs

cs

CS

CS

cs

H

01

r*

00

£&

CO

to

.03

15

%

CJ

V.

03

i±H

2

03

O

.2
"Si

'3

Ui 4->

? c

03
u

c5

_o

S

.y M

•f-

cd

" »2

eft

a

^

't-t

o

U -

o

0

i_

o

o

03

03

15

rj

= —

IO

O

Z

<

U

H

M

<

<

222 w- w- ALPATOV.

Gerstacker (15). The author of the present article was able,
thanks to the courtesy of the curator of the collection of Hy-
nienoptera of the Berlin Zoological Museum, Professor Dr. H.
Bischoff, to examine Pallas's specimen as well as his manuscripts.
Pallas gives in his manuscript the following indication about the
origin of his Caucasian yellow bee: " Ad Caucasum lecta, itemque
ex Hyrcania transmissa fuit." The small size and pronounced
yellow coloration of the specimens preserved in the Berlin Mu-
seum permit us to conclude that Pallas and Gerstacker described
under the name Apis remipcs, the Transcaucasian Persian bee,
but not the north Caucasian darker bees.

Some peculiarities — for instance a much longer tongue — dis-
tinguish Apis mcllifcra remipcs Gerstacker (not Pallas) from
the Italian bee Apis mcllifcra Hgiistica Sp. It is therefore not
correct to identify the A. m. remipcs with the Italian bees (Apis
lii/itstica) as it has been done by G. A. Koschewnikov. Accord-
ing to Gorbatschev (see the map in Fig. 5 taken from his article
(16) ) the prairies and hills of the northern Caucasus and the
valleys of Transcaucasia are populated by a bee of intermediate
type — hybrids in his interpretation. We united our material into
four groups: (a) N. Caucasus bees, (ft) bees from four apiaries
near the coast of the Black Sea — Abchasian, (r) gray Caucasian
mountain bees (A. mcllifcra cancasica Gorb.) and (of) yellow
Transcaucasian bees (Apis mcllifcra remipcs Gerst). Table
VII. shows the frequency distributions and Table VIII. gives
us material for estimating the importance of our differences. The
Apis m. cancasica and remipcs show a pronounced higher number
of hooks than bees of South Russia. Of course such a compara-
tively limited number of colonies from N. Caucasus does not
permit us to draw a perfectly definite conclusion. It is interest-
ing to note that the gray Caucasian bees imported to the United
States (see Table IX.) gave also a high average of the number
of hooks.

Table' IX. n'ives us some data on other European races of bees.
The Italian bees from Italy are characterized by a high number
of hooks. It can be seen that the progeny of Italian queens im-
ported from Italy to X. Caucasus shows also a high number of
hooks. The German black bees, according to our recalculations

HOOKS ON HIXU WING OF HONEY BEE.

K

O

H

u
ft.
o

04

W

c/)

-

h
I]

U

H
Z

04
U
b
ta

Q

o
fc
tn

u

M

l?l

n '0-T-OOO'^-M

\r>

o

M

t"*» IO

I/*, NH

o _u

-r M

fi

H«*S«-,-.««

(N

0

H

o o

C to
00 _L|

|1

M « - 0 ,o C ,0 c, «

C
to

2

CN IO

r- M

S-H

oo O
. »o

crj

c

l«

O C t^ fN Q\ ^- <N

O
to

o

~t o
" -H

C^ "1

. N

O 'O

C

~

_g

w oo -t C oo rf -rf

to

c

M M

o o

.si

< 5

.Q

1-1 rooo Oi^t^-roLO

0
in

o

O to

IN M

^ -H

-r 10

o; 7J

«5

ro t^« M rf ro oo ^

io

o

c c

- n

5 -H

to r^
00 to

i

rf "

M M

to

c

1-4

00 -H

00 IO

c
,J5*

< j; fc

cs oo ^* LO rs M

0

<-

oo o

S -H

to M

od _u

^j

> ctf en

I— I M

O
to

-

t** (N

S-H

It HH

rt

O

*H w ro ^ cs

t
to

M

O

rO Os

o q

oo' _u

C/J

41
«j
CQ

i O

M sC •— ' !>• C (*O C^

O
to

o

M

00 <N

O **

00 O
r^ .

O _u

u

a

3

rt ej

^^ ro ON ON *^ 00 r^« 01 M

M

to

o

M

t^ OO

c' -H

(N 1

l|

rt
_O

------

o

IO

°

oo ro

CO M

t— OO

.§

II

-ff'sasjas*"

C
to

O

1— 1

M 00

10 o

CN 00
M C1)

11

-* .—

0K

Sod
S^"rt

H .OH.,..

O

c

H

0) .
CS M

\o ^?

c

1

^

f.

I

t

'/,

s

.v

o

C

1

a

0

C

2

NP

\O ^-^1 *^J" ^O ^ ^ O O^ OO f""~ O

6

o
2

s

U

224

W. W. ALPATOV.

&!

i

5?

^}-M M rj-mroo M OfO^ro
vO ON »o r~ M ON^tO ^-O\O\oo

C
0

0

0 ==

"3 15

UCn

33

MOi-iONi^f-i-'ONOrOfOO
CNTj-ooOOt^fOi-i

M t-t

6
o

1O

LO

LO

c

ON

oo
ro
O

c

§£

t-i .

v-i 03

^g

-1

QS
»

csoo M 'i'lM ^00 ^tooiorOM
W 1^)00 O ONt^row

f-t

O
0
10

-H

CO

o

t^

I-H
CN

-H •

M

ro
O
<N

00

M

,j5

3"°

rOOO)r~Ot^iOOvo
COOOOI>OiOO O t-

Os
ON

o
I) .

M <
c ,x

^1

u-2

•*tr^O<NO<NvOoO"3-

M TfOO O OO •* M
M

£
•<t

•*
IO

o

•<t

oo

ro

O

5M

^p

-*->

HH

£»

-1
Ofc

W)

Tj-i-ivO ^O rOfOt^o
M ^- oo M o "3" M

h-l

rt-

o
•*

-H

ro

O

o

M
fN

-tt

TJ-
oo

0

CN

vO

rt

3?

M M O •3-ioror~-vOO "i
rOTj-oo t^o Ooooo t^->o

o\
o\

ro

W
03

* s

c3^

tN ONl^O\roOoo romt^
CM 10 C7\ O 00 LO CN

M

ro
r-
M-

10
"">

0

ON

oo
ro

q

j-; QJ
^M
o

w

oot

-1

Ofe

to

•<ti-i<NCNi-iOror^t^i^
w rowOO ONIOOI

M

"fr
r~
Tf

41

M
00

O

M

M

-W

<N

t^

•O

1^
t^

M

r*

.2

3^

OM OOO'^JioOO'S'^OtoO
04 LortOvNO O\^ ror^O\rv}C>i O\\o

t^

O\

lo

CO

« »

(L) <U

c3^

OO'-ttv^|-|0)^f>-(rocN1^tO\Mt^
M rooo r^^tcN roiocs

H CN CN W

10

o\

o\

00

n
o

NO

0

M

q

— CU

•n pq

|O

i

oot

, "O

.0 ^

°fc

en

M I Nu'iroLOi-ioor^iooOooON
M roONO w M "11>H

M CN CN M

0

o
q

M

-H

00

ON

vO

6

CN

-H

ro
rr>
F~
vO
t^

en

_0j
"g

37

^•oocNroooooOM-*O

w\Oi-ii-lOOOtN<NCN

<s

_o

"3 n!
U£

Z«

CNO\\OMMfNMt--wCN

ro oo <N 0) oo PO HI

M M

r^

O
10

t~i

•*

O

^h

ro
ro

q

s "3

o«
'-*-<

oot

: "0
J2 ^

°£
03

lotNTj-MMi-ir^noO'*

M no oo 1-1 M ON co

M I-H

r^
0
10

•fl

UO

00

•*

CN

-H

IO

ON
t^
t^
in

M

CO

_OJ

'5

3"S

M O\T}-M ^ftN ronO r^cN io»oroi-i O\io
ooo roO rotN t^^fl^O O\O t^ \o O\O tN

00

O\

M

0

"o d

^s u
§3

CMO\^-i-ii-HT}-ioooroOO\ooi-iroiH O

CNlOOrOOvO^OlOtNM

0

o\

O

IO

f)

o -n

0 J?

10

O

q

M if

|1

U>4

"b

. "O

•° >

O C

aj

CO

N^)-O\l-ICN^}-O'*rOMt^MMl-l^f(NM

<N 1000 •^•1000 ^ooo COM

M M M HI

1^.

o\
o\

-H

N

O
ro

l-H
0)

-H

10
M
m
N

ro
M'

"^ o5

0 JS

. r

0 o

^ ^

£* i-C

Ooo t^->O lo^ron M o Ooo t^\o lO^fro

y.

s

h

X
w

HOOKS ON HIND WING OF HONEY BEE.

225

•e
<^
s
s

• **

H^j

S
O

U

X
w

>>^

P

O

C •"

rO

IN

0 S

O

O

Q

"o 2

o

-.

ucx

O\

>

o

M

-H

-H

CN

IH

3

c £
o <—
'-*-* rt

-H

o

rO

O

00

O

o

"3

s

•bg

-t

00

o
o

6

CN

00

0

M

6
1

IH

O

1

o

,j^

M

>n

O

CU

q

oo

M <

ro

-H

M

>

m

nl ^

OO

a

•H

u

oo

"3 P

-H

-o

o
oo

0
vO

-

-4-J

hH

cs

O

^

m

10

in

ro

s

o\

X»

M

6

M

00

O

oo

O

1-1
O

oo

r^-

1

IN

6

Z

M

C

a

E

CO

2"

o

q

0

HI

00

-H

-H

M

l_l

3

ro

CJ

•*-* pT[

-H

o

o\
m

O>

O

00

M

M

Tf

CN

Oi

O

rO
O

6

in

3D*

IN

-t
OO

ro
O

O

00
M

M
t-

CN

w

6
1

IH

O

oo

C

N

m

.5

<N

M

CO

t^

~2

CO

3

O

o

ro

" CO

IN

-H

•n

T3 m
•c

-H

rt

oo

ro

i

ro

oo

00

o\

Tf
O\

o\

M

S

i— t

M

O

OCX

O

q\

,_;

q

r^

00

6

IN

"*

M

d

"**

CO

ro

OO

_0j

0

o\

g

O

fn

O _c3

q

M

M

m gj

1— (

-H

-H

§i

-H

o
o

oo

ro

IH

Tf

ON

O

in

ro

o

H

•^r

H

in

T}-

t-—

oo

rO

ro

O

00

q

IN

T}-

>-<

OO*

^

6

"

6

6

rv.

"?

CO

ro

M

_cu

IN

00

'2

{-—

ro

•^

0

o

o cd

TJ-

o

M

1)

h-l

o-c

° a

I-H

-tt

-H

3

is

-H

o

m

oo
m
i-t

ro
O

o

ro
IN

"3
E

0

,*•*

00

O

o\

^j-

o\

0

IH

CN

•b

6

M

0

6

s

N

ro

vO

o
o

O

2

ro

t|

3 J2

o -N

0 o, a

|J

h
f

Sffi

£r?

rt

ca

0.

1?

U

CQ.

•SL

rt

U!

H

O

T3

'§

3Dt

g
U

O 2

cu

., ,

-4-1

2

o

'>

cu

IH

o

3
<j

a!

"3

cu

E

IH

4J

O

E

c

o

*4H

1-1

o

W

•a

+J

«j

JS

"3

3

U

U

226

W. W. ALPATOV.

of Annbruster's data, give a number which corresponds to that
<>f the Middle Russian ones.

Sun lining up now our whole material on European races we
max- say that there is much evidence for an assumption of a high
number of books in southern races in comparison with northern
ones. The Bulgarian group of bees also supports this conclusion.
It would be interesting to test this rule on other castes of bee col-
onies. Unfortunately our material on drones from Russia is very
small (see Table IV.), although it can be seen that the Middle

20%

10%

> — Black bees (USA)
•\ —Italian bees (USA.)
I \\ Italian bees (Italy)

\\
\\

16 17 18 19 20 21 22 23 24 25

IMC. (>. Curves of variation of hooks. Continuous line — black bees in
1. S. A. ; dotted line — Italian bees from Italy; dash line — Italians in U. S. A.

Russian and even South Russian drones have a smaller average
number of hooks than the Bulgarian ones. It would not be wise
to draw any conclusions about the geographical differences in
queens based on such a small number of cases. We have to add
that Middle Russian and South Russian drones give the high de-
gree of variation (C'< ) usual for drones.

HOOKS ().\ HIM) WING OF HONKY BEE.

227

It is well known that at the time of the discovery of the New
World, America had no native bees. The first bees imported to
this country came, according to historical data, from Holland and
England and belonged to the common black bees A. nicllifcra
nicllifcra L. About the middle of the last century the American
beekeepers began to prefer for cultivation the yellow Italian bee,
which is now the dominant race in this country. Thanks to the
help of many beekeepers I have succeeded in examining, from a
considerable number of apiaries, Italian bees of different degrees

200

100

7

\

13

15 16 17 18 19 20 21 22 23 2* 25 26 27 28 29

FIG. 7. Frequency polygon and fitted curve of the variation of the hooks
of the Bulgarian drones.

of development of yellow color as well as pure black bees.
Tables IX. and X. show us the variation of bees acclimatized to
the United States. Firstly, we have to note the great difference
in the number of hooks of black and yellow American bees, sec-
ondly, a little lower number of hooks of Americanized Italian
bees than that of true Italians reared either in Italy or from
Italian queens imported directly from that country. This is il-
lustrated by curves on Fig. 6. The very low average number of
American black bees as compared with our material discussed

228

\Y. W. ALPATOV.

above give us the right to suppose a general decrease of the num-
ber of hooks in the United States as compared with Europe, both
in black and yellow bees. Further investigations need to be made
with special attention to the problem of influence of acclimati-
zation upon physical characteristics in the honey bee.

200

100

16 17 /g 19 20 21 22 23 24 25 26 27 28 29
FIG. 8. Frequency polygon and fitted curve of variation of the worker
bees from Middle Russia.

Our comparatively large material gave us the possibility of de-
termining the character of the frequency distributions. The re-
sults are given in Table X. All distributions are symmetrical
and only one shows a deviation from the normal distribution.
A curve of type VII. was chosen to fit this distribution. The
distribution for the Bulgarian worker bees being symmetrical and
normal in regard to the /?, gave a very high value of the criterion,

HOOKS ON HIND WING OF HONEY BEE.

229

which leads us to the curve of type IV. It was not possible to
calculate the probable errors of the criterion. /?, and /?., being too
close to those characterizing the normal curve of error. There-
fore a normal curve was used for fitting. We used for calcu-
lating the ordinates of the normal curve from Pearson'^ " Tables
for Statisticians and Biometricians." Figs. / and 8 show two of
our curve fittings.

25

FIG. 9. Measurements on the wing. The wing shows the intercubitus
vein not developed. (Microphotograph taken in the Art Department of the
Institute for Biological Research, by Mr. Johansen.)

During the author's residence at Cornell University an attempt
was made to study the influence of undernourishment of larvae
upon the characteristics of adult bees. The experiment consisted
in putting the unsealed brood in an incubator running at 34.5° C.
The brood was taken from a comb approximately one day before
normal sealing. On the following day the cells situated in the
neighborhood of the place from which a piece of comb had been
taken the day before were already sealed by bees and also put
in the incubator in order to provide us with control insects. Bees
normally developed in hives were also collected from the frame of
the hive which gave us material for the experiment. The pieces
of comb with unsealed larvse put in the incubator were covered
with pieces of artificial comb foundation in substitution for the
natural capping bees. The larvse wove cocoons as usual and the
emerging bees were collected in alcohol. The bees emerging from
the unsealed brood evidently suffered from a certain underfeed-
ing in comparison with control bees. Table XL shows that the

230

\Y. \V. ALPATOV.

TABLE XI.

('(INSTANTS OK WlXG MlCASUREM EXTS OF CONTROL AND UNDERFED (iN

LARVAL STAGE) BEES IN MM.

Proximal Length
of Wing
(Meas. N 24).

Distal Length
of Wing
(Meas. N 25).

No.
of
Cases.

Experimental (underfed) bees

4.525 ± -021

4.192 ± .023

46

( Control bee*>

4.696 ± .010

4.353 ± .012

31

experimental bees have a smaller size of wings than the controls.
The characteristics have been measured, as is shown in Fig. 9.
Table XII. shows the average number of hooks in three groups.

TABLE XII.

INFLUENCE OF UNDERFEEDING ON THE NUMBER OF HOOKS AND THE AB-
NORMAL VENATION.

Character.

M ± P.E.

C%±P.E.

Percentage
of Wing
with Abnor-
mal Vein.

Number
of Speci-
mens.

I.

Bees taken from the hive . . .

20.77 ± -16

7-34 ± -56

o.oo —

39

2.

Bees reared in the incubator
from brood normally fed. .

20.60 ± .15

7-33 ± -52

4.44 ±i.37

45

3-

Bees reared in the incubator
from underfed brood

TO. 71 -t .OO

7.30 ± .33

19.82 ± 2.50

116

n

iff. 1-3

1. 06 ± .18

19.82 ± 2.50

R = 5-89

R = 7-93

n

iff. 2-3

O.8o ± .17

15.38 ± 2.85

R = 5.24

R = 5-40

It can be seen that the underfed bees have a smaller number of
hooks than the control bees reared from the sealed brood and
taken directly from the hive. The same is expressed in graphical
torin on curves of the Fig. 10. The experimental bees showed a
quite peculiar type of abnormality in the venation of the first pair
of wings. The abnormality consists in the incomplete develop-
ment of the second intercubitus vein. The percentage expres-
of this abnormality in our three groups shows that the ab-

HOOKS ON 1IIXU \Y!.\<i OF HONEY BEE. 23!

normality occurs also in bees reared in the incubator from
normally sealed broods but reaches a very high tirade of develop-
ment in undernourished bees. The abnormalities in insect wings
have been many times the subject of careful morphological studies.
Our experiment opens a way for studying this problem by means
of the experimental method. It is worth while to note that in
geographical races the southern bees, being usually smaller than

30%

20%

10%

16 17 /$ 19 20 21 22 23 24-

FIG. 10. Continuous line — bees from the hives which gave larvae for
the experiment with underfeeding. Dotted line — variation of hooks in the
group of bees emerged from underfed larvae. Dashed line — control bees
developed from normally fe_d larvae and emerged in the incubator together
with underfed bees.

the northern one (AJpatov, 8), at the same time show an in-
crease in the number of hooks. Our experimental bees gave
the contrary relation. Therefore it is not possible to explain the
smaller number of hooks of the northern bees by the assumption
of an underfeeding of larvae.

232 W. W. ALPATOV.

SUMMARY.

The data presented in this paper show that the average number
of hooks in the honey bee is a characteristic which is differently
developed among single colonies, sexes, castes, and races. As a
general rule the southern races have a large number of hooks in
worker bees and probably in drones. The queens and drones are
more variable in regard to this character than the worker bees.
In this respect the relations differ from those in other social
insects (ants, wasps and termites), where the asexual caste is
the most variable. The experiment with underfeeding of larvae
showed a decrease of the average number of hooks and the
producing of specimens with defective venation — incomplete
second intercubitus vein.

LITERATURE CITED.

i. Alpatov, W. W., and Tjunin, F. A.

'25 Beitrage zur Kenntniss der Variabilitat der Riissellange bei der
Honigbiene (in German, with Russian summary). Rev. Zool.
Russ., Vol. 5, pp. 79-108.
2. Alpatov, W. W.

'25 liber die Verkleinerung der Riissellange der Honigbiene vom Suden
nach Norden bin. Zool. Anz., Bd. 65, pp. 103-111.

3. Alpatov, W. W.

'26 Der Riissel der Kasaner Arbeitsbiene in variationsstatistisher
Bearbeitung (in Russian with German summary). Bull, of
Kasan Agric. Exp. Sta., pp. 23-33.

4. Alpatov, W. W.

'27 On the Variability of the Hooks on the Hind Wings of the Honey
Bee (in Russian). Ptschelowodnoje djelo, January, 1927.

5. Alpatov, W. W.

'27 The Cardinal Problems in the Study of the Caucasian Bee Races
(in Russian). Ptschelowoni Mir, Vol. i, p. 3-8.

6. Alpatov, W. W.

'27 On the Amelioration of the Bee Races (in Russian). Ptschelo-
wodnoje djelo, 8-9, pp. 372-377.

7. Alpatov, W. W.

'28 On the Variability of the Tongue Length of Caucasian Gray Bees
and Its Inheritance (in Russian with English summary).
Ptschelowodni Mir. N i, 2, 3.

8. Alpatov, W. W.

'27 Biometrical Study on Bees of Middle and Southern Russia. Rev.
Zool. Russ., Vol. 7, livr. 4, pp. 31-74 (in Russian with English
summary ) .

HOOKS ON HIND WING OF HONEY REE. 233

9. Alpatov, W. W.

Variability in the Honey Bee Tongue Biometrically Investigated and
Practical Questions Connected with the Problem of the Selec-
tion of the Honey Bee. (In press, Jour, of Econ. Entom.)

10. Alpatov, W. W.

On the Improvement of Bee Races. (In press, Reports of Maryland
Agricultural Society.)

11. Armbruster, T.

'23 Wie untersucht man Bienenstamme und Bienenkreuzungen auf
ihre Farbe? Archiv fur Bienenkunde, Vol. 5.

12. Arnoldi, K.

'26 Studien iiber die Variabilitat der Ameisen. Z. f. Morphologic und
Okologie der Tiere, Vol. 7.

13. Bachmetjew, P.

'10 Analytisch-statistische Untersuchungen iiber die Anzahl der
Fliigelhaken bei Bienen und die daraus hezvorgehende Konse-
quenzen. Z. f. Wiss. Zoologie, Vol. 94.

14. Casteel, D. B., and Phillips, E. F.

'03 Comparative Variability of Drones and Workers of the Honey
Bee. BIOL. BULL., 5, pp. 18-37.

15. Gerstacker, — .

'69 On the Geographical Distribution and Varieties of the Honey Bee.
Ann. and Mag. of N. History. III. ser., Vol. n.

16. Gorbatschev, — .

'16 The Gray Mountain Caucasian Bee (Apis mcliifcra var. Caucasica}
and its Place Among Other Bees. Tiflis. (In Russian with an
English summary.)

17. Kellogg, V., and Bell, G.

'04 Studies on Variation in Insects. Proc. Wash. Acad. Sci., Vol. 6.

18. Kellogg, V.

'06 Variation in Parthenogenetic Insects. Science, Vol. 24, no. 622.

19. Koshewnikov, G. A.

'00-'05 Materials to the Natural History of the Honey Bee, part i
and 2. Moscow, Nachrichten der K. Ges. d. Freunde v. Naturw.,
Anthrop. und Ethnographic, Abt. f. Zool. (In Russian.)

20. Palenitschko, Z. G.

'27 Zur vergleichenden Variabilitat der Arten und Kasten bei den
Ameisen. Z. f. Morphologic und Okologie, Vol. 9.

21. Pearl, R.

'10 Recent Quantitative Studies in Variation in Social Insects. Am.
Nat., Vol. 44, 521.

22. Phillips, E. F.

Variation and Correlation in the Appendages of the Honey Bee.
(In press.)

23. Thomson, E. Y., Bell, J., and Pearson, K.

'09 A Second Cooperative Study of Vespa vulgaris. Biometrika, Vol.
7, 48-63.

24. Thomson, E. Y., Bell, J., and Pearson, K.

'11 A Third Cooperative Study of V. vulyaris. Vol. 8, 1-12.

234

\\ . \V. ALl'ATOV.

_>;. Warren.

'08 Some Statistical Observations on Termites. Biometrika, Vol. 6.

26. Watson, Lloyd R.

'27 Controlled Mating of Queen Bees. Hamilton, Illinois.

27. Wright, A., Lee, A., and Pearson, R.

"07 A Cooperative Study on Queens, Drones and Workers in Vespa
Biometrika, Vol. 5, pp. 407-422.

Vol.LV.

October, 1928.

No. 4

BIOLOGICAL BULLETIN

THE DEVELOPMENT OF THE SPERMATOZOON IN

CAVIA COB AY A i

MARY T. HARMAN AND FRANK P. ROOT.

INTRODUCTION 235

MATERIAL AND METHODS 237

(a) Description of material 237

1. Period with little change in the shape of the cell 237

2. Period of elongation 239

3. Histogenesis of the elongate cell 240

DISCUSSION 241

SUMMARY 244

LITERATURE CITED 244

DESCRIPTION OF PLATES 248

INTRODUCTION.

The development of the spermatozoon in the Mammalia has
been observed in a number of forms but a detailed study has been
made in only a few instances. Among the workers who have
published observations on the development of the mammalian
spermatozoon are: Lenhossek (1898), Meves (1898), Benda
(1897, 1906), Korff (1902), Duesberg (1908, 1920), Jordan
(1911), Oliver (1913), Stockard and Papanicolaou (1918), Gat-
enby and Woodger (1921). There is a general agreement in the
plan of the development but many differences of opinion exist with
reference to the detail. Many of these differences are significant
not only from the development of the spermatozoon itself but also
from their bearing upon other biological problems. Since mam-
mals are bisexual and have not been known to reproduce partheno-
genetically, the continuity of the different parts of the male germ

1 Contribution from the Zoological Laboratory, Kansas State Agricultural
College, No. 100.

16 235

236 MARY T. HARM AN AND FRANK P. ROOT.

cell is of as much significance as that of the female germ cell.
The loss of a part of the nucleus or even a part of the cytoplasm
in the process of transformation of the spermatid into a sperma-
tozoon may affect the theory of the vehicle of the bearers of the
hereditary characteristics.

Cavia cobaya has been used as a subject of investigation for
the development of the spermatozoon as often as any other mam-
mal and the work has been done in as much detail and yet there
is a lack of agreement upon a number of points. All authors
are agreed that the spermatid is a typical one, similar to that de-
scribed for insects and other animals and that the mature sper-
matozoon is composed of at least three parts or regions, the head,
the mid piece and the tail. Also a fourth region, the neck, has
been described by ntany workers. What parts of fhe spermatid
contribute to the formation of each of these regions, of what each
region is composed and whether or not the entire cell is used in
the formation of the spermatozoon are questions upon which
there are significant differences of opinion.

In our study of the development of the spermatozoon of Cavia
cobaya certain things have been impressed upon us as being de-
cidedly different from the observations of other authors. Of
these we shall mention five as the most outstanding : ( i ) Follow-
ing the last maturation divisions the chromatin material goes
through an abortive preparation for division before there is much
change in the shape of the cell. (2) We have found no loss of
cytoplasm or sloughing off as has been described by many authors.
It is true that we find stages when the entire developing sperma-
tozoon is smaller than in previous stages but this seems to be due
to a condensation of the material rather than a sloughing off of
any part of it. This will be discussed in some detail in the body
of the paper. (3) We have not found in any stage a filament
extending out from the cytoplasm. We have diligently looked for
it because we were very anxious to see the nature of this develop-
ment and at what particular time it was first evident. In all of our
observations the axial filament tapers to a blunt point at the ter-
minis. There is no naked end filament even in the fully formed
spermatozoon. (4) The tail is made up of three segments which
are not only shown by the morphological structure but also by the

DEVELOPMENT OF THE SPERMATOZOON. 237

points of breaking as found in hundreds of broken specimens.
(5) As was mentioned in our previous paper, the area of actively
dividing cells are elliptical with the greatest diameter of the ellipse
lengthwise of the tubule. Within this area the cells are generally
in the same stage of development and only occasionally a stray
cell is in some other stage.

It has not been our purpose to describe the origin of the cy'to-
plasmic structures nor to say much about the confused nomencla-
ture of the same. This has been only incidental to our purpose
and we have discussed them only in so far as they contribute to the
development of the spermatozoon. We have used much of the
nomenclature of Bowen when it seemed applicable to our needs.

MATERIAL AND METHODS.

The material used is the same used in our previous paper (Har-
man and Root, 1926). In that paper will be found a detailed de-
scription of the fixing and staining of the material. All drawings
have been made with the aid of a camera lucida and the magnifi-
cations are given in the description of the plates. With one ex-
ception, our drawings could be duplicated from hundreds of cells
in our material. We make this statement to emphasize the fact
that what we are showing is universal and not an exception which
might be attributed to technique. The exception is the bent rod-
shaped cytoplasmic inclusion in Fig. 7 which we have called a
Golgi body.

(a) Description of Material. — We have begun with the changes
which take place in the cell after the last maturation division has
been completed. This is where we stopped in our last paper.
For convenience of description these changes may be divided into
three periods as follows: (i) The period with little change in the
shape of the cell; (2) the period of elongation and (3) histogenesis
of the elongated cell.

i. The Period with Little Change in tJie Shape of the Cell. —
Significant changes take place both in the nucleus and the cyto-
plasm before there is much change in the shape of the cell. At the
end of the last maturation division the chromatin passes through
a typical telophase. It becomes finely granular and a definite nu-
clear membrane is formed. Following this there takes place what

238 MARY T. HARMAN AND FRANK P. ROOT.

we have chosen to call an abortive attempt to divide again. The
chromatin forms into a close network having irregular clumps
and the nuclear membrane nearly disappears, Fig. I. Then the
nucleus increases in size and the chromatin material is in a more
nearly continuous spireme, Fig.. 2. The chromatin clumps be-
come more numerous and prominent. These changes continue
until a compact unbroken spireme is formed, Fig. 3. Then there
is an attempt to form chromosomes, Fig. 4. The chromatin knots
are numerous and the spireme has been separated into irregular
pieces which may be compared to chromosomes but which lack the
smooth contour and the compact appearance of chromosomes.
We have called these masses of chromatin material " chromatin
knots." There remains some trace of the spireme but it is little
more than a suggestion. Following this the chromatin knots be-
come more granular and there is no further indication of a division
of the cell, Figs. 5 and 6. Now the entire cell begins to contract
and to become compact. At first this is more evident in the nu-
cleus than in the cell body. The chromatin material becomes
finely granular and only traces of the spireme are discernible.
The entire nuclues occupies much less space, Figs. 7, 8, and 9.

While these changes have been taking place in the nucleus,
changes have been occurring in the cytoplasm. A number of
spherical bodies varying in size appear in the early spermatid.
These are the Golgi bodies. There is a lack of constancy in the
number and the size of these Golgi bodies. They are found in
the periphery of the cell as well as near the nucleus. Sometimes
they may indent the nuclear wall, Fig. 2. With Heidenhain's
haematoxylin they are stained like chromatin which emphasizes
their spherical form and distinguishes them from the surrounding
cytoplasm in the early spermatids. They are finely granular like
the surrounding cytoplasm but the granules are more closely com-
pact than in the other parts of the cytoplasm. Each Golgi body
has the appearance of a sphere surrounded by a halo.

An idiosome is always near the nucleus. In section it is cres-
centic in shape, with the concave side toward the nucleus, and
stains like the surrounding cytoplasm from which it is distin-
guished by its more homogeneous structure. In older stages it
comes to lie in contact with the nucleus then there is a more den-

DEVELOPMENT OF THE SPERMATOZOON.

239

nite orientation in its position than that of the Golgi bodies and
the nucleus. At least one Golgi body is always near the nucleus
and at the same time near the idiosome, Figs. I, 3, 4, and 5. Some
of the Golgi bodies form a group near the nucleus on the side
opposite to the idiosome.

The idiosome becomes closely applied to one side of the nu-
cleus, Figs. 8 and 9. The idiosphere is in the concavity of the
idiosome between it and the nucleus, NE in Figs. 8 and 9. The
entire cell, both cytoplasm and nucleus, has become smaller and
there is evidence of the beginning of the change in the shape of
the cell.

2. Period of Elongation. — With the diminution of the volume
of the cell there is the beginning of an elongation in the axis de-
termined by the idiosome, on the one side, and the Golgi remnant
on the other. This elongation occurs in the entire cell affecting
the shape of both the cytoplasm and the nucleus, Figs. 10 to 22.
During this time the cell is in intimate connection with the Ser-
toli cell which is at first small but later increases enormously in
size. Fig. 20 illustrates a Sertoli cell with some of the associated
spermatids in an elongated form. The part of the spermatid
destined to become the head is directed toward the base of the
Sertoli cell and the other part toward the lumen of the tubule.
This is true regardless of the stage of development. When the
spermatozoa are freed from the Sertoli cell they are not in a
mixed up mass but are in bundles lying almost parallel, with most
of the heads in the same direction. This could easily be accounted
for by the fact that they have a definite orientation during their
development. While the cell is elongating the idiosphere becomes
embedded in the idiosome. The idiosphere stains more densely
than the idiosome. Thus the idiosome has the appearance of hav-
ing a core. The idiosome and the idiosphere form an elongate
body pointed at its distal end and truncate at its proximal end,
Figs. 13 to 17. Fig. 14 is a surface view while the other figures
show the idiosphere surrounded by the idiosome. The idiosome
and the idiosphere may now be called the acroblast.

The chromatin material in this stage has become finely granu-
lar and can scarcely be distinguished from the cytoplasm either
by its staining reaction or by its structure. The nucleus elongates

240

MARY T. HARMAN AND FRANK P. ROOT.

until it becomes cylindrical, Figs. 14 to 17. A thin coating of
cytoplasm surrounds it and extends in the direction opposite to
the acroblast. Later the nuclear material takes a position to one
side of the cylindrical mass and the cytoplasm forms a flattened
area to the other side extending from the acroblast to the other
end of the cell, Figs. 18 and 19. In these figures the acroblast is
becoming rounded and is beginning to take a position to the side
of the nucleus instead of completely anterior to it as in the earlier
stages. The nuclear material is beginning to become more con-
densed and is spread out over a wider surface. Posterior to the
nucleus there are three fine thread-like filaments which spread
into a somewhat fan-shaped mass in the surrounding cytoplasm.
Associated with these filaments are two areas of cytoplasmic gran-
ules. One area is at the extremity of the filaments and the other
area is near the base of the nucleus, Figs. 19 and 21. Following
this stage, the cytoplasm which is transforming into the tail of the
spermatozoon condenses rapidly and becomes very elongate.

3. Histogcnesis of the Elongate Cell. — In the histogenesis of
the elongate cell the three regions usually recognized in a mam-
malian spermatozoon begin to be evident. At first the nucleus
and the acrosome which make up the head are much longer than
they are wide and become cylindrical and somewhat enlarged at
the free end. The nucleus is now at one side of the cytoplasmic
acrosome and it does not extend entirely to the free end of the
developing spermatozoon. The mid-piece which is occupied
largely by the spiral filament in the adult spermatozoon becomes
granular in regularly arranged clumps, SF, Fig. 23. This is the
region which was occupied by the three thread-like filaments in
Figs. 19, 21, and 22. One of the most noticeable changes is in the
tail region. There is a very rapid condensation of the cytoplasm
which was spread out in a fan-shaped mass to a tapering whip-
like flagellum. The tail is composed of three segments. The first
one is about as long as the mid-piece, the second one in the early
stages is about the same length and the third or terminal one is n.
little longer than the combined length of the other two. It gradu-
ally tapers to a blunt point. We have not found in any stage of
development any unsheathed terminal filament.

As differentiation progresses there is a greater difference be-

DEVELOPMENT OF THE SPERMATOZOON. 24!

tween the sizes of segments one and two of the tail. The second
segment elongates more than the first and tapers more as it in-
creases in length. The segments are recognized by distinct mark-
ings and when the tails of the spermatozoa are broken off, the
break is always at the union of two of these segments. Seldom
is the tail broken from the head at the anterior part of the mid-
piece and practically never is the tail broken off at the posterior
part of the mid-piece. A few of the tails are broken at the end
of the first segment. Most frequently the break is at the distal
end of the first segment, less frequently between the second and
the third segments. We never find the tail broken within any seg-
ment.

Figures 24, 25, and 26 are illustrations of a mature spermatozoon
viewed from different positions. The acrosome forms a hood-
shaped covering to one side and anterior to the nucleus. The head
is broad from side to side, Figs. 24 and 25, but rather thin when
seen from the edge, Fig. 26. The regularly arranged clumps of
cytoplasm in the mid-piece, mentioned above, develop into a dis-
tinct spiral, with the coil always counter-clockwise from the an-
terior part of the mid-piece. The last two coils are almost rings
and might be termed, annulus. There is no annulus separate
from the spiral filament. As is shown in the drawings the coils
are not always regular. They remind one of a spring that has
been put at a tension and the rebound has not been the same in
all regions of the spring. The first four coils of the spiral fila-
ment are inclosed by a thin bladder of cytoplasm.

DISCUSSION.

In the transformation of the spermatid into the spermatozoon
little attention has been given to the behavior of the chromatin
material other than it finally becomes condensed into a more or
less homogeneous mass which appears solid and is stained heav-
ily with nuclear dyes. Meves (1899) has shown the nuclear
material formed into clumps before there has been much change
in the shape of the spermatid. Ballowitz (1891) has also this
clumping of the chromatin material in his drawings. Neither
author has discussed this change nor has mentioned further
changes in the chromatin. They state that the nucleus forms the
greater part of the head of the spermatozoon.

242

MARY T. HARMAN AND FRANK P. ROOT.

In one of the Hemiptera, Bowen (1920) says that "the head
undergoes a characteristic change resulting in what appears to be
a complete vacuolization of the chromatin lining. Then the
chromatin collapses toward the axis of the head, etc."

We have shown that after the last maturation division the
chromatin material passes through changes which are similar to
those in a cell that is getting ready to divide until there is the
breaking up of the chromatin material into clumps. A significant
difference, however, between these changes and the changes previ-
ous to the maturation divisions is that there is no synezesis and
no double thread. We raise the question whether these changes
influence the behavior of the cytoplasm in the process of trans-
formation and thus the attempt at division is aborted or whether
the changes in the cytoplasm arrest the changes taking place in the
nucleus.

The small size of the spermatozoon in comparison with the early
spermatid is recognized by many authors. Some of this differ-
ence in size has been accounted for by a loss in cytoplasm. In the
formation of the spermatozoon of vertebrates, Kolliker as early
as 1856 and la Vallette St. George (1865) described the " slough-
ing off" of the cytoplasm. Later Biondi (1885), Benda (1897),
Hermann (1889), and Neissing (1889 and 1896) agree that there
is a loss in cytoplasm by a sloughing off. Meves and Ballowitz
have shown cytoplasm loosely connected with the transforming
tail part.

This difference in the size of the spermatozoon and the sperma-
tid is recognized in the insects. Montgomery (1911) states that
in Euschistus " no evidence was found for the casting off of any
substance by the sperm."

In the formation of the spermatozoon in Paratctiix, Harman
(1915) did not find any loss of cytoplasm. The cytoplasm con-
densed around the axial filament but there was no indication of a
sloughing off either in the appearance of the cell or the remains
in the follicle.

In our material, the spermatozoon is greatly reduced in size
during the process of transformation, but we have found no evi-
dence in any region of a loss of material. We have shown, Figs.
24, 25 and 26, that a bladder-like structure of cytoplasm is pres-

DEVELOPMENT OF THE SPERMATOZOON. 243

ent in the transformation but that this condenses around a por-
tion of the middle piece and there is no evidence that it is sloughed
off.

Most authors recognize that the greater part of the head of the
spermatozoon is formed from the nucleus of the spermatid and
furthermore, they recognize that this head is much smaller than
the original nucleus. No one has described the loss of nuclear
material. This agrees with our observations. We believe that this
diminution in size is due to a condensation in which the material
appears more compact than in earlier stages.

Meves (1899), Ballowitz (1891), and Duesberg (1910) show
a thread-like filament extending out from the cytoplasm in the
very early stages of development. Meves describes this filament
as arising from one of the centrosomes which gives rise to the
posterior nodule and this filament which in turn becomes the axial
filament. He represents the distal end of this filament as remain-
ing unsheathed and forming the terminal filament. We have
found no unsheathed filament at any stage of development. We
have shown, Figs. 18, 19, 21 and 22, three filamentous structures
which lie deep in the cytoplasm. These filaments are spread out
distally into a fan-shape. Associated with these structures are
two areas of granules. We have not traced the detailed history
of these granules but we have noted that they finally become in-
closed in the cytoplasm which rapidly condenses and with the
associated filaments form the tail of the spermatozoon. There
is a gradual tapering of the tail to a blunt point. This tapering
takes place in the axial filament as well as in the sheath which
encloses it entirely to the distal end.

The tail is made up of three segments as we have shown in Figs.
23, 24, 25, and 26. Early in our study of the mature spermato-
zoon, among mutilated specimens we were impressed with the reg-
ularity of the lengths of the pieces of the tails. These lengths
were quite constant whether the spermatozoa were in bundles,
merely a few together or even if a single spermatozoon was
broken. The pieces were in three different lengths which corre-
sponded to the three segments of the tail. Measurements showed
only a slight variation. It would seem that the tail is weaker at
the points of junctions of the segments.

244 MARY T. HARMAN AND FRANK P. ROOT.

The transformation of the spermatic! into a spermatozoon takes
]»lace in definite areas which are elliptical in shape. The greatest
diameter of the ellipse is always lengthwise of the seminiferous
tubule and the shortest diameter never exceeds two-thirds of the
circumference of the tubule.

SUMMARY.

1. The transformation of the spermatid into a spermatozoon
takes place while the spermatid is closely associated with a Ser-
toli cell and it does not become free in the lumen of the seminif-
erous tubule until the spermatozoon is matured.

2. In the e*arly stages of transformation the cell goes through
a growth period in which the entire cell gets larger and the
chromatin material goes through an abortive preparation as if for
division.

3. During the period of elongation there is a reduction in the
volume of the cell and a rearrangement of its parts.

4. No " sloughing off" or loss of cytoplasm has been observed.

5. The head of the spermatozoon is composed of two parts, the
head proper which arises from the nucleus and the head cap or
acrosome which arises from the idiosome and the idiosphere.

6. There is a cytoplasmic bladder-like structure around the an-
terior part of the mid-piece.

7. The tail is composed of three segments terminating in defi-
nite nodes.

8. We find no indication of an unsheathed terminal filament
either during the transformation or in the mature spermatozoon.

9. The tails of the spermatozoa are always toward the lumen of
the seminiferous tubule.

10. The areas of transformation are elliptical in shape with the
long axis of the ellipse corresponding to the length of the semi-
niferous tubule and the short diameter of the ellipse never exceeds
two thirds the circumference of the tubule.

LITERATURE CITED.
Ballowitz, E.
'86 Zur Lehere von dcr Struktur der Spermatozoen. Anat. Anz.,

Jahrg. i.

'91 \VrikTe Beobachtungen iiber den feineren Bau der Saugethier-sperma-
tozoen. Zeitschr. f. wiss. Zool., Bd. 52.

DEVELOPMENT OF THE SPERMATOZOON. 245

Benda, C.
'87 Untersuchungen iiber den Bau des funktionerenden Samenkanalchens

einiger Saugethiere und Folgerungen fiir die Spermatogenese dieser

Wirbelthiere. Archiv f. mikr. Anal:., Bd. 30.
'97 Neuere Mitheilungen iiber die Histogenese der Saugethiersperma-

tozoen. Verb. d. Physiol. Ges. zu Berlin.
'06 Die Spermiogenese der Marsupialier. Semons Zoologische Forsch-

ungsreisen in Australian. Jena, Fischer.
Biondi, D.
'85 Die Entwicklung der Spermatozoiden. Archiv f. mikr. Anat., Bd.

25-
Bowen, Robert H.

'20 Studies on Insect Spermatogenesis. I. The History of the Cyto-

plasmic Components of the Sperm of Hemiptera. BIOL. BULL.,

Vol. 39-
'22 Studies on Insect Spermatogenesis. II. The Components of the

Spermatid and the Role in the Formation of the Sperm in Hemip-
tera. Jour. Morph., Vol. 37.
'22 Studies on Insect Spermatogenesis. III. On the Structure of the

Nebenkern in the Insect Spermatid and the Origin of Nebenkern

Patterns. BIOL. BULL., Vol. 42.
'24 Studies on Insect Spermatogenesis. VI. Notes on the Formation of

the Sperm in Coleoptera and Aptera, with a General Discussion of

Flagellate Sperms. Jour. Morph., Vol. 39.
'27 Golgi Apparatus and Vacuome. Anat. Rec., Vol. 35.
Duesberg, J.

'08 La spermiogenese chez le rat. Archiv f. Zellforsch., Bd. 2.
'20 Cytoplasmic Structures in the Seminal Epithelium of the Opossum.

Cam. Inst., Washington, Contrib. to Emb. No. 28.
Gatenby, J. B., and Woodger, J. H.

'21 The Cytoplasmic Inclusions of the Germ-cells. Part IX. On the

Origin of the Golgi Apparatus on the Middle-piece of the Ripe

Sperm of Cavia, and the Development of the Acrosome. Quart.

Jour. Micro. Sci., Vol. 65.
Harman, Mary T.

'15 Spermatogenesis in Paratettix. BIOL. BULL., Vol. 29.
Harman, Mary T., and Root, Frank P.
'26 Number and Behavior of the Chromosomes in Cavia cobaya (the

Common Guinea Pig). BIOL. BULL., Vol. Si-
Hermann, F.
'89 Beitrage zur Histologie des Hodens. Archiv f. mikr. Anat., Bd. 34.

Jordan, H. E.

'n The Spermatogenesis of the Opossum (Didclfltys firi/iiiuma) with
Special Reference to the Accessory Chromosome and the Chondrio-
somes. Archiv f. Zellforsch., Bd. 7.
v. Kolliker, A.

'56 Physiologische iiber die Samenfliisigkeit. Zeitschr. f. wiss. Zool.,

Bd. 7.

246 MARY T. HARM AN AND FRANK P. ROOT.

Korff, K. V.
'02 Weitere Beobachtungen iiber das Vorkommen V-formiger Central-

Korper. Anat. Anz., Bd. 19.
Lenhossek, M. V.
'98 Untersuchungen iiber Spermatogenese. Archiv f. mikr. Anat.,

Bd. 51.
Meves, F.
'97 tiber Centralkorper in mannlichen Geschlechtszellen von Schmetter-

lingen. Anat. Anz., Bd. 14.
'98 XJber das Verhalten der Centralkorper bei der Histogenese der

Samenfaden vom Mensch und Ratte. Verh. Anat. Ges., Bd. 14.
'99 Ueber Struktur und Histogenese der Samenfaden des Meerschweinch-

ens. Archiv f. mikr. Anat., Bd. 54.
Montgomery, T. H.

'n The Spermatogenesis of an Hemipteron, Euschistus. Jour. Morph.,

Vol. 22.
Niessing, G.

'89 Untersuchungen iiber die Entwicklung und den feinsten Bau der
Samenfaden einiger Saugethiere. Verh. d. phys. medic. Ges. in
Warzbur N. F., Bd. 52.
'97 Die Betheiligung von Centralkorper und Sphare am Aufbau des

Samenfadens bei Saugethieren. Archiv f. mikr. Anat., Bd. 48.
Oliver, J. R.
'13 The Spermatogenesis of the Pribilof Fur Seal. Am. Jour. Anat.,

Vol. 14.

Papanicolaou, George N., and Stockard, Charles R.
'18 The Development of the Idiosome in the Male Germ-cells of the

Guinea Pig. Am. Jour. Anat., Vol. 24.
v. la Valette, St. George.
'65 Ueber die Genese der Samenkorper. Erste Mittheilung. Archiv f.

mikr. Anat., Bd. I.
Weigl, R.

'12 Vergleichend-zytologesche Untersuchungen iiber den Golgi-Kopschen
Apparat. Bull, de 1'Acad. Scient. Cracovie.

248 MARY T. HARMAN AND FRANK P. ROOT.

EXPLANATION OF PLATES.

All the drawings were made with the aid of a camera lucida, a 1.9 oil-im-
mersion objective and a number 6 compensating ocular at table level. Figure
20 was enlarged two diameters. All other drawings were enlarged four and
one half diameters. The reproductions were reduced one half from the
original.

PLATE I.

FIG. i. Spermatid soon after the last maturation division. G, Golgi
bodies ; N, nucleus ; /, idiosome.

FIG. 2. Spermatid showing increased size. G, Golgi body.

FIG. 3. Spermatid with the chromatin in the form of a spireme. /, idio-
some ; G, Golgi body.

FIGS. 4, 5, AND 6. Spermatid showing an abortive attempt to form
chromosomes. I, idiosome ; G, Golgi body.

BIOLOGICAL BULLETIN, VOL. LV.

PLATE I.

-/v

— /

•

f^-.'S^p' Jfe i":'^v."'-'-'SSp*i

'-,^.

' - VJ

•

IP**

"r-£

-

J

•

MARY T. HARMAN AND FRANK P. ROOT.

17

250

MARY T. HARMAN AND FRANK P. ROOT.

PLATE II.

FIG. 7. Spermatid showing the chromatin finely granular and the begin-
ning of the contracting of the entire cell.

FIG. 8. Spermatid showing the idiosome closely applied to the nucleus
and the appearance of the idiosphere. /, idiosome ; NE, idiosphere ; G, Golgi
body.

FIGS. 9 AND 10. Spermatids showing a great reduction in size. I, idio-
some; NE, idiosphere.

FIGS, ii AND 12. Spermatid showing the ovoid shape which is the be-
ginning of the elongation. /, idiosome; NE, idiosphere; G, Golgi body; C,
cytoplasm.

FIGS. 13 AND 14. Spermatid showing the beginning of the elongation of
the idiosome and the idiosphere. /, idiosome ; NE, idiosphere ; N, nucleus ;
C, cytoplasm; G, Golgi body ; A, acroblast.

FIGS. 15, 16 AND 17. Spermatids showing the elongation of the nucleus
and the spreading out of the cytoplasm in a fan-shape. A, acroblast, N,
nucleus ; C, cytoplasm.

BIOLOGICAL BULLETIN, VOL. LV.

PLATE II.

*"•-.: •"• •

•• • - if

• n \ •/'

'"•'. • V

8

12

\Mf-'"'\

/l/-

1

\. ~ - —{

i

MARY T. HARMAN AND FRANK P. ROOT.

252

MARY T. HARMAN AND FRANK P. ROOT.

PLATE III.

FIGS. -18 AND ig. Spermatids showing appearance of filaments from the
nucleus and the extension of the acroblast to the side of the elongated nu-
cleus. A, acroblast ; N, nucleus ; C, cytoplasm ; F, filaments ; G, Golgi body ;
Gr, granules.

FIG. 20. Sertoli cell with some of the associated spermatids in an elon-
gated form. S, spermatids; N, nucleus; AC, spermatogonial cell.

FIGS. 21 AND 22. Spermatids, a continuation of the development shown
in Figs. 18 and 19. A, acroblast; N, nucleus; F, filaments; C, cytoplasm;
Gr., granules.

BIOLOGICAL BULLETIN, VOL. LV.

PLATE III.

19

L--S

AC-*

lH

, •"-:••''

'

20

...

MARY T. BARMAN AND FRANK P. ROOT.

22

254 MARY T. HARMAN AND FRANK P. ROOT.

PLATE IV.

FIG. 23. Spermatid almost transformed, viewed from one edge. N,
nucleus; SF, spiral filament in formation; A, acrosome; LT, tail segment;
NO, node.

FIGS. 24, 25, AND 26. Mature spermatozoa. Fig. 24 viewed from con-
vex surface, Fig. 25 from side angle and Fig. 26 from edge of head. A,
acrosome ; N, nucleus ; R, residual cytoplasm ; SF, spiral filament ; T, tail ;
NO, node.

BIOLOGICAL BULLETIN, VOL. LV.

PLATE IV.

A

i-T

l-NO

R

-SF

\-NO

MARY T. HARMAN AND FRANK P. ROOT.

STUDIES ON THE SECONDARY SEX CHARACTERS

OF CRAYFISHES. VIII. MODIFIED THIRD

ABDOMINAL APPENDAGES IN MALES

OF CAMBARUS VIRILIS.

C. L. TURNER,
ZOOLOGICAL LABORATORY, NORTHWESTERN UNIVERSITY.

The first and second abdominal appendages of males are ha-
bitually modified in Cambarus for the purpose of copulation.
The individual parts of the first pair of appendages are fused and
twisted and lie compactly in a groove on the ventral side of the
thorax. The appendages of the second abdominal segment re-
semble the typical swimmeret in general plan (Figs. I and 6).
However, the protopodite is elongated and heavier, the basal un-
segmented portion of the endopodite is likewise reinforced, bearing
a conspicuous triangular shoulder. The terminal segmented por-
tion of the endopodite is much reduced. The remaining swimmer -
ets are unusually quite typical.

There is apparently only one published record of a modified
third abdominal appendage. Moenkhaus, Proceedings of the
Indiana Academy of Science, 1903, pp. in and 112, describes a
specimen of Cambarus virilis bearing such a modification as fol-
lows : " The first and second pairs of appendages were modified
in the usual way and in no way differed from corresponding
appendages in the normal male of the same species. The addi-
tionally modified third pair resemble in plan almost exactly the
second pair. The exopod and the segmented flabellum of the
endopod are much less reduced and much more extensively pro-
vided with feathered setae than the second pair. They are about
the same size and in position converge and fit against the sec-
ond pair of appendages much in the same manner that these do
against the first. Whether they were in any way functional I
am, of course, unable to say." Another specimen with a modi-
fication similar to but not so fully developed as the one described
by Moenkhaus was collected by Dr. H. J. Van Cleave of the
University of Illinois and appears in his collection.

255

256 C. L. TURNER.

Since crayfishes are in such common use as laboratory subjects
it seems likely that any considerable occurrence of this aberrancy
would have been noted and described. The writer has examined
thousands of crayfishes during the past seven years, always with
the object of finding peculiarities in the secondary sex characters
and while large numbers of specimens have been found in which
other aberrancies occurred, not one was found with this type of
peculiarity until the lot described came to light.

A collection of several hundred specimens which had been taken
from the Fox River between Green Bay and DePere, Wisconsin,
during the summer of 1927, was being used in the Zoology Lab-
oratory at Northwestern University. A specimen having pecu-
liar appendages was discovered by chance and the writer then ex-
amined the entire lot. Forty-six of a total of three hundred and
forty-two males were found which had third abdominal appendages
modified somewhat like those of the second abdominal appendages.
No other peculiarities were noted among the males, but one fe-
male in seventy possessed a pair of copulatory hooks on the third
walking legs like those of the male. The latter type of aber-
rancy is the most common and it is surprising to find a type that
is apparently rare in greatly superior numbers.

DESCRIPTION OF SPECIMENS.

The male specimens with the modified third abdominal ap-
pendages are about thirteen and a half per cent, of the total num-
ber examined. They range in length from 79 to 107 mm.
Twenty-eight are second form and eighteen are first form males.
A fairly complete series is represented in the aberrant appendages.
In some, the third abdominal appendages varies only in the
presence of a slight projection upon the inner border of the
endopodite between the basal unsegmented and the terminal seg-
mented portions (Figs. 2 and 3), while at the other end of the
series the modifications are practically like those of the second
abdominal appendages (Figs. 4 and 5). There is apparently no
relation between the extent of modifications of the appendages
and the size of the animals. In form I. specimens the angles
upon the shoulder of the aberrant appendages are sharper and
stronger than those of form II., but this might have been expected

STUDIES ON SECONDARY SEX CHARACTERS OF CRAYFISHES. 257

since the same is true of the usual modified appendages in normal
form I. and form II. males.

The first and second abdominal appendages are normal in every
respect in all the specimens.

DISCUSSION.

In attempting to find an explanation for the large occurrence
of a rare aberrancy, age, accidental- embryonic development, effect
of environment or peculiar genetic constitution might be sug-
gested at first thought as causal factors. The fact already noted in
this description that size, and therefore age, and degree of develop-
ment of the peculiarities in the appendages are independent would
seem to eliminate age as a factor. Accident might be called upon
to account for a specimen or two but scarcely for so large a num-
ber as is represented here. It has yet to be shown that environ-
ment has played any part in the development of the secondary
sex characters of crayfishes, nor indeed, in modifying them.

Peculiar genetic constitution seems to be the logical factor here.
It has already been shown for other 'aberrant conditions in sex
characters of crayfishes that there is a strong tendency for the de-
velopment of one type of peculiarity in one locality and the pres-
ent case is another instance of the same tendency. It has been
argued in these other instances that the peculiarity might easily
arise and perpetuate itself as a mutation and the explanation is
again offered for the case in hand.

It does not seem likely that this modification has any functional
significance. Specimens more radically modified in other sex
characters have been functioning normally and there is no reason
to believe that this slight peculiarity would make males any more
efficient nor that it would interfere with copulation.

The series offered in the specimens here, from the slightly modi-
fied to the most completely modified may give a clue as to the
evolutionary changes through which the normal second abdominal
appendages came in the course of their development. This is
speculative, of course, but we have here an actual series ranging
from a practically unmodified third abdominal appendage to one
which almost exactly duplicates the normal second. Unless the
highly peculiar second abdominal appendages arose with all their

258

C. L. TURNER.

STUDIES ON SECONDARY SEX CHARACTERS OF CRAYFISHES. 259

pecularities fully formed in one stage it is easy to believe 'that
they arose through a series of changes such as is represented
here. The first stage would be represented by the development of
a low projection on the inner surface of the endopodite between
the unsegmented basal portion and the segmented terminal por-
tion. Subsequent changes would involve an enlargement of this
spur and a molding of it until it had assumed the shape found in
the normal second appendage of the male. Other changes would
involve an elongation and an enlargement of the propodite, and an
enlargement and a strengthening of the basal portion of the endo-
podite together with a reduction of the terminal segmented portion
of the endopodite.

In aberrant females having first abdominal appendages modified
like those of males the second abdominal appendages are also
sometimes modified. Such aberrant females are rare but even
in a small number various degrees of modification are shown in
the second appendages. These second abdominal appendages are
identical in their structural peculiarities with the third abdominal
appendages described here and are similar also in that they show
various stages of development.

EXPLANATION OF FIGURES.

Note: All figures are drawn to the same scale.

FIG. i. Unmodified left third abdominal appendage of normal male.

FIG. 2. Left third abdominal appendage of aberrant male measuring
88 mm. Callosity on endopodite is showing first stage of development.

FIG. 3. Left third abdominal appendage of aberrant male measuring
82 mm. Callosity on endopodite much larger.

FIG. 4. Right third abdominal appendage of aberrant male measuring
80 mm. The protopodite is longer, the basal portion of the endopodite
elongated and the shoulder upon the endopodite is more prominent.

FIG. 5. Left third abdominal appendage of aberrant male 103 mm. in
length. Modifications almost equal to those of the normal second abdom-
inal appendage.

FIG. 6. Left second abdominal appendage of normal male measuring
98 mm.

18

NATURAL HISTORY OF SHIPWORM, TEREDO NAVA-
LIS, AT WOODS HOLE, MASSACHUSETTS.

B. H. GRAVE,
WABASH COLLEGE.

From the Marine Biological Laboratory, Woods Hole, Mass.

SECTION I. OCCURRENCE.

The common species of shipworm at Woods Hole, as identi-
fied by Kofoid and Clapp, is Teredo navalis. The date of its
first appearance in this region is not known. Verrill lists it in
his ' Invertebrate Animals of Vineyard Sound and Adjacent
Waters " (1871). Whatever its history in American waters may
have been, it is now known to occur throughout the entire North
American coast from Alaska to Labrador.1 The present study
has been carried on during the past four years and in that time no
other species has been collected. It is known, however, that
Utinkia fiinbriata occurs in this region, although in comparatively
small numbers. During the year this work was first undertaken
it was difficult to obtain Teredo in sufficient numbers for satis-
factory study, but this is not an indication that the species is not
abundant in New England waters. The reason for an apparent
scarcity is that shipworms are inaccessible, being, for the most
part imbedded in piles and permanent structures. Subsequently,
by putting out suitable timbers during one summer to be studied
the next, it has been an easy matter to obtain Teredo in abundance.
Lobster pots2 and 2X4 stakes have been found to be the most
convenient. If these timbers are exposed to the water during
the latter part of the summer they are found to contain sex-
ually mature worms by the beginning of the breeding season the
following June. The 2X4 stakes give best results if exposed

Nelson, '22, speaks of an infestation of Teredo navalis in Barnagat
Bay, New Jersey, as a sudden outbreak. He is probably in error in think-
ing that this species arrived so recently on the New England coast.

2 Lobster pots are constructed of small slats about the size of ordinary
plasterer's lath, 2 in. broad and y2 in. in thickness.

260

NATURAL HISTORY OF SHIPWORM. 26!

during July or early August, but the smaller timbers are liable
to complete destruction before winter if put out early in the
summer.

Teredo do not grow large in small timbers such as are used
in the construction of lobster pots, but are easily removed from
such small strips of wood, thereby facilitating study. The size
attained depends upon the degree of crowding. To ascertain the
size to which Teredo will grow, it is necessary to supply larger
pieces of wood and 2X4 stakes are excellent for the purpose.
With a drawing knife it is possible to expose the entire burrow
in a few minutes because Teredo tunnels with the grain of the
wood, usually within half an inch of the surface. A study of such
stakes has shown that Teredo larvae attack the wood in great
numbers at the mud line but less and less abundantly from the
bottom to the surface of the water. Three fourths of the Teredo
burrows in an exposed timber occur within two or three feet of
the mud line. Very few are found more than four feet above
the bottom.

SECTION II. ANATOMY, PHYSIOLOGY AND BEHAVIOR.

The anatomy of Teredo has been accurately described by sev-
eral early investigators and more recently the shell and digestive
tract have received attention by Miller and Lazier, whose ad-
mirable work is published in four papers. It is sufficient here to
say that the shipworm has the structure of an ordinary lamelli-
branch in which the body is much elongated and in which the
bivalve shell is highly modified in adaptation to the burrowing
habit. In one particular my observations are not in agreement
with those of Miller. He attributes the formation of the rings
of growth, the rasping ridges, and denticles of the shell to altera-
tion or fluctuation in the food supply which, according to his
conception, results in corresponding periods of slow and rapid
growth. This may account for the annual rings of growth of
certain mollusks and has been so interpreted, but it could hardly
account for the rings and ridges on the shell of this young ani-
mal which adds two rings per week in the early stages of its de-
velopment. These sculpturings of the shell which adapt it to
burrowing are undoubtedly due to the action of little tongues

262 B. H. GRAVE.

of mantle tissue which are pushed up over the edge of the shell
during deposition of the shell material. This process of shell
sculpturing was observed in the large lamellibranch Atrina
rigida (Grave, '09). The peculiar form and pattern of the shell
is specific and is a matter of inheritance, but the building process
is due to the peculiar manipulation of the mantle and not to
alternate periods of starvation and plenty.

The physiology of digestion has been studied particularly in
recent years by Dore, Miller and Potts.

Potts ('24) corroborates the work of Dore and Miller ('22) in
showing that as the shipworm burrows through the wood it swal-
lows the chips and derives some nourishment from them. A
large section of the digestive tract seems to be devoted entirely
to the digestion of wood (the caecum and liver). Potts believes
that wood is the only food of Teredo but Miller shows that the
digestive tract contains diatoms as well as wood. The burrow
mainly serves as a means of protection.

As the Teredo grows it enlarges its burrow proportionately
until at maturity it may be 16 inches in length and have a diam-
eter of Y% of an inch (40 X i cm.). A pile or other exposed
piece of timber may be honeycombed with Teredo tunnels with-
out showing on the surface that it is infested. The only open-
ing of the burrow leading to the outside is the minute pore
through which the young Teredo entered the wood as a meta-
morphosing veliger. Although less than .35 mm. in diameter
and therefore too small to be seen readily by the unaided eye, it is
through this passage that the siphons are protruded to obtain
respiratory currents and food other than wood. The shipworm
feeds upon minute organisms derived from water currents that
pass over its gills for respiration, just as in ordinary lamelli-
branchs. It is in fact an elongated lamellibranch, whose bur-
rowing shell covers only its anterior tip, leaving most of the body
and the siphons unprotected except for the wooden shell-lined
burrow.

CHARACTER OF THE BURROW.

The burrows are always lined by a calcareous substance, except
at the anterior end, where further excavation is taking place.

NATURAL HISTORY OF SHIPWORM.

263

This shell-like material is secreted by the general surface of the
body or mantle. It has been suggested that this lining of the
burow not only makes a smooth surface, but shuts out wood
acids as well as external enemies which might otherwise injure
the soft body of the animal. Even the outer pore-like opening
is lined with this secretion and is divided transversely by a par-
tition, so that the siphons protrude through two minute pores just
large enough to transmit them. While the shipworm is not feed-
ing, or when it is disturbed, the siphons are withdrawn and the
external openings are plugged by two curious horny pallets, as
they are called, situated one on each side of the siphonal region.
See Figure I.

s s

FIG. i. Young Teredo, length 2 cm., age five weeks from metamorphosis;
drawn by camera lucida. S shell, F foot, i. s. incurrent siphon, e. s. ex-
current siphon, p. pallet.

Effect of Adverse Conditions. (Repairing the burrow, etc.)
In case the tunnel is broken by accident, or by the wearing
away of the surface of the wood from any cause, the adjacent
glands secrete shell substance in greater abundance and mend the
breach. The integrity of the burrow is carefully preserved. In
case adverse conditions arise which make the environment diffi-
cult either from enemies or poisons in the water, or from over-
population by its fellows, this shell substance is secreted in the
form of a heavy casing, not only on the sides, but over the an-
terior burrowing end as well. This is the invariable reaction of

264 B. H. GRAVE.

Teredo to adverse external conditions, the most common cause
of which is the crowding of individuals in small timbers. As a
consequence, the wood becomes extremely fragile, a mere shell,
so porous that enemies, such as bacteria and parasitic protozoa,
find entrance and menace the life of the community. Under
these conditions the worms die within the first year. It may be,
too, that wood is an essential part of their diet, but it is more
probable that the trouble is a lack of adequate protection against
adverse conditions and dangers from without.

No Teredo ever molests the burrow of another. When two
come close together they face about and proceed in another di-
rection, thus avoiding each other. When they become so closely
crowded that further burrowing would infringe upon a neighbor,
growth seems to stop. The size attained depends upon the
amount of crowding. As stated above, the Teredo responds 'to
these conditions by greatly thickening the shell lining of its bur-
row on the front as well as on the sides so that the whole is
strongly encased. However, it is at best a brittle affair and para-
sitic protozoa and bacteria are admitted which soon destroy the
occupant. The protozoan Architophrya (a holotrich) is always
abundant in such situations.

It is difficult to see how growth may cease and the animal sur-
vive, but it is perfectly clear that Teredo three months old living
in crowded situations are often less than one fifth as large as
others of the same age growing under better conditions. The
stunted worms, though packed closely together are frequently
all alive and reproducing. As many as seven young Teredo per
square inch have been observed in test blocks although the aver-
age is by no means so high. When these worms all become two
or three inches long, a crowded group results unless they hap-
pened to have entered a large timber which permits of unlimited
expansion.

Shipworms rarely go from one board to another, no matter how
closely the boards are applied to each other. Only two exceptions
to this rule have been observed among the thousands of burrows
studied. They seem to avoid anything that threatens to interrupt
the continuity of their tunnels.

Teredo seems not to orient to gravity since it burrows down-

NATURAL HISTORY OF SHIPWORM. 265

ward about as frequently as upward. The burrow of a single
individual often shows that there is no tropistic response of this
kind. If in tunneling downward a Teredo approaches the end of
the timber, another Teredo burrow or a knot, it may turn directly
about and proceed in the opposite direction, paralleling the first
part of its burrow. By some means it is able to detect any nearby
surface of the wood and avoid it. Two Teredo tunnels may ap-
proach within an eighth of an inch of each other, but they remain
quite separate. They have some sense also which warns them,
when approaching the end of a timber, to face about before reach-
ing the end, retreating usually at a point 5 to 10 mm. from the
tip.

SECTION III. THE BREEDING SEASON.

My interest in Teredo dates from 1922 when the National Re-
search Council suggested the study of the breeding season of this
species and appropriated funds to meet preliminary expenses.
The results of this study were reported at the Washington meet-
ing of the American Association for the Advancement of Science
in 1924, and an abstract was printed at that time. The publication
of the paper as a whole was deferred until the study of various
details could be completed.

The fact that the female carries the young embryos in the gill
chamber for a short time makes an accurate study of the breed-
ing habits a comparatively easy matter. It may be ascertained at
a glance whether a female is carrying embryos or not and the
presence of eggs or embryos in the suprabranchial chamber is
conclusive evidence of recent spawning. A further useful indi-
cator is that of color. The eggs and young embryos are pure
white, but they gradually take on a dark gray color with age.

The first spawning at Woods Hole occurs from the first to the
middle of May, and the last about the middle of October. Dur-
ing 1925 eggs were first obtained on May 15 and these were in
a late cleavage stage when discovered. Two of twenty females
examined had spawned at this date. In 1926 eggs were first ob-
tained on May 16. Two of the twelve females examined had
spawned, and the embryos were in the gastrula stage of develop-
ment. Frequent previous examinations in April and May had
shown no spawning individuals.

266 B. H. GRAVE.

During the fall of 1925 and 1926 special trips were made to
Woods Hole in order to determine the extreme limits of the
breeding season. At this time an effort was also made to learn
how late in the fall veligers were metamorphosing and entering
wood. On September 22, 1926, numerous females, both in Eel
Pond and at the Cayadetta Wharf in Vineyard Sound, were carry-
ing embryos in various stages of development. On October 10,
of sixty Teredos taken from Eel Pond, none were carrying em-
bryos, while five of twenty five taken from the Sound had
quantities of veligers in their gills. The embryos of one of these
were late trochophores or early veligers and repeated observa-
tion on the rate of development in Teredo has shown that these
would normally be carried from ten days to two weeks longer.
None were found carrying embryos on November 4. These and
other data show that the breeding season in Eel Pond ended two
weeks earlier than in Vineyard Sound. The difference in tempera-
ture is apparently the cause of this diversity in 'the duration of
the breeding season, Eel Pond being approximately two degrees
colder during the fall than the deeper water of the Sound.
Kofoid noted a similar difference in the breeding season in vari-
ous parts of San Francisco Bay where wide stretches of shallow
water become several degrees warmer in early spring and cooler
in the fall than the deeper portions of the same body of water.
His estimate of two weeks difference is no doubt conservative.
Observations just completed at this writing show that the first
spawning by Teredo in Eel Pond in 1927 occurred on May I and
in Vineyard Sound on May 12. Spawning occurred in each case
when the water had reached a temperature of approximately 11°
C. (between 11° and 12° C.). Since spawning ceased in Ed
Pond on October i and in Vineyard Sound about October 15 we
have the same variation due to temperature difference and the
total spawning season for Teredo at Woods Hole is shown to
be nearly or quite five months in duration.

It should be explained that the larva has a free swimming
period of approximately two weeks after leaving the supra-
branchial chamber of the mother before it is ready to enter wood.
In accordance with the fact that veligers are carried by the mother
as late as October 20 in Vineyard Sound, one would expect to

NATURAL HISTORY OF SHIPWORM. 267

find that wooden structures are being entered by the metamorphos-
ing veligers until the first of November. The facts, however,
do not bear out this expectation. The last date on which veligers
successfully metamorphosed and attacked wood in Eel Pond was
September 23, whereas larvae were no doubt present until about
October 5. Lobster pots placed in Vineyard Sound on October
10 were entered by metamorphosing veligers. It is certain that
larva? are present in the water in Vineyard Sound until No-
vember i or the last week in October. In other words, larva? are
present in the water at least two weeks after the last ones suc-
cessfully attack wood. The reason for this is not evident. The
cilia of the swimming mechanism of the larva possibly become
less and less active as the water cools, with the result that mor-
tality among the last generation of larvae of the season is high.
In Bugula also the last larva? of the season fail to metamorphose,
but not to so great an extent as is the case with Teredo.

An examination of the gills of a large number of Teredo on
November 4 showed a spotting of these organs as if the last em-
bryos contained had been resorbed. It is quite likely that the
belated ones lose ability to swim and therefore remain inactive
and disintegrate in the gill chamber. (This may not be the correct
explanation of the cause of the failure of the last embryos of the
season to metamorphose.) The larva? of Bugula and those of cer-
tain hydroids continue to metamorphose successfully into No-
vember and the latter into December although dependent upon
cilia for locomotion.

The data in hand indicate that the breeding season of Teredo
at Woods Hole extends from about May 10 to October 10 or
possibly to October 15, a period of five months.

Fecundity.

Teredo is tremendously prolific. Each female spawns three or
four times in a season. The number of eggs produced varies
with the size of the individual and is estimated to be from one to
five millions. At the end of the season the female seems to be
exhausted. Many molluscs survive for several years but Teredo
dies during the second year as test blocks have shown repeatedly.
This unusual fecundity may explain the early loss of vitality.

268 B. H. GRAVE.

As evidence that the female Teredo spawns every four or five
weeks, the following data are offered. Several cases of this kind
were observed.

June 20, 1925. Two large females which were carrying gray
vi-ligcrs, were ready to spawn a second time. The ovaries were
large and distended with eggs which were full size and fertilizable.

June 24, 1925. Two among several females examined had
spawned a second time this season, numerous late veligers mixed
with cleaving eggs were found in the suprabranchial chamber.

Periodicity.

One of the specific objects of this study was to ascertain the
characteristics of the breeding season, whether or not there is a
lunar or other periodicity in the production or shedding of the
gametes. It was made apparent during the first year's study that
no lunar periodicity occurs in the spawning of Teredo. From the
beginning to the end of the breeding season, the water contains
abundant larva? in all stages of development. The records of
examinations of hundreds of stakes and lobster pots indicate that
larvae are abundant in the water ready to attack any exposed
timber each day of the summer. The evidence bearing on this
point is derived from two types of experiments which are here
described in some detail because other workers have stated that
the spawning of Teredo is periodic and that definite broods ma-
ture at definite times.

ist. The following tables show that no periodicity in the
spawning by this species occurs. Of a large number of ship
worms that may be examined at any time during the summer,
some will be found to carry cleaving eggs, some gastrulae, and
some trochophores, some young veligers and some typical veligers,
thus showing that spawning is continuous and not synchronous.

TABLE i.

Teredo EXAMINATIONS 1925, JULY i.
Material frnin Lobster I'ot I'laeed in U'aler .-]»</. 16, 1924. Eel Pond.

!• i-males carrying unspawnecl eggs 4

Females carrying cleaving eggs in gill chamber i

Females carrying young veligers in gill chamber i

I •(•males carrying typical veligers in gill chamber 4

Mature males with active sperm 7

Total !

NATURAL HISTORY OF SHIPNVOKM. 269

TABLE 2.

Teredo EXAMINATIONS 1924, JULY 5.
Material from Lobster Pot Placed in Water Aug. 20, 1923. Cayadctta Dock.

Females carrying mature eggs 5

Females carrying immature eggs I

Females carrying cleaving eggs in gill chamber 7

Females carrying gastrulae in gill chamber 5

Females carrying young veligers in gill chamber 4

Females carrying typical veligers in gill chamber 5

Mature males having motile sperm 7

Immature males 3

Total 37

TABLE 3.

Teredo EXAMINATIONS 1924, JULY 19.
Material from Lobster Pot Placed in Water Aug. 20, 1923. Cayadctta Dock.

Females carrying eggs 8

Females carrying cleaving eggs in the gill chamber 2

Females carrying gastrulae in the gill chamber n

Females carrying young veligers in the gill chamber 3

Females carrying typical veligers in the gill chamber 5

Mature males with active sperm 5

Total 34

TABLE 4.

Teredo EXAMINATIONS 1924, AUG. 10.
Material from Lobster Pot Placed in Water Aug. 20, 1923. Eel Pond.

Females carrying eggs I

Females carrying cleaving eggs in the gill chambers i

Females carrying blastulaa or gastrulae 4

Females carrying early veligers 2

Females carrying typical veligers 5

Females carrying a few veligers in the gill chambers, apparently

spent i

Mature males with abundant active sperm 12

Total 26

These four tables show that spawning takes place at all times
during the month and not synchronously. They show conclu-
sively that there is no lunar or other periodicity such as that
sometimes caused by variations of temperature. Attention is

27O B. H. GRAVE.

called to the fact that the spawning of these animals took place
not in the laboratory, but normally in their natural habitat.

It is also apparent from these tables that there are no " broods "
or special times of infestation of exposed timber. As further
evidence on this point the test blocks (lobster pots) were put out
every ten days during the summer and all became infested with
metamorphosing Teredo larvae almost at once, certainly within a
day or two after exposure, as numerous experiments on rate of
growth show. At Woods Hole the first larvae settle and begin to
burrow toward the end of June (June 20). From that time on
until early fall the water contains a copious supply of swimming
larvae ready to burrow into any exposed wooden structure.

T. C. Nelson in his report for the year 1923, Table 5, page
208, concludes on very meager and insufficient data that one brood
of larvae settled in Barnegat Bay in June and that a second brood
matured some time between July 26 and September 4. The evi-
dence derived from my experiments covering four years show
that there are no broods but rather a continuous entrance of
timbers by larvae maturing throughout the breeding season. The
evidence of many experiments shows that one can not depend
upon green timber or even seasoned 2X4 stakes for such ex-
periments, as they may remain uninfested for weeks for no ap-
parent reason. Seasoned lobster pots, however, regularly became
infested either the day they were exposed to the water or very
soon thereafter. This is possibly due to the horizontal position
of the timbers in the water, as contrasted with stakes standing
vertically. The answer to the question whether Teredo larvae
enter wooden structures in broods at special times or continu-
ously has important practical bearings as well as scientific interest.

It is also apparent from the data of these tables that Nelson's
statement that there are five hundred females to one male, does
not hold for the Woods Hole region. Females outnumber males
but by no means to so great an extent.

Kofoid has shown that the number of larvae in any particular
region depends upon the extent to which infested timber is pres-
ent. Regions far from wooden warves have relatively few
larvae in the water. I was able to show that Teredo is much
more abundant at the Cayadetta Wharf than in Eel Pond, the

NATURAL HISTORY OF SHIPWORM. 271

ratio being approximately 2:1. The distance between these lo-
cations is less than one hundred yards and the difference in
numbers in this case is not due to a difference in the amount of
wood present. The biological conditions in the more or less iso-
lated Eel Pond are clearly different from those of the open waters
of Vineyard Sound because species inhabiting them are differ-
ent to some extent, as shown in another paper (See Bugula).1 A
study of these conditions is contemplated but at the present no
adequate explanation is suggested unless the large amounts of
formalin and other poisons and oils from the supply station seri-
ously affect the Eel Pond water at times. There are, however,
differences in natural conditions. The tidal currents outside, at
any rate, are much stronger than those in Eel Pond.

SECTION IV. EMBRYOLOGY, AND RATE OF DEVELOPMENT.

The extensive contributions of Sigerfoos and Hatschek give
satisfactory descriptions of embryological development so that I
shall avoid duplication and emphasize only facts that are new.

The egg of Teredo is comparatively small and white in color.
It measures in extreme limits from .050 mm. to .061 mm. with
an average diameter midway between these figures. The oviducts
open into the suprabranchial chambers which are extensive and
serve as brood pouches. When the eggs are extruded they are
retained in the suprabranchial chambers for a period of two or
three weeks, during which time they pass through the early
stages of development. When liberated into the sea water they
are typical lamellibranch veligers, vigorous and hardy. A large fe-
male may liberate from 500,000 to 1,000,000 eggs at a single spawn-
ing, so that the gill chambers are tightly packed with embryos
distributed in two parallel rows along the sides of the slender
elongated body. The approximate age of embryos can be esti-
mated by their color since they gradually change from white to
a dark muddy gray during development.

The embryo is not parasitic upon the mother, but the egg will
not develop outside the gill chamber. Ripe eggs were several
times removed from the gonads and artificially fertilized in an

1 Bugula flabcllcta lives readily in Eel Pond but will not thrive in the
adjacent waters of Vineyard Sound, while the reverse is the case with B.
turrita.

272

B. H. GRAVE.

attempt to observe them in development. Development was
initiated but no egg cleaved beyond the sixteen cell stage, and
many stopped at the two, four, and eight cell stages. Develop-
ment in these cases was extremely slow and cleavage was ir-
regular and abnormal. Eggs fertilized at six P.M. had reached
the eight cell stage at 9 P.M. It is probable that development in
this species is normally slow, but this rate can hardly be con-
sidered normal. Very young embryos in the two and four cell
stages were several times found in the suprabranchial chambers
and these when removed developed no better than the artificially
fertilized eggs. In common with artificially fertilized eggs, they
finally became viscid and adhered to the containing dish. It was
found also that blastulae and gastrulse would fare no better.
They failed to develop into swimming larvae. Late trochophores
and early veligers on the other hand continued to develop nor-
mally when removed from the gill- chamber to sea water. Veligers
removed prematurely showed great vigor and swimming ability,
and were several times kept for two weeks in sea water. Velig-
ers withstand much rough treatment and survive in poorly aerated
and even foul water. Some were kept in glass aquaria and fed
on diatoms for three weeks, but to what extent they meta-
morphosed and entered the wood that was provided was not
learned.

The gastrula is invaginate, similar to that of many other mol-
luscs and annelids that produce small eggs with little yolk. The
trochophore is especially interesting because in adaptation to its
parasitic mode of life.it fails to develop a strong protoroch. The
cells which normally develop this larval swimming organ are
undoubtedly present and distributed in a broad equatorial band
similar to that of many molluscs, and they are more extensive
than in most annelids. The protroeh is apparently present and
was described by Hatscheck. I found it either absent or so feebly
developed as to be easily overlooked. The trochophore is pear
shaped or slightly elongated and on the average measures
.059 X -060 mm. in length. As it begins to transform into the
veliger, strong cilia develop on the velum, and the embryo be-
comes motile long before it is ready to be expelled into the sea
water to shift for itself.

NATURAL HISTORY OF SHIPWOR.M. 2J3

Duration of Larval Period.

Sigerfoos failed to find free swimming vdigers in the water
and both he and Nelson speak of the habits and duration of the
larva as being unknown. The larval period from fertilization to
metamorphosis has usually been estimated at about one month.
It is evident, however, that it varies somewhat with temperature
being shorter in tropical and sub-tropical regions than at \Yoods
Hole.

I have frequently found Teredo veligers, in various stages of
development, settling upon horizontally placed boards and Nelson
has more recently taken them in " tow," as well as hovering about
piles ready to settle permanently. In fact, he corroborates the
observations of Harrington that the mature veligers of Teredo
are attracted to wood and wood extracts. The duration of the
free swimming period has never been accurately determined. To
give attention to this phase of the life history publication of this
paper has been delayed until its study could be completed and
verified. The evidence now at hand indicates that the entire
developmental period from egg to metamorphosing larva, is
about five weeks. At least half of this time is required for de-
velopment in the gill of the mother, leaving for the free-swim-
ming period not to exceed two or three weeks. The evidence on
which this conclusion is based is derived in various ways but is
indirect. Since the method and conclusion may be questioned,
the data are explained in considerable detail in the following pages.

In 1925 the first eggs were laid between May 12 and May 15,
while the first young metamorphosed Teredo were found in test
blocks on July 2 and July 5. These young, metamorphosed ship-
worms measured .35 mm. to .5 mm. Evidence collected from
many experiments carried out during the past two years shows
that young Teredo of this size have spent from 15 to 18 days in
the wood, or rather, that they settled and began to metamorphose
and burrow 15 to 18 days previously. If we subtract 15 days
from July 2 or 18 days from July 5, June 17 is the approximate
date when these Teredo ended their careers as free swimming
larvae. Other young Teredo collected on July 7 measured I mm.
and these are known to be three weeks old, or that three weeks

274

B. H. GRAVE.

had elapsed since they settled upon wood. Subtracting twenty-
one days from July 7 gives the date June 17 when metamorphosis
began. The total larval period is therefore between four and five
weeks. Other data collected during 1925 lead to the belief that
the time is more nearly five weeks than four. This method,
though indirect, is accurate, and was repeated many times at the
opening of the breeding seasons of 1925 and 1926. The experi-
ments made to determine the rate of the metamorphosis and rate
of early growth were also repeated many times during the past
two years and are also reliable. They show conclusively that
young Teredo which measure one millimeter are approximately
three weeks old. The spring of 1925 opened unusually warm,
although the winter was severe, so that animals came out of win-
ter hibernation a few days earlier than usual. The effect of
this was shown most markedly in the rapid somatic growth of
many animals, but it also affected to a slight extent the breeding
seasons of most animals. The date of first settling of Teredo
iarvse at Woods Hole is usually about June 20, and the first
spawning about May 10. The variation in the spawning season
from year to year does not usually exceed two or three days but
it may vary more than a week. There is evidence that some ani-
mals begin to breed only when the water rises to a certain tem-
perature. This, however, is by no means a universal rule.

Rate of Groivth.

The veliger of Teredo has the typical form common to lamilli-
branch larvae, but is not so thick or nearly spherical as sometimes
described. Young veligers taken from the gills in an early stage
of development measure on the average .065 x .080 mm. Five
specimens taken from two individuals measured as follows :
.060 x .080 mm. ; .065 x .080 mm. ; .065 x .082 mm. ; .070 x .083
mm. ; .070 x .085 mm. These measurements represent the range
of variation in length and breadth. One of these seen in edge
view measured .082 x .05 mm., and an older one .09 x .05 mm.
Veligers ready to begin their free swimming life, after spending
two or three weeks in the gill of the mother, measure some-
what larger, as the following examples show : .070 x .090 mm. ;
.072 x .090 mm. ; .075 x .085 mm. ; .075 x .088. These measure-

NATURAL HISTORY OF SHIPWORM. 275

ments not only indicate some growth but also that a considerable
variation exists in the relative measurements. The range of
variation in ratio of length to breadth and also in length of hinge
line is great. Veligers fed upon diatoms for one week meas-
ured from .077 x .090 mm. to .081 x .093 mm. The size at- ( ^ I B ft A R Via
tained at the time of metamorphosis was not learned, but Nelson
('23) gives it as .25 mm. in length. I have collected several
hundred young metamorphosing Teredo, which had burrowed
into wood, varying in age from two to three weeks after settling.
These range in size from .35 mm. to I mm. The smaller ones
in two weeks have almost completed metamorphosis and have
from two to three rasping ridges or rings of growth on the shell.
Individuals three weeks old have four rings of growth and a
typical Teredo shell. At three weeks of age the Teredo is prac-
tically spherical and its burrow, when exposed by cutting away
the surface of the wood, is a hemispherical pit. The young worm
now begins to elongate rapidly and at the end of one month its
burrow measures from 5 to 7 mm. in length, and has a diam-
eter of 2 to 2.5 mm. The shipworm when expanded fills its bur-
row so that, in measuring the rate of growth, the size of the
burow may be taken as the correct measure of the enclosed worm.
When the shipworm is removed from its burrow, it contracts to
one half or two thirds of its expanded measure. Tables 5 and 6
show the rate of growth from the egg to adult size, and need
not be described in detail. Measurements were taken every three
or four days, and the rate of increase in size from day .to day
was found to be surprisingly rapid.

It should be noted that the ages given in the tables include only
the time that elapsed from the time of settling. If the age from
the egg is desired, about thirty-five days should be added to these
figures to include the time from fertilization to the end of the
free swimming period. The larval period is excluded in the fol-
lowing description and from the tables.

Growth during the first twenty-five days seems small but when
the minute size of the animal at the beginning is taken into ac-
count, the growth is not slow. From one month to five months
the increases shown during the intervals of three or four days,
between measurements, are seen to be remarkably great. For
19

276

B. H. GRAVE.

TABLE 5.

RATE OF GROWTH OF Teredo navalis (SUMMER).
Measurements of the Largest Burrows of Specified Ages.

Date and
Period of
Growth.

July 16 to July 26. . . .
July 16 to Aug. 3. . .
July 26 to Aug. 13 . . .
July 16 to Aug. 8. . . .
July 16 to Aug. ii..
July 26 to Aug. 20. ..
July 16 to Aug. 13. ..
July 16 to Aug. 16. . .
July 16 to Aug. 18. . .
July 16 to Aug. 23. ..
July 16 to Aug. 28. ..
July 16 to Sept. i. . . .
June 20 to Aug. 20. . .

July 3 to Sept. 6

June 22 to Sept. 3. . . .
June 22 to Oct. 20. ..
June 22 to Dec. i . . . .
July 1923 to July 1924

Age.

10 days

18 "

18 "

23 "

25 "

25 "

28 "

30 "
33

38 "
43

46 "

60 "

65 "

72 "

130 "

160

One year

Size of Burrows in
Length and Widest
Diameter (Metric).

No visible Teredo burrows.
No visible Teredo burrows.
.35 to .5 ram. x .35 to .5 mm.
.35 to .5 mm. x .35 to .5 mm.
.5 to .1 mm. x .5 to i mm.

1 to 1.5 mm. x i to 1.5 mm.

2 to 3 mm. x i to 2 mm.

5 to 7 mm. x 2 to 2.5 mm.
8 to 10 mm. x 2 to 3 mm.
14 to 17 mm. x 3 mm.
35 to 45 mm. x 4 mm.
50 to 57 mm. x 4 to 4.5 mm.
70 to 75 mm. x 4 to 4.5 mm.
80 to 90 mm. x 4 to 4.5 mm.
100 to 120 mm. x 4.4 to 4.8 mm.
140 to 170 mm. x 6.5 to 7 mm.
175 to 200 mm. x 7 to 7.5 mm.
250 to 400 mm. x 7.8 to 9.4 mm.

Approximate

Length in

Inches.

.014 to .02 in.
.014 to .02 in.
.02 to .04 in.
.04 to .06 in.
.08 to .12 in.
.2 to .28 in.
.32 to .4 in.
.56 to .7 in.
1.4 to 1.8 in.
2 to 2.3 in.
2.8 to 3 in.
3.2 to 3.6 in.
4 to 4.8 in.
5.6 to 6.8 in.
7 to 8 in.
10 to 16 in. x
to f in.

Note: — The left hand columns of Tables 5 and 6 represent the time of
exposure of timber to the sea water. Thus, if we consider the first item
of Table 5, timbers were exposed to sea water on July 16 and examined
for Teredo on July 26, making ten days as the maximum age of the infest-
ing Teredo as indicated in the second column of the table. Column three
gives the measurements of the infesting Teredo if any.

The measurements given in these tables are actual cases and not aver-
ages. Many more were measured than are given here but the data given
are considered typical.

In all cases the larval period is omitted. The age from fertilization
may be approximated by adding thirty-five days to the age as given here.

Timbers exposed forty-three days, July 16 to August 28 contained sex-
ually mature worms. Other similar data show that Teredo under favor-
able conditions becomes sexually mature in six weeks at Woods Hole.
The item second from the bottom of Table 5 shows that Teredo does not
reach adult size during the first season (June 23 to December i). No
growth takes place after December i. (See also Table 6.) The last item
of the table shows that adult size is attained in one year (July 1923 to
July 1924). The largest specimen found at Woods Hole measured forty
centimeters in length. It was precisely one year old.

example, the length of the largest burrows at twenty-five days is
1.5 mm., at thirty days 5 to 7 mm., at thirty-eight days 14 to 17
mm., at forty-three days 35 to 45 mm. Davenport claims thai

NATURAL HISTORY OF SHIPWORM. 277

growth in size is partly due to swelling by the absorption of water.
It is easier to account for this phenomenal growth in this way than
to suppose that the change in size represents only protoplasmic
growth and actual cell multiplication.

It has been repeatedly found that Teredo navalis at Woods Hole
leaches sexual maturity and spawns from six to eight weeks after
entering the wood as a metamorphosing larva. The youngest
to spawn were six weeks old, and their burrow measured one
and one half inches (38 mm.) in length. They spawned in
abundance at the age of two months when they measured 2-2^/2
inches (50 to 63 mm.) in length. Sexual maturity is reached
long before adult size is attained, since a fully developed ship-
worm measures from 12 to 16 inches (30 to 40 cm.) in length.

The larvse that metamorphose first in a season almost reach
adult size by December I, the largest ones, measuring from seven
to nine inches (17.5 to 22.5 cm.) in length. During December,
January, February and early March ship worms are practically
dormant and do not grow perceptibly. Then they may be said
to awaken and by the first of July the oldest have reached maxi-
mum size. The largest specimen found at Woods Hole meas-
ured 16 inches in length and % inch in widest diameter
(40x1 cm.). It grew in a 2x4 test take which was in the
water from July i, 1923 to July i, 1924. Others measuring from
12-15 inches are common. Larvse which enter the wood later
in the summer, even to October i, lie dormant over winter and
resume growth the following spring. It was shown that growth
is greatly retarded after the first of November, especially on the
part of the youngest Teredo. Table 6 shows the rate of growth
for the entire year including the winter. It has already been
stated that Teredo reaches adult size in one year and dies during
the second year.

Teredo navalis is said to grow to a slightly larger size in sub-
tropical climates than at Woods Hole. As stated above the largest
specimen found in this northern locality in four years' study
measured forty centimeters in length and one centimeter in great-
est diameter.

278

B. H. GRAVE.

w

H

to

O

IH

d • d • • d d

l-«

•- ' c •- d d — ' "-1 •

^

.** .^H .^ r^

M

d <>? N^ln ID l^™"^10'^

*4-l

— Q " Q 1" 1™ 2 2 rH|°°

CO

«S <-• X -" 0 0 _,„ 0

^^l^^i^, -M -^WJ^ -*-1

•£ &

M|«MJ>< «^Hx « ^JH

•o •£

«;•* "^ ««!•* '*"* "^ o w C

/•— *«

c

M .H^1"^ \OHN O M '"

B

Cd

O ^^0°° M O-210100

H

r^

-t-JOO-^OO-*-1 to ^

^

M

H< •" •" rt|N -^ *J o "H •"

0

M ro "*• ^f O f» M rf|f>

O ^

iS

0 ^

HI

p "§

pj <£^

H £

Z -c<,

^

T3

. E

^

d ^

. £ 6 £ 6 £ E E

O Cj

43 '"

g H £ £ £ Supl |

s 1

•*-> 4_>

M CU

££r-t-t-t-osMro

u-)ir)C>ooooTJ-ro

fe =

tJ •"-

gd ^Of.^^^ooOO

w ^

d cu

••^

'» 4J

52.*i.S}.S}'t4* ^ X ^ X <r> N

?> 3

o ,rt

coco.Q.jO.Qxi'KjrieeeeXX
^^•Jo'S-K

SS '>'>'>£££££££££

s
o ~

M+J
M

ooooo£E252S)oSE
1111'§^c3M'HtNtv<':h?.^

*c

cu<Daj<D<D°oooo

Sj O

en

ooooo"w*J°<soloo'<J*J
^^^^^ voioo M lor^ioco N

" <-J

l<^^-H^-(AH^H rO^^1-' *-< ^ M tN TfM

0 §

M *»

H v!

fe S

** t~

% i
Oij*'

-4-> -j p

*><

ij

o >-,>-.

d, "^

bO

ecu cu
-1-) *J

0

*^

d d

H

M M CN CN rororo^Tt'<3-LocN*>^>

M |> |>

(L>

— *-"

l^i O

>H

VH M

O ""^

4-J '*-i

H M •MMMM'1-11-1 : :MH

^ °
s a

• M 6 ,•,,•,.. M w >>>,

r'iCJ cutj 'CJCJ • — i —

njCUcJ/"\CU,^(^^lc-J^'^":'

O 3

CO

QQQ n^Q'~''""'Q^.hc^>o1

4- 0

0 D,
T) «

QO *JO.2-So<")^nn"4~)"4~>
t!^°H1M00"M00-M-M^S>

.2 w

MM (^MMCNIHIHM

cu 'o

& -M -M o« ft ^ ^ ^^^'^^Q.DH

£ootiti<<<miti%

NATURAL HISTORY OF SHIPWORM. 279

Note: — Table 6 shows, primarily, the amount of growth attained by
Teredo during the first season up to the beginning of the hibernation period
(Dec. i). It shows not only that the earliest larvae of the season fail to
reach maturity (adult size) the first season, but also gives the amount of
growth attained by the later larvae of the season as well.

The first five items of this table show that Teredo larvae which enter the
wood after September i will not attain sufficient size to be detected in the
wood up to Decvember i. Timbers exposed to the water as late as Au-
gust 20 on the other hand are likely to contain Teredo measuring 35 to 45
mm. in length by the end of the growing season (Dec. i).

The last two items of the table show that timbers exposed to the water
between Sept. 12 and Sept. 23 become infested by Teredo and that they
appear in the wood the following July, although they remain too small
during the winter to be detected.

The latest infestation observed at Woods Hole occurring in Eel Pond
was Sept. 23, and in Vineyard Sound Oct. 10. It may sometimes occur
somewhat later than this, since the point was not sufficiently investigated.
Table 6 shows that growth in the late fall is very slow compared with
summer growth as given in Table 5 and data not tabulated show that
practically no growth takes place in Teredo at Woods Hole between Dec-
i and March I. Some observations indicate that the gonads begin to
proliferate extensively before there is detectable body growth in the spring.

SUMMARY.

Teredo navalis occurs in abundance at Woods Hole and vicinity
and has been known there for many years.

The breeding season extends from about May 10 to October
10. Spawning begins in the spring when the water reaches a tem-
perature between 11° and 12° C. Spawning by each female
occurs several times during the season. No lunar periodicity in
spawning occurs and there are no broods caused by synchronous
spawning.

The eggs are retained in the gills of the mother during cleav-
age and early larval development.

The time required for the fertilized egg to complete larval de-
velopment to metamorphosis is approximately five weeks at Woods
Hole. About half of this time is passed in the brood pouch and
half as a free swimming veliger.

When eggs and early embryos are removed from the gills they
do not develop normally.

The trochophore of Teredo is non-motile, having either a feeblv
developed protroch or none.

Teredo navalis reaches sexual maturity in six weeks or two

28O B. H. GRAVE.

months after metamorphosis when it measures four or five centi-
meters in length. It reaches adult size in on year, and dies dur-
ing the second year. The largest specimen collected in four years
measured forty centimeters in length and one centimeter in great-
est diameter.

The rate of growth during the summer months and also during
the winter was determined and tabulated. Certain habits of ship-
worms were also observed and recorded.

APPENDIX.
Practical Measures.

Because of numerous inquiries by lobstermen and owners of
small boats concerning methods of preventing damage by ship-
worms, a series of experiments was made on the effect of drying
upon Teredo. These experiments were not extensive but suffi-
cient to show that simple but effective precautions may be taken.

Teredo larvae first begin to enter wood between June 20 and
June 25 and stop about October 10. Shipworms do not enter
wood at any other time during the year. Little or no damage is
done to wood until it has been in the water one full month. The
largest of the young shipworms are only one fourth of an inch
long at the end of one month but they attain a length of one inch
in six weeks. It is, therefore, advisable to dry lobster-pots and
boats once per month and leave them out of the water exposed
to the sun for one week. This is especially true in July and
August when most of the damage is done. Shipworms in small
timbers are killed by five days' exposure to sunlight, but 2x4
stakes and larger timbers require from a week to ten days for
drying sufficiently to kill all of the worms.

EXPERIMENTS.

(i) Infested lobster-pot lath:

(a) After exposure in air to bright sunlight for I day,
seemed to be dry but some of the infesting shipworms
were still alive.

After exposure in air to bright sunlight for 2 days, some
shipworms alive.

NATURAL HISTORY OF SHIPWORM. 28l

(c) After exposure in bright sunlight for 3 days, all ship-
worms dead.

(2) Infested lobster-pot lath:

(a) After 7 days on shelf in laboratory, all worms contracted

and shrunken, some of which regained plumpness and
normal activity when placed in sea water. Sperma-
tozoa and larvae taken from these shrunken worms
showed activity.

(b) After 10 days on shelf in laboratory — all worms, sperm

and larvae dead.

(3) Infested 2x4 stakes :

(a) After exposure in air to sunlight for 5 days; many

shipworms dead, some living.

(b) After exposure in air to sunlight for 7 days; all ship-

worms dead.

(4) Infested 2x4 stakes :

(a) After exposure in air in shade for 7 days; many worms

dead but some living.

(b) After exposure in air in shade for 10 days; none living.

Note:—li infested 2x4 stakes are exposed in air in the shade but kept
wet some worms may live for several weeks.

BIBLIOGRAPHY.

1. Barrows, A. L.

'17 An Unusual Extension of the Shipworm in San Francisco Bay.
Univ. of Calif. Pub. in Zool., Vol. 18, p. 27.

2. Blum, H. F.

'22 On the Effect of Low Salinities on Tcrcgo navalis. Univ. of Calif.,
Pub. in Zool., Vol. 22, p. 349.

3. Davenport, C. B.

'97 The Role of Water in Growth. Proc. Boston Soc. Nat. Hist., Vol.
28, p. 73.

4. Core, W. H., and Miller, R. C.

'22 The Digestion of Wood by Teredo navalis. Univ. of Calif. Pub. in
Zool., Vol. 22, p. 383.

5. Grave, B. H.

'09 Anatomy and Physiol. of Atrina rigida. Bull. U. S. Bur. Fish.,
1909.

6. Harrington, C. H.

'21 A Note on the Physiology of the Shipworm, Torcdo norz'cgica. Bio.
Chem. Jour., Vol. 15, p. 736.

7. Hatschek, B.

Ueber Entwicklungsgeschichte von Teredo. Arbeit. Zool. Inst. Wien,
Vol. 3, 1881, p. i.

282 B. H. GRAVE.

8. Kofoid, C. A., and Miller, R. C.

'22 The Specific Status of the Teredo of San Francisco Bay. Report
No. 2 on the San Francisco Bay Marine Pilings Survey.

9. Kofoid, C. A.

Reports Nos. i, 2 and 3 on the San Francisco Marine Pilings Survey,
'21, '22, and '23.

10. Lazier, E. L.

'24 Morphology of the Digestive Tract of Teredo navalis. Univ. of
Calif. Pub. in Zool., Vol. 26, p. 455.

11. Miller, R. C.

'22 Variations in the Shell of Teredo navalis. Univ. of Calif. Pub. in

Zo61., Vol. 22, p. 293.

12. Miller, R. C.

'24 The Boring Mechanism of Teredo. Univ. of Calif. Pub. in Zool.,
Vol. 26, p. 41.

13. Nelson, T. C.

'22 The European Pile Worm. Circular 139, N. J. Ag. Exp. Sta.

14. Nelson, T. C.

'23 Marine Borers. Report Dept. Biol. N. J. Ag. Col. Exp. Sta. year
ending June 30, 1923, p. 204.

15. Potts, F. A.

'20 A Note on the Growth of Teredo navalis. Rep. Dept. Mar. Biol.
Carnegie Inst. Wash, year book No. 19.

16. Potts, F. A.

'24 The Structure and Function of the Liver of Teredo. Proc. Cam-
bridge Phil. Soc. (Biol.), Vol. I., p. i, and Jour. Mar. Biol. Assoc.,
Vol. 13, p. 511.

17. Sigerfoos, C. P.

'08 Natural History, Organization and Late Development of the Teri-
dinida? of Shipworms. U. S. Bur. of Fish., Vol. 27, p. 191.

18. Verrill, A. E., and Smith, S. I.

'71 Report upon the Invertebrate Animals of Vineyard Sound and Ad-
jacent Waters. Report of the U. S. Commissioner of Fish and
Fisheries.

STUDIES OF HUMAN TWINS.

I. METHODS OF DIAGNOSING MONOZYGOTIC AND DIZYGOTIC

TWINS.

H. H. NEWMAN,
THE HULL ZOOLOGICAL LABORATORY, UNIVERSITY OF CHICAGO.

INTRODUCTION.

This is the first of a series of studies of human twins based
upon a considerable collection of pairs taken from the environs of
Chicago. These studies have been carried on in collaboration
with Professors F. N. Freeman, K. J. Holzinger, and Mrs. Blythe
Mitchell. The original objective of this research project was
to secure an adequate collection of monozygotic and same-sexed
dizygotic twins about whose diagnosis we could be certain. With
this objective attained, it was proposed to make an intensive
comparative psychological study of the two types of twins to
determine, if possible, the influence of heredity and environ-
ment upon the various mental traits. This is an old and some-
what hackneyed problem, but one that has never been at all sat-
isfactorily solved. It seemed to us, however, that all previous
studies had been inadequate because methods of diagnosing the
two types of twins were unsatisfactory. The one crying need
then was for a satisfactory method of diagnosing monozygotic
twins, and the working out of such a method was assigned to the
present writer.

COLLECTION AND CLASSIFICATION OF MATERIAL.

The objective set by the collaborators in this study was the col-
lection of fifty pairs of identical twins and fifty pairs of fra-
ternal twins. In order to simplify our task, we decided to elimi-
nate the disturbing factor of sex dimorphism, and therefore
confined our study to twins of the same sex, pairs in which the
twins were both boys or both girls.

At first no selection was practised among same-sexed twins,

283

284

H. H. NEWMAN.

but all cases were taken as they came. As each case was com-
pleted an informal vote of the three or four workers present was
taken as to the category (identical or fraternal) to which the
pair belonged. Rarely, if ever, was there any difference of opin-
ion, but in about one tenth of the cases there was some uncertainty
and these cases had to be studied more intensively.

It soon appeared that the collection of identicals and fraternals
was not going evenly, the fraternals being more numerous. If
our preliminary judgments as to their classification were accu-
rate we would need to stop the collection of fraternals and collect
only identicals during the last stages of the period of study.

When the adjudged " fraternals " mounted to fifty-two cases
(consisting of twenty-four male pairs and twenty-eight female
pairs) there were only forty-three :< identicals " (consisting of
twenty-five male pairs and eighteen female pairs). The sex
ratio at that time was very close to normal expectancy : forty-nine
male pairs to forty-six female pairs. The question arose as to
whether the proportion of identicals to fraternals was running
according to theoretical expectancy.

Various methods have been used to determine the proportion
of monozygotic to dizygotic twins. One method involved the
examination of the fetal membranes of considerable numbers of
twin births in institutions where competent observers were able
to secure these important diagnostic data. Spat in 1860 reported
that, in a total of one hundred eighty-four cases of twins ex-
amined as to the membranes, 24.6 per cent, were monozygotic.
Brem in 1891 reported 22.7 per cent, of monozygotic twins out
of one hundred twenty-six twin births. Krahn in 1891 reports
19 per cent, of monozygotic twins among one hundred twenty-
seven twin births, but includes as monozygotic two opposite-
sexed pairs. Tigges found in 1896, 21 per cent, of monozygotic
twins among fifty-two twin births, and Quenzel in 1894 reported
20.4 per cent, of monozygotics among one hundred eighty-one
pairs of twins. These percentages range from 19 per cent, to
24.6 per cent.

A second method used by several investigators for computing
the proportion of identical twins is statistical in character. The
best known of these methods is Weinberg's " differential method."

STUDIES OF HUMAN TWINS. 285

In 1902 Weinberg described his method as follows : " Assuming
that sex is determined at the time of fertilization and that about
half of all zygotes will produce males and the other half fe-
males, it follows that there will be equal numbers of same-sexed
as opposite-sexed fraternal twins. If, therefore, we double the
number of opposite-sexed twins and subtract the product from
the total of all twins, the remainder will represent the number of
monozygotic twins."

Applying this method to large masses of twin data he found
that the percentage of monozygotic twins varies from 23.4 per
cent, to 31 per cent., the percentage differing in different coun-
tries. This agrees rather closely with the percentages determined
on the basis of fetal membranes.

Recently Knibbs (1926) has worked out a formula for com-
puting the number of monozygotic twins in the twin population,
using data taken from the census of Germany. His formula is
as follows : The ratio of monozygotic twins to all twins is
(M + F- - P) -f- (M -f- F + P), where M is the number of $ $
pairs, F the number of 9 5 pairs, and P the number of $ 9
pairs. This method gives 24.4 per cent, of monozygotic twins in
Germany from 1906 to 1911.

Applying Knibbs' method to the extensive twin data for the
United States that is presented by Nichols (234,497 $ $ ;
264,098 $ 9 ; 219,312 $ $ ), we discover that 26.42 per cent, of
this large group are monozygotic and that nearly 42 per cent, of all
same-sexed twins are monozygotic.

The question now arises as to whether our small random col-
lection of ninety-five pairs of same-sexed twins was composed
of the expected number of identical and fraternal pairs. Ac-
cording to our diagnosis there were forty-three pairs of identicals
and fifty-two pairs of fraternals — i.e., 45 per cent, identicals in-
stead of the expected 42 per cent. This is but a small discrep-
ancy and may have two meanings : Either the random selection of
twins has brought in two or three too many pairs of identicals
or else some two or three of the pairs diagnosed as " identicals "
should be classed as " fraternals." It is probable that the former
explanation is correct, for it is very unlikely that the ideal ratio
as determined on the basis of 717,907 pairs of twins would be

286 H- H- NEWMAN.

realized exactly in the first ninety-five cases selected a't random.
In fact, the close approach to theoretical expectation actually
realized is almost too close. The conclusion may then be drawn
from this that our methods of diagnosing identical and fraternal
pairs cannot be far astray.

In order to complete the proposed collection of fifty pairs of
identicals and fifty pairs of fraternals, it was then necessary to
select seven cases of certain identical twins. Two cases of fra-
ternals were eliminated from the fifty-two cases of fraternals
in order to get down to fifty cases. The two cases eliminated
were chosen for the following reasons : In one case one twin had
lost three fingers and his palm was so scarred that no adequate
palm print could be taken; in the other case one of the twins
showed up with an infected hand and no palm print could be ob-
tained. Since, in our diagnosis of monozygosity, the palm prints
were used as highly important criteria, it seems well to eliminate
these two pairs in which the palm print evidence was incomplete.
The two pairs eliminated were unequivocal cases of unlike fra-
ternal twins.

We have now complete data on one hundred pairs of same-
sexed twins, fifty of which have been classed as identicals and
fifty as fraternals. No doubt some of our readers are wondering
how we can speak so confidently about our ability to classify all of
our cases as either identicals or fraternals. It may be said that the
method was slow in taking shape and was arrived at only after
intensive study of the materials.

DIAGNOSIS OF MONOZYGOTIC TWINS.

The majority of workers on human twins seem to have despaired
of arriving at an adequate classification of twins into clean-cut
categories : monozygotic and dizygotic. Years ago Thorndike
found so much difficulty with his cases that he came to the con-
clusion that all twins belong to a single series and have a similar
origin. Lauterbach, 1925, after the study of nearly two hundred
pairs of twins, found himself unable to separate the same-sexed
pairs with any assurance. He tentatively classified 59 per cent,
of the same-sexed twins as monozygotic, a percentage much too

STUDIES OF HUMAN TWINS. 287

high, suggesting that he has included a good many cases of similar
fraternal twins in his " identical " group.

The most recent study of twins is that of A. H. Wingfield
(1928) who studied one hundred two pairs of twins selected at
random from the public schools of Toronto and Hamilton, On-
tario. Taking all pairs of twins as they came there were accumu-
lated seventy-six like-sexed pairs and twenty-six unlike-sexed
pairs. The expectation would be about 65 per cent, of like-sexed
twins instead of about 74 per cent., the number found in this col-
lection. It seems probable, therefore, that some unlike-sexed twins
were overlooked. Wingfield made an attempt to separate the
seventy-six like-sexed pairs into two groups, identicals and fra-
ternals. His method was somewhat precarious. He classed as
" identical " all those which seemed to himself and the teacher to
have a higher degree of physical identity than siblings are likely to
exhibit. " Only those pairs of twins showing practically indis-
tinguishable physical traits, as judged by the teachers in the school
and myself, were included in the identical group. While it is
not absolutely certain that all pairs included in the identical group
had identical heredity, the chances in favor of this being the
case are very great." The fact that he classed as identical over
44 per cent, of all the twins in his group is surprising in view of
the fact that the statistical expectation is only about 26 per cent.
It seems probable then that Wingfield has included among the
" identicals " several cases of similar fraternal twins. This is
further suggested by the fact that he found a coefficient of corre-
lation of only about -j- 0.90 for this group as compared with
-)- 0.95 obtained for our identicals.

That it is possible to develop a method of distinguishing be-
tween identical and fraternal twins is strongly suggested by the
fact that two European twin specialists claim to be able to make
such a distinction with a high degree of infallibility.

Dahlberg (1926), in his monograph on ; Twin Births and
Twins from a Hereditary Point of View," makes this statement:
' The following demands should be satisfied for a diagnosis of
monozygotism for a grown-up pair of twins :

" i. That the appearance of the twins give an impression of
very great resemblance or identity.

288 H- H. NEWMAN.

" 2. That during childhood, neighbors, school-fellows, etc.,
have had difficulties in distinguishing them and have sometimes
confused them.

" 3. That the configuration of the ears does not show great
dissimilarity.

" 4. That the finger prints show a certain high degree of simi-
larity.

' 5. That the anthropological measurements do not show too
considerable differences."

Siemen's method (1927) is somewhat more detailed and exact-
ing. He takes the very sensible view that no single criterion of
monozygotic origin is reliable, but that judgment in doubtful cases
should be based upon identity in as many traits as possible. He
emphasizes the rarity of really questionable cases. Many years
of experience in the study of twins has developed in him such a
degree of confidence in his method of diagnosis that he consid-
ers that he has been able to reach " a certain diagnosis in virtually
every case of twinning."

He finds, as others have found before and since, that the great
majority of all twins are either so completely alike or so markedly
different that there is no question about their diagnosis. A care-
ful study of the certain cases should furnish criteria for diag-
nosing the few doubtful cases. Thus a study of over a hundred
pairs of unquestionable identical twins has resulted in the fol-
lowing " scheme " for diagnosing monozygosity :

A. Traits in which one-egg twins practically always agree and in
which two-egg twins agree only very rarely:

1. Hair color and form.

2. Eye color.

3. Skin color.

4. Downy hair of the body.

B Traits in which one-egg twins differ only within narrow limits
and in which two-egg twins usually differ more
widely.

5. Freckles.

6. Appearance of blood in the skin.

7. Follicular processes.

8. Tongue (furrowed or not) and teeth.

STUDIES OF HUMAN TWINS. 289

C. Traits in which one-egg twins usually, and two-egg twins

rarely show strong resemblance :
9. Form of face.

10. Form of ears.

11. Form of hands.

12. Body build.

13. Mentality.

14. Illness and abnormality.

15. Traits studied by special methods — finger prints, etc.

Our own method of diagnosis has been considerably influenced
by the methods of Dahlberg and of Siemens, especially by the
latter, but is somewhat different from any previously used. Our
effort has been to combine the best features of all known methods.

After our own method was developed and while reading
Wingfield's monograph, the writer noted a reference to a short
note in Science by Taku Komai (1927) entitled "A Criterion
for Distinguishing Identical Twins from Fraternal Twins." The
criterion described has to do with finger prints and palm and
sole prints of twins. " Generally speaking," he says, " the same
hands or feet of the identical twins resemble each other more
closely in their patterns than the two hands or feet of the same
individual." This I have found to be very frequently true, but
the formula needs modification, as will be shown below.

OUR OWN METHOD OF DIAGNOSIS.

The method of identifying monozygotic twins used in the pres-
ent work may now be described in detail. A great deal of atten-
rion has been given to this matter, for we realize that the sound-
ness of our conclusions as to heredity and environment depend
upon the correctness of this diagnosis.

At the beginning, it may be said that in over 90 per cent, of
our cases there was at no time any doubt as to their classifica-
tion. The great majority of one type of twins are so strikingly
similar that their monozygotic origin is obvious. Their resemb-
lance is not confined to gross physical correspondence, but
extends to tones of voice, gestures, and peculiar mannerisms.
One soon becomes sensitized to the intangible correspondences of

290

H. H. NEWMAN.

identical twins and diagnoses them almost at a glance. The great
majority of the other type of twins strike one at once as en-
tirely unlike, often being more different than average brothers
or sisters. About these there is no question after the first glance.
Our ability to diagnose cases improved during the course of our
study and we found that there was no difficulty at all in diag-
nosing the last half of the pairs that presented themselves. Two
of the very early pairs were diagnosed doubtfully that, when
reexamined after a year of experience, offered no difficulty at
all. Two other cases were left uncertain because we allowed
ourselves to be influenced by statements of the mother. About
these cases there should never have been any question had the
mother not been loquacious.

Out of one hundred two pairs of twins there was justifiable
doubt about only six cases. These cases have all been diagnosed
satisfactorily with the possible exception of No. 61, which still
remains slightly uncertain.

The following are our criteria for diagnosing identical (mono-
zygotic) twins.

1. They must be strikingly similar in general appearance in-
cluding various intangible resemblances.

2. They must be essentially identical in hair color, texture and
form.

3. They must have the same shade of eye color and form of
iris.

4. They must have the same skin color and texture (com-
plexion) except when one is more tanned than the other.

5. They must have no marked differences in features ; shape
of ears ; shape, size and arrangement of teeth.

6. They must have hands of the same type and nearly equal
in size.

7. The general microscopic character of the papillary ridges
in fingers and palms must be essentially the same.

8. There must be stronger cross resemblance than internal re-
semblance in one or more of the details of finger and palm pat-
terns.

9. The presence of reversed asymmetry in handedness or hair
whorl in one twin is confirmatory evidence of monozygosity, but

STUDIES OF HUMAN TWINS.

its occasional presence in unlike twins is not to be 'taken as an
indication of monozygosity.

A great deal of stress has been laid upon the diagnostic value
of the palm and ringer patterns. While this criterion alone is
inadequate for certain diagnosis, it is surprising how few mis-
takes were made in our effort to diagnose monozygosity on this
basis alone. In the first forty-two cases in which a judgment
was attempted on the basis of palm and finger prints alone, there
was disagreement in only two cases with the judgment based on
general resemblance. Our method has been to classify all cases
on the basis of the first six criteria and then to check this classifi-
cation by criteria 7 and 8.

PALM AND FINGER PRINTS AS CRITERIA.

The intensive study of palm and finger patterns is perhaps the
best single diagnostic aid. After a scrutiny of the first thirty or
forty sets of palm prints the writer began to notice an important
fact about the palm and finger patterns of strikingly identical
twins : namely, that, instead of showing mirror-imaging of pat-
terns (involving the resemblance of the right hand of one to the
left hand of the other) the two hands of one of the twins were
direct duplicates in major features of the two hands of the other.
Specifically, the right hand of one twin is more like the right
hand of the other than like own left hand, and the left hand of
one twin is more like left hand of other than like own right hand.
Thus cross resemblance between the two twin individuals is
stronger than resemblance between the two hands of the same
individual.

Among twins that are somewhat less alike the same rule holds
in a somewhat modified form. Thus right hand of one twin may
be like right of the other, or left of one like left of the other,
but the close resemblance does not extend to both sides. In still
other pairs of twins in which one is distinctly left-handed, there
is a reversal of asymmetry, so that the right hand of each twin
is like the left hand of the other. In every pair of obviously
monozygotic twins the rule holds that there is stronger cross re-
semblance betzuecn the hands of one twin and those of the other
than between the two hands of the same individual. The same
20

292

H. H. NEWMAN.

is true of ears, teeth, and other structures that show more or
Jess asymmetry, but there is more detail in palm and finger prints
and a more objective method of comparing them. In the case
of the fingers the types of patterns have been formulated in all
cases in order to obtain a qualitative basis of comparison, and the
friction ridges in all patterns (following the method of Bon-
navie, somewhat modified) were counted under binocular so that
a quantitative comparison between the fingers of one hand and
those of another is possible. In both qualitative and quantita-
tive respects the rule that cross resemblance is stronger than in-
ternal resemblance holds, for identical twins.

The studies of palm main line formulae and of the occurrence
and varied expression of the six fundamental primitive patterns
have been greatly facilitated by the study of a paper now in •
manuscript, the work of a considerable group of experts, entitled
" A Study of Error in the Interpretation and Formulation of
Palmar Dermatoglyphies," by Cummings, Keith, Midlo, Mont-
gomery, H. H. Wilder and I. W. Wilder. Professor Cummings,
evidently the guiding spirit of the group in this collaborative
inquiry, has very kindly furnished me with a copy of the manu-
script and has thus made it possible for me to study the palms of
our twins with far greater efficiency than would have been pos-
sible without this assistance.

With few exceptions the same rules of cross resemblance apply
to the palmar main lines and patterns that apply to finger prints.
Most frequently the cross resemblance runs similarly in all four
respects: in qualitative characters of finger patterns, in quanti-
tative values of finger patterns, in palmar main line formulae, and
in the occurrence of palmar patterns. Sometimes the cross re-
semblance is obvious in only three of four respects, sometimes in
two, or only one ; but if it is greater between one hand of one
twin and either the same or opposite hand of the other twin than in
own hands, the rule is considered to hold good.

While it is of importance that the detailed analysis of the
finger and palm characters of this collection of twins should be
published, this is hardly the appropriate place for it. One or
iwo separate papers devoted to a special presentation and analysis
of these data are planned for subsequent publication.

STUDIES OF HUMAN TWINS. 2Q3

At this time we must ask the indulgent reader to accept ten-
tatively our criteria for diagnosing twins. With the publication
of the complete data used in this diagnosis the methods used may
be put to any test that seems necessary.

Applying the criteria of diagnosis above described to the six
pairs of twins about which there was some doubt, three of them
fell readily into the category of identicals and three were classi-
fied as similar fraternals. At the present time the writer feels
quite confident as to the correctness of diagnosis of the whole
collection. The cases 'that might be questioned by some are the
three cases of similar fraternals just referred to. Before dis-
cussing the problems arising out of a study of identical twins,
it seems advisable to devote a few paragraphs to the fraternal
twins, especially to the three cases most difficult to diagnose.

THE DIAGNOSIS OF FRATERNAL TWINS.

Of the fifty-two pairs of fraternal twins in our collection, three
may be classed as " similar fraternals," and twenty as " slightly
similar fraternals," and twenty-nine as ;' unlike fraternals."
None of the pairs show as much resemblance as the least similar
of the identical twins. The only cases that could possibly be at all
in question as to their classification are the three " similar " pairs,
numbered 61, 15, and 74. Let us carefully scrutinize these rather
crucial cases as to the possibility that they might be monozygotic
twins of the less nearly identical sort.

Pair 61. — These girls at first impressed us with their similar-
ity. They were dressed exactly alike, arranged their hair alike
and had very similar coloring. In height there was but three
eights of an inch difference; there were two and three fourths
pounds difference in weight. Head length of A was 13.95 mm.,
of B 14.35 mm.; head width of A was 17.7 mm., that of B was
17.9 mm. The hair of both was in general rather similar, but
that of B was a shade darker, softer, finer and not so heavy.
Eye color was the same in both, a type of hazel. There was no
difference in skin color. Ears of A were higher and narrower
than those of B, and had a shorter lower lobe. A has fuller lips ;
B has the longer, more prominent chin. A holds eyes wide open ;
B has them nearly half closed. Bridge of A's nose more bowed

294

H. H. NEWMAN.

than that of B. The teeth of the two differ rather sharply, the
upper arch of B being narrower and the teeth crowded and ir-
regular, while those of A are regular.

The finger print formulae are decidedly different :

Left Hands. Right Hands.

I, 2, 3, 4, 5 i. 2- 3, 4, 5

A— U, R, A, W, W A— U, U, U, U, U

B-W, U, R, U, U B— W, R, U, U, U

The quantitative values of the finger prints are :

A — right hand 24 A — left hand 28

B— right hand 38 B— left hand 25

All four palm main line formulae are different and the patterns are
also different.

Left Hands. Right Hands.

A— (9.8.5".5') B.O.O.O.O. A— (11.9.7 -5.') B.O.O.L.O.

B— (9.8.s".3 ) A.0.0.0.0. B— ( 9.7.5" -3 ) C.O.O.O.O.

Both are equally right-handed and both have clockwise hair- whorl.

In spite of a superficial rather close resemblance, then, there is
no indication that these twins have had a monozygotic origin. This
was the most difficult case to diagnose, but there seems now no
doubt that these twins are dizygotic in origin.

Pair 65. — This case was somewhat puzzling because the two
girls are both rather peculiar in appearance and are similar in
many peculiarities.

In height A is 57^2 inches, B 56% inches. In weight, A is
II31/4 pounds, B is m1^ pounds. Head length of A is 14.5 mm.;
that of B is 14.4 mm.; head width of A is 17.7 mm.; that of B is
17.1 mm. Hair of both is the same in color and texture; eye
color of both is of the same shade of blue; B has a lower brow
and a sullen expression about the eyes, while A has a contented
expression. The skin is somewhat more florid in B. The ears of
the two differ greatly, B having much longer lower lobe. The
hands differ in shape, those of A being broader and thicker. B
has shorter, more turned-up nose, a distinctly wider mouth, fuller
lips and fatter face. The teeth differ radically, the upper arch of
B being wider and straighter across the front and with wider
teeth.

The finger print formulae read as follows :

STUDIES OF HUMAN TWINS.

295

Left Hands.

i, 2, 3, 4, 5
A— W, R, W, W, U
B— W,W, U, W, U

Right Hands.
I, 2, 3, 4, 5
A— W, W, W, W, U
B— W, W, W, W, U

The quantitative values of the finger prints are :

A — right hand 53
B — right hand 52

The palm formulae are as follows :

Left Hands.

A— (11.7.7. 3 ) A.O.O.O.D.
B— ( g.8.5".5') A/B.O.O.O.D.

A — left hand 54
B— left hand 60

Right Hands.
-(11.9.75') A.O.M.O.D.

B— (11.8.7.5') O.O.M.L.O.

In several respects there is a little more resemblance between right
palm and fingers of the two than to their respective lefts, but this
does not extend to details. On the whole these 'two girls make an
entirely different impression. One has a rather pleasing, happy
expression, the other a sullen, lowering expression. The fact that
B is ambidextrous in both finger and wrist tapping suggests that
she might be the left-hand component of a monozygotic twin pair,
but there are too many differences between them to permit such a
diagnosis.

Case 24- — These girls have many traits in common, but show
also some extreme differences. A's height is 59 inches ; B's is
53^4 inches. A's weight was 70^2 pounds; B's 66^/2 pounds. A's
head width is 14.1 mm.; B's 13.5 mm. A's head length is 17.6
mm.; B's is 17.1 m.m Hair color, texture and crown whorl same
in both. Eye color of both a gray brown, but A's eyes are dis-
tinctly grayer and B's browner. B's ears are distinctly larger and
wider although her head is considerably smaller. A's eyes are
wider spaced than B's. A's nose is larger, longer and different
in shape. B's teeth are crowded and overlap in front, while A's
are straight.

Finger print formulae :

Left Hands.

i, 2, 3, 4, 5

A— W, R, R, U, U :

B— U, R, U, U, U

Quantitative values of finger patterns

A — right hand 44
B — right hand 27

Right Hands.

A— W, R, U, U, U
B— U, A, A, U, U

A — left hand 30
B— left hand 26

296 n- n- NEWMAN.

Palm formulae:

Left Hands Right Hands.

A— (n.77-3) O.O.O.O.D. A— (11.9.7-3) O.O.O.L.D.

B— (11.7.7.3) O.O.O.O.O. B— (11.7.7.3) A.O.O.O.O.

Here again the palm formulae suggests a closer resemblance than
actually exists, in that we have the same pattern for the two left
hands, but the two palms of B also have the same pattern and
are far more similar in detail. Nowhere is there stronger cross
resemblance than internal resemblance. On the whole there can
be no doubt that these are fraternal twins.

Apart from these three cases there are no decidedly similar
twins among the fifty-two pairs in our collection. Twenty pairs
are designated as " slightly similar "' fraternal twins and the re-
maining twenty-nine cases are designated as " unlike " fraternal
twins. The slightly similar fraternals show merely the degree
of resemblance common among siblings, while the unlike frater-
nals seem to show hardly as much resemblance as do average
siblings. Even the three cases of decidedly similar fraternals,
except for their identity in age, are no more alike than are occa-
sional siblings. On the whole then, there seems to be nothing
about these fifty-two cases out of accord with their classification
as fraternal, or dizygotic, twins. Hence there is now no ground
for doubting the validity of our classification of the one hundred
two pairs of twins used in this study, into the two categories,
monozygotic and dizygotic.

SUMMARY.

1. The original objective of these studies was the study of the
roles of heredity and environment in determining mental capaci-
ties of various sorts.

2. The first essential was to learn how to diagnose with cer-
tainty the two types of twins, monozygotic and dizygotic.

3. Only about 25 per cent, of all twins are monozygotic. Col-
lections that depart widely from this figure have probably been
incorrectly diagnosed.

4. Only about 42 per cent, of same-sexed twins are monozy-
gotic.

5. The method of diagnosis used in this study combines the best

STUDIES OF HUMAN TWINS. 2Q7

features of the methods of Dahlberg, Siemens, and Komai. Cer-
tain refinements of technique are added, the details of which are
explained in the text.

6. Out of a collection of one hundred two pairs of same-sexed
twins, only six pairs caused any difficulty, three of which are now
classified as monozygotic and three as dizygotic.

7. The details concerning the three " similar f raternals " are
presented and the reasons for their diagnosis as dizygotic twins
are given.

8. The result is that we have now a collection of fifty pairs of
monozygotic and fifty-two pairs of dizygotic same-sexed twins
accurately diagnosed. These are to be used for further biological
and psychological study.

BIBLIOGRAPHY.

1. Bonnevie, K.

'24 Studies of Papillary Patterns of Human Fingers. Jour. Genetics,
Vol. XV., No. i.

2. Dahlberg, G.

'26 Twin Births and Twins from a Hereditary Point of View. Stock-
holm.

3. Knibbs, G.

'28 Multiple Births, their Characteristics and Laws Mathematically
Considered. Jour, and Proc. Roy. Soc., N. S. Wales, Vol. 59.

4. Komai, T.

'27 A Criterion for Distinguishing Identical Twins. Science, Vol.
LXV., No. 1681.

5. Lauterbach, C. E.

'25 Studies in Twin Resemblance. Genetics, Vol. X., No. 6.

6. Newman, H. H.

'15 The Biology of Twins, Chicago, 1915. The Physiology of Twin-
ning. Chicago, 1923.

7. Siemens, H. W.

'27 The Diagnosis of Identity in Twins. Jour, of Heredity, Vol. 18,
No. 5.

8. Thorndyke, E. L.

'05 Measurement of Twins, Archives of Philos., Psychol. and Scientific
Methods, No. I.

9. Weinberg, W.

'02 Probleme der Mehrlingsgehurtenstatistik. Zeitschr. f. (it-nurtsh.

u. Gynak., Bd. 47.
10. Wingfield, A. H.

'28 Twins and Orphans. The Inheritance of Intelligence. London and

Toronto.

STUDIES OF HUMAN TWINS.

II. ASYMMETRY REVERSAL, OF MIRROR IMAGING IN IDENTICAL

TWINS.

H. H. NEWMAN,

INTRODUCTION.

One of the most striking phenomena observed among monozy-
gotic 'twins is that of the reversal of asymmetry between the in-
dividuals of a pair. Among armadillo quadruplets the present
writer (Newman, 1916) found numerous instances in which a
band or scute doubling occurred on the left side of one twin and
on the right side of the other. Such cases call to mind the fact
'that in human double monsters (Siamese twins) situs inversus
viscerum occurs in many cases. The same type of asymmetry
reversal was noted by Swett and by Morrill in double-headed
fish embryos. In separate identical twins in man it has been
noted that the incidence of left-handedness in one twin of a pair
is very much greater than among fraternal twins or in the gen-
eral population of single individuals. Asymmetry reversal in the
direction of crown whorl of the head hair seems to have about
the same incidence in monozygotic twins, dizygotic twins, and
single individuals as has left-handedness. These two expressions
of asymmetry have been studied intensively in the present in-
vestigation and their significance will be discussed in some detail
later.

HANDEDNESS AN EXPRESSION OF ASYMMETRY.

As an introduction to this study it seems well to examine the
phenomena of handedness as it is found among human beings.
In the first place, there are two distinct kinds of handedness:
that which is genetically determined and that which is the result
of twinning and therefore epigenetically determined.

Genetic handedness is evidently transmitted in such a way that
any given zygote will give rise, when no twinning occurs, to
a right-handed or left-handed single individual. There seems,
however, to be varying degrees of right- or left-handedness.
The majority of individuals, apparently about eighty per cent.

298

STUDIES OF HUMAN TWINS. 299

of single individuals, are definitely righ-handed ; about four per
cent, definitely left-handed, and the remaining sixteen per cent.
partially left-handed or ambidextrous. The incidence of right-
and left-handedness is about what one would expect if right-
handedness is a dominant Mendelian unit character and left-
handedness recessive. The ambidextrous individuals and those
showing lesser degrees of left-handedness may be heterozygous
individuals in which the dominance of right-handedness is in-
complete.

The other type of left-handedness, quite different in origin and
heritability, is that which results epigenetically as the result of
the twinning. Such left-handedness, being a somatic modifica-
tion would not be hereditary: it would be merely an expression
of asymmetry reversal due to the development of a whole in-
dividual from a half embryo which had already become more or
less differentiated in a left-handed direction before the separation
into twins has taken place.

Thus in genetic right-handed embryos which undergo twinning
after some asymmetry has been established, the left-hand half
embryo would be the superior one and would give rise to a right-
handed individual, since right-handed superiority is due to left-
sided superiority in the brain. Conversely, in a genetic left-
handed embryo, the right side would be superior and the left
side the inferior side, in which case the left-handed individual
would retain the genetic asymmetry and the right-handed individ-
ual would exhibit asymmetry reversal.

In embryos genetically ambidextrous the right and left sides
would be equal and would produce twins both of whom would
be ambidextrous.

PREVIOUS DATA ON HANDEDNESS IN TWINS.

A good deal of attention has been paid by various authors to
the peculiar incidence of left-handedness in twins. Siemens
(1924) found in thirty-seven pairs of identical twins twenty-six
cases both right-handed, ten cases in which one was right-handed
snd the other left-handed, and one case where both were left-
handed. In a later paper the same writer reported on a larger
number of identical twins (the total number not given) in which
21

3QO H- H- NEWMAN.

there were twenty-one cases where one was right- and the other
left-handed and three cases where both were left-handed.

Weitz (1924) found among eighteen pairs of identical twins,
seven pairs composed of a right- and a left-hander, ten pairs both
right-handed, and one pair both left-handed.

Dahlberg (1926) reports for sixty-nine pairs of identicals
fifty-three pairs both right-handed, twelve pairs one left-handed,
and four pairs both left-handed. Adding the three sets of cases
together, -we have one hundred and twenty-four cases of iden-
tical twins divided as follows :

89 pairs, biilh right-handed, 7 I .X per cent.

29 pairs, one left-handed, 23.4 per cent.

6 pairs, both left-handed, 4.8 per cent.

Dahlberg has also studied the incidence of left-handedness in one
hundred and twenty-eight pairs of dizygotic twins. The follow-
ing figures indicate his results :

in pairs, both right-handed, 86.7 per cent.

16 pairs, one left-handed 12.5 per cent.

i pair, both left-handed, 0.8 per cent.

It will be seen that the incidence of left-handedness among
identical twins is over twice as great as among fraternal, or four
times as great in proportion to the number of zygotes involved,
for a pair of identical twins involves only one zygote. Even
among fraternal twins, the incidence of left-handedness is rela-
tively high as compared with the general population, which is
reported by Jones (1918) to be about four per cent. Jones' esti-
mate, however, is probably much too low and takes account of only
the most complete cases of left-handedness.

Lauterbach (1925) reports among fifty-seven same-sexed twins
(not distinguished as to monozygotic or dizygotic origin) twenty
pairs in which one was left-handed, about 35 per cent, of all cases.
This is a higher incidence of left-handedness than any previously
reported, especially when it is taken into consideration that the
group examined consists of both identical and fraternal twins.

The most recent data on handedness in twins is furnished by
Verschuer (1927). He found one or more left-handed individ-
uals in 26.8 per cent, of seventy-nine pairs of identical twins and

STUDIES OF HUMAN TWINS.

301

in 26.3 per cent, of the thirty-eight pairs of fraternal twins. They
were distributed as follows :

Identical Twins. Fraternal Twins.

58 both right-handed.

28 both right-handed.

one uSi

10 one right- the other left-
handed.
o both left-handed.

right- the other left-
handed.

5 both left-handed.

i one right-handed, the other
ambidextrous.

The percentage is rather low as compared with those of others,
particularly those of Lauterbach and the present writer, but the
difference is probably due to the inclusion of only the cases of
complete left-handedness. The percentage of pairs showing left-
handedness among fraternal twins is exceptionally high and not
in accord with the findings of others. Possibly the relatively
small number of cases may be the cause of this discrepancy.
Even more probable, it seems to me, is the inclusion among
fraternal twins of a few of the least similar identical twins among
whom left-handedness is common.

CRITERIA OF HANDEDNESS.

It is by no means a simple matter to diagnose left-handedness.
There are many cases, of course, where the twins are (or were at
an earlier period) obvious left-handers, but there are also many
cases where congenital left-handedness is obscured by training
the right hand and suppressing the left. Such cases often result
in a sort of ambidextrality in ordinary manipulations. In our
work we have used as a test of handedness speed in tapping with
wrist and fingers. In all cases of complete left-handedness the
tapping tests confirm the left-handed diagnosis. It appears to be
safe then to use the tapping tests to reveal native left-handedness
obscured by right-hand training or various degrees of partial left-
handedness.

A good many cases of partial left-handedness were revealed by
tapping tests. Among identical twins, in addition to the eleven
pairs showing complete left-handedness, there were thirteen pairs
(both of whom considered themselves right-handers) in which
some degree of left-handed superiority was revealed in one or
both members of the pair. In three of these pairs both members

-Y

<*{*
L I B R A R Y ' C

-2Q2 H- H- NEWMAN.

were shown to be partially left-handed, and in two pairs both
members were definitely ambidextrous.

Among fraternal twins, in addition to six pairs in which one
individual was completely left-handed, there were five pairs in
which one individual was partially left-handed, two pairs in
which both were partially left-handed, and two pairs in which one
individual was right-handed and the other ambidextrous.

Assuming that all these cases represent grades of left-handed-
ness we have added to the seventeen pairs showing complete left-
handedness twenty-two pairs showing partial left-handedness, a
total of thirty-nine out of one hundred pairs in which some degree
of left-handedness appears in one or both members of a pair-.
This high percentage would be much like that found by Lauter-
bach (35 per cent.) if we omitted the cases of ambidextrality.

CROWN WHORL AN EXPRESSION OF ASYMMETRY.

As is well known, the head hair at the crown twists or whorls
in either a clockwise or a counter-clockwise direction. The great
majority of individuals show clockwise hair-whorl, and therefore
clockwise asymmetry may be considered as the normal and coun-
ter-clockwise asymmetry as the reversed asymmetry. Various
writers have called attention to sporadic instances of reversed
crown whorl, and a few cases have been described for identical
twins.

Only one writer, however, has thus far made a systematic study
of crown whorl in twins. Lauterbach (1925) in a study of re-
semblances and differences in twins has presented some very in-
teresting data. Out of fifty-seven pairs of same-sexed twins
there occurred fifteen pairs in which one or both twins showed
counter-clockwise hair-whorl. In one of these cases both twins
were counter-clockwise. This means that about twenty-six per
cent, of the pairs of same-sexed twins showed asymmetry re-
versal in hair-whorl. In addition to these, there were six cases
showing double crown in which one half of the whorl has a clock-
wise and the other a counter-clockwise direction. These cases
are possibly comparable to ambidextrality in handedness and
should probably be listed as a form of asymmetry reversal. Add-
ing these six cases, the percentage of pairs showing more or
less reversed hair-whorl among same-sexed twins comes to nearly

STUDIES OF HUMAN TWINS. 303

37 per cent., not unlike the percentage of left-handedness in the
same set of twins, which was 35 per cent.

In our own collection of one hundred pairs of same-sexed
twins there are in all ninety-five pairs in which it was possible
'to determine the hair-whorl. In five pairs (three identicals and
two fraternals) the kinky or closely matted character of the hair
rendered diagnosis of hair-whorl extremely difficult or impos-
sible. Among the identicals there were twenty pairs showing
some form of asymmetry reversal in crown-whorl. In fifteen
pairs one twin showed clockwise and the other counter-clockwise
whorl, in three pairs both twins were counter-clockwise, and in
one pair one twin had a double crown and the other a clockwise
whorl. The remaining twenty-seven diagnosed pairs showed clock •-
wise hair whorl in both twins.

Among fraternal twins there were but five pairs having any
form of asymmetry reversal in hair-whorl. In four of these pairs,
one twin was counter-clockwise, and in one pair one twin had a
double crown and the other a clockwise whorl.

As in the case of handedness, there are doubtless instances of
incomplete asymmetry reversal that are not recognizable. Prob-
ably some of the crowns diagnosed as slightly clockwise or in-
definite may be cases of partial asymmetry reversal.

Crown-whorl has one advantage over handedness as a criterion
of reversal of asymmetry in that it is not subject to modification
by training and is therefore a somewhat surer sign of asymmetry
reversal than is left-handedness.

THE RELATION BETWEEN HANDEDNESS AND CROWN WHORL.

In only ten pairs of our identical twins do we find reversed
asymmetry of any sort (either left-handedness, counter-clockwise
hair-whorl, or both) in both twins of a pair. In eight of these
cases (73, 25, 23, 87, 43, 38, 7, 27) both twins of a pair are left-
handed or both have counter-clockwise hair whorl. It would
seem natural to assume that all such pairs have been derived
from zygotes, genetically left-handed. But what shall we do
with the other two cases (13 and 72) in which one twin of each
pair is plainly left-handed and the other clearly counter-clock-
wise in hair whorl? Since both of these indications are valid

304 H- H- NEWMAN.

criteria of reversed asymmetry there seems no escape from the con-
clusion that these two pairs also are derived from genetically left-
handed zygotes.

THE INCIDENCE OF ASYMMETRY REVERSAL IN OUR OWN COL-
LECTION OF TWINS.

When the present study began, the writer was keenly on the
lookout for evidences of asymmetry reversal in identical twins.
The expectation was that the more strikingly identical the twins
were, the more evidence of asymmetry reversal would be present.
Before the study was half over it seemed certain that this expecta-
tion was not to be realized. In fact, the very opposite of this
appeared to be true, namely, that the least evidence of asym-
metry reversal appears among those twins that are practically
indistinguishable, while the twins that are less nearly identical
show the most evidence of reversal of asymmetry.

In order to test out this conception, the writer tried to arrange
the fifty pairs of identical twins in the order of their closeness of
resemblance, including resemblances in features, height, weight,
headsize, finger prints and palm prints. After this was done,
Mrs. Blythe Mitchell, the one who has had the most intimate and
prolonged acquaintance with the twins, was asked to rearrange
the cases according to her impression as to the degrees of identity.
On the whole, there was a very close agreement, no case being
changed more than a few places up or down in the series. Using
the photographs, we worked over all cases together and arrived
at the arrangement shown in Table I., which is not intended to
be exact, place for place, but certainly represents a real grouping
in the sense that the first five pairs are more similar than the
second five, the second five than the third, the ninth five than the
tenth. Within groups of five the order might be more or less
shifted as the criteria of resemblance are not entirely objective,
but depend to a large extent upon one's judgment of degrees of
facial resemblance. In the following table asymmetries in hand-
edness, crown-whorl, and head dimensions are given for the
fifty pairs arranged in fives, beginning with the most alike and
ending with the least alike. In this table R and L indicate defi-
nite right- and left-handedness, / indicates partial left-handedness,

STUDIES OF HUMAN TWINS.

305

A indicates ambidextrality ; -(- indicates clockwise, or the com-
mon type of hair-whorl; and - - indicates counter-clockwise hair-
whorl. Double hair-whorls are indicated by (-f- -).

TABLE I.

t

,

T)

£ •

r*

^2

.2 d

_d

1

Hande
ness.

1 o
££

u£

J

0 •£
'<s> S
pq

Remarks.

62

A

cf

R

+

17.4

15-5

B

cf

R

+

17.2

15.2

98

A

9

R

+

17.7

i3-9

B

9

R

+

17-5

14.1

63

A

j

R

+

lS.2

14-3

B

cf

R

—

18.1

i3-9

40

A

cf

R

(+-)

17.85

15.0

B

cf

R

+

18.1

15-5

3

A

cf

R

?

20.0

14-75

Negroes. Crown-whorl

B

cf

R

?

19.7

i4-3

could not be made out.

9

A

9

R

_

17.0

13.6

A shows partial asymmetry

B

9

R

+

17.7

13-7

reversal in crown.

8o

A

9

R

+

I8.7

13-5

B

9

R

+

18.1

13-2

67

A

j

R

+

18.9

14.6

B

cf

1

+

18.7

14.1

B left-handed in wrist tap-

ping.

55

A

cf

R

+

ig.2

14.8

B

cf

R

+

19.1

14.4

35

A

tf

R

—

18.55

15.0

A shows partial asymmetry

B

cf

R

+

18.55

iS-i

reversal in crown.

96

A

cf

R

—

18.10

13-9

B

o*

R

+

18.15

13-9

*73

A

9

1

—

17.1

13-35

A, incompletely reversed

B

9

A

—

17-5

13-5

crown; left-handed in

finger tapping.

B, completely reversed in

crown; nearly ambidex-

102

A

9

R

—

17.8

15.0

trous.

B

9

R

+

18.1

15.0

*25

A

cf

R

—

18.3

14.2

B

cf

R

—

17.9

13-9

30

A

9

R

+

18.8

13-7

B

9

R

+

18.7

13-55

306

H. H. NEWMAN.

TABLE I. (Continued.)

^

_g

T-J

•s g

> *r

O _r*

•o -S

ctf b/3

N "2

Remarks.

dj *Z,

^

f/1

s a

1-1 ^

VjH "

W <U

rt C

c

a

m

*23

A

9

A

+

16.2

14-25

A more left-handed than B.

B

9

A

+

16.1

14.0

94

A

9

R

+

18.4

13-5

B

9

L

+

18.15

13-2

68

A

9

R

—

17-45

14-25

B

9

R

+

17-5

14.4

49

A

9

1

—

18.0

14.6

.

B

9

R

+

18.2

14.1

*I3

A

9

R

—

17.0

14-5

B

9

L

+

17.0

14.6

78

A

d1

L

+

18.45

14.7

B

d1

R

+

18.40

14.7

*87

A

d1

A

—

19-95

13-9

A more decidedly left-

B

d1

A

+

19.7

14.0

handed.

*43

A

d1

1

19.6

15-7

Bats left-handed naturally.

B

cT

1

+

19.1

15-3

Left-handed in finger tap-

ping.

*38

A

9

1

—

19.4

15-5

Left in both wrist and

finger tapping.

B

9

1

—

19.1

15-5

Left in finger tapping only.

79

A

d1

R

+

17-35

13-9

B

d1

L

+

18.1

13-3

* 72

A

d1

L

+

19-15

14-5

B

d1

R

—

19-25

15.0

99

A

d1

R

+

18.2

14-5

B

d1

R

+

18.25

14-5

33

A

d1

L

+

18.05

14.9

B

d1

R

+

18.05

15-25

53

A

d1

R

+

19.0

13-9

B

d1

L

+

18.8

13-4

44

A

d1

L

+

18.5

14-3

B

d1

R

+

19-5

13-9

2

A

9

L

?

17-45

14-45

B

9

R

?

17-95

14-35

91

A

9

R

+

18.0

13-8

B

9

R

+

17.9

13-9

IOO

A

d1

R

+

17-65

13-6

B

d"

R

+

17.2

14.0

STUDIES OF HUMAN TWINS.

307

TABLE I. (Continued.}

-a

.

'C °

'^

*

4i °5

TJ to

c u

H

v c

N ii

Remarks.

*

H

rt C

£

o£

a 5
^

55 a

pq

101

A

d1

R

+

18.6

15-5

B

d1

L

+

18.9

iS-5

70

A

d1

R

+

18.5

15-6

B

d1

L

+

18.1

15-3

37

A

d1

R

+

19-15

14-75

B

d1

1

+

19-35

14.65

Slightly left-handed in

wrist tapping.

34

A

d1

R

+

iS.8

14.8

Inclined to be ambidex-

trous.

B

d1

1

+

19.0

14.8

Slightly left-handed in

finger tapping.

28

A

9

R

—

16.8

13-8

B

9

R

+

17.7

13-9

*7

A

d1

1

_

17-3

iS-7

Strongly left-handed in

finger tapping.

B

d1

1

+

17.6

iS-5

Strongly left-handed in

finger tapping.

6

A

9

R

+

17.5

13-9

B

9

R

+

17.2

13-7

97

A

9

R

+

17-85

16.0

B

9

R

+

18.0

16.0

17

A

9

R

+

17-85

14-45

B

9

R

+

17-75

14.65

14

A

9

R

—

17.2

14-45

B

9

R

+

17-45

14-45

15

A

d1

R

+

19.65

14-6

Both ambidextrous as

B

d1

R

—

19-45

14.9

babies.

69

A

d1

R

+

18-75

14.9

B

d1

R

+

18.9

14-3

24

A

d1

R

4-

17-45

14.65

B

d1

1

+

17.4

14.65

Slightly left-handed in

playing marbles.

18

A

d1

R

+

18.6

13-8

B

cT

R

+

18.8

13-8

*27

A

d1

R

—

19-25

13-5

B

d1

1

—

19-3

13-7

Slightly left-handed in

wrist and finger tapping.

4i

A

9

L

p

17.26

15.2

Hair whorl could not be

B

9

R

p

17.6

15-6

determined.

60

A

9

R

+

17-6

14.1

B

9

1

+

17-5

14-15

Slightly left-handed in

finger tapping.

3o8

H. H. NEWMAN.

HANDEDNESS IN RELATION TO DEGREES OF RESEMBLANCE.

In this table there are listed twelve pairs of twins one member
of which is fully left-handed and, in addition to these, there are
eleven cases that show partial left-hancledness in one or both in-
dividuals of the pair. Besides the twenty -three cases showing
some degree of left-handedness, there are two cases in which
both members of the pair are classed as ambidextrous. Thus in
exactly fifty per cent, of our pairs of identical twins there is some
degree of left-handedness.

It is significant that the first case in the series to show complete
left-handedness is seventeenth out of fifty. There are only two
cases of partial left-handedness among the fifteen most strikingly
similar set of twins, while some degree of left-handedness be-
comes the rule rather than the exception from the sixteenth to
the end of the list.

CROWN WHORL IN RELATION TO DEGREES OF RESEMBLANCE.

The incidence of asymmetry reversal in crown hair whorl fol-
lows the same general lines as does left-handedness. In the first
ten pairs there is but one case (No. 63, in third place) that shows
true counter-clockwise hair-whorl. Two other cases (No. 9,
in sixth place, and No. 35, in tenth place) show a mixed hair-
whorl partly clockwise and partly counter-clockwise. There is
also one case of a double hair-whorl, one whorl being clockwise,
the other counter-clockwise (No. 40, in fourth place). The most
frequent incidence of counter-clockwise hair-whorl in one twin
occurs among the middle grade twins, neither the most alike or the
most different. This is true also of left-handedness, and such :\
correspondence in the incidence of two forms of asymmetry re-
versal must have some real significance.

THE RELATION BETWEEN HANDEDNESS AND HEAD SIZE.

In the following study both left-handedness and counter-clock-
wise hair-whorl are taken to be equivalent criteria of either genetic
or epigenetic reversal of asymmetry. For the present we shall
omit from consideration the ten pairs of twins (Nos. 73, 25, 23,
13, 87, 43, 38, 72, 7 and 27) that were diagnosed as derived from
zygotes genetically left-handed. These are starred in the list.

STUDIES OF HUMAN TWINS. 309

Before discussing the relation of head-size to handedness it
should be said that there is undoubtedly some inaccuracy in the
figures for head dimensions. Repeated measurements of the same
head rarely give exactly the same result. Dahlberg has calculated
that the average error in head measurements is about 0.5 mm.
It seems probable that our own errors were at least as great as
this, and probably greater. Hence differences of no more than i
mm. may be ignored or considered as without significance.

Glance with me down the list of forty pairs of identical twins
not previously diagnosed as derived from genetically left-handed
zygotes. In all, there are seventeen pairs in which one twin may
be classed as right-handed, the other left-handed, and in which
there is a significant difference in head size. In thirteen of these
pairs (63, 9, 67, 102, 68, 33, 53, 44, 2, 70, 28, 14 and 41) the right-
handed individual, derived from the superior side of the embryo,
has a distinctly larger head.

The four other cases (49, 79, 101 and 34) reverse this condition,
the left-hander has the larger head, though case 34 is ambiguous
in that one twin is slightly left-handed in tapping and the other
nearly ambidextrous and may therefore belong with the list of
ten diagnosed as derived from a left-handed zygote. The other
three cases (49, 79 and 101) are valid exceptions. Let us con-
sider these cases carefully. What would happen in the case of a
genetically left-handed zygote if one of the twins underwent
asymmetry reversal? Obviously the reversed twin would be a
right-hander, and should have the smaller head. This interpre-
tation appears to fit cases 49, 79 and 101. It would be strange if
some cases such as these did not occur in view of the existence of
genetically left-handed zygotes.

This hypothesis, that head size is correlated with handedness,
may be checked still further by examining the ten pairs of twins
diagnosed as derived from genetically left-handed zygotes. Of
these, eight show a significant difference in head size. These eight
cases deserve individual attention :

Pair No. 73. — This is a confusing case. Twin A shows left-
handedness in finger tapping and has a partially reversed hair-
whorl ; twin B is practically ambidextrous in tapping and has a
well-defined counter-clockwise hair-whorl, the only really posi-

-2IO H. H. NEWMAN.

tive indication of left-handedness present in the pair. This twin
(B) has the larger head.

Pair No. 25. — In this pair both twins are right-handed and
both have counter-clockwise hair-whorl. It is impossible to decide
which of these has been derived from the superior side or whether
they are derived from a right-handed or left-handed zygote, for
the handedness and hair-whorl completely contradict each other.

Pair No. 23. — Both twins are ambidextrous, and both have
clockwise hair-whorl. Twin A, with the larger head, is more
nearly left-handed than B.

Pair No. 87. — Twin A, while ambidextrous, tends to be more
left-handed than B, and has counter-clockwise hair-whorl; twin
B is ambidextrous and has clockwise hair-whorl. Evidently A is
the left-handed (superior) individual, and he has the larger head.

Pair No. 43. — Both twins are partly (probably natively) left-
handed. Twin A has counter-clockwise hair-whorl, twin B clock
wise. A, the more distinctly left-handed twin, has the larger head.

Pair No. 38. — Both twins are partially left-handed and both
have counter-clockwise hair-whorl. A is left-handed in both wrist
and finger tapping; B, only in finger tapping. A, the more left-
handed, has the larger head.

Pair No. J2. — Twin A is strongly left-handed but has clockwise
hair-whorl ; twin B is right-handed but has counter-clockwise hair-
whorl. It is impossible to say which individual should be diag-
nosed as from the superior side, since the two criteria seem to be
of equal value. Of the two the reversed hair-whorl is somewhat
safer as a criterion, and it happens that the twin (B) with the
counter-clockwise hair-whorl has the larger head.

Pair No. 27. — Twin A is right-handed ; B, slightly left-handed
in wrist and finger tapping. Both have counter-clockwise hair-
whorl. Twin B, the partially left-handed member of the pair,
has the larger head.

All of these eight cases except pair 25, which is neutral, support
the conclusion that the twin derived from the genetically superior
side (the right side in these cases) of the embryo has the larger
head.

One other class of cases remains to be dealt with, those in which
a significant difference in head size exists without any complete

STUDIES OF HUMAN TWINS. 31!

asymmetry reversal in handedness or hair-whorl. There are nine
such pairs (62, 3, 80, 55, 30, 6, 97, 69, 18). In all but two of

»

these cases (3, 55, 30, 6, 97, 69, 18) one twin was definitely
more right-handed than the other and the more right-handed in-
dividual has the larger head in all pairs. In pairs 62 and 80,
both twins are equally strongly right-handed and offer neutral
evidence. Instead of weakening the general theory, then, all of
these cases, where varying degrees of difference in right-handed-
ness but no true left-handedness occur, tend strongly to support
it. There is beyond question a strong correlation between hand-
edness and head size. With very few exceptions indeed, the twin
having the larger head shows evidence of having been derived
from the genetically superior side of the embryo; from the left-
hand side in twins derived from zygotes genetically destined to
form right-handers, and from the right side of zygotes destined
to form left-handers.

TWINNING AND THE ASYMMETRY MECHANISM.

The data just presented have given rise to a theory that seems
to rationalize for the first time the peculiar incidence of reversal
of asymmetry in twins. It is well known that in some groups
of animals, notably those characterized by a striking degree of
determinate cleavage, bilateral symmetry and asymmetry are es-
tablished in the undivided zygote before or at the time of the
first cleavage. In those forms, on the other hand, that show a
strong tendency toward indeterminate cleavage, notably the ver-
tebrates and echinoderms, symmetry and asymmetry are not
definitely fixed until considerably later in development. The
writer's work (Newman, 1924) on asymmetry reversal in the
starfish indicates clearly that asymmetry is fixed before the time
of gastrulation, for no reversal of asymmtry could be induced in
embryos older than late blastulse.

There are also indications among the vertebrates that asym-
metry is established prior to or during gastrulation. Thus in the
nine-banded armadillo, the only case of twinning among mammals
where the stage at which twinning occurs is definitely known, it
has been found that the first step in the twinning process usually
precedes the period at which symmetry and asymmetry are es-

312

H. H. NEWMAN.

tablished and that the second step in twinning takes place during
the process of the establishment of the axis of symmetry. By
analogy, we may infer that twinning in man takes place in close
association with, and possibly as an aberration of, the process of
establishing and fixing the relations of symmetry and asymmetry
in the embryo.

Now, since no biologic processes takes place with the same
clock-like precision in all specimens, we may suppose that the
twinning act in some cases is consummated during relatively early
stages of the establishment of symmetry and asymmetry, and that
in other cases it is established later. In the cases in which twin-
ning occurs relatively late, the establishment of a single bilateral
individual may have gone so far that complete twinning is im-
possible. This is probably the case in all partial twinning, re-
sulting in conjoined twins and double monsters. In such twins
one of the most striking features is the occurrence of profound
reversal of asymmetry, as expressed in more or less complete
situs inversus mscerum.

If then, we may assume that conjoined twins with the most
extreme reversal of asymmetry in the inferior component, repre-
sents one end of the series of twins, it is natural to assume that
the opposite end of the series is represented by cases in which
twinning is consummated before any asymmetry is fixed. In such
cases the twins would be derived from two equivalent primordia
which had not yet been differentiated into right- and left-hand
sides. When, later, asymmetry comes to be established in these
two genetically equivalent and still undifferentiated embryos, it
should follow the same course in both and each should develop
the same asymmetry as the embryo would have done had it not
undergone twinning. Thus, if the original embryo was genet-
ically a right-hander, two right-handed twins should result; simi-
larly, if the original embryo was genetically a left-hander, two
left-handed twins should result — a condition not uncommon
among twins, but hitherto unexplained. In such twins we would
expect a high degree of same-sided asymmetry in such details as
palm and finger prints, ear shape, dentition, handedness, hair-
whorl, etc. Moreover, since the two twins are derived from two
primordia that have not yet become differentiated as right- and

STUDIES OF HUMAN TWINS. 313

left-hand components, the two resulting twins would be expected
to be very strikingly similar, more similar than would be twins
separated after asymmetry had been more or less fixed in the
embryo from which they are derived.

Thus the earlier twinning occurs with respect to the establish-
ment of asymmetry, the more similar should he the resultant twins
and the less should they show such evidences of reversal of asym-
metry as left-handedness and counter-clockwise hair whorl. This
explains why these criteria of asymmetry reversal are rarely pres-
ent in the most strikingly similar twins and are increasingly com-
mon among identical twins that are less similar.

If this theory be sound, and there is much evidence in its favor,
we have discovered another mechanism, not classifiable as envir-
onmental, that operates to make identical twins different. This
factor, the asymmetry mechanism, may be the main, if not the
only, factor responsible for observed differences between identical
twins reared together. Consequently it would be quite unsafe to
infer that any differences between such twins are due to differ-
ences in environment or in training. On the other hand, once
we have established the average degree of difference between
identical twins reared together, we should be able to use this as
a base line in determining to what extent, in cases of identical
twins reared apart, the differences in environment have operated
to increase the physical or mental difference.

This theory goes far to explain why some, but not all, pairs of
twins show left-handedness and counter-clockwise hair-whorl in
one twin of a pair ; why there should be occasional cases in which
both twins of a pair are left-handed or have counter-clockwise
hair-whorl ; why there should be various degrees of incomplete
asymmetry reversal as the result of separation of twins prior to
complete establishment of asymmetry. The establishment of
asymmetry is a progressive process and takes some time to be-
come fully fixed. Hence we may expect to find that twinning
early in the process will result in little if any signs of asymmetry
reversal in one of the twins, and that twinning occurring late in
the process will result in extensive reversal of asymmetry in one
of the components.

In brief, this theory seems to clear up many if not all the

314

H. H. NEWMAN.

formerly baffling asymmetry situations found in twins. It lacks
experimental confirmation, but this must be so from the nature of
the material. Yet the data themselves almost speak out the theory
of their own accord.

SUMMARY.

1. Reversal of asymmetry in monozygotic twins expresses itself
in varying degrees, ranging from complete situs inversus viscerum
in conjoined twins to left-handedness or counter-clockwise hair-
whorl in separate twjns.

2. There are two kinds of handedness : genetic and epigenetic.
Genetic right- and left-handedness have about the incidence, re-
spectively, of dominant and recessive allelomorphs. Epigenetic
left-handedness (or in genetic left-handers, right-handedness)
results from twinning, the inferior side having an asymmetry
opposite to that of the superior side.

3. Arranging fifty pairs of identical twins in the order of their
closeness of physical resemblance, it is found that there is very
little evidence of asymmetry reversal among the most similar
twins, while the less similar twins show a high degree of it.

4. Clockwise hair-whorl has about the same incidence as right-
handedness, and counter-clockwise hair-whorl that of left-hand-
edness.

5. Varying degrees of partial left-handedness and of ambidex-
trality are revealed by tapping tests.

6. Ten pairs of identical twins show asymmetry reversal in
both members of a pair and are therefore diagnosed as derived
from genetically " left-handed " zygotes ; three pairs showing
asymmetry reversal in but one twin should probably be classed
as " left-handers " ; the remaining thirty-seven pairs are believed
to be derived from right-handed zygotes.

7. There is a very close correlation between head size and
handedness. The twin derived from the superior side of the em-
bryo nearly always has a significantly larger head.

8. The reason why many but not all identical twins show asym-
metry reversal in one twin is that the epigenetic establishment of
asymmetry takes place sometimes before and sometimes after
twinning. If it takes place before twinning the twins will show

STUDIES OF III "MAX TWINS. 315

a high degree of asymmetry reversal: if it takes place after the
twinning the twins will both show the same asymmetry and be in
other respects more alike than when the establishment of asym-
metry precedes twinning; if it takes place during the twinning
process the twins will show varying degrees of asvmmetry re-
versal in one individual and varying degrees of close resemblance
in physical and mental characters.

REFERENCES.
Dahlberg, G.

'26 Twin Births and Twins from an 1 Invditary Point of View.

Stockholm.
Jones, W. F.

'18 A Study of Handedness. Vermilion, S. D.
Jordan, H. E.

'ii The Inheritance of Left-handedness. Anicr. Breeders' Magazine,

2: 19-29.
Lauterbach, C. E.

'25 Studies in Twin Resemblance. Genetics, Vol. 10.
Meirowsky, E.
'27 Zwillingsbiologische Untersuchungen. Arch. f. Rasscn und < u -scll-

schaftsbiol., Bd. 18.
Verschuer, O. von.
'27 Der Vererbungsbiologische Zwillingsforschung. Ergeh. Innerui.

Med. u. Kinderheit., Bd. 31.
Newman, H. H.
'16 Heredity and Organic Symmetry in Armadillo Quadruplets. Bun,

BULL., XXX.

'17 The Biology of Twins. Chicago.
'21 The Experimental Production of Twins and Double Monsters in tin-

Larvre of the Starfish Pateria. Jour. Exper. Z<>"1.. Vol. 33.
'23 The Physiology of Twinning. Chicago.
Siemens, H. W.

'24 Die Bedeutungs der Zwillingspathologie fur die aetiologischo
Forschung elautert an Beispeil der Linkhandigkeit. Setzungsber. •!.
Ges. f. Morph. u. Physiol. in Munich, Jhg. 35.
Weitz, W.

'24 Studien an eineiigen Zwillingen. Zeitsch. f. Klin. Med., Bd. 101.

Vol.LV.

November, 1928.

No. 5

BIOLOGICAL BULLETIN

SEX DIFFERENTIATION IN GONADS DEVELOPED

FROM TRANSPLANTS OF THE INTERMEDIA'!!;

MESODERM OF AMBLYSTOMA.

R. R. HUMPHREY,

DEPARTMENT OF ANATOMY, SCHOOL OF MEDICINE,
UNIVERSITY OF BUFFALO.

INTRODUCTORY AND HISTORICAL.

From his studies on parabiotic twins in A mblystoma punctatum
Burns ('25) is led to the conclusion that in this species there may
occur a complete reversal of sex previous to the period of sex-
differentiation. Embryos joined in pairs in early stages should,
by the laws of chance, be combined in the proportion of
i cf cf : i cf 9 :i 9 cf : i 99. Instead of this expected ratio,
Burns obtains exclusively one-sexed pairs, in the proportion of
44 cf cf to 36 99. This result he is inclined to interpret as a
i : i ratio. Having no evidence that the two-sexed pairs had
been eliminated through selective mortality, Burns postulates
that half the pairs reared must have been, originally, 9 cf
combinations; in these pairs, from a condition of near-equilibrium
as regards sex, one or the other sex, he assumes, had eventually
gained the ascendency, so that at sex-differentiation the gonads
of the two members of the pair were identical. Since the sex-
ratio found was approximately i cf cf : i 99, Burns infers that
there can be no prepotency constantly favoring either male or
female, since in this event a 3 : I ratio favoring the prepotent
sex would be expected.

The more recent studies of Witschi ('27) on frog embryos
joined in parabiosis show that in these amphibia the early sex
reversal assumed by Burns does not occur. Witschi finds in 56

21

R. R. HUMPHREY.

pairs the following combinations: cf cf , 16 pairs; cf 9 , 17 pairs,
with 7 of the females undergoing sex-reversal ; 9 cf , 10 pairs, with
4 of the females undergoing sex-reversal; 9 9 , 13 pairs. This
approximates very closely the expected ratio of i cf cf : i cf 9 :
i 9 cf : i 99, and shows conclusively that there could have
been no sex-reversal previous to the time of sex-differentiation.
From the fact that in many of the two-sexed pairs the females
were found undergoing sex-reversal, while a female united with a
male undergoing reversal was never found, Witschi concludes
that the male is always dominant in the sex-reversal which
finally occurs. Though he believes that the independent sex-
differentiation in the individuals of genetically two-sexed pairs
favors the theory of localized sex-differentiators ("lokalisierte
Innenfaktoren," probably comparable to Spemann's "Organisa-
toren"), he states that in the later sex-reversal of the female of
the pair, "the cooperation of hormones is not improbable."

The method of parabiosis used by Burns and Witschi has
certain obvious disadvantages. If used with a species in which
the zygotic sex-determination can be completely reversed previous
to sex-differentiation, as is possibly the case in Amblystoma, there
can be no certainty regarding the original state of any one-
sexed pair examined after sex-differentiation has taken place.
In drawing conclusions as to the occurrence of sex-reversal in
these one-sexed pairs, one must depend entirely upon the sex-
ratio obtained. Further, if the death rate among pairs joined in
parabiosis is high, the possibility of a selective mortality cannot
be entirely eliminated, even though evidence in favor of it may
be scanty or lacking. Hence there is no absolute proof that the
one-sexed pairs found at sex-differentiation were not all of this
character genetically at the time they were joined; the proof of
sex-reversal, therefore, remains inconclusive.

The method of parabiosis is relatively advantageous if used
with a species in which an early reversal of sex does not occur
(as Rana sylvatica; Witschi, '27). In such a species, pairs pre-
served at a suitable period in development would show the
actual progress of sex-reversal in one member of the pair. If,
however, the reversal becomes complete, all pairs killed at later
periods would be found to be one-sexed. Although sex-reversal
could be confidently asserted for a species of this type as a result

SEX DIFFERENTIATION IN GONADS. 319

of the disappearance of the two-sexed condition observable in
younger pairs, the identity of any of the originally mixed pairs
could be established in adult animals only with great difficulty
if at all.

In the spring of 1926 the writer undertook to transplant the
intermediate mesoderm of Amblystoma from one embryo to a
latero-ventral site in another in order to determine the fate of
the primordial germ cells included in such grafts. Among the
seven embryos surviving the implantation was one in which at
forty-four days after operation the germ cells of the graft were
found to have given rise to a gonad of considerable size
(Humphrey, '27). This suggested the possibility that such
grafts, if allowed to develop until after the period of sex-
differentiation of the host, might be found to contain gonads
which had likewise undergone sex-differentiation. The donor
serving as the source of the transplant, and the host into which
it was engrafted, though selected at random long before sex-
differentiation had occurred, must in many cases be unlike in sex.
Since the donor could be reared, its sex could be determined
from the gonad it possessed, and since donor and host were not
joined, the sex-differentiation in either could not be influenced by
the other, except in so far as the graft might be able to modify
the sex-differentiation of the host. If, then, after sexual differ
entiation the gonad of the graft were found to agree in type
with that of the host regardless of the sex of the donor, the fact
of an early sex-reversal would be established beyond question.
If, on the other, hand, the gonad of any graft differed in type
from that of the host, agreeing with that of the donor, it would
show conclusively that sex-reversal previous to sex-differentiation
had not occurred. By the method of grafting, therefore, it
seemed possible to obviate certain difficulties inherent in the
method of parabiosis. The donor furnishing the graft would
undergo sex-differentiation according to the factors present in
the egg at fertilization; its gonad could be compared directly
with the gonad developed in a transplant removed during the
germ-layer stage and grown in a host of the opposite sex. Con-
clusions as to the occurrence of sex-reversal, therefore, could be
drawn from comparison of structures rather than by reasoning

320

R. R. HUMPHREY.

from sex ratios in which the factor of selective mortality might
possibly be involved.

MATERIAL AND METHODS.

The removal of the intermediate mesoderm (preprimordia of
gonad and mesonephros) of Amblystoma and its implantation into
another embryo is a relatively simple operation, the technique
for which has been outlined elsewhere (Humphrey, '27). During
the operating season of 1927, 180 such implantations were
carried out. The graft always included a large part of that
region of the intermediate mesoderm in which it had been found
that primordial germ cells develop (i.e., the territory of the
seventh to the seventeenth somites, approximately) ; in addition
it included parts of the adjacent axial and lateral mesoderm,
together with the overlying ectoderm.

Following operation, the host receiving the transplant and the
donor furnishing it were reared to the age of fifty days or over—
i.e., until after the beginning of morphological sex-differentiation.
At autopsy of the host the graft derivatives were found, as a
rule, attached to the inside of the ventral or lateral body wall.
In the donor, at autopsy, the gonad was always very small or
entirely lacking on the right, the side from which the transplant
was invariably taken in the embryo.

RESULTS.

Of 180 pairs (donor and host) only 49 or 27 per cent, of the
.total, were reared to the age of 50 days or over. This, however,
does not indicate an actual mortality of 73 per cent, in the
grafted animals, since 25 additional hosts were reared to the age
of 50 days or more, although the donors which had furnished
them transplants had died in early stages of development.
Several hosts were also killed before reaching the age of 50 days,
in order to study the development of the gonad and other
structures in the graft; these hosts were always those of pairs
from which the donor had already died from operative injury
or other causes. In all, 74 grafts were recovered after sex of
the host had become distinguishable. Of these grafts, 40 con-
tained a gonad, the sex of which was determinable with a fair
degree of certainty in 33 cases. In the remaining grafts the

SEX DIFFERENTIATION IN GONADS. 32!

gonad was small with few germ cells and no features permitting
it to be classified as either ovary or testis.

Unfortunately for this study, the majority of the gonads which
developed were in homoplastic transplants in Amblystoma jeffer-
sonianum. In this species, instead of the expected i : I sex-
ratio, the animals reared in the laboratory in 1927 were in the
proportion of 56 females to 19 males, essentially a ratio of 3 : I.
As a result of the predominance of females, donor and host were
both of this sex in an excessive number of cases. In only two
instances were donor and host unlike in sex and in these, un-
fortunately, the gonad of the graft was in each case of somewhat
atypical structure due to unfavorable environmental factors.

To the writer's knowledge a sex-ratio such as the one here
reported for Amblystoma jeffersonianum has not been previously
recorded for this species. Whether it is to be explained on the
basis of a selective mortality among operated animals, or whether
it is due to an induced reversal of sex in certain males resulting
from nutritive disturbance or other alteration of environmental
factors, or whether an excess of females is a normal condition in
this species or at least in its local strain, cannot be positively
stated. It is worthy of note that in Amblystoma maculatum
(punctatum) reared in the laboratory under identical conditions
and after similar operative procedure, the sex-ratio is apparently
quite normal. The collection of large numbers of A. jeffer-
sonianum larvae from local ponds and a study of their sex-
ratio has not been possible. The few specimens picked up near
ponds after metamorphosis have been found to be females in
the great majority of cases.

A second feature of interest noted particularly in this species
is the occurrence of spermatocyte stages in the testes of males
60 to 80 days of age. This cannot be due to the presence of a
graft furnished by a female, since spermatocytes are no more
frequent in hosts than in donors. Though Burns ('25) makes no
mention of spermatocytes in A. maculatum of similar age, the
writer has encountered such stages occasionally in this species
as well as in A. jeffersonianum. In the latter, however, they
occur in a higher percentage of the males examined, and usually
in greater numbers than in A. maculatum. In neither species,
were the spermatocytes found in stages later than the pachytene

322

R. R. HUMPHREY.

condition of the heterotypic prophase. Though the presence of
heterotypic prophases in males renders these stages of little
value as a criterion of sex when considered alone, it may be noted
that their number in the male is small as compared with the
number of other germ cells, and that they were not found in the
diplotene or later stages characteristic of the oocytes of amphibian
females.

Of 56 Amblystoma maculatum reared in the laboratory in 1927
30 were females and 26 males. These numbers give an approxi-
mation of the expected I : I ratio. In this species, however,
the majority of the transplants used were furnished by very
young donors (stages 21 to 25 *) and but few gonads developed.
In only two cases in which the sex of the donor was known to
differ from that of the host was a gonad present in the graft.
In one of these two the gonad was small and of the indifferent
type, while in the second it was of a type combining features of
both ovary and testis.

From the above it may be seen that relatively little evidence
bearing upon the problem of sex-reversal was obtainable from
grafts the donors of which had survived to sex-differentiation.
But in several cases in which the donor had died before reaching
this period, the transplant furnished by it was found to contain
a gonad differing in sex type from that of the host in which the
graft had developed. In these cases it would appear that donor
and host must have been unlike in sex, but that the gonad of the
graft had differentiated in a fashion determined by the organiza-
tion of the transplant previous to its isolation from the donor.
These cases may now be described in some detail.

No. 211. — Transplant from A. jeffersonianum of stage 29
implanted in A. maculatum of stage 25. The donor died 18 days
after operation. The host, killed 61 days after operation, proved
to be a female. A section of the ovary is shown in Fig. i. The
central ovarian cavity is well developed, and the germ cells are
peripheral in position. Their nuclei are largely in the heterotypic
prophase stages characteristic of the early urodele ovary, although
few in number or lacking in the testis, as a rule, until a much
later period of development. The graft removed from this host

1 The stages referred to throughout this paper are those of Harrison's series of
standard stages.

SEX DIFFERENTIATION IN GONADS. 323

included a fairly large gonad of testicular type (see Fig. 2). No
central cavity is present. The germ cells are somewhat uniformly
scattered through the organ, intermingled with numerous smaller
cells which constitute the 'sex cords' (anlagen of duct system),
and the stromal and sustentacular elements of the testis. The
germ cells are all in spermatogonial stages; heterotypic prophases
are entirely lacking.

No. 284. — Transplant from A . maculatum of stage 30 implanted
in host of same species and stage. The donor was killed by the
host 37 days after operation. The host, autopsied 58 days after
implantation of the graft, is a female. Although sex-differ-
entiation had but recently occurred, the ovary has the charac-
teristic central cavity and peripheral oocytes with nuclei in
heterotypic prophase (see Fig. 3). The gonad found in the
transplant is a pear-shaped testis attached by a stalk to the
surface of the graft mesonephros. It lacks the central cavity
characteristic of the ovary, and shows the more uniform distri-
bution of germ cells typical of the young testis (see Fig. 4). No
heterotypic prophases are present, all germ cells being in sper-
matogonial stages.

No. 244. — Transplant from A. jeffersonianum of stage 31,
implanted in host of same species and stage. Donor presumably
devoured by host at about 31 days after operation. The host,
autopsied 61 days after operation, is a male; a section of one
testis is shown in Fig. 5. As is frequently the case in males at
this stage of development, the testes of this animal show a few
cells in the spermatocyte stage, but the germ cells are distributed
in the fashion characteristic of the testis, and no central cavity is
present. For comparison with the testis of the host a section
of the gonad of the graft is shown in Fig. 6. This gonad must be
interpreted as an ovary in an early stage of sex-differentiation.
Although no central cavity is yet present, the germ cells are
arranged in a layer around the periphery of the gonad and are
for the most part oocytes in earlier stages of the heterotypic
prophase. By comparison of Fig. 6 with Figs. I and 3 (ovaries
of fairly early stages of differentiation) it will be readily appreci-
ated that this graft gonad is ovarian in nature. The differences
between it and the graft gonads of Figs. 2 and 4 (testes) are

324

R. R. HUMPHREY.

clearly evident from the photographs, and need no further
comment.

The three cases above described show clearly that a gonad
developing in a graft need not agree in sex type with the gonad of
the host. It may be logically inferred that in these three cases
the sex-differentiation of the graft gonad was determined by the
organization in the implanted mesoderm previous to its removal
from the donor embryo.

In a few cases in which both donor and host lived until after
sex-differentiation and were found to be of unlike sex, a gonad
was present in the graft. These cases, however, are less satis-
factory than the preceding, since the gonad of the graft is either
in an early stage of sex-differentiation or is of atypical structure.
Three such cases will now be described.

No. 207. — Transplant from A. jeffersonianum of stage 29
implanted in A. maculatum of stage 25. The host, killed 61
days after operation, is unquestionably a male, although a few
germ cells in heterotypic prophase are found in one of the testes.
A section of the testis is shown in Fig. 7. The donor, a female,
was not killed until 78 days after operation. The gonad shown
in Fig. 8 is therefore more advanced in development than the
testis of Fig. 7. The gonad found in the graft is small and in
an early stage of sex-differentiation. Although no central cavity
is present, the germ cells tend to take a peripheral position.
Of the 95 germ cells present, 38 are in early stages of the
heterotypic prophase. Considering all its structural features,
this gonad should be classed as an ovary. In the peripheral
arrangement of its germ cells, and in the high proportion of
these cells found in heterotypic prophase, it is clearly similar
to the gonad of the donor rather than to that of the host.

No. igo. — Transplant from A. jeffersonianum of stage 33
implanted in host embryo of same age and species. The host,
killed 64 days after operation, is a female (see ovary in Fig. 10).
The donor, killed at the same age as the host, is a male (see
Fig. n). The gonad of the graft is atypical in structure in
that an unusual amount of stroma is present, in the form of a
mucous type of connective tissue (Fig. 12). It may nevertheless
be classed as testis rather than ovary. The germ cells, though
frequently included in the covering epithelium, are predominantly

SEX DIFFERENTIATION IN GONADS. 325

scattered through the central part of the organ. No central
cavity is present. Sex cords (anlagen of duct system of testis)
are recognizable as groups or strands of smaller cells, in some
sections extending a third or more of the length of the gonad.
The germ cells are for the most part spermatogonial in type,
only three or four of the several dozen present being in heterotypic
prophase, and none of these having the characteristics of growing
oocytes. Though of atypical structure, this gonad cannot be
considered as undergoing transformation from testis into ovary.
Aside from the abundance of mucous connective tissue, its
structural features are clearly similar to those of the testis in
the donor. Atypical gonads of the same general appearance may
develop in grafts from a male donor implanted in a male host.
The peculiarities of structure exhibited are therefore due, prob-
ably, to the action of local environmental factors rather than to
the activity of sex hormones secreted by the gonads of the
host.

No. 188. — Transplant from A. jeffersonianum of stage 33
implanted in host of same age and species. Both donor and host
were killed 64 days after operation. The host is a female, the
donor a male (see Figs. 13 and 14). The gonad of the graft is an
atypical structure difficult to classify (see Fig. 15). Neither
typical ovarian cavity nor testicular duct system is recognizable.
The germ cells are predominantly peripheral in location, although
frequently scattered or in masses deeper within the stroma.
In one instance a mass of germ cells lies in a cavity, with no
apparent attachment to other tissues of the gonad; these cells
show marked degenerative changes. The cells at the periphery
of the gonad frequently exhibit a grouping or 'nesting' com-
parable to that of young oocytes in a normal ovary. Though
for the most part in heterotypic prophase (several hundred such
cells must be present) these germ cells seem never to progress
beyond the pachytene stage. If the gonad were actually ovarian,
some few at least of these cells might be expected to pass through
the diplotene stage and then enlarge as growing oocytes. This
has been found to occur in those atypical gonads which have
developed in grafts from female donors. In this gonad, how-
ever, no growing oocytes are present, numerous pyknotic and
fragmenting nuclei indicating the degeneration of the germ cells

326 R. R. HUMPHREY.

during the pachytene stage rather than their continued de-
velopment.

While it might appear at first glance that the features exhibited
by this gonad have resulted from the action of the hormones of
the host, it is highly probable that many of its peculiarities are
referable to the growth potentialities of the implanted tissue as
modified through local environmental influences. The donor
furnishing this particular transplant exhibits an unusual number
of spermatocytes in its one (left) gonad. Four such cells may
be recognized in the section shown in Fig. 14 (at left). Pre-
sumably the tissue implanted possessed the potentiality for
developing a gonad in which unusual numbers of heterotypic
prophases would have appeared precociously, even without an
endocrine stimulus from a female host. As to local environ-
mental conditions, it may be noted that the gonad was attached
by a very delicate fold of tissue, and was apparently poorly
vascularized. The latter condition alone would be unfavorable
to the development of a gonad of normal histological structure.

In addition to the graft gonad above described (No. 188) two
other specimens exhibit features which might possibly be in-
terpreted as modifications due to the action of sex-differentiating
hormones. In one of these the graft gonad consists of a central
core of testicular character overlaid by a cortex ovarian in type.
This structure resembles the modified testes described by Burns
('28) as resulting from the action of ovarian hormones. The
position of this particular graft in the body of the host, however,
is such that some of the primordial germ cells of the host may
actually have entered into the make-up of the graft gonad. If
this be the case, this structure must be regarded as a 'mosaic'
gonad derived from two preprimordia of unlike sex-potentialities
rather than as a testis undergoing sex-reversal due to the
endocrine influence of a female host. It is significant that graft
gonads developing in sites sufficiently far ventral to exclude the
possibility of actual contribution of host germ cells generally
show no indication of sex-reversal (see Figs. 2, 4, and 6).

Among those cases in which only the host survived until the
period of sex-differentiation are seven in which the gonad of
the graft agrees in type with those of the host. While a reversal
of sex in these few cases cannot be positively excluded, it is

SEX DIFFERENTIATION IN GONADS. 327

rendered exceedingly improbable by the fact that in five other
cases the gonad of the graft is of opposite sex from those of the
host. Examples of this latter group have already been described
(Nos. 211, 284, and 244; Figs, i to 6).

DISCUSSION.

The outstanding feature of the results described in the pre-
ceding pages is the apparently independent sex-differentiation of
the gonads which develop in grafts. Although in one or two
cases such a gonad has been modified in a fashion suggesting an
influence from sex hormones of the host, in no case is a complete
early reversal of sex clearly indicated. So far as can be deter-
mined from cases in which the sex of the donor is known, the
primary sex-differentiation in the gonad of the graft always
proceeds in a fashion determined by the sex of the donor. In
five cases in which the sex of the donor is not known, the gonad
is of opposite sex from that of the host. In four of these. cases,
gonads with the features characteristic of a testis have differ-
entiated in grafts implanted in female hosts, while in the fifth
an ovary has developed in a graft implanted in a male.

It is difficult to reconcile these findings with the conclusions
reached by Burns ('25) from his studies on the sex of parabiotic
twins in Amblystoma. Burns finds that the sex of the two
members of any pair is always the same. From this he is led to
infer that complete reversal of sex has occurred in one member
of all two-sexed pairs, such reversal being accomplished before
sex differences in the gonads become morphologically distin-
guishable. He assumes that when embryos of unlike sex are
joined in parabiosis there results a condition of close balance or
unstable equilibrium, which is broken if one animal of the pair
gains a slight advantage, presumably through earlier or more
abundant output of sex-differentiating hormones. All hormones
being mingled in the blood stream, and neither sex being con-
stantly prepotent, either the male or the female hormone may
become dominant. Such domination being established before
the onset of morphological sex-differentiation, the phenomena of
this period will be identical in the two members of any parabiotic
combination, or essentially so. The twin which has undergone
reversal thus differentiates directly without first exhibiting the

328

R. R. HUMPHREY.

sexual characters to be expected from its genetic constitution.
Under these conditions, a reversal cannot be detected by study of
developmental stages of the gonads but must be inferred from the
absence of two-sexed pairs after morphological differentiation has
been completed, unless it be assumed that all such pairs have
been eliminated through a selective mortality.

In discussing his results, Burns considers the possibility that a
'selective' mortality has operated to eliminate all heterogeneous
(male-female) pairs, permitting only homogeneous pairs to sur-
vive. While this explanation cannot be positively rejected,
Burns regards the occurrence of a selective mortality as highly
improbable. Although the death rate among his operated
animals is very high (about 77 per cent.), he believes that it is
possible to explain it without postulating a physiological incom-
patibility of the sexes so profound as to induce the death of all
two-sexed pairs. Witschi ('27) has demonstrated that no such
incompatibility exists in the frog, since he finds the expected
number of mixed pairs at metamorphosis of his parabiotic
animals. If we assume that among Burns's experimental animals
there was likewise no selective mortality eliminating mixed pairs,
we are forced to conclude that parabiosis induces an early sex-
reversal in one member of every two-sexed pairs.

If sex-reversal in parabiotic twins in Amblystoma be assumed to
have occurred in the manner postulated by Burns, it would be
logical to expect a reversal of sex in the gonad of a graft implanted
in a host of opposite sex from that of the donor. The bulk of
the transplant is small compared with the entire body of the host,
and the gonad to which the graft gives rise is but a fraction of
the size of the host's own gonads. Under these conditions there
should exist no state of near-equilibrium as regards sex. If sex-
differentiating hormones are produced previous to morphological
sex-differentiation, those of the host should always, from their
greater abundance, be able to dominate the differentiation of
the gonad in the graft;2 the latter, therefore, should always

2 The gonad of the graft is often somewhat retarded in development as com-
pared with those of the host, possibly, in some cases, because of inadequate nutrition.
Such retardation of its development should favor modification of. the graft gonad
by the gonads of the host, assuming that sex differentiating hormones are poured
into the circulation when the gonads reach a certain stage in their differentiation.

SEX DIFFERENTIATION IX GONADS. 329

agree in type with the gonads of the host. Yet the gonad of a
graft is clearly able to develop as ovary in a male host, or as
testis in a female. In none of my animals could sex-reversal be
demonstrated as having preceded the primary sex-differentiation.

Since the extent to which a hormone may modify an embryonic
structure probably depends in part upon the period of develop-
ment at which it is introduced and the time during which it is
allowed to act, these conditioning factors may well be compared
for parabiotic twin and graft.

In Burns's experiments, Amblystoma embryos were joined in
parabiosis at about stage 28 of Harrison's series. In my own
experiments many of the grafts were implanted at this or even
earlier stages. In none of the cases considered in this paper
was either donor or host more advanced in development than
stage 34 at the time of operation. In neither the parabiotic
twins at the time of union nor in the host receiving an implant
has the blood yet begun circulation. While it is probable that
the blood streams of embryos joined in parabiosis are in com-
munication from the time the circulations of the two first become
established, my observations indicate that the graft becomes
vascularized at a correspondingly early period in its development.
In short, the sex-modifying influence of the host upon the graft
should be exerted fully as early as the influence of an embryo
upon its parabiotic twin, assuming that this influence is mediated
through the activity of substances transported by the blood.

As regards the actual time elapsing between operation and
autopsy, the advantage appears to lie with the parabiotic twins.
Burns states that among the pairs of his series even the best did
not show sex-differentiation until seventy days, while the general
average required considerably longer (eighty to ninety days)
for sex to become clearly distinguishable. In my own animals
sex was usually determinable without difficulty at fifty days
after operation. The longer indifferent period in the parabiotic
twins doubtless results chiefly from growth retardation due to
difficulties in feeding. In any event, it greatly increases the
period over which one animal is subjected to the influence of the
other before morphological sex-differentiation occurs. Possibly
in this prolonged indifferent period the physiological state of
the gonads in one animal may be so altered through the influence

330

R. R. HUMPHREY.

of its opposite-sexed twin that at the time morphological sex-
differentiation finally occurs the gonads of the two animals
differentiate in identical fashion. In my own experimental
animals the shorter indifferent period may be insufficient to
effect such a physiological reversal in the gonad of the graft,
which in consequence differentiates as determined by the genetic
constitution of the donor. In the parabiosis experiments of
\Vitschi the indifferent period (in Rana sylvatica) is likewise
short, which may possibly explain the fact that sex-reversal of
the female follows rather than precedes the primary morpho-
logical differentiation of the gonad.

It is also possible that conditions attendant upon development
of the graft may render it less subject to hormone influence
from the host than is a parabiotic twin to the influence of its
mate. Since the graft usually becomes well vascularized, how-
ever, it would seem that the nutritive materials and hormones of
the host's blood stream should be as readily available for the
gonad of the graft as for the host's own gonads. As has been
stated before, sex-differentiating hormones of the host should be
but little diluted by antagonistic hormones secreted in the graft.
Moreover, the removal of the graft from its natural environment
in the donor while in a germ-layer stage and its implantation
into an essentially foreign situation should, if anything, disturb
the action of local factors affecting sex-differentiation, and
facilitate the modification of this process through hormones
produced by the host. It would seem that in a graft the de-
veloping gonad has been removed from both the endocrine and
environmental influence of the donor and subjected to the
influence of the host in a far more complete fashion than the
gonads of one parabiotic twin can be brought under the influence
of the other embryo of the pair.

From comparison of the conditions acting upon parabiotic
twin and graft, we may conclude that two, at least, possibly
have significance in determining the difference in the results
obtained. First, the greater time required for morphological sex-
differentiation in parabiotic twins may permit an influence of
one animal upon the other such as would not be possible in the
case of a graft gonad differentiating in from half to two-thirds of
the same period. Secondly, the fact that in one case (parabiosis)

SEX DIFFERENTIATION IN GONADS. 33!

the gonad has remained undisturbed in the organism, while in the
other its preprimordium has been implanted in an ectopic
situation in another individual, may possibly explain the different
way in which it reacts preceding or during sex-differentiation.

The results obtained by the writer in Amblystoma are not with-
out parallel from experimental work on other vertebrates.
Willier ('27), from his study of the differentiation of chick gonads
implanted in the chorio-allantoic membranes of either male or
female hosts is led to the conclusion that "the course of sex-
differentiation in the chick embryo is apparently not determined
by the action of sexual hormones circulating in the blood stream."
He believes that "hormonic sex-differentiating factors of the host
embryo are either absent, or if present, they are ineffective in
the modification of the engrafted sexual glands." Witschi ('276)
reaches similar conclusions from one of his latest studies on sex-
differentiation in Rana temporaries. He finds that the implanta-
tion of a large graft of adult frog testis in tadpoles of this species
does not "exert the least influence upon the larval and early post-
larval development of the gonads." In both frog and chick,
therefore, the indifferent gonads are found to undergo their
primary sex-differentiation apparently unmodified by sex hor-
mones from outside sources. In cattle, too, recent studies may
be interpreted as showing that even when the chorions of two-
sexed twins are fused at a very early period, the gonad of the
female co-twin first begins to differentiate as an ovary, and only
later undergoes modifications leading to the production of the
characteristic free-martin gonad (Lillie, '23; Bissonnette, '28).

That the vertebrate ovary in situ may be modified in its
development subsequent to its primary sex-differentiation is
apparent from the studies of Lillie ('17) and others on the free-
martin, or from the cases of sex-reversal in parabiotic frogs
reported by Witschi ('27a). That these same gonads would
have undergone a comparable modification if implanted as grafts
in a host of the opposite sex has not been actually demonstrated.
According to Willier, no modification of engrafted gonads of the
chick is demonstrable after a period of nine days on the host
embryo. It is conceivable, however, as W'illier states, that the
transplantation of the embryonic sexual glands into chicks after
hatching might yield different results than when these same

332 R- R- HUMPHREY.

glands are implanted on the membranes of embryonic hosts.
Greenwood ('25) has reported the development of spermatic
tubules in grafts of the left ovary taken from chicks two to four
days after hatching and implanted in young chicks of the same
age. It would appear probable, therefore, that isolation and
implantation of a gonad (or its preprimordium) do not necessarily
prevent the modification of that gonad through the action of
sex-hormones of the host: i.e., there remains possible an inhibition
of growth, or an induction of growth, in those parts (as for
example the medullary cords of the bird's ovary) which have
retained their embryonic capacity to react in a specific fashion to
growth stimuli.

The grafts described in this paper were in no case left im-
planted in the host for a period longer than seventy days.
Although in none of the grafts recovered had the gonad under-
gone a complete reversal of sex previous to its primary differ-
entiation, it is possible that in one or two cases it had undergone
some slight modification which might be ascribed to the action
of sex hormones of the host. Whether a complete reversal of
sex might have occurred had the graft remained implanted for a
longer period is problematic. From grafting experiments recently
reported by Burns ('27) it is evident that sex-reversal in the
gonads of Amblystoma is not complete even after periods of from
fifty to seventy-six days in a host of the opposite sex. Burns
transplanted gonads from larval stages, just before and just
after the beginning of morphological sex-differentiation, into
older larvae in which sex-differentiation was more advanced.
Since several of the grafts showed an admixture of the charac-
teristics of the two sexes, it is possible that a complete reversal of
sex might eventually have been effected.

Whether or not complete reversal of sex in Amblystoma may
occur subsequent to morphological sex-differentiation, a reversal
of sex preceding this period does not appear to be effected in
gonads developed in grafts, when such grafts are implanted in an
ectopic situation, such as the latero-ventral body wall. Whether
implantation of the graft into its normal site would insure reversal
of the gonad as postulated for animals joined in parabiosis still
remains a question. The writer now has in progress an extensive
series of experiments to test this point.

SEX DIFFERENTIATION IN GONADS. 333

SUMMARY AND CONCLUSIONS.

1. An area of mesoderm which included the preprimordium of
the gonad was transplanted from one Amblystoma embryo to
another at stages 21 to 34. Such transplants, when taken from
donors older than stage 25, gave rise to a gonad in a high per-
centage of cases. This gonad was ectopic in position, being
attached to the inside of the lateral or ventral body wall, and
was always far smaller than the normal gonads of the host.

2. Morphological sex-differentiation .occurred in the grafts at
from fifty to sixty days after implantation. All grafts were
removed and fixed within seventy days. In several cases, gonad^
of testicular type were recovered from female hosts. In two
cases gonads of ovarian type were found in grafts implanted in
males.

3. In two cases in which donor and host were of opposite sex
the gonad of the graft was modified in such fashion as to suggest
an influence from sex hormones of the host. In no case, however,
was the sex of the graft gonad completely reversed previous to
the period of morphological sex-differentiation.

4. It may be concluded that gonads developed in ectopic grafts
of the gonadal preprimordia undergo their primary morpho-
logical sex-differentiation according to the organization of the
graft at the time of its removal from the donor.

5. If sex-differentiating hormones are produced by the host
previous to morphological sex-differentiation, they are apparently
incapable of bringing about reversal in the gonad of the graft.
The possibility of reversal at a later stage of development is not
excluded, since no grafts were allowed to develop for periods
longer than seventy days.

6. The failure of the gonad in a graft to undergo sex-reversal
previous to its morphological differentiation is in marked con-
trast to the complete reversal which appears to occur in parabion-
(cf. Burns '25). 3

3 Studies completed while this paper was in press indicate that the graft ovui i<--
of Figs. 6 and 9 possibly owe certain features of their structure to the action of the
testicular hormones of the host. These studies show that the developing ovary is
readily modified if subjected to the continued influence of a testis resident in th«-
same host, and that one of the first perceptible indications of this modification is tin-
absence of the characteristic central ovarian cavity. These studies will be repoi t< >1

in a separate paper.

22

•534 R- R- HUMPHREY.

BIBLIOGRAPHY.
Bissonnette, T. H.

'28 Notes on a 32 Millimeter Freemartin. BIOL. BULL., Vol. 54, pp. 238-253.
Burns, R. K.

'25 The Sex of Parabiotic Twins in Amphibia. Jour. Exp. Zool., Vol. 42,

pp. 31-89-

'27 -Some Results of the Transplantation of Larval Gonads in Urodele Am-
phibians. Anat. Rec., Vol. 37, p. 163.
Greenwood, A. W.

'25 Gonad Grafts in the Fowl. Brit. Jour. Exp. Biol., Vol. 2, pp. 469-492.
Humphrey, R. R.

'27 The Fate of the Primordial Germ Cells of Amblystoma in Grafts Implanted
in the Somatopleure of Other Embryos. Anat. Rec., Vol. 35, pp. 40-41.
Lillie, F. R.

'17 The Free-martin; a Study of the Action of Sex-hormones in the Foetal

Life of Cattle. Jour. Exp. Zool., Vol. 23, pp. 371-452.
'23 Supplementary Notes on Twins in Cattle. BIOL. BULL., Vol. 44, pp.

47-77-
Willier, B. H.

'27 The Specificity of Sex, of Organization, and of Differentiation of Embryonic

Chick Gonads as Shown by Grafting Experiments. Jour. Exp. Zool.,

Vol. 46, pp. 409-465.
Witschi, Emil.

'270 Sex-Reversal in Parabiotic Twins of the American Wood-Frog. BIOL.

BULL., Vol. 52, pp. 137-146.
'27^ Testis Grafting in Tadpoles of Rana temporaria L. and its Bearing on the

Hormone Theory of Sex Determination. Jour. Exp. Zool., Vol. 47,

pp. 269-294.

336 R- R- HUMPHREY.

PLATE I. EXPLANATION OF FIGURES.

All figures on this plate are photomicrographs. Magnification 145 X.

FIG. i. Ovary of host No. 211, Amblystoma maculatum, killed 61 days after
implantation of graft at stage 29. The central ovarian cavity is well developed,
and the germ cells peripheral to it are chiefly oocytes in heterotypic prophase.
Compare with Fig. 2.

FIG. 2. Testis of graft recovered from host No. 211. Note the absence of a
central cavity. The germ cells are uniformly distributed, and none are in
heterotypic prophase. Compare with the ovary of the host (Fig. i). This testis
was attached to the body wall by a slender stalk not included in this section.

FIG. 3. Ovary of host No. 284, Amblystoma maculatum, killed 50 days after
implantation of graft at stage 30. Ovarian cavity, peripheral arrangement of
germ cells, and abundance of heterotypic prophase stages, as in Fig. i. Compare
with graft gonad of Fig. 4.

FIG. 4. Testis of graft recovered from host No. 284. Note absence of central
cavity and heterotypic prophases, and the uniform distribution of the germ cells.
Compare with the ovary of the host in which this testis developed (Fig. 3). The
slender stalk attaching the testis to the mesonephros of the graft is not included in
this section.

FIG. 5. Testis of host No. 244, Amblystoma jeffersonianum, autopsied 61 days
after implantation of graft at stage 31. This gonad exhibits the scattered arrange-
ment of germ cells and the absence of a central cavity noted in the testes of Figs.
2 and 4. Compare with graft gonad shown in Fig. 6.

FIG. 6. Ovary of graft recovered from host No. 244. Although the central
cavity is not yet developed, the germ cells are peripheral in position and are for
the most part in heterotypic prophase. This gonad thus resembles an ovary
(see Figs, i and 3) rather than the testes of the host from which it was recovered
(see Fig. 5).

BIOLOGICAL BULLETIN, VOL. LV.

PLATE I.

R. R. HUMPHREY.

338 R- R- HUMPHREY.

PLATE II. EXPLANATION OF FIGURES.

All figures on this plate are photomicrographs. The magnification is 121 X
except for Figs. 8 and 9, in which it is 162 X.

FIG. 7. Testis of host No. 207, Amblystoma maculatum, killed 61 days after
implantation of graft at stage 25.

FIG. 8. Left ovary of donor No. 207, Amblystoma jeffersonianum, killed 78
days after furnishing graft (at stage 29) for implantation in host No. 207. Due
to the age at which this animal was killed, the ovary is advanced in development as
compared with those of Figs, i and 3.

FIG. 9. Gonad of graft recovered from host No. 207. Though retarded in its
differentiation, this gonad is apparently an ovary, since its germ cells are peripheral
in arrangement, and a large proportion of them are in heterotypic prophase stages.
Compare with Figs. 7 and 8.

FIG. 10. Ovary of host No. 190, Amblystoma jeffersonianum, killed 64 days
after implantation of graft at stage 33.

FIG. ii. Left testis of donor No. 190, Amblystoma jeffersonianum, killed 64
days after furnishing graft for implantation in host No. 190.

FIG. 12. Gonad of graft recovered from host No. 190. Though atypical in
structure, due to the presence of mucous connective tissue, this gonad is apparently
a testis. No central cavity is present, the germ cells are scattered, and but very
few of them are in heterotypic prophase. Compare with gonad of donor (Fig. n).

FIG. 13. Ovary of host No. 188, Amblystoma jeffersonianum, killed 64 days
after implantation of graft at stage 33.

FIG. 14. Left testis of donor No. 188, Amblystoma jeffersonianum, killed 64
days after furnishing graft for implantation in host No. 188.

FIG. 15. Gonad of graft recovered from host No. 188. It lacks a central
cavity, but has its germ cells predominantly peripheral in position, and frequently
in groups or "nests" as in the ovary. Although many of its germ cells are in
heterotypic prophase, this is true also of the testis of the donor. This gonad is
possibly a testis modified by reason of its development in a graft in a female host.

BIOLOGICAL BULLETIN, VOL. LV.

PLATE II.

Wfc

i !

13

R. R. HUMPHREY.

I14

15

ON THE PROPERTIES OF THE GONADS AS CON-
TROLLERS OF SOMATIC AND PSYCHICAL
CHARACTERISTICS.

XI. HORMONE PRODUCTION IN THE NORMAL TESTES, CRYP-

TORCHID TESTES AND NON-LIVING TESTIS GRAFTS

AS INDICATED BY THE SPERMATOZOON

MOTILTTY TEST.1

CARL R. MOORE,

HULL ZOOLOGICAL LABORATORY, THE UNIVERSITY OF CHICAGO.

I. INTRODUCTION.

Advances in the study of the internal secretions are very often
largely dependent upon the development of successful indicators
for the substances concerned. Since the studies of Brown-
Sequard, innumerable attempts have been made to increase our
knowledge of the internal secretions of the sex glands, and
indeed vast stores of information have been accumulated through
these investigations. The chief difficulty in many of these
attempts and especially in attempts to obtain the hormone
principle in extractions, has been the lack of an applicable
indicator of the substances sought for isolation.

A tremendous step forward in the study of the female hormone
was the demonstration of the details of the oestrous cycle as
indicated by vaginal smears first by Stockard and Papanicolau
('17) in the guinea pig and later by Long and Evans ('22) in the
rat. By the vaginal smear method, one is enabled to determine
the presence or absence of substances concerned with the regula-
tion of the oestrous cycle. With such a useful indicator, the
advances made in the study of the internal secretions of the
ovary have been indeed marked.

On the male side, however, the situation has been a less happy
one from the standpoint of real advancement. In some species

1 This investigation has been aided by a grant from the Committee for research
in problems of sex of the National Research Council; grant administered by Prof.
F. R. Lillie.

339

340

CARL R. MOORE.

of the bird, notably certain breeds of the domestic fowl, the
male feather pattern, behavior, and head furnishings have
afforded a good criterion of testicular presence and activity
though many details were insufficiently known until of late to
make conditions as well understood as was needed. The ex-
tensive work of Domm ('27) on the brown leghorn breed has
given a much greater appreciation of the many pitfalls that
present themselves in this field (for a review of the extensive
literature on this subject, see Domm's paper).

\Yhen we approach the study of the internal secretions of the
gonads in the male mammal, however, a careful analysis will
show the marked absence of useful criteria to indicate the
activity of the testis hormone, operating over limited periods of
time. It is true that there is the sex impulse, supposedly
entirely under the control of the internal secretions of the testicles,
but many things lead us to believe this supposition to be
erroneous. A castrated male theoretically should lose its attrac-
tion for the female, but I have repeatedly utilized guinea pigs
castrated at 30 days of age as testers for the period of female
acceptance for some months after castration. Stone ('27) has
recently reported that young male rats castrated at the age of
three months will continue to copulate with females for periods
of four, five and even eight months. And it is reported that the
eunuch, though castrated early in life, will years afterward have
not only an attraction toward the female but experiences a
degree of satisfaction in this association.

It is likewise true that the growth of the penis, seminal vesicles
(when present), prostate, etc., are to a large extent dependent
upon the internal secretions of the testis but not only have the
variations in such structures proven so great as to make an
assay of a given experimental procedure difficult and often
impossible, but also, if castrations are made on adult animals,
to be followed by such procedures as testis transplantations,
injections or other possible approaches, the question of the
condition of these structures as representing a balance between
postoperative regression or possible stimulation from the materials
or conditions utilized often presents insurmountable barriers.

Such other indicators for testicular internal secretions as
individual body weight, body length, fat deposition, hair coat,

ON THE PROPERTIES OF THE GONADS. 34!

and pugnacity as have been utilized by other workers serve
often to mislead the investigator due to the lack of specificity
of the indicator (for further criticisms of this phase see my
papers, '21 and '22).

During the course of a study of the physiology of the scrotum
or its heat regulating effects on the generative tissues of the te^ti-
(Moore, '240, 6; '26, '27, and '28; Moore and Quick, '24) a
possible, fairly satisfactory indicator for the internal secretions
of the testis in the differential survival of the capacity for
motility of epididymal spermatozoa was discovered by accident.
The same conditions were also discovered by Benoit ('26) a
little earlier, in the course of his beautiful work on the histology
and cytology of the epididymis. The application of this "sper-
matozoon-motility" test for the testis hormone has been under
investigation in this laboratory for longer than three years.
The principle of the test may be expressed in details for the
guinea pig.

When both testes of an adult guinea pig are removed from the
animal, leaving the inferior portion of the epididymides, con-
taining their millions of spermatozoa, in the normal scrotal
position, one finds that the spermatozoa gradually lose their
capacity for motility when these are suspended in physiological
saline solution. A lessened capacity for motility is evident
within a few days after testis removal and seldom can one see
any degree of motility in the spermatozoa after a period of
twenty-three days following the operation. However, if instead
of removing both testes, one is allowed to remain normal, the
single, opposite, epididymis will contain spermatozoa that show
motility when suspended in saline solution for a period of sixty-
five to seventy days (Moore, '28). The difference between the
23 days retention of the capacity for motility when both testes
have been removed and that of 65 days when one testis has
remained, has been proven to be an expression of the internal
secretion of the testis (for further details of this reaction see
Moore, '28).

While it is freely admitted that the spermatozoon motility
reaction has many limitations we have found it very useful and it
will continue to be useful until a better hormone indicator has

342

CARL R. MOORE.

been discovered. In the following pages a few items of in-
formation with respect to its usefulness will be presented.

II. HORMONE PRODUCTION BY NORMAL TESTES.

Utilizing the spermatozoon motility reaction in the guinea pig
as described above, I have attempted to study hormone pro-
duction in the normal testis to learn more concerning its action
upon the life of spermatozoa when it is removed from the animal
through castration at varying periods during the possible life of
the mature germ cell. It has been indicated, for example, that
under the influence of the full hormone compliment of one
testis, the spermatozoon life, as shown by its capacity to exhibit
motion in physiological saline solution, gradually wanes until
after approximately 65 to 70 days it no longer responds to this
stimulus. Should we, for example, wish to supply the hormone
by testis transplantation or by injection of material supposed
to contain it, we should be able, if possible, to test the effect in
the shortest time within which the reaction will indicate any
effect. What influence, therefore, does hormone supplied by a
testis in situ exert when it acts for ten, fifteen or twenty days?
The following procedure will present the method employed.

Young adult guinea pigs are operated under ether anaesthesia
through a low mid-abdominal incision and one testis withdrawn
into the field of operation. The testis is carefully separated
from the inferior pole of the epididymis (tail portion), the
internal spermatic vessels are ligated and the testis, along with
the head and body of the epididymis removed. The remaining
(inferior) portion of the epididymis, connected with its vas
deferens, is then carefully replaced in the scrotum — a necessary
precaution (see Heller, '29). Ten or fifteen days later the
opposite normal testis is removed entire through a scrotal
incision. At selected intervals after the second operation, the
animals are sacrificed, the isolated epididymis finely hashed with
scissors in a small quantity of physiological saline and examined
immediately with the microscope for spermatozoon motility.
To properly express gradations in motility * signs have been
employed in which the normal movement is expressed by ****;
the barest vibratile movement on the part of a few spermatozoa
(perhaps I in 10,000 will contract weakly with little or no trans-

ON THE PROPERTIES OF THE GONADS.

343

lation) is rated *. Where no movement can be detected the
observation is designated o.

Table I. will serve to illustrate the observations on the motility
of spermatozoa obtained from the isolated epididymis of animals
whose normal testis was allowed to remain for 10, 15, 20, 30,
and 40 days after the epididymis to be tested had been isolated.

TABLE I.

UNILATERAL EPIDIDYMAL ISOLATION; OPPOSITE TESTIS REMOVED SUBSEQUENT TO
ISOLATION AS INDICATED IN DIFFERENT GROUPS (GUINEA PIG).

Excess Life

beyond 23

Animal.

Epididymal

Isolation.

Killed.

Days after
Isolation.

Motility.

Days At-
tributed to

Hormone

Effect.

(delayed removal i<

) days)

5i7

i-3i

3-i

30

*

7 days

5i8

i-3i

3-3

32

*

9 days

520

i-3i

3-5

34

o

521

i-3i

3-5

34

o

-

(delayed removal 15 days)

406

6-10

7-1 1

3i

**

8 days

407

6-10

7-14

34

***

ii days

409

6-10

7-20

40

0

409

6-10

7-20

40

**

17 days

410

6-1

7-24

44

**

20 days

411

6-1 1

7-29

48

o

412

6-1 1

7-29

48

o

4i3

6-1 1

7-29

48

o

414

6-1 1

7-29

48

o

415

6-1 1

7-29

48

*

25 days

(delayed removal 20 days)

441

IO-2I

11-29

39

*

1 6 days

442

10-21

11-29

39

*

1 6 days

443

IO-2I

12-3

43

*

20 days

444

IO-2I

12-3

43

*

20 days

445

10-21

12-6

46

o

446

10-21

12-6

46

o

(delayed removal 30 days)

450

10-25

12-8

44

**

21 days

447

10-25

12-17

5

*

30 days

451

10-25

I2-2O

56

*

33 days

452

10-25

12-24

60

0

(delayed removal 40 day~

469

12-3

1-25

53

*

30 days

472

12-3

1-25

53

o

470

12-3

1-27

55

0

473

12-3

1-27

55

*

32 days

474

12-3

I-3I

59

o

344

CARL R. MOORE.

To understand the observations recorded it must be re-
membered that when an epididymis is isolated from its testis
and both testes are removed from the animal, the spermatozoa
contained within an epididymis retain their capacity to show
motility for a period of 23 days; this we may call the basic life
period and realize that they will live for this period without any
hormone being produced by the testicle.

From the table it can be seen that under "delayed removal 10
days" the spermatozoa were observed to show motility for 32
days or nine days longer than expected, had both testes been
removed at the time of epididymal isolation. We see, therefore,
that the hormone supplied by the normal testicle for a period
of ten days before its removal, actually extended the life of the
spermatozoon nine days. Similarly, hormone supplied for fifteen
days extended the retention of the capacity for motility for a
similar length of time (actually slightly longer since in animal
415 a few sperm were seen to move slightly on the 48th day after
isolation or an increase of 25 days above the basic 23 days
expected). When the normal testis was allowed to remain 20
days after epididymal isolation, motile capacity was increased 20
days beyond what it would have been had both testes been
removed at the first operation. Hormone supplied by the
normal testis for 30 days permitted retention of spermatozoon
motility up to 56 days or 33 days longer than the natural life
without hormone being supplied. A hormone supply from the
normal testis for forty days increased the sperm life by little
more than thirty days. But it must be remembered that as we
add to the length of time after operation we gradually approach
the natural limits of spermatozoon life even with a full compli-
ment of hormone; this limit is 65 to 70 days. We could not
therefore expect the relative progressive effectiveness to continue
much beyond a 4O-day normal testis retention because of the
approach to the maximum period of persistence of sperm under
a continuous hormone influence. In an earlier paper I have
emphasized that even utilizing the greatest possible care in the
selection of standard animals for operation and in doing the
operation itself, there is an individual animal variability that
cannot be eliminated; at best we can only make an approach
toward quantitative relationships.

ON THE PROPERTIES OF THE GONADS. 345

Utilizing the basic 23 days as the approximate maximum of
retention of the capacity for motility on the part of spermatozoa
when no hormone is being supplied (and in scores of observations
I have never observed motility for periods above 23 days) we
see that one can actually detect the influence of the testis
hormone when it is supplied for only ten days. Due to the
individual animal variability, I would consider attempts to read
reactions more finely as decidedly unprofitable. In fact, to err
on the safe side, I have arbitrarily chosen to regard any tested
substance or condition involving gonads as negative unless the
capacity for motility is retained for thirty days or longer.

It is of interest to examine the data of the above table with
reference to what they may tell us of hormone production and
storage. When a ten day hormone supply by the normal testis
shows an effect of ten days in the reaction, and likewise when
spermatozoon motility is extended 15, 20, and 30 days beyond
the basic expectations in conditions wherein the testis was
present for 15, 20 and 30 days after epididymal isolation, one
must conclude, I believe — (i) that hormone secretion is a
continuous process and (2) that the hormone is not stored within
the body. When hormone is supplied by the normal testis, for
10 days, the reaction indicates an effect for the same length of
time.

III. HORMONE PRODUCTION IN CRYPTORCHID TKSTES.

It has long been known that man or the domestic mammals
may experience a failure of testicular descent into the scrotum.
Such animals, although always sterile, nevertheless possess their
full compliment of secondary sex characters; they are spoken of
as Cryptorchid individuals. The undescended testes of such
animals have long been known to lack a germinal epithelium;
the gametogenetic function of the testis is deficient but its
internal secretory effects are not visibly diminished. It is now
known that a normal testis removed from the scrotum and
confined within the abdomen very rapidly loses its germinal
epithelium and assumes within a month or two, almost identical
characteristics to those testes that have never descended. The
cause of the degeneration of testes confined within the abdomen

has been found to be the warmer environment of the abdomen
23

346

CARL R. MOORE.

and the function of the scrotum has thus been seen to be that of
a local thermoregulator (for details of this work see Moore,
I924&, 19246, and 1926; Moore and Quick, '24).

Regarding the amount of hormone produced by such a cryp-
torchid testis, occurring normally or artificially made, little
is known. It could be assumed perhaps that a smaller
quantity of hormone might be required to produce or to maintain
the secondary sex characteristics than would be required to
maintain completely normal male conditions. Lipschutz and
his co-workers have maintained that in the rabbit I per cent,
of the normal amount of testicular tissue is sufficient to maintain
all the secondary sex characters ('22). On the other hand, it
has been assumed by some investigators that any condition
leading to an "Apparent increase in interstitial cells" whether
by testis transplantation, X-rays, vasoligation (this latter is the
basis of the contentions underlying the ideas of the Steinach
rejuvenation hypothesis) or any other means, presages an in-
creased production of hormone. The argument proceeds from
the assumption — First, that the apparent compensatory hyper-
trophy of interstitial cells is real (see discussion Moore, '240;
Bascom, '25) ; and second, that the hormone is produced ex-
clusively by the Leydig cells. Some authors have gone so far
as to speak of castrated males, bearing testis grafts containing
appreciable amounts of interstitial tissue, as "supermales," an
implication that I consider without any basis of fact.

In order to gain any new information possible regarding the
quantitative aspects of hormone production by cryptorchid
testes, I have utilized the guinea pig in the following manner:
Young adult guinea pigs have been operated so that one testicle
was removed from the scrotum into the abdomen and the
inguinal canal closed to prevent scrotal redescent. Four months,
and five months, later a second operation was made to isolate
the normal epididymis from, and to remove, the normal
testicle. We thus have an isolated epididymis with its sper-
matozoon content to use as a test for the hormone produced by
the opposite degenerate four or five months experimental cryp-
torchid testis.

Table II. is a record of observations made on eleven animals
in which one testis was confined to the abdomen for four months

ON THE PROPERTIES OF THE GONADS.

347

and upon six animals where hormone supply came from a testis
confined in the abdomen for five months.

TABLE II.

UNILATERAL CRYPORCHIDISM FOUR MONTHS; NORMAL TESTIS REMOVED FROM

EPIDIDYMIS, AND EFFECT OF FOUR MONTHS CRYPTORCHID TESTIS

ON SPERMATOZOON MOTILITY DETERMINED.

Isolation of

Days

Wt. of

Animal.

Normal

Killed.

since

Motility.

Cryptorchid

Epididymis.

Isolation.

Testes.*

393

Oct. 13

Nov. 25

43

***

Not recorded

394

Oct. 13

Dec. 5

53

**

0.26 gms.

402

Oct. 17

Dec. 8

52

***

0.424

396

Oct. 13

Dec. 12

60

*

O.2I

397

Oct. 13

Dec. 12

60

o

0.175

398

Oct. 13

Dec. 17

65

*

0.158

400

Oct. 13

Dec. 1 7

65

o

0.130

403

Oct. 17

Dec. 21

65

#

Not recorded

404

Oct. 17

Dec. 26

70

*

0.170 gms.

416

Oct. 17

Dec. 31

75

*

0.255 "

4i"

Oct. 17

Jan. 7

82

0

0.205

UNILATERAL CRYPTORCHID FIVE MONTHS; SUBSEQUENT
TREATMENT SIMILAR TO ABOVE.

475

Apr. 28

June 26

59

**

0.153 gms.

476

Apr. 28

June 30

63

*

0.13 "

477

Apr. 28

July 6

69

o

Not recorded

478

Apr. 28

July 6

69

*

Not recorded

479

Apr. 30

July 12

73

o

Not recorded

481

Apr. 30

July 12

73

*

0.095 gms.

* Testis weight, without epididymis or fat body.

It will be seen from Table II. that spermatozoa within the
isolated epididymis have been observed to retain their capacity
to show movement on proper stimulation for seventy to seventy-
five days. Since the normal testicle, actively carrying on its
spermatogenetic activity supplies only sufficient hormone to
permit the sperm to live for the same length of time, we must
conclude that the hormone producing capacity of a non-gameto-
genetic, degenerative, or cryptorchid testicle as measured by the
spermatozoon motility test is the equivalent of the normal
testis.

Figure i is introduced to show the histological character of
the testicle after abdominal confinement (experimental cryp-

348

CARL R. MOORE.

torchidism) for a period of seven months; the microphotograph
shows a portion of a section from the testicle of animal No. 481.
This testicle had been confined within the abdomen for a period
of five months, when the epididymis of the opposite testicle was
isolated for the subsequent spermatozoon-motility test and the
normal testicle removed from the animal. Reference to Table II.

FIG. I. Photomicrograph of portion of 7 months cryptorchid testicle (animal
no. 481) showing shrunken seminiferous tubules separated by interstitial tissue.

will serve to recall that the test epididymis contained a few living
spermatozoa 73 days after epididymal isolation and these few
exhibited very weak motility on suspension in saline solution.
When the animal was sacrificed on the 73d day after epididymal
isolation the testicle had been confined in the abdomen for a
period slightly longer than seven months. The weight of the
organ, after removal of its attached epididymis, was 0.095
grams. Since the average weight of eight normal testicles,
without the epididymis, removed from similar sized animals and
at the same time of the year, was 1.7 grams (1.34 minimum wt.—

OX THE PROPERTIES OF THE GONADS.

349

2.06 maximum) it will be appreciated that the weight of this
cryptorchid testis represents 2.8 per cent, of the total testicular
weight of the normal animal. Had the epididymides been
included in this weight the percentage of the normal testicular
weight represented by this cryptorchid testis would have been
considerably less; the spermatozoon and secretion mass within
the normal epididymis being very much greater than the slight
fluid content of the cryptorchid epididymis. It can be concluded,
therefore, that the cryptorchid testis representing 2.8 per cent.
of the normal testicular mass was producing sufficient hormone
to maintain the life of spermatozoa in the isolated epididymis for
the same period as would the hormone produced by two normal
testes.

Figure I shows that the seminiferous tubules of the seven
month cryptorchid testis were very much reduced in caliber and
consisted of a basement membrane, somewhat thickened, and a
few Sertoli nuclei and reticulum; the tubules were rather widely
separated by interstitial tissue. Fig. 2 is a drawing of the

.-1C

FIG. 2. Drawing of tubule marked off by white lines in Fig. 2. bm, basement
membrane; ic, interstitial cell; S, Sertoli nucleus.

tubule marked off by white lines at the upper part of Fig. i.
The thickened basement membrane is more clearly shown and
the character of the contents of the tubule indicates an absence
of any germ cells; the nuclei that are visible are believed to be
Sertoli nuclei. Careful microscopic study has failed to bring to
my attention any cell that appeared different from those repre-
sented in this figure and it is for that reason that I believe no

350

CARL R. MOORE.

germinal cells were present in this testis, at the time of its
removal.

IV. HORMONE PRODUCTION BY TESTIS GRAFTS.

The question of the function of testis transplants must of
necessity be considered under at least two categories: (i) The
function of grafts which have become successfully incorporated
into the host organism and remain as living masses of testicular
tissue, and (2) the function of such masses of testis tissue trans-
planted into various parts of the host organism, which by reason
of host resistance to the transplant or because of too great a
mass of tissue for vascularization, dies and is resorbed or sloughed
out of the incorporation bed often with pronounced suppuration.
It is known that living testis grafts can be obtained and that
they will function. This question, along with the presentation
of personal observations, has been reviewed by me at some
length (Moore, '26). The question of the function of testis
tissue transplanted into a host organism under conditions that
have been so unfavorable as to prevent its retention and growth
has been dealt with most usually by the clinician. For various
reasons many cases of transplantation in man of human testis
tissue or testicular tissue from another mammal such as the ram,
boar, monkey or deer, have been done. The effects reported
are so all-embracing that discredit of all effect is engendered
(for discussion of this work see Moore, '26). In general it may
be repeated that the effects reported have been expressed in
terms of the subjective feelings of the patient — whether he may
feel better after remaining quietly in bed for a week or longer
after the operation; or whether after the suggestions and dis-
cussion of the question and the anticipations of the operation
and its outcome, he has a greater desire for coitus; or whether
the patient feels that he can walk more sprightly or feels that he
can climb a stairs two steps at a time instead of the customary pre-
operation 'One step. In short, such evidence is worthless from
the scientific point of view.

In order to study by objective means the question of the
function of such non-living testis transplants, I have utilized the
guinea pig as the experimental animal and the spermatozoon
motility test as an indicator of effectiveness. In an earlier paper

OX THE PROPERTIES OF THE GONADS. 351

(Moore, '28) I included a few observations then at hand and
have since given additional attention to the problem. The
method employed, in brief, is the bilateral isolation of the
epididymides along with removal of the testes from the animal.
The two testes removed were replaced immediately in an es-
pecially prepared subcutaneous incorporation bed made by
tunneling under the skin, with some destruction of skin muscula-
ture and a general scarification of the particular region. Each
testicle, cut into two parts, was placed in a separate implantation
bed, one on each side of the mid-ventral line of incision.

In addition to the subcutaneous transplantation of the animal's
own two testes at the time of epididymal isolation, a few cases of
multiple transplantations were studied with the idea that perhaps
a small amount of hormone might be liberated from the intro-
duced tissue which if introduced more than once would con-
ceivably show a greater effectiveness. Accordingly at the time
of bilateral epididymal isolation, two, one-fourth testes, were
introduced subcutaneously at the time of the first operation as
well as on the 3d and 5th day following. In each animal,
therefore, six transplantations were made, the aggregate amount
of tissue transplanted being one and one half testes. The
observations on four such experiments are recorded in Table
lllB.

Since an arbitrary choice of thirty days after operation for
effectiveness to be registered was made, animals were sacrificed
close to this period for the study of the spermatozoon content
of the isolated epididymides. From section II. it will be seen
that motility of spermatozoa at this time would indicate the
equivalent of effectiveness of the normal testis present for seven
to ten days. Too much uncertainty surrounds the application
of this test to make it profitable to attempt readings at an earlier
date.

Table III. presents some of the observations recorded.

Among the nineteen animals whose isolated epididymides were
studied for spermatozoon movement between the 25th and 36th
day after autoplastic transplantation of two testes, only two
animals have shown any movement of spermatozoa and in each
case (animals No. 320, No. 456) observed on the 3ist day the
motility was the weakest possible for a positive reading. Amid

352

CARL R. MOORE.

TABLE III.

A. THE EFFECT OF NON-LIVING TESTIS GRAFTS ON SPERMATOZOON
MOTILITY (GUINEA PIG).

Animal.

Date-
Operation.

Date-
Killed.

Days
after
Opera-
tion.

Motility.

3<>4

4- 6-27

5-12-27

36

o

many non-mot.

sperm

365

4- 6-27

5-12-27

36

0

370

4- 8-27

5-12-27

34

0

« « (I

11

371

4- 8-27

5-12-27

34

o

1 t( < t

it

372

4- 8-27

5-12-27

34

o

I It tt

n

373

4- 8-27

5-12-27

34

o

< It «

it

320

11-22-26

12-23-26

3i

*

(i n 1000 weak

mot.)

436

10-15-27

11-15-27

3i

*

(i in 1000 weak

mot.)

435

10-15-27

11-15-27

3i

0

347

3- 8-27

4- 7-27

30

o

many non-mot.

sperm

376

4-15-27

5-15-27

30

o

377

4-15-27

5-15-27

30

0

« t i ti

*

439

10-20—27

11-19-27

30

o

440

10-20-27

11-19-27

30

0

it i 4 .

1

437

10-20-27

11-19-27

30

0

455

10-26-27

11-25-27

30

o

it t it

*

457

10-27-27

11-25-27

29

o

( t t t(

*

458

1 0-2 7-2 7

11-25-27

29

o

" ' **

*

453

10-25-27

11-19-27

25

o

t ( t i «

'

B. MULTIPLE GRAFTS AFTER BILATERAL EPIDIDYMAL ISOLATION.

Animal.

Transplantation Days after
Epididymal Isolation.

Killed, Days after
Testis Removal.

Motility.

553

ist, 3d, 5th

30

0

554

ist, 3d, 5th

30

0

555

ist, 3d, 5th

30

*

549

ist, 3d, sth

32

o

the field of millions of spermatozoa, here and there, an individual
cell could be seen to show a weak contraction, perhaps a weak
vibratile movement every thirty seconds; a rough estimate of
i in 1000 was made to give a relative notion of the quantity
of spermatozoa capable of movement. In contrast to this,
eleven animals observed a shorter period of time after operation
(25 to 30 days) were all negative; no spermatozoon movement
could be seen. One animal (No. 555) receiving six transplanta-
tions of one fourth of one testis at three different operations
subsequent to testis removal, showed a few spermatozoa capable
of weak movement on the 3Oth day after operation, whereas

ON THE PROPERTIES OF THE GONADS. 353

two others on the same day and one on the 32d day after operation
failed to show any spermatozoa capable of exhibiting movement
despite the fact that quantities of normal looking sperm were
present.

These results show, therefore, that subcutaneous transplanta-
tion of testes provided such a small amount of hormone (if any
at all) that its effect was less than the effect of a normal testis
remaining in situ for seven to ten days after operation.

It is difficult or impossible, as pointed out above, to prevent
individual animal variation. Whether the two positive readings
on the 3 ist day are to be explained on this basis of more virile
spermatozoa or as an indicator of some hormone effect cannot be
stated. But since all operations were done alike, and equivalent
masses of tissues transplanted, it would seem as if animals
sacrificed earlier (between 25 and 30 days) would have given as
strong or a stronger reaction than these three. In any event,
should we attribute the results to hormone production and
express it as a positive effect of the transplantations, the mildness
of the reaction would still be evident. At best it is a questionable
indication of hormone production.

The transplanted tissue reactions have been characteristic in
all cases. A few days after transplantation the graft site, con-
siderably removed from the line of skin incision, is swollen and
decidedly reddened; the elevation caused by the transplanted
tissue, at first scarcely visible, becomes approximately the size
of a pigeon's egg or larger. It is typically an inflammatory
reaction. Ten days after operation the swelling may be almost
as large as three days after transplantation and an active pus
discharge may be noted. In many cases, the pus spreads toward
and escapes through the healing incision but in many cases
erosion of the skin over the site of transplantation may provide
an escape for the discharge through the new opening. Pus is
often seen exuding from such areas up to the termination of
the experiment (30 days). In some cases the transplanted mass
has so completely sloughed by the termination of the experiment
as to be invisible excepting as the site of transplantation may be
marked by scar tissue. In other cases small encapsulated
masses of pus are present.

It is evident from these observations and considerations that a

354

CARL R. MOORE.

mass of testicular tissues undergoing autolysis after transplanta-
tion gives little or no evidence of having liberated hormones into

the organism.

V. DISCUSSION.

Our three years experience with the spermatozoon-mo til ity
reaction as an indicator for the testis hormone have increased our
confidence in the test as a dependable objective test for hormone
production. Readily admitting that it lacks several desirable
qualities to make it entirely adequate for many different ap-
proaches to the subject it must still be recognized as a valuable
means of studying hormone production.

Relative to our interests here under discussion, we realize for
the first time that the hormone produced by mammal testes is
not stored within the body of the organism and the internal
secretions of this organ are thus brought into line with such
other organs producing internal secretions as parathyroids,
pituitary, ovary, etc. In the ovarian follicle it has been realized
that a temporary storage, perhaps at the site of production,
does occur, but that the body does not ordinarily store it up for
future release is shown by the failure of recurrence of cestrous
in spayed females. Removal of the testis eliminates the hor-
mone source and there is no evidence that any appreciable
quantity is retained within the organism. This is especially
emphasized when one realizes that a ten day hormone output
by a normal testicle expresses itself with an increase of ten days
in the length of sperm life (as indicated by their motility) ;
similar additional periods can in general be detected by the
reaction.

When the aspermatogenetic or cryptorchid testicle is studied
it is indeed interesting to learn that a testicle reduced in weight
to 0.095 grams produces as much hormone as two normal testicles
carrying on spermatogenetic activity and weighing approxi-
mately 3.4 grams. These cryptorchid testes had originally
produced sperm, but due to experimental elevation into the
abdomen their germinal epithelium had undergone dissolution
and removal. Sections show the typical picture of natural
cryptorchid testes in that the seminiferous tubule outlines are
reduced to small cavities with the characteristic single-celled
layer of Sertoli elements; between the tubules the interstitial

ON THE PROPERTIES OF THE GONADS. 355

cell masses present the typical picture of "apparent interstitial
cell hypertrophy." Such anatomical characteristics have been
sufficient stimulus for some writers to designate such an animal
as a "supermale" but I have never been able to see adequate
justification for the introduction of such a term.

The fact that such cryptorchid testes, having a fraction of the
weight of the testicular mass in normal male animals, generated
the same quantity of hormone, as do two normal testes (judging
from the spermatozoon motility test) suggests again speculation
as to the actual elements producing the hormone.

It must be admitted that of all possible elements within the
testis, the Leydig cells appear to have the weight of evidence in
their favor as being the source of origin of the internal secretions.
However, since no one has ever satisfactorily eliminated all other
elements such as the general connective tissue, but more es-
pecially the cells of Sertoli, there remains the same debatable
conditions regarding the actual source of origin. \Yhen one
views the structure of such degenerate testes as these six and
seven month cryptorchid testes (cryptorchid four and five
months before epididymal isolation and two months of the
experiment) and realizes that each is functioning in producing a
full hormone quotient (again judging by the test employed) one
is certainly inclined to favor the Leydig cells as the source of
origin and to minimize the apparently inactive cells of Sertoli.
The question however is not yet settled despite the suggestiveness
associated with the Leydig cells.

The chief interest connected with the transplanted testis
materials was to see if the spermatozoon motility test would
reveal the liberation of any hormone during the process of
autolysis of the tissue. Since certain writers had reported such
imaginary effects from testis transplantation, it was thought
possible that during the breakdown of the incorporated tissues
some action might be detected. The results of my investigations,
however, have failed to show the liberation of sufficient hormone
to be detectable. Despite the capability of the test to reveal
hormone action for a period of ten days by a normal testicle it
becomes evident that should any hormone effect be derived from
transplantation of two entire guinea pig testicles, its effect must
be less than that of the normal ten-day production period.

356

CARL R. MOORE.

As a further defense of the capability of the spermatozoon
motility test to indicate the presence of substances produced by
the testicle I may mention that subcutaneous injections of lipoid
extracts of the bull testicle, prepared by McGee, and injected
by me into guinea pigs whose testes had been removed from
their epididymides, resulted in prolonging the life of epididymal
sperm to the 54th day after operation (Moore and McGee, '28).
Therefore, had the transplanted testis masses been liberating
hormone into the host organism, we should have been able to
detect it by the test under discussion. Such a result certainly
lends no credence to the idea expressed by others that similar non-
viable testis grafts in man are sufficiently effective to be noticeable
for a period of approximately two years, or again that such
transplantations are able to aid in the cure of asthma, tubercu-
losis, myopia, or the host of other debilities attributed to its effect.

The evidence is very direct that as soon as the hormone
producing tissues are removed from the organism the lack of the
substance produced begins to be detectable in a very short time.
No storage within the body for future utilization is evident.

SUMMARY AND CONCLUSION.

The spermatozoon motility test has been found capable of
detecting the production of testis hormone for a period as short
as ten days.

The hormone produced by the testicles is not stored within the
animal body.

An experimental cryptorchid testicle of five months duration,
having a weight of approximately one-tenth of a gram, produces
as much hormone (indicated by the spermatozoon-motility test)
as do two normal testicles at the height of their spermatogenetic
activity. The experimental cryptorchid testis is by weight
approximately 2.8 to 3.5 per cent, that of the normal testicular
mass.

Autoplastic subcutaneous transplantation of two testes results
In the liberation of hormone in such small amounts (if at all) as
to be scarcely detectable : if any hormone is liberated by these non-
viable testis transplants, the effect upon the animal is no greater
than the effect of a ten day normal hormone production. So far
as present means will indicate, there is no storage or cumulative
effect of the hormone.

ON THE PROPERTIES OF THE GONADS. 357

BIBLIOGRAPHY.
Benoit, J.

'26 Recherches anatomique, cytologique, et histo-physiologiques sur les voies
excretrices du testicule chez les mammiferes. Arch, d'anat. D'physiol.
et D'embryol., 5: 173-412.
Bascom, K. F.

'25 Quantitative Studies of the Testis. Some Observations on the Cryptorchid

Testes of Sheep and Swine. Anat. Rec., 30: 225-241.
Domm, L. V.

'27 Ne\v Experiments on Ovariotomy and the Problem of Sex Inversion in

the Fowl. Jour. Exp. Zool., 48:
Heller, R. E.

'29 New Evidence for the Function of the Scrotum. Physiological Zoology

(in press).
Lipschutz, Ottow and Wagner.

'22 On the Hypertrophy of the Interstitial Cells of the Testicle of the Guinea
Pig under Different Experimental Conditions. Proc. Roy. Soc., 93:
132-142.
Long, J. A. and Evans, H. E.

'22 The OZstrous Cycle in the Rat and it? Associated Phenomena. Memoirs

Univ. of California, Vol. 6.
Moore, Carl R.

'21 On the Physiological Properties of the Gonads as Controllers of Somatic
and Psychical Characteristics. IV. Gonad Transplantation in the
Guinea Pig. Jour. Exp. Zool., 33: 365-389.

'22 V. The Effects of Gonadectomy in the Guinea Pig on Growth, Bone
Lengths, and Weight of Organs of Internal Secretion. BIOL. BULL., 43:
285-312.
'240 VI. Testicular Reactions in Experimental Cryptorchidism. Amer. Jour.

Anat., 34: 269-316.
'24^ VIII. Heat Application and Testicular Degeneration, the Function of

the Scrotum. Amer. Jour. Anat., 34: 337-358.
'26 IX. Testis-graft Reactions in Different Environments (Rat). Amer.

Jour. Anat., 37: 351-416.
'27 A Qualitative Test for the Testis Hormone. Proc. Soc. Exp. Biol. and

Med., 24: 847.
'28 X. Spermatozoon Activity and the Testis Hormone. Jour. Exp. Zool.,

50: 455-494.
Moore, Carl R., and Lemuel C. McGee.

'28 On the Effects of Injecting Lipoid Extracts of Bull Testes into Castrated

Guinea Pigs. Amer. Jour. Physiol., 87 (in press).
Moore, Carl R., and Wm. J. Quick.

'24 The Scrotum as a Temperature Regulator for the Testes. Amer. Jour.

Physiol., 78: 70-79.
Stockard, C. R., and G. N. Papanicolau.

'17 The Existence of a Typical CEstrous Cycle in the Guinea Pig with a Study
of its Histological and Physiological Changes. Amer. Jour. Anat., 22:
225-264.
Stone, Calvin P.

'27 The Retention of Copulatory Ability in Male Rats following Castration.
Jour. Comp. Psychology, 7: 369-387.

INITIATION OF DEVELOPMENT IN ARBACIA. V

THE EFFECT OF SLOWLY EVAPORATING SEA-WATER

AND ITS SIGNIFICANCE FOR THE THEORY

OF AUTO-PARTHENOGENESIS.*

E. E. JUST,

ROSENWALD FELLOW IN BIOLOGY, NATIONAL RESEARCH COUNCIL.

In 1901 Hunter published results of experiments which showed
that uninseminated eggs of Arbacia exposed to sea-water con-
centrated by evaporation develop on return to normal sea-water.
The present writer has been able to confirm these results though
his method differs from Hunter's. The sole reason for reporting
the findings here presented is that they lead to some interpreta-
tions of significance for Lillie's fertilizin theory of fertilization.
The work was done during several summers spent at the Marine
Biological Laboratory, Woods Hole, Mass.

THE EXPERIMENTS.

Normal uninseminated eggs of Arbacia — free of perivisceral
fluid, of high fertilizin content, and capable of giving one hundred
per cent, fertilization and cleavage — show a small per cent, of
cleavage and of abnormal blastulse that do not rise to the surface,
if after having lain in a small volume of normal sea-water for
one or more hours, they are removed to a larger volume of
normal sea-water. Two conditions are important for this method
of initiating development in the egg of Arbacia. First, it is
best to use fairly dense egg suspensions. The less dense sus-
pensions prolong the time of exposure necessary to initiate
development. Secondly, it is indispensable for the experiment
that the dish containing the eggs be left uncovered to insure evapo-
ration. A concentration of I cc. of "dry" eggs plus 99 cc. of
normal sea-water was the least dense suspension successfully
used. In some cases it was necessary to leave such a suspension
uncovered for twenty-four hours before transfer to the larger
volume of sea-water; but it was clear here that evaporation was

* From the Marine Biological Laboratory, Woods Hole, Massachusetts, and
the Department of Zoology, Howard University, Washington, D. C.

358

INITIATION OF DEVELOP M K. XT IX ARBACIA. 359

responsible since suspensions of this kind always gave better
results when placed in larger dishes with greater surface for
evaporation. And if, moreover, a I cc. suspension of eggs in
100 cc. of sea-water be poured on a glass plate thereby insuring
greater evaporation, the results were indeed striking. How-
ever, I am here interested mainly in the results obtained with
smaller volumes of eggs and of sea-water.

In all the experiments it was first ascertained that the eggs to
be used were in optimum fertilizable condition by trial insemina-
tions for the estimation of their capacity to separate normal
membranes. First, the eggs were carefully collected uncon-
taminated by perivisceral fluid, washed in four changes of 200 cc.
of sea-water, and allowed to settle. The supernatant sea-water
was decanted and a highly concentrated bulk of eggs thus
obtained. For each experiment the eggs were from one female.
These were divided into two equal lots whenever their bulk
permitted; one lot was placed in an open dish, the other in a
glass vial closed with a ground glass stopper. At intervals of 30
minutes a drop of eggs from each lot was removed to 250 cc. of
normal sea-water and their development observed. The per
cent, of cleavage was as carefully counted as possible, though
the count is often made difficult because of the number of
cytolyzed eggs. With further development complete cytolysis of
eggs makes the counting of "swimmers" more difficult and of
doubtful value since these counts cannot take into consideration
eggs that have disappeared through complete disintegration.
Moreover, many of the eggs exposed to evaporation develop with
their blastomeres separated because the eggs do not possess
membranes. In such cases, two " micro-blastula? " counted may
have developed from one egg or each from a different egg, its
fellow mass of blastomeres having disappeared completely
through disintegration. I therefore early abandoned attempts
at making accurate counts and merely noted the presence or
absence of "swimmers."

I wish to emphasize that in not one single experiment did I
ever find an egg in the stoppered vial that showed the least sign
of development. That these eggs were not impaired I deter-
mined by inseminating them — in normal sea-water in the case

360

E. E. JUST.

of highly concentrated suspensions, or in the vials in the case of
the less concentrated suspensions.

One other point before we consider the experiments in detail.
The reader appreciates the fact that the rate of evaporation
varied from day to day. I made no attempt to control this
variation. It is also obvious that the rate of evaporation depends
upon the volume of solution employed — smaller volumes evapo-
rating more rapidly than larger. Finally, the vessels used make
a difference; in my experiments I used either shallow dishes,
with a large surface for evaporation, or for the greatest volumes
of solutions employed glass plates, 30 x 30 cm. For volumes up
to 4 cc. Syracuse watch glasses served admirably.

There now follow a summary (Table I.) of the first type of
experiment and a brief comment for the purpose of elucidation.

TABLE I.

THE EFFECT OF SLOWLY EVAPORATING SEA-WATER ON THE UNINSEMINATED EGGS

OF Arbacia AS SHOWN BY THE PER CENT. OF THEIR DEVELOPMENT

ON RETURN TO NORMAL SEA-WATER. EXPERIMENTS

ON EGGS OF 45 FEMALES

Per Cent, of

Per Cent, of

No.

Bulk of Con-
centrated

Volume of
Sea-water

Cleavage.

"Swimmers."

Eggs (in cc.).

(in cc.).

Exp. No.

i

2

3

4

5

Exp. No. i.

i

O.I

0.9

8

II

7

14

20

9

2

0-5

o-5

7

O

ii

14

3

5

3

0.6

1.2

6

12

4

9

o

4

4

i

I

0

6

9

7

2

o

5

i

I

10

7

8

4

14

13

6

i-5

3

21

i?

23

27

18

23

7

2

2

0

3

4

o

5

o

8

2-5

5

13

6

18

21

19

15

9

3

3

14

24

S

O

6

IO

Eggs from the same females in stoppered vials: No cleavage, no "swimmers
Same volumes of eggs and sea water in each case except as follows: No. 7, 0.5 cc.
of eggs + 0.5 cc. sea-water; No. 8, 0.5 cc. of eggs + 0.5 cc. of sea-water; No. 9,
0.5 cc. of eggs + i cc. of sea-water.

The data given in Table I. are for eggs exposed to slowly
evaporating sea-water for two hours. This one length of exposure
is arbitrarily taken for the purpose of simplicity, instead of
presenting the results of each 3O-minute exposure. In some

INITIATION OF DEVELOPMENT IN ARBACIA. 361

instances the per cent, of development was greater after a longer
or a shorter exposure; the results of the two-hour exposure is
very nearly the average of all exposures made. Though the
per cent, of development in no experiment is high, yet it shows
that the evaporating sea-water does initiate development. I do
not regard this as an efficient method for experimental partheno-
genesis: it has been very suggestive, however, for other lines of
my work.

Eggs in sea-water protected against evaporation never show
indication that development is initiated. This statement is
certainly superfluous for suspensions of uninseminated eggs of
Arbacia that are ordinarily employed as controls, as all worker-
know. Of the more dense egg suspensions it might be that
lack of oxygen or CO2 concentration makes initiation of develop-
ment impossible. The fact that such eggs from such suspension
fertilize on return to larger volumes of sea-water does not meet
this possible objection. However, I might repeat that some
suspensions made of I cc. of eggs plus 99 cc. of sea-water exposed
to slowly evaporating sea-water showed initiation of development
whereas similar suspensions in stoppered vials never did.

Eggs that show initiation of development as the result of
exposure to evaporating sea-water never separate membranes,
their cleavage is irregular, and the blastomeres tend to fall apart.
Many eggs do not cleave and of these some reach the monaster
stage with rhythmical dissolution and re-formation of the nucleus.
All uncleaved eggs on insemination separate membranes, cleave,
and reach the pluteus stage.

The abnormal swimming forms developing from these eggs
subjected to treatment with slowly evaporating sea-water never
swim at the surface, but merely rotate on the bottom of the
dishes; among them are micro- and mega-"blastuke" —i.e.,
swimming forms developed from blastomeres that have fallen
apart and those developed from two or more cleaving eggs.
It is this fact that makes difficult the counting of swimmers;
hence, the reader will note that except for the first experiment
(Table I.) and for one experiment described below (Table II.),
I give no counts, but simply note with a + or o sign their presence
or absence.

I interpret these experiments to mean that these eggs in

24

362

E. E. JUST.

evaporating sea-water are by such evaporation exposed to
hypertonic sea-water. It is the hypertonicity that is responsible
for the initiation of development and not the mere crowding of
the eggs since equivalent volumes of eggs from the same females
and of sea-water protected against evaporation do not give any
evidence of initiation of development after transfer to larger
volumes of sea-water. These eggs as noted above had been
thoroughly washed before exposure to evaporation; they would
nevertheless continue to produce fertilizin — but so would the
eggs protected against evaporation. If fertilizin production,
therefore, were responsible for the initiation of development we
might expect that at least the highly concentrated eggs in
stoppered vials would show some signs of cleavage and farther
development. And, what is more, the use of "egg water"
instead of normal sea-water does not increase the per cent, or
improve the development. Table II. gives the results of a
typical experiment on eggs exposed to evaporating "egg water."
Drops of eggs from both the uncovered and the stoppered lots
were returned at half hour intervals to 200 cc. of normal sea-
water. The percentages given are those of eggs having had a
two-hour exposure to the evaporating "egg water." This experi-
ment was made five times.

TABLE II.

THE EFFECT OF SLOWLY EVAPORATING EGG WATER ON THE UNINSEMINATED EGGS

OF Arbacia AS SHOWN BY THE PER CENT. OF THEIR DEVELOPMENT

ON RETURN TO NORMAL SEA-WATER. EXPERIMENTS

ON THE EGGS OF 9 FEMALES.

No.

Bulk of Con-
centrated Eggs
(in cc.).

Volume of Egg
Water (in cc.).

Per Cent, of
Cleavage.

Per Cent, of
"Swimmers."

i

0-5

i-5

9

7

2

0-5

2-5

13

ii

3

i

i

7

o

4

i

i

18

14

5

i

2

n

5

6

2

I

3

5

7

2

2

IS

12

8

2

3

12

10

9

2-5

5

9

6

Equivalent volumes of eggs from the same females and of "egg-water," except for
No. 9 where 0.5 cc. of eggs and i cc. of "egg water" were used, in stoppered vials
gave no trace of development after return to normal sea-water.

INITIATION OF DEVELOPMENT IN ARBACIA. 363

It would appear from a study of Table II. that there is no
advantage in substituting "egg water" for sea-water. As a
matter of fact, other experiments with "egg water" gave inferior
results. In addition, one gains the impression that exposure to
evaporating "egg water" causes more eggs to separate blasto-
meres, and that there are more micro- and mega-"blastulae."
This I did not properly investigate, i.e., by running experiments
on lots of concentrated eggs from the same females, one lot
exposed to evaporating sea-water, one to stoppered sea-water,
one to evaporating "egg water," and one to stoppered "egg
water" counting both the eggs that showed blasto meres falling
apart and the micro-" blastuke." However, some older un-
published observations made independently by Lillie and by
the writer may be cited. These showed that "egg water'
actually possesses a deleterious effect on development. Thus, I
found that if eggs from the same female be divided into two lots,
one suspended in sea-water and the other in strong "egg water'
before or after insemination, the development of the eggs in
"egg water" are markedly inferior to that of the eggs in normal
sea-water as measured by the per cent, and normality both of
cleavage and of plutei. Lillie also has commented on the adverse
effect of "egg water" in other ways on eggs. There is indeed no
reason why this should not be true and several reasons why it
should. "Egg water" is not simply sea-water charged with
fertilizin — it contains products of metabolism of the uninsemi-
nated eggs, even though metabolism is at a low level; this would
be especially true of eggs highly concentrated in strong "egg
water," which perhaps also contains more bacteria than normal
sea-water.

Glaser likewise notes that "addition of the extracts ["egg
water"] in certain concentrations to normally fertilized eggs,
resulted in a retardation of development; normal blastulse
instantly slowed their movements, and underwent a noticeable
increase in volume when subjected to the extracts. Similar
observations were made on the larvae of Arenicola whose rate
of movement was also slowed down, to be followed instantly by
an outflow of their yellow pigment and a slight reversible aggluti-
nation." Unfortunately, however, Glaser's method of preparing
his egg extracts — by removing the eggs directly from the ovaries

364 E- E- JUST.

into twice their volume of sea-water — is open to objection since
he must have carried over some peri visceral fluid. The peri-
visceral fluid alone may have been responsible for his results.

The following experiment was also made ten times: eggs from
one female were placed (i) in sea-water exposed to evaporation,
(2) in sea-water in a stoppered vial, (3) in "egg water" exposed
to evaporation and (4) in "egg water" in a stoppered vial; at
30 minute intervals drops of eggs were removed from each of
the four lots to dishes each containing 200 cc. of normal sea-
water. I give now the summary of one long experiment because
it shows the results with varying concentration of eggs from one
female :

No. i. 10 drops of eggs + 90 drops of uncovered sea-water gave 18 per cent.

cleavage, + + "swimmers."
No. 2. 10 drops of eggs + 90 drops of sea-water in a stoppered vial gave o per

cent, cleavage, o "swimmers."
No. 3. 10 drops of eggs + 90 drops of uncovered egg water gave 6 per cent.

cleavage, + "swimmers."
No. 4. 10 drops of eggs + 90 drops of egg water in a stoppered vial gave o per

cent, cleavage, o "swimmers."
No. 5. 20 drops of eggs + 80 drops of uncovered sea-water gave 27 per cent.

cleavage, + + "swimmers."
No. 6. 20 drops of eggs + 80 drops of sea-water in a stoppered vial gave o per

cent, cleavage, o "swimmers."
No. 7. 20 drops of eggs + 80 drops of uncovered egg-water gave 8 per cent.

cleavage, + "swimmers."
No. 8. 20 drops of eggs + 80 drops of egg water in a stoppered vial gave o per

cent, cleavage, o "swimmers."
No. 9. 30 drops of eggs + 70 drops of uncovered sea-water gave 31 per cent.

cleavage, + + "swimmers."
No. 10. 30 drops of eggs + 70 drops of sea-water in a stoppered vial gave o per

cent, cleavage, o "swimmers."
No. ii. 30 drops of eggs + 70 drops of uncovered egg water gave n per cent.

cleavage, + "swimmers."
No. 12. 30 drops of eggs + 70 drops of egg water in a stoppered vial gave o per

cent, cleavage, o "swimmers."
No. 13. 40 drops of eggs + 60 drops of uncovered sea-water gave 21 per cent.

cleavage, + + "swimmers."
No. 14. 40 drops of eggs + 60 drops of sea-water in a stoppered vial gave o per

cent, cleavage, o "swimmers."
No. 15. 40 drops of eggs + 60 drops of uncovered egg water gave 17 per cent.

cleavage, o "swimmers."
No. 16. 40 drops of eggs + 60 drops of egg water in a stoppered vial gave o per

cent, cleavage, o "swimmers."

INITIATION OF DEVELOPMENT IN ARBACIA. 365

In this experiment because of the rapidity of evaporation on
this particular day the eggs were removed to normal sea-water
after one hour. The experiment reveals that the effect of
evaporating "egg water' is certainly not superior to that of
evaporating sea-water in causing initiation of development. It
shows also as other experiments cited show that more concen-
trated suspensions do not yield markedly higher percentages of
development than less concentrated ones.

On the whole I think that the evidence which I have submitted
indicates that eggs exposed in uncovered dishes develop because
of an increasing hypertonicity due to evaporation. Further,
the evidence indicates that "egg water" is not necessary for
this effect; indeed, "egg water" appears to be less efficacious if
not actually more harmful in some small degree than normal
sea-water. If this evidence be accepted, Glaser's work on auto-
parthenogenesis must be questioned. A brief discussion of
Glaser's work and its significance for the fertilizin theory in
the light of the work which I herein report now follows.

DISCUSSION.

In 1914 Glaser reported for eggs of Arbacia and Asterias a
type of initiation of development due to exposure to "egg water"
for which he chose the name, auto-parthenogenesis. Glaser's
procedure was as follows: "Standard secretion ["egg water"]
was prepared by adding to a certain number of "dry' ripe
ovarian eggs, double their volume of sea-water. At the end of
ten minutes, during which the eggs were slightly agitated at
intervals, the suspension was centrifuged, and the eggs cast down.
After 100 revolutions the supernatant fluid was carefully decanted
and set aside for use.

"Ripe eggs were then shaken, usually from the ovaries of a
single individual, into a small quantity of fresh sea- water, and
to i cc. of a concentrated suspension of these was added I cc.
of the secretion. In this mixture the eggs were allowed to stand
2 hours, when cleavages were usually found in all the dishes."

And further: "Many experiments were tried varying the con-
centration of the secretion as well as the time of exposure. My
records indicate cleavages at higher concentrations as well as
lower, and also in less than two hours, but the greatest number

366 E. E. JUST.

was always obtained when I volume of the concentrated egg
suspension was exposed for 2 hours to I volume of the standard
secretion. If at the end of this time the supernatant fluid is
poured off and replaced by fresh sea-water, free swimming
blastulse will be found within 24 hours. In one case only did
development proceed to the pluteus stage."

As I have stated above, Lillie was never able to repeat this
observation made by Glaser. Nor was I until by chance I
observed the extent of evaporation that had taken place in two
cc. of egg water put in a Syracuse watch glass one hour before.
Deliberately repeating this chance observation on eggs suspended
in either "egg water" or sea-water through several seasons I have
obtained initiation of development in Arbacia eggs provided the
"egg water" or sea-water be allowed to evaporate. I am there-
fore constrained to believe that Glaser's auto-parthenogenesis is
a hypertonic effect due to evaporation.

Glaser has also reported what he calls an improved method of
auto-parthenogenesis. Says Glaser: "Loeb's improved method
of artificial parthenogenesis consists in following the treatment
with parthenogenetic agents, by an after treatment with hyper-
tonic sea-water, 8 cc. of 2.5 M NaCl plus 50 cc. of sea-water.
It seemed likely, therefore, that a better yield of larvae could
be secured if eggs, after having been subjected to the action of
the secretion for two hours, were afterwards treated with the
hypertonic solution for forty minutes. This surmise proved
correct." The proof offered is the outline of a typical experi-
ment showing the development in two sets of eggs both of which
were exposed to hypertonic sea-water after treatment with the
egg secretion. There are, it seems to me, two objections to this
experiment.

In the first place, in the improved method of artificial partheno-
genesis Loeb typically used butyric acid which alone is not
capable of causing development of the egg beyond the separation
of the vitelline membrane and formation of a monaster around
the egg nucleus; according to Glaser, the egg secretion which he
used causes development at least to the blastula stage without
separation of membranes. Moreover, when one uses butyric
acid one must replace the acid sea-water with normal sea-water
and allow a certain time to elapse before beginning the treatment

INITIATION OF DEVELOPMENT IN ARBACIA. 367

with the hypertonic sea-water; Glaser exposed his eggs to the
egg secretion and at once transferred them to the hypertonic sea-
water. There is here, therefore, no similarity between the
improved method of Loeb and that of Glaser.

Secondly, and this is far more serious, Glaser does not tell us
to what extent there is an improvement through the after treat-
ment with the hypertonic sea-water; he gives no information
concerning the development of two lots of eggs from the same
female, one with and one without hypertonic sea-water after the
exposure to the egg sea-water. Obviously, Glaser should have
set up an experiment on four lots of eggs from the same female,
one an uninseminated control in normal sea-water, one exposed
to hypertonic sea-water alone, one to egg water alone, and our
to hypertonic solution after a treatment with "egg water."

In the same communication Glaser also described auto-
parthenogenesis in eggs of Asterias. For this he used either I
or 2 volumes of maturing Asterias eggs plus one of "egg water'
and obtained fertilization membranes, cleavage, and "much
gastrulation." I would suggest that this result was due in part
to CO2, which in Asterias eggs initiates development, and to
hypertonicity.

Glaser's "hetero-parthenogenesis" is the effect of Arbacia
"egg water" on Asterias eggs. Here again he used I volume of
"egg water" (from Arbacia eggs) to i volume of Asterias eggs.
The foreign "egg water" gave fertilization membranes and
numerous cleavages. I venture the opinion that the initiation
of development was due to one, two or a combination of three
factors: CO2, hypertonicity, and the foreign perivisceral fluid
which from Glaser's method of procuring the Arbacia "egg
water" must have been present.

On the basis of my findings and the possibility that these
adverse criticisms of Glaser's work be correct, I suggest that auto-
parthenogenesis is an initiation of development due to hyper-
tonicity of either "egg water" or sea-water. If this be true
Glaser's criticisms of Lillie's fertilizin theory based on his findings
are without foundation.

368 E. E. JUST.

LITERATURE CITED.

Glaser, Otto.

'14 On Auto-parthenogenesis in Arbacia and Aslerias. BIOL. BULL., 26, pp.

387-409.
Hunter, S. J.

'01 On the Production of Artificial Parthenogenesis in Arbacia by the Use of
Sea-water Concentrated by Evaporation. Amer. Jour. Physiol., 6, pp.
177-180.
Lillie, Frank R.

'14 Studies of Fertilization. VI. The Mechanism of Fertilization in Arbacia.
Jour. Exp. Zool., 16, pp. 523-590.

INTRACELLULAR HYDRION CONCENTRATION

STUDIES.

I. THE RELATION OF THE ENVIRONMENT TO THE pH OF PROTO-
PLASM AND OF ITS INCLUSION BODIES.
ROBERT CHAMBERS,

LABORATORY OF CELLULAR BIOLOGY, DEPARTMENT OF ANATOMY, CORNELL I
VERSITY MEDICAL COLLEGE, NEW YORK CITY, AND THE ELI
RESEARCH DIVISION, MARINE BIOLOGICAL LABORA-
TORY, WOODS HOLE, MASSACHUSETTS.

Recent micrurgical investigations (i, 2, 3) on the colorimetric
determination of the protoplasmic pH have emphasized the need
of studying the relation between the pH of the protoplasm of a
living cell and that of its environment. Of the acids and bases
which affect the pH of the environment some penetrate living
cells while others apparently do not. This has been demon-
strated by the change in color of cells stained with indicators.
For example, with the use of neutral red it has been shown by
previous investigators (4, 5) that living cells are readily per-
meable to CO2 and NH3 but not to HC1 nor NaOH. This fact
that the color of the intracellular stain can be readily shifted to
the acid or the alkaline side suggests that the intraprotoplasmic
pH can be changed very easily by environmental conditions, a
conclusion which is at variance with experiments which indicate
that protoplasm has a marked buffering power. Thus, when
solutions of indicators, both in the acid and the alkaline states
of their color ranges, are injected into living cells the colors
quickly shift to those characteristic of a constant pH (6.9±o.i).
This has been found true for such varied types of cells as the
ameba (i, 6), marine ova (2, 3), and various tissue cells of the
frog and the mammal (6). In addition, there is the significant
result that the localized increase in intraprotoplasmic acidity,
caused by mechanical injury is almost immediately neutralized
as long as no cytolysis results (i, 2, 3, 6).

In view of these facts it was considered advisable to test
further the constancy of the intraprotoplasmic pH, to discover

369

370

ROBERT CHAMBERS.

whether this pH can be shifted appreciably without detriment
to the cell and to obtain evidence, if any, of localized variations
in the intracellular pH.

The purpose of the experiments described in this paper is to
determine whether the intraprotoplasmic pH can be shifted by
exposure to COa or to NH3 and whether the reaction to indicators
of such intracellular structures as granules and vacuoles are
comparable to those of the optically homogeneous protoplasmic
matrix.

Before dealing with the actual experiments performed it is
necessary to describe the manner in which the protoplasm
becomes colored with neutral red and with the other dyes used.
When cells are stained with neutral red or certain other basic
dyes, the dye accumulates in or on the intracellular granules and
vacuoles while the hyaline protoplasmic matrix remains colorless.
This occurs not only when cells are stained by immersion in a
solution of the dye but also when the dye is injected directly
into the cell. In the latter case the color appears at first diffuse
but gradually the granules and vacuoles take up more and
more of the color until none of it can be detected in the hyaline
cytoplasmic matrix. On the other hand the acid dyes used,
e.g., brom cresol purple, phenol red and cresol red, do not pene-
trate from the environment into the cells. When injected,
however, they quickly spread through the cytoplasm giving to
its hyaline matrix a more or less permanent and diffuse coloration
(i, 2, 3, 6).

The fresh water Amceba dubia and the unfertilized eggs of the
starfish, Asterias forbesii, and sanddollar, Echinarachnius parma,
were used in these experiments. The amceba and the eggs were
colored with the dyes either by the immersion method or by the
microinjection method. Both methods were also used simul-
taneously on the same cell. The cells were then immersed in
various acid and alkaline solutions and the color changes noted.
For a study of the effect of NH3 and CO^ the cells were suspended
in hanging drops of water from the roof of a special form of

moist chamber which was closed except for narrow inlet
and outlet tubes (7). The hanging drops were then charged

with either CO2 or with NH3 by passing the moist gas through
the chamber.

INTRACELLULAR HYDRION CONCENTRATION STUDIES. 371

i. EFFECT OF ACIDS AND BASES ON AMEB.E COLORED BY TIII;
INJECTION OF ACID INDICATORS ONLY.

Amebae were injected with 0.4 per cent solution of brom
cresol purple, phenol red and cresol red (8). These indicators
were selected because they change color within the pH ranges
tested (i, 3). Amebae, injected with brom cresol purple, are
uniformly blue (the alkaline range), with phenol red, a pale
orange yellow (approaching the acid range). These findings
accord with those already published (6) from which the pH of
the freshwater ameba was placed at 6.9 ± o.i.

Amebae, colored by the injection of the above-mentioned dyes,
were immersed in solutions of HC1 (pH 5.5), NH4C1 (pH 5.5),
CO2 charged water (pH 5.5), NaHCO3 (pH 8), NH4OH (pH 8)
and NaOH (pH 8). The acidity of the first three solutions is
sufficient to cause the indicators to take on the yellow color of
their acid ranges, while the alkalinity of the last three solutions
is sufficient to give to brom cresol purple the purple blue, and to
phenol red and cresol red the bright red color of their alkaline
ranges. It was found that the immersed amebae all maintained
their original colors as long as they remained alive. The color
of those which rounded up and died changed to that characteristic
for the pH of the environing medium.

These results indicate, either that there is no penetration of
the acid or of the alkali from the solutions used, or that the proto-
plasm is sufficiently buffered to neutralize the acid or the alkali
which does penetrate.

2. EFFECT OF ACIDS AND BASES ON CELLS STAINED WITH

NEUTRAL RED AND INJECTED WITH ACID

INDICATORS.

a. Amoeba dubia.

Since the permeability of cells to certain acids and bases can be
demonstrated by the change in color of neutral red, amebae were
immersed in a solution of neutral red until various intracellular
inclusions took on a red color. These amebae were then injected
with solutions of the indicators which color the cytoplasm dif-
fusely. On immersing these doubly colored amebae into the
various acid and alkaline solutions the following results were
obtained :

372

ROBERT CHAMBERS.

In accordance with the previous experiment it was found that
immersion produced no change whatever in the diffuse coloration
of the hyaline cytoplasmic matrix. On the other hand, the
inclusion bodies which were stained with neutral red quickly
became yellow in the solutions containing the NH3 (NH4OH
and NH4C1) and bright red in those containing CO2 (NaHCOs
and CO2 charged water).

These results imply that the pH of the hyaline cytoplasm does
not change even when sufficient NH3 or CO2 penetrates to change
the color of the intracellular inclusions. In other words, the
pH of the intracellular inclusions can be shifted readily by the
presence of CO2 or of NH3 in the environment while that of the
protoplasmic matrix remains constant.

b. Unfertilized Eggs of the Sanddollar (Echinarachnius parma}
and the Starfish (Asterias forbesii).

The protoplasm of these eggs is uniformly crowded with
granules or macrosomes practically all of which ultimately stain
a deep rose red with neutral red. The eggs were allowed to
remain in sea-water containing neutral red only long enough to
stain a small percentage of the granules. The eggs were then
washed, transferred to hanging drops of sea-water in the moist
chamber and injected with the indicator solutions. In the same
chamber were placed, as controls, other hanging drops of sea-
water colored with the same indicators. Ammonia gas was
then passed through the chamber until the hanging drops became
sufficiently saturated with ammonia to change the color of the
control drops.

The color of the eggs was noted when the dyes in the control
drops had assumed colors indicating a pH more alkaline than 8.4.
In every case the color of the granules, stained with neutral red,
changed from red (acid) to yellow (alkaline) while the diffuse
coloration of the indicators in the hyaloplasm of the eggs per-
sisted in registering the originally recorded pH of 6.8 ± o.i (3).

An experiment giving striking color contrasts is one in which
three dyes, neutral red, phenol red and cresol red, were used for
the purpose of detecting simultaneously the pH changes in the
cytoplasm, the cytoplasmic granules, and the sea-water sur-
rounding the eggs. It is to be remembered that neutral red

INTRACELLULAR HYDRION CONCENTRATION STUDIES. 373

which stains the granules is red at a pH more acid than 6.8 and
yellow at a pH more alkaline than 7.4. Phenol red which colors
the hyaloplasm is yellow at a pH more acid than 6.8 and red at a
pH more alkaline than 7.4, and cresol red which was used for
the environing sea-water is yellow at a pH more acid than 7.8
and red at a pH more alkaline than 8.0. The experiment was
the following: Eggs, stained with neutral red, were immersed in
a hanging drop of sea-water colored with cresol red and were
then injected with phenol red. The result was a striking picture
of yellow eggs containing scattered red granules and surrounded
by a medium of yellow sea-water. Ammonia gas was then
passed through the chamber until the cresol red in the sea-water
changed from yellow (acid) to red (alkaline). As soon as this
occurred the cytoplasmic granules, stained with the neutral red
turned yellow (alkaline) while the hyaloplasm maintained the
original yellow (acid) color of the phenol red. The result was
now a picture of uniformly yellow eggs standing out against a
background of red sea-water. Carbonic acid gas was then
passed through the chamber until it displaced the NH3 in the
hanging drops. As a result the original colors returned, viz.,
the sea-water again became yellow, the cytoplasmic granules
turned from yellow to red but the cytoplasm itself remained
yellow.

Since the cytoplasm has a pH of 6.8 ± o.i (3) which is in the
acid range of phenol red the above experiment is not suited for
detecting a possible effect of the CO2 on the cytoplasmic pH.
For this purpose it is necessary to use brom cresol purple (yellow
at a pH more acid than 6.0 and blue at a pH more alkaline than
6.2) which, upon injection, colors the hyaloplasm blue. These
eggs were immersed in a hanging drop of sea-water colored blue
with the same dye. The hanging drop was suspended in the
hermetic chamber through which moist CO2 gas was made to
stream until the sea-water became charged with CO2 sufficiently
to change its color from blue to yellow. The eggs in the yellow
water kept their original blue color.

These experiments indicate that NH3 and CO2, both of which
penetrate the protoplasm and affect the pH of the intracellular
granules, do not shift the pH of the hyaloplasm as measured
by the indicators.

374

ROBERT CHAMBERS.

3. EFFECT OF CO2 AND OF NH3 ON AMEB/E WHOSE CYTOPLASM

AND INCLUSION BODIES ARE COLORED

WITH THE SAME INDICATOR.

A possible error in the previous experiments lies in the fact
that the coloration of the cytoplasmic inclusions and of the
hyaline cytoplasm were not made with the same dye. For
example, neutral red, which colors the cellular inclusions, is a
basic dye, while the dyes used for producing a diffuse coloration
are acidic. It is conceivable that this may be responsible for
their difference in reaction to the penetrating CC>2 or NH3.

To meet this objection it was found that methyl red could be
used. Methyl red has already been used as a vital stain for
plant protoplasm (9) and is a pH indicator, being red at a pH
more acid than 5.0 and yellow at a pH more alkaline than 5.4.
Immersion of amebae in an aqueous solution of this dye stains
the hyaline cytoplasm, its various inclusions and the nucleus
an intense yellow. Amebae colored in this way were placed in a
moist chamber in hanging drops of the yellow aqueous solution
of methyl red. Moist CO2 gas was then passed through the
chamber until the hanging drops turned from yellow to red.
When this occurred it was found that the yellow stained inclusions
of the ameba had also become red while the cytoplasm and
nucleus remained yellow. Ammonia vapor was now passed
through the chamber whereupon the color of the hanging drops
and of the intracellular inclusions quickly changed back to
yellow.

These experiments with methyl red clearly demonstrate the
penetration of CO2 into the living ameba l as registered by the
change in color of the intracellular inclusions. The hyaline
cytoplasm and the nucleus, however, maintain their original
color and give no evidence of a change in pH.

1 The neutral red method is not very favorable for detecting the penetration of
CO» into cells since the granules stained with neutral red under normal conditions
already have the rose red color characteristic for the acid range of the dye. On
the other hand, methyl red under normal conditions stains the intracellular granules
the yellow color of its alkaline range. Upon exposure to CO2 the color of the
granules changes to red, which is as decided an evidence for the penetration of
the CO2 as is the neutral red method for the penetration of NH3.

INTRACELLULAR HYDRION CONCENTRATION STUDIES. 375

4. THE EFFECT OF PENETRATING ACIDS AND BASES
ON THE NUCLEAR pH.

The nuclei of immature starfish eggs were used in these experi-
ments. The nuclei of different eggs were colored with cresol
red, neutral red and phenol red by the microinjection method
after which the eggs were exposed to CO2 and to NH3. In every
case the color within the nuclei of living eggs remained constant
irrespective of the color changes of the granules in the sur-
rounding cytoplasm. In other words, the nucleus was found to
be sufficiently buffered so that the intranuclear pH of 7.6-7.8
(3) remains unchanged. When the egg disintegrates by crushing
or tearing, the nucleus undergoes changes (3) and loses all
buffering action. The persisting spherical nuclear remnant is
then immediately susceptible to acid and alkali changes in its
environment.

SUMMARY.

The presence of CO2 or of NH3 in the aqueous medium sur-
rounding living cells readily changes the pH of the intracellular
inclusions which stain with neutral red but does not change the
pH of the protoplasmic matrix nor of the nucleus as long as the
cell is alive.

BIBLIOGRAPHY.

1. Needham, J., and Needham, D. M.

'25 The Hydrogen-ion Concentration and the Oxidation-reduction Potential
of the Cell-interior: A Micro-injection Study. Proc. Roy. Soc., B.
98, 259.

2. Needham, J., and Needham, D. M.

'26 The Hydrogen-ion Concentration and Oxidation-reduction Potential of
the Cell-interior before and after Cleavage: A Micro-injection Study
on Marine Eggs. Proc. Roy. Soc., B, 99, 173-199-

3. Chambers, R., and Pollack, H.

'27 Micrurgical Studies in Cell Physiology. IV. Colorimetric Determination
of the Nuclear and Cytoplasmic pH in the Starfish Egg. Jour. Gen.
Physiol., 10, 739-755-

4. Harvey, E. N.

'14 The Relation between the Rate of Penetration of Marine Tissues by
Alkali and the Change in Functional Activity Induced by the Alkali.
Publ. Carneg. Instit., Wash., No. 183, 131.

5. Jacobs, M. H.

'20 The Production of Intracellular Acidity by Neutral and Alkaline Solutions
Containing Carbon Dioxide. Amer. J. Physiol., 53, 457.

'22 The Influence of Ammonium Salts on Cell Reaction. J. Gen. Physiol.,
5, 181.

376

ROBERT CHAMBERS.

6. Chambers, R., Pollack, H., and Hiller, S.

'27 The Protoplasmic pH of Living Cells. Proc. Soc. Exp. Biol. and Med.,
24, 760-761.

7. Cohen, B., Chambers, R., and Reznikoff, P.

'28 Intracellular Oxidation-reduction Studies, I. J. Gen. Physiol, II, 585.

8. Clark, W. M.

'25 The Determination of Hydrogen Ions. Williams and Wilkins Co..
Baltimore, 2d ed.

9. Schaede, R.

'24 Uber die Reaktion des lebenden plasmas. Ber. d. bot. Ges., 42, 219.

INTRACELLULAR HYDRION CONCENTRATION

STUDIES.

II. THE EFFECT OF INJECTION OF ACIDS AND SALTS ox THE
CYTOPLASMIC pH OF Ama>ba dubia.1

PAUL REZNIKOFF AXD HERBERT POLLACK

In a previous communication (i) from this laboratory the
pH of the cytoplasm of Amceba dubia was reported to have a
value of 6.9 ±0.1. To determine whether any permanent
variations in the intracellular pH could be artificially produced,
solutions of acids salts and simple salts having toxic actions
were injected by the micrurgical technique (2) into amebre
previously colored with indicators.

EXPERIMENTAL.

The ameba and methods used in these experiments were the
same as those described previously (2). The hydrion indicators
(3) employed were thymol blue, brom phenol blue, brom cresol
green, methyl red, chlor phenol red, brom cresol purple, phenol
red, and orange III. Of these phenol red was the most exten-
sively used. The advantages of this dye are twofold. It is the
least toxic of all the indicators and is the most valuable one in
experiments of this type since its useful range covers the normal
cytoplasmic pH. The other indicators were used in extreme
changes of pH.

In the case of each solution, at least 10 ameba? were used, and
for critical concentrations from 25 to 50. Small quartz cover
slips were employed in these experiments. They were attached
by means of water films to the ordinary long glass cover slips.
On the quartz slips were placed amebae in a drop of their medium,
varying in reaction from pH 5.8 to 7.5, a drop of indicator, and

1 From the Laboratory of Cellular Biology, Department of Anatomy, Cornell
University Medical College, New York City, and the Marine Biological Laboratory.
Woods Hole, Massachusetts.

2 Expenses connected with this investigation were in part drtrayril from a
grant by the Ella Sachs Plotz Foundation.

377

PAUL REZNIKOFF AND HERBERT POLLACK.

a drop of the solution the effect of which was to be tested. The
pipettes used were made of pyrex glass and were rinsed several
times in distilled water and then in solutions of the substances
to be injected. The dyes were injected into the amebae which
were permitted to recover. After recovery the next solutions
were introduced.

As a control, the degree of injury caused by the insertion of
the pipette was determined. The method employed was to
note any change in color of the previously injected dye from
the possible formation of acid associated with injury (4). The
simple introduction of a pipette into an ameba was found to
give no indication of acid production. If the pipette, however,
stirred up the cytoplasm so vigorously that the injured area was
ultimately discarded, a distinct acidity was produced. When
death occurred in the presence of those dyes covering the range,
a pH of about 5.5 was indicated, unless the pellicle surrounding
the mass was broken in which case the color was rapidly
washed out.

Solutions of HC1 (pH 2) when introduced into an ameba,
which is colored an orange-yellow with previously injected
phenol red, cause an immediate and intense yellow coloration of
the injected area. If the injected region is not irreparably
injured by the acid the pH of the area reverts within a few
seconds to that of the normal cytoplasm. Usually, however,
the injected portion is injured to such an extent that it is pinched
off in a manner previously described (2) after which the yellow
color (acid) of the discarded sphere gradually changes to that
indicative of the pH of the environment. In time the color
entirely washes out. The unaffected remnant of the ameba
retains its orange-yellow color.

When CaCl2 is injected in concentrations stronger than M/2OO
the phenomenon of solidification and pinching off is accompanied
by distinct evidence of acid production. If a solution of M/2OO
CaCl2 is injected into amebae colored with phenol red, the flash
of yellow color indicating acid production rapidly returns to
that of the normal pH, provided the injected area is not dis-
carded. If the color does not revert within a few seconds the
affected portion is pinched off.

1NTRACELLULAR HYDRION CONCENTRATION STUDIIs 379

In only three cases out of several hundred did the maintenance
of a localized acid reaction persist for as long as a minute after
HC1 or CaCl2 had been introduced without subsequent pinching
off. To investigate further this condition in which a localixed
acid reaction is maintained for an appreciable time with subse-
quent complete recovery solutions of M/^2 Aids were introduced
into amebae colored with phenol red. Such a concentration of
A1C13 causes a solidification of the injected portion but this
region is not infrequently reincorporated after being almost dis-
carded (5). Of at least 50 cases only one showed a delay of two
to three minutes in the return of color from yellow (acid) to the
original orange-yellow after the affected area had been re-
incorporated. In every other test the reversal of color was
immediate if reincorporation occurred or, if the area was dis-
carded, its color remained yellow.

The introduction of solutions of MgClo of pH 6.5 in con-
centrations of M/3O and stronger into amebae previously injected
with phenol red causes an immediate shift to yellow, indicating
acid production. The cell breaks and the color diffuses out.
When an M[6o solution of MgClo is injected the yellow color
reverts rapidly to that normal for healthy cytoplasm and the
ameba recovers.

In order to determine the degree of .acid production by the
injection of CaCl2 and MgCl (pH 6.6) amebae were injected
with this salt after having been colored with thymol blue, orange
III, methyl red, brom phenol blue and brom cresol green. All
these dyes were injected with the exception of methyl red (6).
Amebae were stained with methyl red by immersing them into
5 cc. of distilled water into which a few drops of a 0.4 per cent,
aqueous solution of methyl red were placed. With methyl red
a distinct red is produced when either CaCl2 or MgClo is injected
into amebae. With orange III the yellow color persists. This
places the reaction of the acidified portion of the cell between
pH 4.0 and 4.6. It is difficult to determine a more exact pH
value because the color changes with brom phenol blue are not
sufficiently distinctive within the critical range. These results
show that the acid produced by injection of CaCl2 or MgCl2 is
more marked than the acid of injury which was found to be

about pH 5.5.

25

380 PAUL REZNIKOFF AND HERBERT POLLACK.

When NaCl and KC1 (pH 7) are injected into amebse colored
with phenol red, no immediate change in color occurs. If the
concentration of these salts is lethal (2) the rounded amebae
gradually take on the color indicative of the pH of the environ-
ment. Injection of non-lethal concentrations of these salts
results during the quiescent period in a slight shift in color
toward that suggestive of the pH of the surrounding medium
whether this be acid or alkaline. But as soon as the ameba
recovers the color reverts to the normal orange-yellow.

The change in color of the discarded spheres after HC1 and
CaCl2 had been injected or of the dead ameba when lethal
amounts of NaCl or KC1 were introduced is due to the penetration
from the environment. It is quite obvious that any uncontrolled
changes in the environmental pH would be confusing. Therefore
it was necessary to take precautions to obviate this factor.
In preliminary experiments, when ordinary cover slips were
used the medium increased in alkalinity markedly during its
contact with the coverslip. The use of pyrex glass or coverslips
coated with balsam or collodion did not prevent this change in
hydrogen ion concentration. To maintain a constant pH of the
environment a buffer calcium acetate solution l of pH 6 was used.
In this ameba? were immersed and the various salts and acids
injected. In this case the dead spheres and the amebae killed
with NaCl or KC1 took on the color representing the reaction
of the surrounding medium, viz., pH 6. With quartz cover
slips, which do not affect the pH of solutions coming into contact
with them, the medium remained constant and the discarded
spheres and dead amebse assumed the reaction of any environing
medium into which they were placed. These results show that
the injection of the individual chlorides are ineffective in changing
the intracellular pH except when toxic concentrations were used.

DISCUSSION.

In the marine egg the production of acid due to injury is
much more easily manifested than in the ameba. As previously
shown (4) a localized flash of color indicating acid production is
apparent in the starfish egg if the needle is introduced abruptly

1 We wish to thank Dr. William Perlzweig for the preparation of this buffer
solution.

INTRACELLULAR HYDRION CONCEN TK.VI H ).\ STUDIES. 38!

into the interior. This change is not evident in the ameba unless
the mechanical trauma is vigorous enough to cause death of the
disturbed part. This difference points to a greater susceptibility
to injury of the egg or an increased buffering power of. the ameba
which may in turn be due either to an increase in the amount of
buffer present or to a greater ease in diffusion of buffers through
the cell. This faster rate of mobilization of buffers in the ameba
as compared to the egg is suggested by the constant flow of cyto-
plasm of the ameba in contrast to the relatively ' quiescent
cytoplasm of the egg.

The production of acid when CaCl2 or MgCl2 is introduced
into the ameba may be due to the production of insoluble Ca
or Mg salts wTith the liberation of free acid. Aub and Reznikoff
(7) have suggested such an explanation for the effect of Pb salts
on red blood cells. Ca may also unite with the carbonate and
phosphate to form insoluble salts with the production of free
acid.' This acid formation is evident until some alkali diffuses
into the solidified mass and neutralizes the acids present. Such
an explanation does not preclude the possibility also of the for-
mation of a Ca or Mg organic compound.

CONCLUSIONS.

1 . The cytoplasm of the living Amoeba dubia shows considerable
buffering power to pH changes induced by the injection of salts
and buffers.

2. If HC1, injected into the ameba, is immediately buffered by
the cytoplasm no toxic effect results. If the quantity injected
is too great to be buffered, the affected portion of the cell dies
and is discarded.

3. CaClo, MgCl2 and A1C13, injected into amebac colored with
indicators, give colorimetric evidence of the production of acid
greater in amount than can be explained by acid produced by
mechanical injury. Unless this color reverts immediately to
that indicative of normal cytoplasm, the affected portion is
discarded in the case of CaClo and A1C13 and the entire cell
dies in the case of MgCl2.

4. Upon death permeability changes occur so that the dead
mass of the ameba quickly assumes the hydrogen ion concen-
tration of the environment.

382 PAUL REZNIKOFF AND HERBERT POLLACK.

BIBLIOGRAPHY.

*

1. Chambers, R., Pollack, H., and Hiller, S.

'27 The Protoplasmic pH of Living Cells. Proc. Soc. Exp. Biol. and Med.,
xxiv, 760.

2. Chambers, R., and Reznikoff, P.

'26 Micrurgical Studies in Cell Physiology. I. The Action of the Chlorides
of Na, K, Ca, and Mg on the Protoplasm of Amoeba proteus. J. Gen,
Physiol., viii, 369.

3. Clark, W. M.

'25 The Determination of Hydrogen Ions. Williams and Wilkins Co..
Baltimore, 2d ed., 81.

4. Chambers, R., and Pollack, H.

'27 Micrurgical Studies in Cell Physiology. IV. Colorimetric Determina-
tion of the Nuclear and Cytoplasmic pH in the Starfish Egg. J. Gen.
Physiol., x, 739.

5. Reznikoff, P.

'26 Micrurgical Studies in Cell Physiology. II. The Action of the Chlorides
of Lead, Mercury, Copper, Iron, and Aluminum on the Protoplasm of
Amoeba proteus. J. Gen. Physiol., x, 9.

6. Chambers, R.

'29 Intracellular Hydrion Concentration Studies, I. The Environment and
the pH of Cytoplasm and of Inclusion Bodies. Biol. Bull., LVI, 3£9.

7. Aub, J. C., and Reznikoff, P.

'24 Lead Studies III. The Effects of Lead on Red Blood Cells. Part 3.
A Chemical Explanation of the Reaction of Lead with Red Blood
Cells. J. Exper. Med., xl, 189.

INTRACELLULAR HYDRION CONCENTRATION

STUDIES.

III. THE BUFFER ACTION OF THE CYTOPLASM OF Amoeba dubia
AND ITS USE IN MEASURING THE pH.

HERBERT POLLACK,

LABORATORY OF CELLULAR BIOLOGY, DEPARTMENT OF ANATOMY, CORNELL UNI-
VERSITY MEDICAL COLLEGE, NEW YORK CITY.

Recent determinations of intracellular pH have been made by
noting the color of indicator dyes injected directly into the
protoplasm (i, 2, 3, 4, 5). The recorded value was found by
comparing the results of injecting a series of overlapping dyes.
The color of the dye, whose range was found to include the pH
of the cytoplasm, was compared with known standards projected
optically into the field of the microscope.

While attempting to determine the buffer action of the cyto-
plasm it was found that an indirect method could be used to
check the results obtained from the direct color comparisons.

It is known that a drop of a solution at a certain pH added
to another buffer solution containing an indicator dye will
cause a momentary localized change in the color providing the
reactions of the two solutions are different. The closer the pH
values of the two solutions are to one another, the less marked
will be the color change. When they have the same pH there will
be no change in color. It is possible to take advantage of this
fact in measuring the intraprotoplasmic pH by injecting a series
of solutions of known pH into cells colored by the previous
injection of indicator dyes. As will be brought out later, this
technique is only approximate but serves to check wide deviations
from direct tint comparisons.

It has been shown that M/4 solutions of mono-sodium phos-
phate may be injected with no toxic effect (6), and that the
potassium ion has about the same toxicity as the sodium ion on
injection (7). Hence the Clark buffer solutions (8) whose
KH2PO4 concentration is M/2O should be non-toxic from the
point of view of salt concentrations, and any toxic effect must
be due to the buffered hydrogen ion concentration.

383

384 HERBERT POLLACK.

Amoeba proteus and Amoeba dubia were used in this study
since their pH has been determined by direct tint readings.

The amebae were injected with brom cresol purple and phenol
red which were the indicators whose ranges cover the pH as found
by previous work. The colored amebae were then injected with
the phosphate buffers from pH 5.6-8.0 and observations made
on the changes in color.

When buffer solutions of 5.6, 5.8, 6.0 were injected into
amebae colored with brom cresol purple, a temporary but distinct
yellow flash was produced. Those above 6.2 produced no color
change with this indicator which is already blue in the cell.
Buffer solutions of pH 6.2 and 6.4, when injected into amebae
colored orange yellow with phenol red, gave temporary yellow
flashes. With the same indicator, solutions having a pH of 6.6,
6.8 or 7.0 showed no color change. Those whose pH was 7.2
and above showed reddish flashes in the orange yellow colored
cytoplasm.

This shows that the pH of the amoeba is not less than 6.6
and not greater than 7.2. This is in accord with the results
obtained in this laboratory in previous investigations and not
with those obtained by the Needhams (2). They also used the
microinjection technique with direct color comparison for reading
the pH values. Their value for the cytoplasmic pH of the
amoebae was 7.6, as was Pantin's, who used the neutral red vital
staining technique (9).

As for the Needhams' results it must be remembered that they
were using a European species and also that they report the
amebae died within five minutes after injection. In the investi-
gation reported in this paper the amebae were allowed to recover
fully after the injection before treatment with the buffer solutions.
With a proper injection of phenol red and brom cresol purple,
amebae can be kept alive and apparently normal for at least two
days (4). Frequent checks on the color by direct comparison
with standard buffers showed no change during that time. The
amebae colored with phenol red maintained the same orange
yellow tint (pH 6.9 ± o.i) as long as they were kept under
observation. On the other hand moribund and dead amebae
take on the pH of the environment which is usually alkaline
when the ordinary glass coverslips are employed without proper

INTRACELLTLAR IIVDRION CONCENTRATION Sl( DIES. 385

precautions (5). As for the value obtained by Pantin (8) tin-
inefficacy of neutral red staining has been shown (3, 10).

The interesting fact is that regardless of the pH value of
the buffer solution injected the return of color of the indicator
present to its usual one is quite rapid and constant. If, how-
ever, sufficient buffer was put in to change the pH of the cell,
the cell died. These facts emphasize two important point-
relating to intracellular hydrogen ion concentration. One, that
the cytoplasm has a considerable buffering power, and t\v<>
that when the pH of the cytoplasm is changed, t,he cell dies.

BIBLIOGRAPHY.

1. Needham, J., and Needham, D. M.

'25-'26 The Hydrogen Ion Concentration and Oxidation-reduction Potential
of the Cell Interior before and after Fertilization and Cleava •<:•: A
Micro-injection Study of Marine Eggs. Proc. Roy. Soc. London,
Series B, XCIX, 173.

2. Needham, J., and Needham, D. M.

'26 Further Micro-injection Studies on the Oxidation-reduction Potential of
the Cell Interior. Proc. Roy.-Soc. London, Series B, XCIX, 383.

3. Chambers, R., and Pollack, H.

'27 Micrurgical Studies in Cell Physiology, IV. Colorimetric Determination
of the Nuclear and Cytoplasmic pH in the Starfish Egg. J. Gen.
Phys., X, 739-

4. Chambers, R., Pollack, H., and Killer, S.

'27 Protoplasmic pH of Living Cells. Proc. Soc. Exp. Biol. Med., XXIV,
760.

5. Reznikoff, Paul, and Pollack, H.

'29 Intracellular Hydrion Concentration Studies. II. The Effect of In-
jection of Acids and Salts on the Cytoplasmic pH of Amoeba dubi,i.
Biol. Bull., LVI, 377-

6. Reznikoff, Paul, and Chambers, R.

'27 Micrurgical Studies in Cell Physiology. III. The Action of CO2, and
Some Salts of Xa, Ca, and K on the Protoplasm of Amoeba dnbia. J.
Gen. Physiol., X, 731-

7. Chambers, R., and Reznikoff, P.

'26 Micrurgical Studies in Cell Physiology. I. Action of the Chlorides
of Na, K, Ca, and Mg on the Protoplasm of Amoeba proteiis. J. Gen.
Phys., VIII, 369-

8. Clark, Wm. M.

'22 The Determination of Hydrogen Ions. Williams and \Vilkins Co.,
Baltimore, Md.

9. Pantin, C. F. A.

'23 On the Physiology of Ameboid Movement. J. Marine Biol. A

XIII, i, 24.
10. Chambers, Robert.

'29 Intracellular Hydrion Concentration Studies. I. The Relation of ilu-

Environment to the pH of Protoplasm and its Inclusion Bodir-.

Biol. Bull., LVI, 369-

THE EFFECTS OF CHANGES IN MEDIUM DURING

DIFFERENT PERIODS IN THE LIFE HISTORY

OF UROLEPTUS MOBILIS AND

OTHER PROTOZOA.

LOUISE H. GREGORY.
3. THE EFFECTS OF YEAST EXTRACTS.

The effect of vitamines on the vitality of protozoa has been a
subject of but little investigation. In 1917 Calkins and Eddy (i)
reported no effect of treating paramecia with pancreatic vitamine
extracted with Fuller's earth. In 1918 Lund (2), working with
yeast extracts, found that if Paramecia had been starved before
being fed with boiled yeast their size and speed of oxidation were
increased but not the cell division. In 1919 Chambers (3)
reported a slight increase when the animals were fed yeast,
especially ground yeast, and in the same year Flather (4) obtained
similar results with the unpolished rice. All of these experiments
were upon Paramecium, which is not a favorable subject for
investigation unless pure lines are established and endomixis
watched, for a change in the vitality may be due to a reorganiza-
tion of the protoplasm rather than to a change in the environ-
ment. Abderhalden and Kohler in 1919 (5) reported a slight
stimulation of Colpoda cucullus when treated with yeast extracts
but the evidence is not decisive.

Through the courtesy of Professor W. H. Eddy and Dr. Ralph
Kerr, of Teachers College, I have had placed at my disposal the
following yeast extracts, (i) Alpha bios No. 223 extracted by
Professor Eddy in 1924 (6). (2) Beta bios isolated in 1928 by
Dr. Kerr (7). (3) Gamma bios a residue substance similar to
bios II reported by Lucas and Miller (8) in 1924. These three
substances were prepared as indicated in Table I., which has
been compiled by Dr. Kerr.

TABLE I.

THE SEPARATION OF YEAST AUTOLYZATE INTO VARIOUS Bios FACTORS.
I. Preliminary Fractionation.

1. Make autolyzed yeast 66 per cent, alcoholic by volume to precipitate proteins.

Filter.

2. Filtrate from i. Add hot saturated baryta so long as an immediate precipitate

386

EFFECTS OF CHANGES IN MEDIUM.

387

forms. Add alcohol as necessary to maintain 66 per cent, strength. Filter.

Save ppt. for 8.
3. Filtrate from 2. Contains alpha bios and some gamma bios. Neutralize

immediately with sulfuric. Adjust to pH 4.7. Precipitate with iron sol.

and discard pptate.
4 Filtrate from 3. Adjust to pH 5.3 Precipitate with iron sol. Filter. Save

nitrate for 7. Precipitate contains all the alpha bios.

II. Isolation of Alpha Bios.

5 Ppt. from 4. Work up with water
and refilter to remove water
washings. Dissolve ppt. in 30
per cent, sulfuric. When solu-
tion is complete dilute with water
and neutralize with baryta. (Fe,
SO4, ions removed as Fe(OH)s
and BaSO*). Filter by suction
and discard ppt. With baryta
and sulfuric remove quantita-
tively all Ba, Fe and SC>4 ions.

6. Filtrate from 5. Evaporate to dry-
ness and recrystallize from hot
95 per cent, ethyl alcohol. Puri-
fied product has melting point
223° C.

III. Concentration of Gamma Bios.
7. Start with nitrate from 4 Evapo-
rate to small volume. Add sul-
furic to make 5 per cent, by
weight. Filter if necessary and
discard ppt. Now add phospho-
tungstic dissolved in 5 per cent,
sulfuric so long as any ppt. forms.
Filter and discard filtrates. Ppt.
contains gamma bios.
15. Start with phosphotungstates from
7 and 12. Decompose with ba-
ryta in the usual way. Filter.
Make filtrate decidedly alkaline
to litmus. Add alcohol to 80
per cent, by volume. Filter and
discard ppt. if any. Free filtrate
of Ba and SO4 quantitatively.
Filtrate now contains a product
not yet purified but which sug -
gests Miller and Lucas* bios II.
We designate it here as gamma
bios.

IV. Isolation of Beta Bios.

8. Start with ppt. from 2. Wash with alcohol. Then stir washed ppt. repeatedly

with water and filtrate by suction so long as the water is colored. Neutralize
the filtrates immediately with sulfuric. Refilter and discard pptates and
residue.

9. Filtrate from 8. Treat with hot saturated Ag2SC>4. Filter and discard ppt.
10. Filtrate from 9. Treat with hot saturated acid mercuric sulfate. Filter.

Discard ppt.

n. Filtrate from 10. Free from Ag and Hg ions with HjS. Remove excess SO«
with baryta. Save clear filtrates. Evaporate at 40° C. to small volume.

12. Filtrate from n. Make 5 per cent, sulfuric by weight and extract five times

with ether of equal volume discarding ether extract. Add phosphotungstic
acid in 5 per cent, sulfuric until no further ppt. forms. Filter. Add the
ppt. to 7. (See III, 7 above.) Make filtrate slightly alkaline with baryta
and refilter. Discard this ppt.

13. Filtrate from 12. Neutralize with sulfuric and evaporate to small volume.

Make 80 per cent, alcoholic by volume and again ppt. with baryta. Filter
and discard filtrate.

388 LOUISE H. GREGORY.

14. Precipitate from 13. Free of Ba with sulfuric. Evaporate to a thick sirup
at 40° C. Dehydrate by stirring and grinding with dry acetone to a fine
white powder. Filter nearly to dryness on suction filter but leave enough
acetone to make a moist mass. Transfer acetone-wet product to vacuum
desiccator and here free of acetone by suction. Product is Beta bios.

I wish to express my appreciation to Professor Eddy and Dr.
Kerr for their interest and helpful suggestions.

The work with the bioses was begun in 1926-27 at the same
time when experiments with di-sodium phosphate were being
conducted on Uroleptus mobilis in order to determine any vari-
ations in response according to the age of the protoplasm.
Since then Dallasia from pure lines of Professor Calkins,
Stylonychia, and Pleurotricha have been used in addition to
Uroleptus mobilis. Whenever possible the material was taken
from pure lines started from an exconjugant or cyst so that the
age of the protoplasm was known. In the case of Stylonychia
conjugation did not occur and the material was obtained from
a single individual isolated from the wild culture. The methods
used in all experiments were the same as those of earlier papers
and as usual the rate of division is considered an indication of
the vitality of the protoplasm.

In earlier papers (9, 10), results of experiments have been
reported which indicate that the protoplasm of Uroleptus mobilis
varies in its response to treatment. Beef extracts and di-
potassium phosphate cause an increase in the division rate
only when the protoplasm is mature while di-sodium phosphate
causes an increase in the division rate of cells of all ages but
the greatest increase occurs in the mature cells. Experiments
with di-sodium phosphate have been continued and will serve as
an additional control in the majority of the experiments with
the bioses.

I. Experiments with Alpha Bios.

Three series of Uroleptus mobilis were used in these experi-
ments. Various concentrations of alpha bios were tried and
finally a concentration of .05 mg. per cc. was fixed as the best.
The bios solution was added to the normal hay flour medium
daily and controls were carried on in the normal medium and
also parallel experiments were conducted at the same time with a
medium containing di-sodium phosphate in the Packard (n)
concentration of .M/yooo. The results are shown in Table II.

KFKKCTS OF CHANGES IN MEDIUM.

389

TABLE II.

EFFECTS OF DI-SODIUM PHOSPHATE AND ALPHA Bios ON THE DIVISION
RATE OF Uroleplus mobilix.

Amount of Variation from Control in

Division Rate per Line in

Series

Age in

ro-day Periods.

No.

Gen.

*

Di-sodium Phosphate

Alpha Bios

Series.

Series.

139

19

+ 3-8

—

64

+ 3-4

—

125

+ .6

- 3-o

175

+ .6

- 2.4

225

+ 4-4

+ 3-0

240

+ 1.6

- 2.6

275

+ 3-4

+ 3-0

140

19

+ -4

—

60

+ 2.4

+ 1.8

117

- 3-0

- 3-2

157

— i.o

-4-8

218

+ 3-4

+ 2.O

227

+ 4.2

+ .6

141

6

+ 1.2

- i.S

36

- I.O

— 2.2

66

+ 1.8

— 2.2

125

+ 1.6

- 2.4

180

+ 4-2

+ 1.9

192

+ 4-4

+ 2.4

234

+ 3-2

+ 4-0

243

- .6

- 4.0

As in former experiments the sodium phosphate caused a
stimulation of the vitality throughout the life history of Series

139 and practically throughout the life of Series 141. Series

140 was the least vigorous and died out in the 227th genera-
tion after showing instability throughout its life. The greatest
stimulation however in all three series appeared during maturity.

Alpha bios failed to act as a stimulant save in the 225th and
275th generation of Series 139, three times at slightly irregular
intervals in Series 140 and only slightly after the iSoth generation
in Series 141 save in the 234th generation when there was a
slightly higher division rate than that of the control or sodium
series. Thus alpha bios with few exceptions has a depressing
effect on the vitality of these three series of Uroleplus.

These results may be due to at least two factors: (i) too acid a

390

LOUISE H. GREGORY.

condition of the medium, (2) a general lowering of the vitality
of Uroleptus mobilis. Undoubtedly the protoplasm was weaken-
ing as it did not respond as vigorously to sodium stimulation as it
did in the experiments of 1926. On the other hand the H ion
concentration of the alpha bios medium was slightly lower than
that of the normal medium (7.2). Since trial experiments with
beta bios known to be more acid, resulted in a decided lowering
of the vitality and since it seemed uncertain whether any bios
would cause a definite stimulation of an animal cell, experiments
were conducted in which a yeast extract containing all the bioses
was used and in one series di-sodium phosphate was added to
the yeast extract medium to increase the alkalinity.

II. Experiments with Harris Yeast Extract.

Four experiments with Uroleptus at varying ages and two with
Dallasia were carried on in which the animals were kept (i) in a
normal medium to which was added daily yeast extract of a
concentration of .01 mg. per cc., (2) in a normal medium to
which was added di-sodium phosphate, (3) in the same medium
as in (2) with the addition of the yeast extract and (4) in normal
hay flour medium as the control series. These results are shown
in Table III.

TABLE III.

EFFECTS OF YEAST EXTRACT ON THE DIVISION RATE OF

Uroleptus mobilis AND Dallasia.

Series
No.

Age in
Gen.

Amount of Variation from Control in Division
Rate per Line in lo-day Periods.

Sodium
Phosphate
Series.

Yeast
Extract
Series.

Sodium Phos-
phate Yeast
Extract Series.

Uroleptus
146. . . .

143
142
141

15
30
84
250

died
— 2.0

- 3-4
o.o

— .2

-3-6

- 2.8

- 5-2

died

+ -4

- 2.6
- 2.8

Dallasia
I

75

75

- 3-4

+ 3-2

+ 5-0

+ 7.2

+ 9-0
+ 10.4

2

EFFECTS OF CHANGES IN MEDIUM.

391

The experiments with Uroleptus show practically no stimu-
lating effect of the yeast extracts. The protoplasm was too
weak to respond and the entire race died out shortly afterwards.
The two experiments with two different series of Dallasia both
in the 75th generation, showed a definite increase in division
rate in all the experimental series. It was especially marked in
the yeast sodium hay-flour medium when the rate was 10.4
divisions higher than that of the normal control series and 12.4
higher than the sodium hay-flour series for the same period.
Since the yeast extracts caused a marked stimulation of the
vitality of Dallasia both with and without the addition of sodium
to the normal medium the fractional extracts of the yeast were
then used.

III. Experiments with a Neutral Salt of Beta Bios and

with Gamma Bios.

Dallasia, Pleurotricha sp. and Stylonychia sp. were treated
with the two bioses using the same methods as above. These
three animals differ in their normal rate of cell division. Dallasia
when young undergoes from 3-5 divisions daily, Pleurotricha like
Uroleptus not more than 1-2 divisions and Stylonychia divided
every other day. Stylonychia may have been more mature as in
this series no conjugations occurred and the age is unknown.
The results of the experiments are shown in Table IV.

TABLE IV.

THE EFFECTS OF BETA AND GAMMA Bios ON THE DIVISION RATE.

Series

XT/-*

Age in

C^f^rt

Amount of Variation
Division Rat
lo-day

from the Control in
2 per Line in
Periods.

Beta Bios Series.

Gamma Bios Series.

Dallasia

2 ....

115

- .8

+ 4-8

130

+ -2

+ 4-8

Pleurotricha

90
110

+ 5.0 (ist 10 da'
died (2d )

+ 10.2 (ist 10 days)
+ 1.6 (2d " " )
+ 4-8

Stylonychia

—

+ 3.8 (ist 10 days)
+ 4-8 (2d )

392

LOUISE H. GREGORY.

Beta bios apparently had no effect on Dallasia when older
gamma bios, however acted as a definite stimulant increasing the
division rate, 4.8 divisions per line in 10 days. Pleurotricha was
stimulated in the goth and noth generation and showed a
marked response to gamma bios. Stylonychia had its division
rate almost doubled in the gamma bios solution and this effect
continued for twenty days. When Pleurotricha was stimulated
for twenty days the effect died out during the second ten-day
period. This may be correlated with the variation in norma1
vitality of the two species.

IV. Experiments with Alpha, Beta and Gamma Bios.

Finally experiments were conducted to compare the effects of
the three bioses on Pleurotricha and Stylonychia, when added to
the normal medium and in a few experiments to the sodium
medium. The results are shown in Table V.

TABLE V.

EFFECTS OF ALPHA, BETA, AND GAMMA Bios ON THE DIVISION RATE
OF Stylonychia AND Pleurotricha.

Amount of Variation from the Control in the Division Rate
per Line in lo-day Periods.

Alpha
Bios
Series.

Beta
Bios
Series.

Gamma
Bios
Series.

Na2HPo4
Series.

NajHPo4
Alpha
Bios.

Na2HPo4
Beta
Bios.

Na2HPo4
Gamma
Bios.

Stylonychia

+ 1-4

+ 4-2

o.o

• + 4-4

+ 3-2

Pleurotricha
115 gen +10.2
130 gen. died. .

+ II- 4
+ 4-8

+ IO.2

+ 8.0

-I- 8.2
- .8

died

9.0

+ 16.0

+ 12.8

In these experiments, Stylonychia quickened its division rate
in all media save that of the Gamma Bios, where the division
rate equalled that of the control. Pleurotricha in the H5th
generation showed a definite stimulation in all media, especially
in that with the HNa2PO4 and gamma bios. In the second
experiment the division rate of the control dropped to 5.4
divisions per line in 10 days while the experimental series kept
a much higher level, the climax being reached with 18.2 divisions
per line for the same lo-day period in the series kept in normal

EFFECTS OF CHANGES IN MEDIUM. 393

medium to which HNa2PO4 and Gamma Bios solution had been
added.

SUMMARY AND CONCLUSION-,.

These preliminary experiments indicate that while the proto-
plasm of Uroleptus mobilis was usually depressed when treated
with yeast extracts due probably to its weakened condition, that
of Dallasia, Pleurotricha and Styloiiychia were definitely stimu-
lated by the addition of fractional extracts of yeast to the normal
medium. Alpha bios in general causes the least effect and gamma
bios the greatest increase in division rate. \Yhile there is usually
an increase in the protoplasmic activity when sodium phosphate
is added to the medium already containing the bios solution,
this may not be -due to an increased alkalinity as the variation
in Hydrogen ion concentration were not more than .I-.2 of a
point. The explanation may lie in an increase in the per-
meability of the cell allowing a far more reaching effect of the
bios solution. The age and general characteristics of the proto-
plasm must also be taken into consideration and further experi-
ments are planned with pure lines of varying ages.

The fact of a sudden marked increase in the division rate of a
protozoan cell when treated with these yeast extracts brings to
mind the theory of Burrows in which the rapid growth of cells
and formation of tumors is assumed to be due to a lack of balance
between vitamines in the cells. The relation of these extracts
to vitamines has yet to be proved. They are however stimu-
lating substances to yeast cells and to certain animal cells, their
effects varying according to the age and general conditions of
the protoplasm.

BARNARD COLLEGE,
June, 1928.

REFERENCES.

1. Calkins, G. N., and Eddy, W. H.

'17 Soc. Exp. Biol. and Med., Vol. XIV., 162.

2. Lund.

'18 Am. Jour. Phys., Vol. XLVIL, p. 167.

3. Chambers, B.

'19 BIOL. BULL., Vol. XXXVI., p. 82.

4. Flather.

'19 BIOL. BULL., Vol. XXXVI., p. 54-

5. Abderhalden and Kohler.

'19 Arch. f. ges. Physiologic, Vol. CLXXVI., p. 209.

394 LOUISE H. GREGORY.

6. Eddy, W. H., Kerr, and Williams.

'24 Jour. Am. Chem. Soc., Vol. XLVI., 2846.

7. Kerr, R. W.

'28 Soc. Exp. Biol. and Med., Vol. XXV., 3847.

8. Lucas and Miller.

'24 Jour. Phys. Chem., Vol. XXVIII., 1180.

9. Gregory, L. H.

'25 BIOL. BULL., Vol. XLVIII., No. 3.

10. Gregory, L. H.

'26 BIOL. BULL., Vol. LI., No. 3.

11. Packard, C.

'26 Jour. Cancer Research, May.

12. Burrows.

'26 Soc. Exp. Biol. and Med., Vol. XXIV., 3240.

Vol LV December, 1928 No. 6

BIOLOGICAL BULLETIN

INSECT METABOLISM.

THE ANAEROBIC METABOLISM OF AN INSECT (ORTIIOPTKKA).

JOSEPH HALL BODINE,

ZOOLOGICAL LABORATORY, UNIVERSITY OF PENNSYLVANIA.

That insects can live anaerobically for varying periods of time
has been repeatedly pointed out (Winterstein, 1921; Lee, 1924,
1925; Willis, 1925; Davis and Slater, 1926, etc.). When de-
prived of oxygen they enter into a state closely resembling
anesthesia. Various methods for the withdrawal of oxygen have
been experimentally employed, such as replacing the air by the
gases hydrogen, carbon dioxide, nitrogen, etc.; by evacuating
the vessel in which the insects are contained and by immersing
the insects in water. The results produced by all of these
methods closely resemble each other and if the deprivation of
oxygen has not been too long the insects recover and appear quite
normal. A state of anaerobiosis thus produced in insects offers
rather unique conditions for studying the gaseous exchange
of an organism during oxygen lack.

The present work deals with the rates of oxygen consumption
and the blood pH changes in grasshoppers under normal as well
as anaerobic conditions.

MATERIAL AND METHODS.

The grasshoppers, including individuals of the following
species, M elano plus differ entialis, Melanoplus fenuif-nibnini, and
Chortophaga viridifasciata, were hatched and raised under
laboratory conditions and fed lettuce. Organisms of known and
varied ages were used in the experiments. Oxygen determina-
tions were made by means of the modified Krogh Manometer

395

396

JOSEPH HALL BODINE.

(Bodine and Orr, 1925), immersed in a Freas constant temperature
water bath maintained at 25° ±0.1° C. Animals were subjected
to anaerobic conditions largely by immersion in water at 25° C.
for varying periods of time. Immersion in water was found to
give results identical with those produced by the gases hydrogen,
carbon dioxide or nitrogen.

Animals were first put in the manometers and their normal
rates of oxygen consumption determined. After removal from
immersion in water they were again quickly put in the same
manometer and their rates of oxygen consumption during re-
covery noted. By such a procedure a continuous record of the
oxygen consumption of the organism was obtained except for
the actual period of immersion in water.

Blood pH determinations were made by micro-colorimetric
methods (Bodine, 1925). By means of fine capillary pipettes
blood was easily obtained from minute punctures made by fine
needles in the lateral abdominal wall of the animal.

f 6 7 ft 9 /»

TIME //y HOURS

FIG. i. Curve showing the effect of immersion in water for 75 minutes on the
rate of oxygen consumption of a male, nymph, Melanoplus differ entialis. Space
within arrows indicates the period of immersion of the animal; points on curve, the
rates of oxygen consumption before (which is taken as 100 per cent.) and after im-
mersion. Abscissa, time in hours.

JNSI-XT MKTABOLISM.

397

RESULTS.

Oxygen Consumption.

The normal rate of oxygen consumption for each organism
was determined over a period of an hour or more until a constant
rate was obtained. The animal was then removed from the
manometer, placed in a glass tube, the open ends of which wen-
CD vered with wire gauze, and immersed in water at 25° ('. to a

TIME iK HOURS

FIG. 2. Curve showing the effect of immersion in water for i5<j minutes <ni t he-
rate of oxygen consumption of a male, nymph, Mdatwplus diffcn-nliuli^. Space
within arrows indicates the period of immersion of the animal; points mi < urve,
the rates of oxygen consumption before (which is taken as 100 per cent.) ami alter
immersion. Abscissa, time in hours.

depth of 1 80 mm. All air bubbles were removed from tin-
surface of the animal and the ends of the tube by gentle shaking.
The animal becomes motionless within a very short time after
immersion and remains so throughout the immersion period.
After immersion, the organism is quickly removed from the glass
tube, dried on filter paper and returned to the same manometer
originally used to determine its normal rate of oxygen consump-
tion. The recovering animal is left in the manometer and its

•

rate of oxygen consumption followed until complete recovery

26

398

JOSEPH HALL BODINE.

occurs. Since the general response of all organisms is essentially
the same, only typical experiments will be presented.

Figures I, 2, 3 and 4, in which the rates of oxygen consumption
are expressed in terms of the normal rates (100 per cent.), show
graphically the changes in the rates of oxygen consumption pro-
duced by exposures to lack of oxygen. An examination of these

100

kl

3

"

CHOR. V/RID

(,0 Min.

•« -*.-*-«- tf- -a- -A .

/ 2 * t e 9 "»

T/rtE Ifi HOURS'

P"IG. 3. Curve showing the effect of immersion in water for 60 minutes oji the
rate of oxygen consumption of a male, numph, Chorlophaga viridifaciata. Space
within arrows indicates the period of immersion of the animal; points on curve, the
rates of oxygen consumption before (which is taken as 100 per cent.) and after im-
mersion. Abscissa, time in hours.

figures further shows that when the animal is readmitted to
oxygen after immersion its rate of oxygen consumption increases
considerably over the normal rate or that characteristic for the
animal before deprived of oxygen. This excess oxygen taken up
by the organism can be shown, in carefully controlled experiments,
to be approximately equal to the amount the organism would
have taken up normally during the period it was deprived of
oxygen. In other words, it seems that the grasshopper when
deprived of oxygen or existing anaerobically, goes into debt for
oxygen in a manner quite similar to that pointed out for the

INSECT METABOLISM.

399

cockroach (Davis and Slater, 1926) and for heavy physical work
in man or for isolated muscle (Hill, 1922).

The length of exposure to lack of oxygen that can be with-
stood by different species of grasshoppers varies as pointed out
below. Some species have been found to successfully withstand
as high as 7 hours immersion in water. The rates of oxygen con-
sumption during recovery, as indicated in Figs, i, 2, 3 and 4,

CHOR

no ft,n

.

9 6

TIME //Y HOURS

e

/o

FIG. 4. Curve showing the effect of immersion in water for 120 minutes on the
rate of oxygen consumption of a male, nymph, Chortophaga viridifasciata. Space
within arrows indicates the period of immersion of the animal; points on curve, the
rates of oxygen consumption before (which is taken as 100 per cent.) and after im-
mersion. Abscissa, time in hours.

seem to be greatly influenced by the length of the immersion
period. Animals immersed for 60 to 120 minutes usually recover
in a typical manner as shown in these figures. When the im-
mersion period is lengthened, however, there is a strong tendency
for the rates of oxygen consumption to return to normal in an
extremely slow fashion as shown in Figs. 2 and 4. This slowness
in recovery to a normal rate of oxygen consumption is probably
correlated with the physiological condition of the organism as well
as with the fact that the exposure might be just a sub-lethal one
for the animal.

400

JOSEPH HALL BODINE.

There also appears to be a marked difference in the rates of
recovery in the different species. Melanoplus differ entialis
seems better able to reversibly withstand long immersion than
Chortophaga viridifasciata, as indicated in Figs. 2 and 4. Age is
also an important factor, since younger individuals withstand
and recover from long immersions better than older ones.

The relations between length of immersion in water and re-
covery time for the different species of grasshoppers examined are

K

f

I

CHOR. v/x/a
VEL. 01 rr.

tOO IfO 20O 25O 3OO

LEVGTH Or EXFDSURE-

35O

fD

FIG. 5. Curves showing relation between average mean recovery time and
length of immersion in water in young adult grasshoppers of three species, Melano-
plus femur rubrum, Melanoplus differentialis and Chortophaga viridifasciata. Each
curve based on several hundred observations. Recovery time indicates return of
"turn-over or righting" reflex.

graphically shown in Fig. 5. From a study of this figure it is
evident that a linear relationship exists between length of im-
mersion and recovery time. That the causes of the anaesthetic
condition produced by lack of oxygen are doubtless due to the
carbon dioxide and lactic acid produced within the organism
seems reasonable when a comparison is made between the results
obtained by immersion in water and those obtained by subjecting
the organism to carbon dioxide (Willis, 1925), as indicated in
Fig. 6. These curves are qualitatively similar. In view of
similar evidence gained from studies on anaerobiosis of mam-
malian muscle (Hill, 1922), such an hypothesis seems reasonable
in explaining these phenomena in grasshoppers. As a matter of

I.SMCCT METABOLISM.

401

fact, Lee (1924) has shown that injection of carbonic acid or
lactic acid into a grasshopper produces results quite similar to
the present ones on immersion of the animals.

FIG. 6. Curves showing comparisions between average mean recovery times
and rates and length of immersion in water and exposure to carbon dioxide in
Melanoplus differ entialis. Carbon dioxide data from Willis (1925).

BLOOD pH.

As previously pointed out the normal pH of the blood of the
grasshopper, M. differ entialis, seems to be 6.8. (Bodine, 1926.)
A careful study has been made of the blood of individuals through-
out the present immersion and recovery experiments. Fig. 7, in
which are plotted together results of experiments carried out for
different periods of time, shows graphically the pH changes oc-
curring in the blood during the anaerobic and recovery periods.
A progressive lowering in pH values with length of immersion
occurs and seems to reach a minimum at about 5.8. Below this
minimum the animal seems irreversibly affected. Upon re-
covery, a slow blowing off of acids (HzCOs), probably occurs and
the blood then gradually returns to its normal pH value. An
interesting fact, indicated in Fig. 7, is that no marked changes in
blood pH occur in immersed animals when removed from the
water until after respiratory movements have become established.
Upon careful examination of the animal it is found that initial
respiratory movements upon recovery are extremely slow and of
great depth. As recovery progresses the movements become
more regular and normal. These changes in blood pH strongly

4O2

JOSEPH HALL BODINE.

suggest that during anaerobiosis large quantities of acids,
carbonic and lactic, are produced and that recovery consists
largely in their elimination by the organism.

As pointed out by Davis and Slater (1926), who have found
similar results for the anaerobic metabolism of the cockroach,
insects seem to be extremely favorable material with which to
further elucidate the changes occurring during the anaerobic
existence of organisms and the question of energy formation under
such conditions.

7.0
(,.8

I

f-7

VELDIFF;

/oo /fo

Tint in

zoo

FIG. 7. Curve showing the effect of immersion in water for 120 minutes on the
pH of the blood of a male, nymph, Melanoplus differ entialis. Space within arrows
indicates period of immersion of the animal; points on curve, actual pH determina-
tions of blood before and after immersion. Broken portion of curve is composite,
being made up from pH determinations of the blood of individuals immersed for
periods of time ranging from 50 to 120 minutes.

SUMMARY AND CONCLUSION.

1. Rates of oxygen consumption in grasshoppers before and
after immersion in water (lack of oxygen) have been determined.

2. During oxygen lack, grasshoppers build up an oxygen debt.
When readmitted to oxygen an increased rate of oxygen consump-
tion occurs and an excess of oxygen, approximately equal in
amount to that which the organism would have taken up normally
during the period it was deprived of oxygen, is consumed.

IN si ( I MKTABOLISM. 403

3. During anaerobiosis Mood pH falls. I'pon recovery pll
values slowly return to normal.

4. It is suggested that the chemical change responsible for the
anesthetic condition accompanying anarrobiosis is the production
of an excess of acid, carbonic and lactic, and that recovery con-
sists in their elimination.

LITERATURE CITED.
Bodine, J. H.

'25 Physiology of the Orthoptera. Hydrogen Ion Concentration of the Blood
and Alimentary Tract of Certain Orthoptera (Grasshoppers). BIOL.
BULL., Vol. XLVIII., pp. 79.
Bodine, J. H.

'26 Hydrogen Ion Concentration in the Blood of Certain Insects (Orthoptera).

BIOL. BULL., Vol. LI., pp. 363.
Bodine, J. H., and Orr, P. R.

'25 Respiratory Metabolism. Physiological Studies on Respiratory Metab-
olism. BIOL. BULL., Vol. XLVIII., p. i.
Davis, J. H., and Slater, W. R.

'26 The Aerobic and Anaerobic Metabolism of the Common Cockroach (P.

Orientalis). Biochemical Jour., Vol. 20, pp. 1167.
Hill, A. V.

'22 Mechanism of Muscular Contraction. Physiol. Reviews, Vol. 2, pp. 310.
Lee, M. O.

'24 Respiration in Orthoptera. Amer. Jour. Physiol., Vol. 68, pp. 135.
Lee, M. O.

'25 On the Mechanism of Respiration in Certain Orthoptera. Jour. Exp.

Zool., Vol. 41, pp. 125.
Willis, J.

'25 Effects of Carbon Dioxide. Effects of Different Tensions of Carbon Di-
oxide on Certain Orthoptera (Grasshoppers). BIOL. BULL., Vol. XLVIII..
pp. 209.
Winterstein, H.

'21 Handbuch d. vergleichenden Physiologic, Jena.

THE PULSATORY RHYTHM OF THE CONTRACTILE
VESICLE IN PARAMECIUM.

FRANCIS E. LLOYD AND J. BEATTIE.
I.

It became apparent to the senior author a year ago that there
are discrepancies between the more recent, and therefore pre-
sumably the most correct, accounts of the behavior of the con-
tractile vesicle in Paramecium and the objective facts as ap-
prehended by him. As to these facts both the present authors
have found themselves in agreement, and it was therefore
determined to make an investigation of said behavior by such
refined means, by way of control of direct and unaided vision,
as are available. Two methods were used, that of recording
visual observations of critical points in the contractile cycle on a
rotating drum and, still better, that of making motion pictures at
normal speed, viz: 16 per second, without any lapse. It is not
easy to get a Paramecium to remain within the field of vision long
enough to take a motion picture so as to get a continuous view of
the contractile vesicle for two or three cycles of movement. We
have however succeeded by making use of very slight compression
between cover slip and slide — a method to which there is some,
but we believe not wholly justified objection — and by surrounding
the animals with a suspension of Chinese ink, a time honored
method for demonstrating the expulsion of fluid from the con-
tractile vesicle, as used by Carter (1861) and by Jennings (1904)
and by many others before and since. It happens that a rather
thick suspension of this insoluble pigment impedes the move-
ments of the animal, so that the chances for observing a relatively
quiet one, with the contractile vesicles in full view, is much in-
creased without, we think, in the least affecting the behavior of the
vesicles themselves. We suspect that in Chinese ink there is an
admixture of some aromatic substance which may act as a mild
depressant, narcotic or otherwise, but of this we have no sure
knowledge. However, the animals can live for several hours

404

CONTRACTILE VESICLE IN I'ARAMI.i HM.

405

under a cover-glass in a thick suspension of Chinese ink and ap-
pear quite undamaged. We have no doubt at all events that the
behavior of the vesicles is normal. The motion pictures produced
by the senior author have been shown l in illustration of this
paper; in the accompanying plate we present a few excerpts from
one of the films to serve present purposes.

The point of departure of this study may be better appreciated
by considering in the first place the latest pronouncement on the
subject, that of Nassonow (1924). In the text which follows we
shall speak simply of the vesicle 2 (-- contractile vacuole) and the
canals (= rays, canaliculi, radial canals). The general topog-
raphy of the apparatus is so wrell known that we may take
familiarity with it for granted. Nassonow says :

"After the emptying and the complete disappearance of the
vesicle there become visible in its immediate vicinity the 5-7
canals. The ends directed toward the center of the vesicle are
strongly swollen and no continuity between them is to be seen.
The swollen ends approach each other, flow into each other and
form a new vesicle, into which the fluid of the canals now flows.
Thereupon the canals quite disappear and only later do new
canals begin to be formed in their place the ends of which after
the emptying and disappearance of the vesicle suffers enlarge-
ment and in this manner complete the cycle " (Nassonow, IQ24,3
p. 454). Nassonow then goes on to recall the views of others,
including that of Stempell (1914) in regard to the existence of a
membrane, with which Nassonow was particularly concerned, ap-
parently acquiescing completely with this author in respect of the
progress of the cycle of behavior. We therefore quote Stempell
also as follows:

1 Winnepeg meeting of the Royal Society of Canada, May, 1928.

2 It seems to us that Claparede and Lachmann chose the better terminology,
and we follow them, with Pritchard.

3 Among the figures illustrating the paper by Nassonow occur two which wo
may remark in passing to be capable of precisely the opposite interpretation to
that given by him. His Fig. 40 is labelled "Diastole of the excretion apparatus"
while Fig. 42 is labelled "Systole of the excretion apparatus." If by excretion ap-
paratus he means the vesicle then figure 40 represents early systole and figure 42
early diastole. If however he means the canals then his labelling is correct. It is
not easy to understand his exact meaning. It is certain that diastole and systole
of the vesicle ar,e not synchronous with those of the canals.

406 FRANCIS E. LLOYD AND J. BEATTIE.

'The end-products of metabolism collect in dissolved condition
in certain places in the protoplasm, namely, in a canal- or space
system (probably a branched one) the exits of which run towards
the two pulsating vesicles as afferent canals. As soon as the
vesicle is emptied ('Sobald die Vacuole sich entleert') the ends of
these afferent canals swell up to form ' Bildungsvacuole' since
the fluid flows hereinto as to the place of minimal pressure, and
is here dammed up. As a result of this pressure delicate proto-
plasmic valves open and permit the volumes of fluid which have
collected in the canal-ends to flow together in the vesicular space,
on which, after this is filled, the valves promptly close. Since
the fluid now collected in the vesicle has a high osmotic pressure,
there results a lasting addition of water to the fluid already held
in the vesicle by diffusion through the semipermeable membrane
formed ad hoc. As soon as the pressure of the fluid in the vesicle
has reached a certain height, that is, has become higher than that
of the external water pressure, a second valve at the apex of the
papilla-like dome of the pellicula opens and there results from the
pressure of the protoplasm and of the surface tension of the
vesicular drop a complete emptying of the vesicle, whereupon the
process is repeated in the same way" (Stempell, 1924, p. 460).
Here it may be remarked that Fortner (19260 and b} and v. Gelei
(1926) without inquiring into the validity of the above view,
proceeded on the assumption of its truth.

With regard to Stempell's ideas, as above expressed, there
can be no doubt since he has furnished us with a diagram.
From both description and diagram we learn that Stempell does
not entertain at all the idea that there is any flow of fluid from the
vesicle into the canals. Nassonow's idea is identical, we believe
we are right in saying, but his diagram, taken from Putter (1903),
might be interpreted otherwise, as witness Figs. 5, 6 and 7, which
show the canals enlarging before the contraction of the vesicle.
Whether this enlargement results from backflow from the vesicle,
or from the collection ("ponding back" as Carter expressed it)
of fluid derived from the surrounding protoplasm, is the question
with which we are concerned. We are now in a position in the
second place to examine the view of earlier observers of the same
phenomenon.

Felix Dujardin (1841) made a drawing, reproduced in his Plate

COXTRACTII.K VKSICI.K IN I'\R\M|.( HM. 407

8, Fig. 6a and 6b in which the radiating vesicles ("taken by Khren-
berg for seminal vesicles") are seen in the condition just before
the systole of the vesicle, the canals being expanded. Dujunlin
does not, of course, have anything to say about the matter, but
his drawing could not be correct if the canals do not fill before the
systole of the vesicle. The question above indicated therefore
recurs, whence the fluid which fills the canals.

As to this there was no doubt in the minds of Claparede and
Lachmann (1858) (Lachmann 1857 for 1856). It was these
who held the view that the apparatus is the homologue of
the circulatory apparatus in the more differentiated animals and

<5tC(jndj. f 2s 3 tj. &• 6 & y /a //

it was consistent with their view that there was no opening afford-
ing an exit for the fluid to the outside. Their unfortunate error
in this seems to have led to a general condemnation of their win >lr
conception and thereby their critics, in overlooking what they
did see, fell into an error as grievous, namely, in failing to see
that the canals are in the first instance filled at the expense of the
vesicle, as we hold. Lachmann 's description (1857) will serve
our turn at the moment.

He says (1857, p. 224), maintaining that the thin area of the
body wall over the vesicle is only a thin place fit for diffusion and
with no opening, that when the vesicle is fully expanded the
canals are fine lines. By the sudden contraction of the vesicle,
however, the canals instantly swell into pyriform spaces close to
the contractile vesicle, which has disappeared. During the slow
reappearance of the vesicle, the canals gradually decrease and
they have again been reduced to fine lines by the time the
vesicle has become fully inflated.

It must be clear that Lachmann believed that the swelling of
the canals is synchronous with the early period of systole of the

4o8 FRANCIS E. LLOYD AND J. BEATTIE.

vesicle. Carter (1861) does not, we think, correctly take his
meaning when he says: "Claparede and Lachmann have said
that the fluid of the vesicle is returned into the vessels on the
systole or contraction of the vesicula because the sinuses and
vessels become filled immediately afterward" (italics ours) as this
is not what Lachmann said.1

It is in this connection that Carter suggests that the swelling of
the canals into the characteristic pyriform is due to the "ponding
back" of the fluid which flows through the canals into the vesicle
for the short time that the latter empties itself, like the ventricles
of the heart but in the other direction (1861, p. 282). We may
here remark, what we shall endeavor to show to be true, that the
rate of swelling of the canals does not consist with the idea that
the fluid reaches the lacunae by diffusion through the walls, the
rhythm of diastole and systole in these being of the same character
as in the vesicle ; and, if the rhythm of the vesicle can be under-
stood only when it is admitted that the fluid of the canals gushes
into it, the same must be admitted for the canals, but in the
opposite sense.

Somewhat earlier, and in contrast to Lachmann and Claparede
Lieberkiihn (1856), while agreeing with them as to time relations
of vesicular systole and canalar diastole, saying that "a little
before we observe the commencement of the systole, the vessels
begin to expand slowly, etc.," simply denied that there is any
backflow. The interest here is obviously the correct observation
in regard to time relations in question. Spallanzani also believed
that the canals become empty as the vesicle fills, and do not re-
appear until some time after it has contracted and that therefore
"The fluid with which the vesicula is distended comes through
the sinuses, but is not returned by them to the body" (through
Pritchard, 1861).

J. Miiller (1856) appears, according to Claparede and Lach-
mann (1858, p. 51), to have taken the same view of the time rela-
tions. We transcribe their summary of his views, since we have
been unable yet to see Miiller's original paper. This author
distinguishes in the behavior of "central circulatory apparatus"
of Paramecium two partial systoles which alternate with each

1 We have not been able to see Claparede's paper, but it appears that these two
observers, Claparede and Lachmann worked in harmony, sharing each other's views.

CONTRACTILE VESICLE IN I'AKAM KCIfM. 409

other — the systole of the vesicle, then the systole of the fusiform
or pyriform swellings. The latter coincides with the diastole of
the vesicle. Lieberkiihn had already observed that "un instant
avant le systole des vesicules les rayons se ren fluent considerable-
ment." Miiller explains the phenomenon by showing that the
vesicle contracts, diminishing insensibly in volume in the instant
which precedes systole and forces at once a part of its content-
into the "rays of the star." Then the systole of the vesicle takes
place, which produces a further swelling of these rays.

We cannot refrain from mentioning, in passing, the work of
Wrzesniowski (1869), who studied Enchelyodon, Trachelopliyllnni
and Loxophyllum, (but was however chiefly concerned with the
question of the absence or presence of a contractile membrane),
because there is some evidence in his results which point to the
presence of a contractile vesicular apparatus similar to that of
Paramecium, though the author himself, if he adhered to the
original account, would deny this. One point may be mentioned,
however, namely, that a series of small vesicles is formed on the
surface of, and from the contractile vesicle during early systole,
and these, upon growth, run together later to form a new con-
tractile vesicle (not the old one reextended). This view of
Wrzesniowski's seems to be strongly linked with his conviction
that the vesicle is formed de novo and totally lacks a membrane
in any but the sense of molecular physics as Khainski (1911)
would express it.1

We pass to the year 1883 when Maupas attacked the subject.
According to him the systolic movement of the vesicle is sudden
and rapid. A little before it happens the canals commence to till
in the form of elongated pears at a little distance from the point
where they open into the vesicle. Maupas' account indicates a
high degree of meticulous care in observation He goes on to
remark for example that the systole of the vesicle takes place
more often before the pyriform swellings (of the canals) have at-
tained their full size. In spite of the fact that he correctly ap-
prehended the time relations involved he pronounces for the VH-NV
that the canals are simple afferent conduits and sententiously

1Samuelson observed in 1857 that the single globular vesicle in Glaucoma
scintillans when, it contracts forces the fluid into others which appear temporarily
around it.

410 FRANCIS E. LLOYD AND J. BEATTIE.

remarks "I have never seen the liquid of the vacuole reenter
them." This would indeed be difficult and his failure cannot be
charged to his discredit, for at all events he very correctly de-
scribes the at first irregular contours of the vesicle during the
early stages of its diastole when, under systole of the canals, these
empty themselves into the vesicle. Maupas was on the side of
the non-membranists.

It will be seen that these earlier observers, while disagreeing in
regard to the afferent-efferent nature of the canals, support a
majority view which, as we believe, correctly describes the time
relations between the behaviours of the vesicle and the contrib-
utory canals. It is therefore a curious fact that later observers,
as we have already shown at the outset of the paper, siding with
the view of the solely afferent nature of the canals have in some
way been led to overlook the true time relations.

Closely connected with the general trend of inquiry above out-
lined is the parallel inquiry into the nature of the membrane lining
the vesicular cavity It will be easily apprehended that very
convincing evidence has been so difficult to obtain that only
recently has Miss Howland (1924) favored the view that a proper
membrane in the morphological sense is present constituting the
branching cavity composed of the central vesicle and its con-
tributory canals. She succeeded in isolating the membrane with
little distortion by micro-dissection from an animal (Paramecium
catidatum) treated with a strong solution of alizarin blue. This
author expressed some doubt of her interpretation based on the
possibility that the dye had coagulated the surface material of the
vesicle and so produced an artefact. In the same year Nassonow
presented evidence based on the method of osmication which
would convince even the elect were it not for a doubt similar to
that expressed by the former author. We venture to think that a
weak link exists in the chain of his argument. We are not here
concerned with this author's views of the homology of the con-
tractile vesicle with the Golgi apparatus although we subscribe to
the general view supported by Nassonow that the pulsating
vesicle is a true organelle of morphological value, as Lachmann so
long ago held. With regard to earlier observers it will boot us
little to bring forward the details of their views, a summary of

CONTRA* III. !•; VESICLE IN I'AKAMKM I'M . 4! I

which will he found in a paper by Taylor (1923). Portlier, by
compressing animals in a hypertonic solution of cant- sugar. \va>
able to set free the apparatus surrounded by protoplasm and in a
state of approximate diastole. Their behavior he argues un-
qualifiedly postulates the impermeability of the membranes; but
these membranes he believes arise ad hoc, that of the vesicle at
the completion of each systole affording the new membrane for
the papilla pulsatoria. Without further discussion of this matter
from the historical point of view we may be permitted to remark
that had the true time relations in the cycle of events not been
lost sight of, the protagonists of the "non-membranous" view
would have suffered pause.

II.

No special technique is required to demonstrate the phasic
activity of the contractile vacuole and canaliculi in Paramecinm.
Care must be taken that the cover slip over the preparation does
not press untowardly on the animal, otherwise the pore to the
exterior may be blocked and the contractile vesicle fail to dis-
charge in the normal manner, and at normal rate.

After the preparation has been made it is well to allow some
minutes to elapse before the preparation is examined as it is a
hopeless task to attempt to observe the contractile vesicle in one
single animal, while the animals are in rapid motion immediately
after they have been placed on the microscopic slide. In a short
time the animals settle down to feed, and it is then possible to
watch a whole group and to pick out one animal for observation.
It is possible also to trap the animals in a very fine capillary tube
and so limit their movements except round a longitudinal axis.
A better method, but open to the objection of an abnormal en-
vironment, is to mix finely ground China ink with the mounting
medium. This appears to impede the movements and so far a>
one can see there is no interference with the normal cycle of
events within the contractile vacuole system.

After close observation for a few minutes the following series of
changes can be seen The contractile vesicle will be observed as
a highly refractile almost spherical droplet lying in the m<»t
superficial part of the cytoplasm. When the animal rolls over on
its side it will appear that at one point there is a close attachment

412 FRANCIS E. LLOYD AND J. BEATTIE.

of the vacuole to the pellicle. At this time the vacuole when
viewed from the side will appear as three quarters of a sphere with
a conical apex attached to the pellicle. When the animal rolls
so that the vacuole is observed from above with careful focusing a
bright minute ring will be seen in the center of a small clear area
in the pellicle. This is the pore through which the vesicle expels
fluid to the exterior. The vesicle gradually enlarges and in doing
so changes its shape from the conico-spherical form to a perfect
sphere. Enlargement after the spherical shape has been attained
is slow and very small in amount as to linear dimensions. Sud-
denly at the end of diastole the vesicle appears to get smaller
(Plate I, Figs. 3 to 4) and at the same moment, not afterwards,
radiating canals appear surrounding the vacuole (Plate I, Fig. 3).
Seen from above the inner ends of these structures are separated
from the vacuole by a distinct area of protoplasm. Seen from the
side the bulbous or pear-shaped ends of the canals are observed
to lie in the most superficial layer of cytoplasm and to be con-
tinued more distally into the deeper parts of the cytoplasm as
fine canals.

After this phase, which can only be interpreted as a distinct
diastole of the canaliculi caused by systole of the vacuole and not
merely as a damming back of liquid attempting to flow into the
vesicle, the vacuole suddenly contracts (Plate I, between Figs. 4
and 5) and expels the remaining contents to the exterior. There-
fore systole of the contractile vesicle consists of two distinct
phases :

(a) First, an early systolic phase during which the contractile

movement of the vacuole is slow and diastole of the canals
rapid (Plate I, Fig. 3, 4).

(b) Second, a later period during which the vacuole expels the

remainder of its contents to the exterior (Plate I, Fig. 5).
The behavior as thus set forth has been displayed graphically
in the accompanying diagram, in which, to some extent provi-
sionally, we have attempted to express the time relations seen in
the rhythm of the contractions and expansions of vesicle and
canals, while the volume relations are avowedly inexact, but ap-
proximate. Time is plotted on the abscissa, and the volume of
the canals and of the vesicle on the ordinates, the total volume of
the vesicle being taken as one. The hatched areas are bounded

CONTRACTILE VESICLE IN PAKAMKCIUM. 413

by the curve of diastole and systole of the canals; the areas
bounded by the curves for the vesicle are left blank.

There is no doubt that there is a discharge of vesicular con ten t -
to the exterior. Jennings showed this first convincingly as has
been stated above and we have been able to confirm his observa-
tions and to make a motion picture of the process.

There has been some doubt expressed as to whether or not the
pore through which the vacuole discharges can admit fluid from
the surrounding medium. We have found no evidence to support
this theory. All our observations go to show that after the
vesicle has discharged its contents reconstitution of the vacuolar
space takes place by the discharge into the collapsed cavity of
the fluid contained in the canals (Plate I, Figs. 1-2; 6-8).
Discharge of the contents of the canals into the vesicular space
takes place within one second after the completion of systole of
the vacuole. The canals do not however discharge simultane-
ously but by careful observation one is able to make out that
first one canal may discharge into the collapsed vesicular region
which then forms an irregular angular cavity 1 soon followed by
another and then by the remainder. When the last canal has
discharged the space is seen to be conico-spherical as described
above. It is possible to analyse the discharge of the canaliculi
into the vacuolar space only by study of the motion picture film.

After the reconstitution of the vesicle enlargement takes place
and this phase of diastole of the vacuole occupies the longest
period of the cycle of events. One notices that the conico-
spherical form persists for quite a time (almost three quarters of
diastole) before the spherical form is assumed. Once the vesicle
becomes spherical systole of the structure takes place within a
second or two.

The cycle of events occupies normally about eight seconds.
Records which we have made show that in fresh specimens cycles
of seven and one fifth seconds were common. \Ye have observed
cycles which required ten seconds for completion. When the
cycle lengthens it is the diastolic period which is chiefly prolonged.
When the animal is compressed gently it is possible to occlude
the pore and so prevent the second phase of systole taking place.
The first phase, i.e. diastole of the canals takes place but then is

1 Beautifully recorded by Nassonow, Fig. 42.
27

*

414 FRANCIS E. LLOYD AND J. BEATTIE.

no discharge to the exterior. In a short time the canals reappear
and so the cycle goes on. The vacuole continues to enlarge and
before very many minutes the pellicle ruptures and the proto-
phism is extruded carrying with it in some cases the entire con-
tractile vesicle. The vacuole may be seen lying as a spherical
body in the surrounding fluid. We have not observed any sign
of a canal when the protoplasm is examined after bursting.

When neutral red is used in solutions of one part to four hundred
or higher concentrations it is frequent to observe the gradual con-
traction of the cytoplasm from the pellicle and the formation of a
peri-cytoplasmic space rilled with fluid. In one specimen of
which a photograph is shown (Plate I, See Fig. 9) the cytoplasm
in contracting pulled a fine cone of pellicle downwards. At the
apex of this was attached the pulled-out contractile vesicle
which extended as a conical cavity through the peri-cytoplasmic
space to the dimple in the pellicle. The actual interface between
the vesicle and the surrounding fluid could be seen. The apex of
the dimple was the pore through which the vacuole discharges.
It also would appear to show that the vacuole when it discharges
to the exterior is not reconstituted de novo in the old site but rather
that there is something of a permanent nature — a vesicular
membrane into which is discharged the contents of the canaliculi
when the vesicle is reformed.

Miss Rowland, as we have already said, has been able to isolate
the vesicular membrane from preparations treated with alizarin
blue.

By way of summary of the above we draw attention to the
following important facts.

Diastole of the vesicle falls into two phases — an early rapid and
a later slow one.

The early rapid phase is due to the systole of the canals during
which their fluid content is forced into the vesicle.

i

The later slow phase of diastole of the vesicle is due to further
distention by diffusion of water into the vesicle.

Systole of the vesicle falls similarly into two phases, an early
slow phase during which the fluid is forced into th'e canals
(diastole of these) and a later rapid phase during which the re-
mainder of the vesicular fluid is forced through the spore into the
surrounding medium.

CONTRACTILE VESICLE IN PARAMECIUM. 415

It appears that early diastole of the vesicle is synchronous with
the systole of the canals ; and that early systole of the vesicle is
synchronous with diastole of the canals. During early diastole
of the vesicle, this is partly filled with fluid from the canals,
This is the residual fluid plus that which has in the interim en-
tered by diffusion into them. During early systole of the vesicle
the canals are partly filled with fluid from it — this we may speak
of as the residual volume. That volume which is discharged l>y
the vesicle is the overplus accumulated by diffusion into the
vesicle and canals during their diastolic periods.

Viewed thus, the mechanism is one in which a certain quantity
of fluid of relatively high osmotic pressure is retained in the
canals, derived by them from the central vesicle, and which is at
once put into service to withdraw water from the body into the
pulsatory apparatus. Thus an important feature of Stempell's
view receives support, even though his conception of the methods
of working of the apparatus is incomplete. If it depended solely
upon diffusion for filling, from the completely collapsed state to
the completely replete, it could, in our opinion not work so
rapidly and efficiently.

REFERENCES.
Carter, H. J.

'61 Notes and Corrections on the Organization of Infusoria, etc. Annals &

Mag. Nat. History, ser. 3, 8: 281-290.
Claparede, Ed., and Lachmann, Johannes.

'57-'6i Etudes sur les infusoires et les rhizopodes. Mem. de 1'inst. Nat.
Genevois 5: (for 1857) 1-260, pi. 1-13, 1858; 6: (for 1858) 261-482, pi. 1-24.
1859; 7: (for 1859-60) 5-291, pi. 1-13, 1861. Vol. 5 contains the general
discussions.
Dujardin, Felix.

'41 Infusoires, etc., Histoire naturelle des zoophytes. Suites a Buffon, Paris.
Fortner, H.

'263 Ueber die Gesetzmaessigkeit der Wirkungen des Osmotischen Druckes
physiologisch indifferenter Loesungen auf einzellige, tierische Organismen,
Biol. Centralbl. 45: 417-446.
'26b Zur Frage der discontinuirlichen Excretion bei Protisten, Arch. f. Protist-

enk. 56: 295-320.
v. Gelei, J.

Nephridialapparat bei den Protozoen. Biol. Centralbl., 45: 676-683.
Rowland, Ruth B.

'24 Dissection of the Pellicle of Amoeba verrucosa. Journ. Exp. Zool., 40:

263-270.

'24 On Excretion of Nitrogenous Waste as a Function of the Contractile
Vacuole. Ibid., 40: 251-250.

41 6 FRANCIS E. LLOYD AND J. BEATTIE.

'24 Experiments on the Contractile Vacuole of Amoeba verrucosa and Para-

mecium caudatum. Ibid., 40: 251-262.
Jennings, H. S.

'04 A Method of Demonstrating the External Discharge of the Contractile

Vacuole. Zool. Anz., 27: 656-658.
Lachmann, C. F. J.

'57 On the Organization of the Infusoria, especially the Vorticellae. A. & M.
N. H. 19: ser. 2, 113-128; 226-241. (Translated from Miiller's Archiv. p.
240, 1856.)
Lieberkuhn, N.

'56 Contributions to the anatomy of the Infusoria. A. & M. N. H. 18: ser. 2,

319. (Translated from Miiller's Archiv. Jan. 1856.)
Maupas, £.

'83 Etude des infusoires cilies. Archives de Zool. Exp. et Gen., i: 634.
Miiller, J.

'56 Beobachtungen an Infusorien. Monatsbericht der Berliner Akad., p. 393.
Nassonow, D.

'24 Der Exkretionsapparat (Kontraktile Vakuole) der Protozoa als Homolog des
Golgischen apparats der Metazoazellen. Arch. f. mikr. Anat. u. Entwick.
mech., 103: 437-482.
Putter, A.

'03 Die Reizbeantwortung der ciliaten Infusorien. Zeitschr. f . allgem. Physiol.,

3: (Heft i).
Pritchard, A.

'61 A History of Infusoria. 4 Ed. London.
Schmidt, O.

'53 Lehrbuch der Vergleichenden Anatomic, 1853 Froriep's Notiz., Vol. 9:

(through Pritchard).
Stempell, W.

'14 Die Funktion der pulsierenden Vacuole. Zool. Jahrb., 34: 437-478.
Taylor, C. V.

'23 The Contractile Vacuole of Euplotes, an Example of Sol-gel Reversibility of

Cytoplasm. Journ. Exp. Zool., 37: 259-290.
Wrzesniowski, A.

'69 Ein Beitrag zur Anatomie der Infusorien. Arch. f. mikr. Anat., 5: 25-48.

FRANCIS E. LLOYD AND J. BEATTIE.

EXPLANATION OF PLATE.

FIGS. 1-8. Eight episodes from motion picture (photomicrographic) of Para-
mecium caudatum. The animal was slightly compressed between slip and cover;
the periodicity was slightly slower then normal therefor. The position in the film
is indicated for each. Exposure 16 per second. Enlargements at constant distance.

FIG. i. Foot 887 frame 16. Mid-diastole. Canals are emptying into vesicle.

FIG. 2. 883-1. Late diastole. Traces of canals visible.

FIG. 3. 880-16. Early systole of vesicle; canals beginning to fill.

FIG. 4. 876-10. Mid systole of vesicle which is now smaller; canals nearly
filled.

FIG. 5. 873-15. Systole of vesicle complete, canals full.

FIG. 6. 871-10. Mid-diastole (somewhat later than Fig. i); canals emptying
into vesicle.

FIG. 7. 870-3. Later diastole; canals nearly disappeared.

FIG. 8. 868-9. Diastole complete; canals empty. One canal persists longer
than the others: note that it occurs in Figs, i, 2, 5—8.

FIG. 9. An animal treated with neutral red (see text), showing the vesicle
pulled away from the pellicle, and dimpling it by pulling on the pore rim. The result
follows from the shrinkage of the cytoplasm.

BIOLOGICAL BULLETIN. VOL. LV.

PLATE I.

FRANCIS E. LLOYD AND J. BEATTIE.

OBSERVATIONS ON HYDRA AND PELMATOIIYDRA

UNDER DETERMINED HYDROGEN ION

CONCENTRATION.

W. L. THRELKELD AND S. R. HALL,
UNIVERSITY OF VIRGINIA.'

Much has been written recently concerning reduction, de-
differentiation and resorption in Hydra. It is generally conceded
that reduction in hydra is accompanied by a loss of tentacles.
The literature enumerates the following causes by which hydras
lose their tentacles. N. Annandale ('07) observed, in studying
Hydra orientalis, that during the hot season of the year this
species has but four tentacles while during the cold season it has
six tentacles. G. Entz ('12) observed that an infection with
Amoeba hydroxena may lead to a degeneration of tentacles.
Reynolds and Looper ('28) have come to the conclusion that this
parasite is responsible for the degeneration of the tentacles.
Certain ciliates recorded by E. Reukauf ('12) and P. Shultze ('13)
also caused the loss of tentacles. E. Shultz ('06) observed that
hunger set up a process of dedifferentiation within the tentacles.
Huxley and DeBeer ('23) observed that adverse environmental
conditions accelerate dedifferentiation and resorption of the
tentacles of Obelia and Campanularia . They also found that
this process of dedifferentiation and resorption might involve not
only the tentacles but also part of the zooid. Berninger ('10)
found that, in response to inanition, hydra lost its tentacles.
Finally Kepner and Jester ('27) also observed that the loss of
tentacles was brought about in response to inanition. This loss,
according to them, was accomplished by ingestion of the tips of
the tentacles through the mouth. This may occur, but undoubt-
edly is not the usual method, as Hyman ('28) indicated.

It is a well known fact that the concentration of the hydrogen
ion medium that bathes the protoplasm or protoplasmic tissue

1 These investigations were carried on under the direction of Professor W. A.
Kepner. Acknowledgments are due Mr. Carl H. McConnell of this laboratory, for
the preparations of the photomicrographs.

419

42O \V. L. THRELKELD AND S. R. HALL.

has a profound effect upon it, therefore it seems strange that no
attempts have been made to account for reduction, dedifferentia-
tion and resorption on the basis of such environmental conditions.
The following observations and results have been obtained
through an effort to determine whether or not the concentration
of the hydrogen ion is an important factor with reference to the
three above mentioned phenomena.

METHODS AND MATERIALS.

Filtered spring water in 300 cc. portions kept in thoroughly
cleansed glass dishes was used as a culture medium.

Very dilute solutions of N/2O sodium hydroxide and of hydro-
chloric acid were used in quantities to adjust the pH of the solu-
tions. The colorimeter method was used for the pH determina-
tion of the solutions and LaMotte color standards were employed
for color matching. Tests, adjustments and observations were
made every twenty-four hours except where otherwise indicated.
The temperature was maintained between 18 and 22° C. During
these investigations frequent examinations were made of both the
culture and of the animals for protozoa which might have been
responsible for reduction. None were found except where
stated. Observations were made with a dissecting binocular of a
magnification of twenty diameters. These observations were
supplemented by histological preparations.

At first distilled water was tried as a culture medium with the
idea that a more accurate determination could be made of the
hydrogen ion concentration. Various deleterious factors enter
into the use of such a medium so it was discarded. In the sub-
sequent experiments, filtered spring water was used.

The terms reduction, dedifferentiation and resorption, as used
by other authors and us, may be defined as follows: Reduction
is a uniform decrease in surface area in which process the ecto-
derm, mesoglea and endoderm remain intact and maintain a
normal position in relation to each other. Dedifferentiation and
resorption represent a dual phenomenon which involves a local
reduction of surface. The presence of this dual phenomenon in
the tentacles is indicated by a thickening and knobbed appearance
at the tips of the tentacles.

OBSERVATIONS ON HYDRA AND PELMATOHYDRA. 421

EXPERIMENTAL.

Culture i. — Four Pelmatohydra oligactis (Pallas), were placed
in a culture medium consisting of distilled water and NaOH was
added to maintain a constant pH of 7.8. At the end of a period
of six days there was much apparent reduction and resorption
of the tentacles in all specimens. One polyp was fed on the sixth
day and one on the seventh. At this point the experiment was
terminated through an accident.

Culture 2. — Four Pelmatohydra oligactis were placed in a
culture medium of distilled water. The culture maintained a
pH of 6.8 without the addition of either hydrogen or hydroxyl
ions. These polyps disintegrated in five days.

Culture j. — Four Pelmatohydra oligactis were placed in a
culture medium consisting of distilled water. This culture
maintained a pH of 7.0 which was fatal to the polyps in five days.
At this phase of our observations we came to the conclusion that
we were imposing other factors than the controlled pH repre-
sented, upon the hydras in using distilled water. A change in
osmotic pressure was undoubtedly involved when distilled water
was used instead of spring water. So, from this point on, spring
water was employed as the medium in which to keep the observed
polyps.

Culture 4. — Four Chlorohydra viridissima (Pallas) were taken
from spring water which tested pH 7.6. They were normal in
every respect. The pH of the second lot of spring water was now
maintained at 6.6. The only change being made here was using
a second glass dish similar to the one in which the pH tested 7.6
and in the pH now being 6.6. In five days, six of the polyps had
disintegrated and the remaining one had undergone advanced
dedifferentiation and resorption. It was placed in filtered spring
water of pH 8.6 in an effort to bring about regeneration but it
disintegrated in a few hours. This result, together with general
observations made on various cultures, in the laboratory, in
which the polyps displayed marked dedifferentiation and re-
sorption, indicates that the acid condition of the medium in-
duces dedifferentiation and resorption. Our observation upon a
lower hydrogen ion concentration (higher pH) proved to be little
more instructive as seen by the following culture.

422 W. L. THRELKELD AND S. R. HALL.

Culture 5- Six Pelmatogydra oligactis were isolated in filtered
spring water the pH of which was maintained between 7.6 and
8.2. On the 8th day all of the hydras appeared perfectly normal;
however, on the 9th day, all except one had disintegrated. The
one remaining hydra showed no apparent reduction or de-
differentiation and resorption of the tentacles. This hydra was
sectioned and its histology appears later in the paper.

On several occasions similar results were obtained when the
pH was held within the range from pH 7.8-8.0. It appears that
the first ten days represent a critical period when the polyps are
exposed to inanition. After the loth day has passed we have
had uniform results as the following observations indicate.

Culture 6. — Four Chlorohydra viridissima, in which some re-
sorption was displayed, were isolated in filtered spring water
pH 6.6. This water was over Elodea which had been previously
boiled. The Elodea was separated from the polyps by a double
thickness of cheese-cloth spread over the bottom of the container.
The Elodea was removed after six days and spring water alone was
used. As indicated above, these hydras were in a somewhat re-
sorbed condition. The pH of this culture was varied, first de-
creasing the concentration of the hydrogen ions after the first
two days up to 7.6, then increasing to 7.0, then again decreasing
to 7.8. A pH of 7.8 was maintained for the last thirteen days.
Immediately following these changes in pH, we observed the
physiological aspect of the polyps. It was seen that the greater
the concentration of the hydrogen ions the greater was the degree
of dedifferentiation and resorption in the polyps. If the concen-
tration of the hydrogen ions was lessened the hydras returned to
normal. Two of the four hydras survived for a period of twenty-
three days. One of these was sectioned (its histology is referred
to later in the paper) and the other was lost during a transfer for
examination. On the nineteenth day a green hydra, with much
resorbed tentacles and bearing gonads, was introduced into this
culture. In two days this hydra had gained its normal appear-
ance but its gonads had partially disappeared. It was fed and
placed in an aquarium containing food where it developed into a
fine vegetative specimen apparently normal. In this last
specimen the change from laboratory culture water to filtered
spring water must have been a factor as well as the change in pH.

OBSERVATIONS ON HYDRA AND I'KI.M A I ( )ll VDK A . 423

This does not however lessen the significance of the reaction of the
other individuals of culture 6, wherein only the pH concentration
has been the factor involved.

Culture 7. — Six Chlorohydra viridissima in a slightly resorbed
condition were placed in filtered spring water without Elodea
the pH of which tested 8.6. After the first two days the pH was
maintained at 7.8 until this experiment was terminated. On the
fourteenth day one hydra was sectioned. At the end of a period
of twenty-four days three hydras remained. They were much
reduced in size but their tentacles were apparently normal.
On the twenty fifth day they were placed in an aquarium con-
taining food where they lived for several days and attained nearly
normal size. At this point our observations on these animals
ceased.

These most interesting cases (cultures 6 and 7), in which the
polyps that had been reduced and in which apparent dedifferentia-
tion and resorption had taken place at a hydrogen ion concentra-
tion above the optimum, were restored to a completely normal
condition when subjected to hydrogen ion concentration at or
near the optimum. This undoubtedly indicates that food is not
necessary for the regeneration of hydra, but regeneration depends
rather upon the hydrogen ion concentration of the culture water.
Kepner and Jester ('27) record one hydra which had lost all <>t
its tentacles and without the presence of food the lost tentacles
were replaced by regenerated ones in eight days. As the culture
medium was frequently changed it is probable that a favorable
pH was accidentally maintained. Hyman ('28) records the
same phenomena when she says: "Depressed specimens may be
caused to regenerate if the water is replaced by culture water'
(page 78). Huxley and DeBeer in working with Obelia and
Campanularia were unable to cause the regeneration of dedifferen-
tiated and resorbed tissue.

Culture 8. — Eight Pelmatohydra oligactis were isolated in
filtered spring water the pH of which was maintained for the
first two days at 8.4 and for the remainder of the period it \\.i-
kept at pH 7.8. On the tenth day three hydras had completely
disintegrated without displaying reduction, dedifferentiation
and resorption. On the i/th day, Halteria appeared in the
culture. These were not abundant, about ten being found in the

424 W. L. THRELKELD AND S. R. HALL.

field of the binocular dissecting microscope. As all the hydras
appeared in the same condition one was sectioned. These sec-
tions showed no Halteria present within coelenteron or the food
vacuoles. But menatocysts were present in the epithelio-
muscular cells of the endoderm and within the coelenteron, hence
the histology indicates that resorption had taken place. This
resorption was so slight that it is overlooked by examination of
the living polyps under a dissecting microscope. The culture
medium was changed, so as to have water free of protozoa, and
the observations continued. On the twenty third day one hydra
was sectioned (its histology is referred to later). On the twenty
fifth day the remaining three hydras were given bits of liver
which they readily accepted. Thus indicating that they were
not in a "depressed" condition as described by Hyman ('28).
They were placed in an aquarium containing food where they were
observed for several days. No indication of "depression"
became evident during these observations nor was there any
evidence of it at the time the observations ceased.

In order to determine wherein the optimum range of hydrogen
ion concentration for the medium lay, both green and brown
hydras were exposed to varying degree of hydrogen ion concentra-
tion ranging from pH 5.2-8.0 and the time recorded when all
hydras had disappeared in each culture. The result of this
experiment is given in the following table.

Four more cultures were run, with both green and brown
hydras, one with a pH of 7.8, the other at pH 8.0. All the
polyps in these cultures were alive at the end of a period of
twenty four days.

This indicates that the optimum hydrogen ion concentration
lies near pH 7.8. And further hydrogen ion concentration is an
important factor in the determination of dedifferentiation and
resorption; for, in the same medium (filtered spring water) with
only the concentration of hydrogen and hydroxyl ions altered, we
have been able to either induce or inhibit dedifferentiation and
resorption. This does not support the later part of Hyman ('28)
page 93, paragraph 2, Biological Bulletin volume LIV, January
1928, number I in her explanation of the phenomenon of depres-
sion when she says that "it is induced by transfer to clean fresh
water." It is quite evident that, if two different lots of hydra

OBSERVATIONS ON HYDRA AND I'KLMATOHYDRA.

425

TABLE I.

THE x MARK INDICATES THE DAY OF THE DEATH OF THE LAST HYDRA IN THE

CULTURE.

Existence in D;iys.

2

3

4

5

6

7

8

0

IO

1 1

12

13

X

Brown hydra in pH 5.2 ...

X

X

X

X
X
X
X
X

X
X
X
X

X
X
X
X

X
X
X
X

X

X

X

X

X

Green " " 5.2 ...

Brown " " 5.4

Green 5.4. . . .

Brown " " 5.6. ..

Green 5.6. . .

Brown " " 5.8. .

Green " 5.8 . .

Brown " 6.0. .

Green ' " 6.0

Brown ' "6.2

Green ' "6.2

Brown 6.4

Green 6.4

Brown ' ' 6.6

Green ' ' 6.6

Brown ' ' 6.8. . . .

Green ' ' 6.8

Brown 7.0 . .

Green 7.0

Brown 7.2

Green ' ' 7.2

Brown ' ' ' 7.4.

Green ' ' ' 7.4.

Brown 7.6

Green 7.6 . .

taken from the same culture or aquarium are placed in identical
spring water cultures save for the concentration of the hydrogen
ions and favorable reactions are repeatedly to be noted in the
culture of low hydrogen ion concentration while unfavorable
reactions are always to be noted in the culture of high hydrogen
ion concentration, undoubtedly the pH of the culture must be a
strong factor in determining this difference in the reaction-

HISTOLOGY.

The histology of reduction, dedifferentiation and resorption in
Hydra has been observed by E. Shultz ('06) and \Y. Rehm ('25).
Huxley and DeBeer ('23) also described histologically dediffer-
entiation and resorption in Obelia and Campanularia. Our obser-
vations are almost in exact agreement with those of the above.

Studies on reduction, dedifferentiation and resorption always

426 W. L. THRELKELD AND S. R. HALL.

involve the histology of the animal. Hydra is a diploblastic
animal having only an ectoderm and endoderm. The ectoderm
presents in its vegetative condition, epithelio-muscular, inter-
stitial, cnidoblastic and nerve cells. The endoderm, on the other
hand, is made up of epithelio-muscular, glandular, interstitial
and scattered or isolated nerve cells. In the ectoderm there is
no great local specialization or differentiation into regions.
The ectoderm, however, shows three distinct regions: (i) the
oral two-thirds in which there are scattered gland cells and a
general covering of epithelio-muscular cells that are heavily
charged with absorbed alimentary products; (2) a basal third
that has few if any gland cells and in which the epithelio-muscular
cells are usually highly vacuolated, except for those at the basal
disc and (3) the endoderm of the tentacles. In this third region
there are no gland cells and the epithelio-muscular cells are
highly vacuolated. Thus it appears that the endoderm of the
highly active or moving tentacles resembles that of the relatively
quiet basal third of the body proper.

Dedifferentiation and resorption have been referred to fre-
quently above. This has been denned as a dual phenomenon
which involves a local reduction of surface. We take the presence
of ectodermal elements (nematocysts being the most easily
recognized) within the coelenteron or endoderm as evidence that
dedifferentiation and resorption have taken place.

The question now remains: How is the surface reduced locally,
and how do ectodermal elements gain their entrance into the
coelenteron? As this phenomenon is most often seen in the
tentacles, we have studied it there. In response to adverse en-
vironmental conditions, the cells at the tips of the polyp's
tentacles coalesce or become dedifferentiated. The ectoderm is
apparently affected first. Here the dedifferentiated cells, pre-
paratory to resorption, group themselves into rounded or
spheroidal masses. (Fig. i-A.} Nematocysts as well as numer-
ous cell-fragments may be seen within these aggregates. Ob-
viously there must be some change in the non-living mesoglea as
well as the living endoderm before resorption of the modified
ectoderm can proceed. Dedifferentiation, therefore, starts in the
endoderm. These cells, apparently, break away from the walls
of the tentacles and soon assume a globular form (Fig. I, B).

OBSERVATIONS ON HYDRA AND PELMATOH YDKA. 427

They migrate down the lumen of the tentacle (Fig. i, 5).
Now the mesoglea breaks or is resorbed (Fig. i, C) and the endo-
dermal elements apparently have little trouble in finding their
way to the coelenteron. The cellular masses of ectoderm,
spheroidal in shape and often with contained nematocysts,
together with the above mentioned dedifferentiated endo-epi-
thelial masses, may be found in the coelenteron as far down as
the basal disc. Thus the surface of the tentacle is decreased.
To use the language of Huxley and DeBeer ('23) in describing a
similar phenomenon in Obelia and Campanularia, "The ecto-
dermal cells may be compared with that of a rear guard, retreat-
ing yet always maintaining an unbroken front." These histo-
logical details serve as a final criterion for determining whether
dedifferentiation and resorption have taken place. But with the
aid of low magnification, one can see that, as resorption proceeds,
the tips of the tentacles increase in diameter, and finally appear
knobbed and the involved area becomes darker and darker,
The endodermal cells lining the tentacles are normally highly
vacuolated. These cells, however, appropriate relatively much
food during the later stages of resorption.

It is certain that this dedifferentiated and resorbed tissue is
used as food by the animal because nematocysts in various stages
of digestion may be found in the epithelio-muscular cells in all
parts of the endoderm. This confirms Kepner and Jester ('23)
in their minor claim that the ingested parts were used as food;
but Kepner and Jester were misled by the occasional biting off of
the tentacles. Dedifferentiation and resorption are the usual
reaction.

Since it was seen that both the cells of the ectoderm and the
endoderm of the tentacles were almost exactly like those of the
lateral walls of the basal one third of hydra, dedifferentiation and
resorption was looked for in this basal region. It was found to
occur in the case of the sectioned hydra recorded in culture
numbers (Fig. 2, A). No explanation is offered for deditfrivntia-
tion and resorption being found in the basal disc in this and no
other case. It was noticed, however, that in this case resorption
was not found in the tentacles. Resorption has not been re-
ported before as occurring in the basal region prior to its inception

in the tentacles and peristome. All other writers state that it

28

428 W. L. THRELKELD AND S. R. HALL.

starts at the tentacles and proceeds towards the base. The
peristome is affected, according to them, after the tentacles have
been removed. But this specimen showed dedifferentiation only
in the basal region.

Green hydra reported in culture number 6 which was carried
twenty-three days without food, showed histologically only slight
resorption.

Rehm ('25) says that at the end of twenty one days the body of
hydra subjected to inanition was reduced to a mere rounded
form, which he calls, following Will and other investigators,
" Reductionskorper " (§371). At other places he refers to these
rounded hydras as presenting planula-like pictures ("planula-
ahnliches Gebilde, der Reductionskorper") (§382). We have
carried brown hydra for twenty three days within the optimum
hydrogen ion concentration. This polyp showed so little de-
differentiation and resorption that they could only be detected
histologically. Under low magnification the living polyp, though
reduced in size, appeared to be complete and have no broken
surface. The brown hydra, as recorded in culture number 8,
which was sectioned after sixteen days of inanition within the
optimum hydrogen ion concentration, presented, while living,
no evidence of dedifferentiation and resorption under low magni-
fication. However, the histology of this animal shows frequent
nematocysts in the coelenteron hence slight dedifferentiation and
resorption must have taken place during the seventeen days of
inanition. Examination on this day under the dissecting micro-
scope disclosed no difference in appearance between the remaining
hydras and the one sectioned. On the twenty third day another
hydra from this culture was sectioned. From the histology of
this polyp, it is seen that dedifferentiation and resorption which
were shown in the histological examination of the hydra sectioned
on the iyth day not only has ceased but the resorbed tissue has
been digested by the polyp sectioned after twenty three days of
inanition within the optimumrange of hydrogen ion concentration.
Similar phenomena have been observed for green hydras. For a
green hydra, which had suffered 14 days of inanition at optimum
hydrogen ion concentration showed slight dedifferentiation and
resorption; while a second green polyp, from the same cultuie
sectioned after twenty three days of inanition at optimum hydro-

OBSERVATIONS OX HYDRA AND I'KI.MATOH YDKA. 429

gen ion concentration, showed no evidence of dedifferentiation
and resorption.

Thus it appears that during inanition at optimum hydrogen ion
concentration a crisis is reached after about two weeks. During
this crisis slight dedifferentiation and resorption make their
appearance. The resorbed material may supply sufficient
nourishment to tide the polyp, now reduced in size, through a
long period before a second crisis develops and compells the de-
differentiation and resorption of more tissue.

SUMMARY.

1. The optimum range of hydrogen ion concentration for both
Hydra viridissima and Pelmatohydra oligactis lies within the range
pH 7.8 and 8.0.

2. Polyps allowed to develop pronounced dedifferentiation and
resorption in a high hydrogen ion concentration (low pH) were
induced to completely restore their lost parts when the medium
was altered to be within the optimum range of pH.

3. Hydras carried within the optimum range of pH were sub-
jected to periods of inanition as great as twenty five days without
showing any external evidence of dedifferentiation and resorption
at the end of this period.

4. Histological preparation of polyps, kept for long periods
without food at the optimum hydrogen ion concentration,
show slight evidence histologically of dedifferentiation and re-
sorption at a critical period. This critical period appears some-
where between ten and seventeen days after inanition within the
optimum range of pH. Such microscopic dedifferentiation and
resorption are not progressive; for after this critical period has
passed no further histological evidence of dedifferentiation and
resorption has been observed.

(b) This microscopic dedifferentiation and resorption usually
appear at the tips of the tentacles; but in one case we have seen
it involve the basal third of the polyp and not the tentacles.

5. Hydras subjected to long periods of inanition within the
optimum range of pH accept food readily. There is, therefore,
no evidence of depression given by these polyps.

6. Dedifferentiation and resorption are induced rather by un-
favorable hydrogen ion concentration than by inanition.

430 W. L. THRELKELD AND S. R. HALL.

LITERATURE.
Annandale, N.

'07 Seasonal Variations in Hydra orientalis. Jour, and Proc. Asiatic Soc.

Bengal, N. S., 111.
Berninger, von Julius.

'10 Uber Einwirkung des Hungers auf Hydra. Zool. Anz., Bd. 36.
Entz, G.

'12 Uber eine neue Amobe auf Susswasser — Polpen (Hydra oligactis Poll).

Arch. Protistenk., Bd. 27.
Huxley, J. S. and G. R. DeBeer.

'23 Studies in Dedifferentiation. IV. Resorption and Differential Inhibition in

Obelia and Campanularia. Quart. J. Mic. Soc., Vol. 67.
Hyman, Libbie H.

'28 Miscellaneous Observations on Hydra, with Special Reference to Repro-
duction. BIOL. BULL., Vol. LIV.
Kepner, W. A. and Jester, P. N.

'27 The Reaction of Hydra to Inanition. BIOL. BULL., Vol. 52, pp. 173-84.
Kepner, W. A., and Miller.

'28 A New Histological Region in Hydra oligactis. BIOL. BULL., Vol. 54.
Marshall, Shema.
'23 Observations on the Behavior and Structure of Hydra. Quart. I. Mic. Soc.,

Vol. 67.
Rehm, W.

'25 Uber Depression and Reduktion bei Hydra. Zeitsch. f. Morphol. u. Okol.

d. Thiere, Vol. 3, pp. 358-88.
Reukauf, E.

'12 Selbstumstulpung and Armanputation durch ein Wimperinfusor (Prorodon

teres) bei Hydra Fusca. Zool. Anz., Bd. 39.
Reynolds, B. D., and Looper.

'28 Infection Experiments with Hydramaeba hydroxena (nov. gen.) Jour, of

Parasitology. Vol. XV pp. 23-31.
Schutz, E.

'06 Ueber Reductionen. II. Ueber Hungerersheinumgen bei Hydra Fusca.

Arch. Entw. Mech., Bd. 21.
Schutze, Paul.

'13 Hypertrophic der Tentakeln von Hydra oligactis Poll, infolge massenhaften
Befalls mit Kerona Pediculus O. F. M. Zool. Anz., Bd. 42.

432 W. L. THRELKELD AND S. R. HALL.

PLATE I.
Explanation of Figures.

FIG. i. Longitudinal section of the free end of a tentacle of Pelmatohydra
oligactis which had been starved twenty-four hours in spring water at pH 6.8. This
shows the inception of dedifferentiation and resorption. The mesoglea has broken
down at end of tentacle. Rounded masses of coalesced ectodermal cells are forming
(A). Similar rounded masses of coalesced endodermal cells are forming (B);
at B' we see a mass of coalesced endodermal cells having migrated towards the
lumen of the tentacle; at C a mass of coalesced ectodermal cells is passing through
the region of the broken down mesoglea. X70O.

FIG. 2. A longitudinal section involving a part of the basal disc of Pelmato-
hydra oligactis. (Culture number 5.) This specimen had been starved nine days
within optimum hydrogen ion concentration. The inception of dedifferentiation
and resorption is shown at A; BGC, basal disc glands cells; E, endodermal cells;
L, lateral ectodermal cells. X70O.

BIOLOGICAL BULLETIN, VOL. LV

PLATE I.

W. L. THRELKELD AND S. R. HALL.

THE OCCURRENCE OF NUCLEAR VARIATIONS IN
PLEUROTRICHA LANCEOLATA (STEIN).

REGINALD D. MANWELL.i
SCHOOL OF HYGIENE AND PUBLIC HEALTH, JOHNS HOPKINS UNIVERSITY.

The occurrence of variations from the accepted type among the
protozoa has received much attention in recent years, and a
number of such cases have been reported, both of the artificially
induced and spontaneously appearing sort. Most of the former
have been of the "enduring modification" type, that is they per-
sist throughout a longer or shorter period of vegetative division,
but are eventually lost when conjugation or endomixis takes
place. The latter may be divided into two classes. The first
group would include the true mutations, of which the tetraploid
Chilodon described by MacDougall (1925) is probably one of the
best authenticated examples. In this case the mutation, which
consisted in the possession of twice the usual number of chromo-
somes, combined with unusual size and certain other minor char-
acteristics, persisted through both conjugation and division.
To the second group would belong all other departures from
normal, such as the production of monsters, the amicronucleate
.condition in infusoria, and various other unusual physiological
and morphological characters which persist through division
but tend to revert to normality eventually. Examples of this
kind of variation are quite numerous. Among them may be
mentioned the amicronucleate Oxytricha studied by Dawson
(1919), the race of Paramecium which possessed extra contractile
vacuoles (Hance, 1917), the rapidly-dividing race of Didinium
reported by Mast (1917), and the sudden appearance of an
Arcella having double characteristics described by Reynolds
(1923). Since the latter investigator found that these abnormal
characteristics could be diminished until a completely normal
condition was reestablished, or increased by selection of suitable

1 From the Department of Protozoology, School of Hygiene and Public Health,
Johns Hopkins University, and the Marine Biological Laboratory, Woods Hole,
Mass.

433

434 REGINALD D. MANWELL.

individuals this last variation evidently belongs with those found
by Jennings (1920) and Root (1918) to exist in Difflugia and
Centropyxis, with this difference, however — the former occurred
suddenly, while the latter were of lesser degree and appeared more
gradually. More recently Dawson (1924) has reported the oc-
currence of a peculiar form of Paramecium aurelia which has been
carried in culture for several years since. The abnormal char-
acter in this case consists of a "notched" condition which is
definitely heritable, at least in ordinary asexual division.

The present paper deals with variations in the number of both
micro- and macronuclei in Plcurotricha lanceolata. Pedigreed
cultures of this ciliate, which is a hypotrich belonging to the
family Oxytrichidae, were maintained for 18 months and studied
mainly from the standpoint of the cytological changes occurring
during conjugation and division, as described in a previous paper
(Manwell, 1928).

The normal animal is shown in Fig. I. It will be noted that
it possesses two nuclei of each sort, and according to Stein (1858)
who first described both the species and genus, the presence of
two macro- and two micronuclei is a generic character. About
two months before the culture was discontinued however, and
while to all appearances it was in a very vigorous condition with
division taking place very actively, individuals possessing only one
macronucleus were noticed in some of the stained preparations.
The micronuclear condition varied ; in some cases there was only
one and in others there were two as in normal individuals.
Animals possessing the normal macronuclear complex but with
three micronuclei have also been observed, and such changes are
indeed not very uncommon, not only in Pleurotricha but in
Oxytricha and other ciliates containing more than one micro-
nucleus. But no individuals have been observed with only one
macronucleus and more than two micronuclei. Fig. 2 shows an
individual possessing but one nucleus of each sort in division, and
in Fig. 3 a similar individual, differing only in having two micro-
nuclei, may also be seen dividing. The next two figures show
later stages in the division of such individuals, and in Fig. 6 a
unimacro- and micronucleate animal is shown just after division.

From these figures it can be seen that division takes place in
exactly the same way as it does in individuals having the normal

NUCLEAR VARIATIONS IX PLKTROTRKTIA I.AN< 'KOI .ATA . 435

nuclear complex, and that the variations are heritable, at least in
ordinary vegetative fission. To settle this point still more
definitely several lines were started from individuals possessing
hut one nucleus of each sort and followed for 10 days. At the
end of that time these subcultures were lost by accident and other
circumstances made it necessary to conclude the experiment, but
stained preparations made from each generation showed clearly
that the reduced number of nuclei was being passed from one
generation to the next.

A careful examination of stained preparations has been made
in an effort to discover whether the abnormal nuclear complex
was accompanied by any other morphological changes, but ap-
parently there were none. During the early stages of division
however (about the stage shown in Fig. 2) it was frequently pos-
sible to distinguish animals possessing but one macronucleus from
normal individuals in the same culture in a similar stage, for the
bodies of the former were definitely broader about 1/3 of the way
back from the anterior end and then tended to become narrower,
while in the normal animals the entire middle third of the body
was of a fairly uniform width. If there were any differences in
size they were in favor of those individuals possessing but one
nucleus of each sort.

No evidences of conjugation among these abnormal individuals
was ever observed, but since as previously reported, conjugation
occurred but rarely in all the cultures from start to finish of the
experiment, not much stress can be laid on this point. Encyst-
ment was also not observed. Consequently it cannot be said
whether such a variation as this would survive endomixis and
conjugation, although it seems probable that in some cases at
least, unimicro- and macronucleate conjugants might produce
similar individuals.

In view of the work of Baitsell (1914), and the fact that conju-
gation in this species has been shown to result, at least when it oc-
curs under cultural conditions favorable to vegetative division, in
almost 100 per cent, mortality (Manwell, 1928) the question of
the occurrence of such morphological variations as herein described
becomes of some practical importance. For obviously, if under
favorable conditions multiplication by fission can continue in-
definitely, then such changes might be perpetuated for a very

436 REGINALD D. MANWELL.

long time in nature, as well as in artificial cultures. And if this
is so account should be taken of the fact in the description of
genus and species, since the number of nuclei, especially of the
macronuclei, is a conspicuous character. If asexual reproduction
can continue indefinitely then the sudden appearance of changes
of the kind described would, for practical purposes, have the
value of a mutation.

The occurrence of abnormal micronuclear conditions has been
reported a number of times before, particularly with respect to
the total absence of a micronucleus, and the presence of one or
two supernumerary micronuclei is not very uncommon in
species ordinarily possessing two or more, as already noted, but
apparently the number of macronuclei is a much more constant
character. The only instance in which a variation in the latter
has been reported, to the author's knowledge, at least, is that
given by Calkins (1926). Here he states (p. 579) that in early
cultures of Uroleptus mobilis the number of macronuclei was al-
most uniformly 8, but as the age of the cultures increased indi-
viduals with a greater number of nuclei became common, until
finally the number was nearly always 14 or 15.

SUMMARY AND CONCLUSIONS.

In a pedigreed culture of Pleurotricha lanceolata, a species o,
hypotrich normally possessing two macro- and two micronuclei
individuals with only one macronucleus and one or two micro-
nuclei suddenly appeared, at a time when division was rapid and
the culture apparently very vigorous.

That the difference in nuclear number was heritable, at least in
asexual multiplication, was shown from stained preparations and
pedigreed lines, and the fact that it has been shown that this
species will live and divide normally apparently indefinitely under
favorable conditions, without conjugation, makes it probable
that such variations as have been described would continue for a
very long time, and that animals with such peculiarities may be
common in nature as distinct varieties.

BIBLIOGRAPHY.
Baitsell, G. A.

'14 Experiments on the Reproduction of the Hypotrichous Infusoria. II. A
Study of the So-called Life Cycle of Oxytricha fallax and Pleurotricha
lanceolata. Jour. Exp. Zool., Vol. 13, pp. 211-234.

NUCLEAR VARIATIONS IN PLKUROTRICII A I.ANCKOLATA. 437

Calkins, G. N.

'26 Biology of the Protozoa. Philadelphia. 623 pp.
Dawson, J. A.

'19 An Experimental Study of an Amicronucleate Oxytricha. I. Study of the
Normal Animal with an Account of Cannibalism. Jour. Exp. Zool., Vol. 29,

No. 3. PP- 473-513-
'24 Inheritance of Abnormality of Form in Paramecium aurelia. Proc. Soc.

Exp. Biol. and Med., Vol. 22, pp. 104-106.
Hance, R. T.

'17 Studies on a Race of Paramecium Possessing Extra Contractile Vacuoles.

Jour. Exp. Zool., Vol. 23, No. 2, pp. 287-327.
Jennings, H. S.

'20 Life, Death, Heredity and Evolution in the Protozoa. Boston, 233 pp.
MacDougall, M. S.

'25 Cytological Observarions on Gymnostomatous Ciliata with a Description of
the Maturation Phenomena in Diploid and Tetraploid forms of Chilodon
uncinatus. Quart. Jour. Mic. Sci., Vol. 69 (new series), Pt. 3. PP- 361-384-
Manwell, R. D.

'28 Conjugation, Division and Encystment in Pleurolricha lanceolata. BIOL.

BULL., Vol. 54, No. 5, May, pp. 417-463.
Mast, S. O.

'17 Mutation in Didinium nasutum. Amer. Natur., Vol. 51, pp. 35i-3°o.
Reynolds, B. D.

'23 Inheritance of Double Characteristics in Arcella polypora. Genet., Vol. 8,

pp. 477-493-
Root, F. M.

'18 Inheritance in Asexual Reproduction of Centropyxis aculeata. Genet.,

Vol. 3, pp. 173-199-
Stein, F. R.

'59 Der Organismus der Infusionsthiere. Leipzig. Abdruck i, 206 pp.

438 REGINALD D. MANWKLL.

EXPLANATION OF THE FIGURES.

Magnification X 550; all drawings made with camera lucida

PLATE I.

FIG. i. A typical vegetative individual.

FIG. 2. An individual with one macronucleus and one micronucleus in a
moderately early stage of division.

FIG. 3. A division stage similar to the above in an animal having two micro-
nuclei, but only one macronucleus.

FIG. 4. A more advanced stage in an individual similar to the above.

FIG. 5. The final stage of division in a unimicro- and macro-nucleate individual.

FIG. 6. A daughter individual just after fission.

BIOLOGICAL BULLETIN, VOL. LV.

PLATE I.

•£5 //

_; [ K "''v' .— '
•-?'•> -^-

Op-< i-

^

REGINALD D. MAMWELL

OBSERVATIONS ON THE LIEE HISTORY AND

PHYSIOLOGICAL CONDITION OF THE

PACIFIC DOG FISH

(SQUALUS SUCKLII).

J. P. QflGLEY.1

Incidental to an investigation of the reactions oi .SV/
sucklii to variations in the salinity of the surrounding medium (i)
observations were made regarding the life history and physio-
logical condition of this fish.

The fish were captured during the months of June, July and
August of 1926 from the Straits of Georgia in the vicinity of
Departure Bay, Vancouver Island, B. C. They were taken on a
set line, the hooks of which were baited with pieces of salted
herring. Most of the fish were obtained at a depth of about 30
meters, and they were generally caught near kelp beds. A
sample of water taken at a depth of 30 meters in the region where
many of the fish were taken was found by Lucas (2) to have tin-
following characteristics; pH 8.4, temperature 10.3° C., density
1. 0218, oxygen content 4.41 cc. per liter, sodium chloride content
27.37 gm- Per liter.

Weight of Fish. — It was found that many of the factor^ as-
sociated with the weight of the fish could be emphasized by
grouping the fish according to weight as has been done in Table I.
Examination of this table shows that with the fish of lighter
weight the two sexes are nearly equally represented, the number
of males being slightly greater. As heavier fish are considered,
the relative number of males shows a marked iiiriva-r, then a
sudden decrease so that in the weight divisions above 4,000 gram>
the males are entirely absent.

These results probably indicate that male fish with body weight
over 4,000 grams do not exist in this locality during the Bummer.
It cannot be definitely stated that the figures obtained with li^h
of lighter weight indicate the relative proportion in which tin-

1 From The Pacific Biological Station, Nanaimo, H. C., and Tin- Department »i
Physiology and Pharmacology, Fnivcrsity of Alberta, Kdmonton, Alberta.

439
29

44<>

J. P. QUIGLEY.

TABLE I.

\\Vinht

Limit -

Number

<it Fish
i il.l.iiunl.

Number of

Percentage.

Average
Length
(Cm.).

Average
Increase
in Length.

M.il' 3.

Females.

Males.

Females.

300 ,i<;9- . .

12

7

5

58

42

39-9

joo [99

15

9

6

60

40

43-6

3-8

500 500

[6

ii

5

69

31

45<7

2.1

ooo 009. . .

13

8

5

62

38

48.4

2.7

0 709- - •

5

5

o

100

0

52.5

4.1

,Xoo S<;o. .

7

4

3

57

43

53-8

i-3

000-999. . .

5

5

0

IOO

0

54-7

0.9

i .DOM i ,499 .

22

20

2

91

9

60.3

5-6

1.500 i. <><><)

I I

9

2

82

18

69.2

8.'9

2, OOO-2, <)<i<j

30

26

4

87

13

74-9

5-7

3,000-3,999

13

5

8

38

62

83.3

8.4

4.OOO \.<)<>'i

10

o

16

o

IOO

90.5

6.2

5,000 s,999

c6

o

16

0

IOO

91.6

i.i

6,000-6,999 .

4

o

4

0

IOO

95-5

3-9

7.000-7,999 .

i

o

I

o

IOO

99.0

3-5

i\v<> sexes occur, although such probably is the case. Since the
lish were taken on a set line hunger or greed might conceivably be
a factor in determining whether or not fish would take the bait.
The stomach of fish captured usually contained much food, a
l.ict which indicates that feeding for this fish is determined more
by the availability of food than by hunger.

Out of 219 fish captured, 128 (58 per cent.) were males. Craigie
(3) examined the fish obtained in the same region during July
and August, 1925, and found that among 76 specimens 44 (60
per cent.) were males, while during December of 1925 by examin-
ing 1 17 specimens he found 47 (40 per cent.) males.

As was to have been expected, there is a comparatively definite
relationship between weight and length of fish. The increase in
length is rather steady though not entirely uniform as heavier fish
are compared with those of lighter weight. It could not be shown
thai sex altered the relation of weight and length. There was a
slight though inconstant indication that nonpregnant females
were longer than pregnant females of the same weight. The
longest fish captured measured 99 cm., the shortest 35.5. The
lu-a\ -icst Ii-h weighed 7,550 grams and the lightest 300 grams.
When increasing their weight 100 grams the smaller fish made an
increase in length of approximately the same magnitude as did
the larger fish when making a weight increase of 1,000 grams.

LIFE HISTORY OF PACIFIC DOG FISH. 441

Pregnancy and Embryos. — Of the females captured, 43 per cent,
carried embryos large enough to be readily noted in a cursory
inspection. The lightest fish having embryos weighed 3,440
grams and was 85 cm. in length. These figures give an approx-
imate minimum limit of the size of the mature female. Among
the 50 females captured with a weight equal to or above 3,440
grams, 39 (78 per cent.) carried embryos.

Ford (4) quotes the conclusion of several investigators that
Squalns acanthias breeds throughout the year and of other
investigators that this species breeds only during certain periods.
The results of his own investigations support the latter conclusion
and tend to show that near Plymouth, England, specimens ready
for birth would not be found earlier than the end of August.
I found specimens of Squalus sucklii embryos at all times during
the summer which ranged through all the sizes from the smallest
to those with the umbilical scar healed completely and apparently
ready for birth. This observation naturally suggests that in the
vicinity of Nanaimo, Squalus sucklii breeds at all times of the
year.

In any one parent, the embryos were of the same general size.
A set of developing eggs was always found in females carrying
embryos. The number of embryos obtained from 16 fish varied
between 3 and II with an average number of 6.87. Although it
could not be definitely stated that none of the embryos had been
lost from the mother in the course of capture it is believed that this
was a rare occurrence. No embryos were lost after the mother
was taken from the set line and in most cases egg capsules still
unruptured wrere obtained. In an examination of Squalus acan-
thias Ford (4) found that females of this species could carry as
many as n embryos but the greatest number of pregnant fish
carried only 3. In Squalus sucklii I found that embryos of both
sexes usually occurred in the same uterus but there was no
relation between the number of either sex, e.g. in one fish I found
6 females and I male, in another 3 males and no females. Of the
embryos obtained 50 per cent, were males. This figure is to be
contrasted with that previously noted for the fish of small size
taken on the set line where a preponderence of males existed.
A blue shark, Prionace glance, (identified by Professor J. R.
Dymond) received at the Pacific Biological Station, August 19,

442 J- P- QUIGLEY.

1926, was found to have 11 females and 8 male embryos all the
same size nearly ready for birth.

Constitution of Shoals. — Throughout the period fish were being
taken, the specimens obtained on any set line usually consisted of
both sexes in approximately equal numbers and of all sizes.
The conclusion was reached that the shoals consisted of both sexes
and all sizes of fish or else the line had been visited within a few
hours by several different shoals. It was also noted that the
largest fish were usually taken at a greater depth (very near or
actually on the sea bottom) than the smallest and it may be that
the composition of shoals is in part determined by size. From
his study of Squalus acanthias, Ford (4) concluded that for this
species the mature males and females each form separate shoals
while these shoals in turn are distinct from those composed of
immature males and females together. I obtained fish in the
same region throughout the summer. It is therefore likely that
certain shoals inhabit this region during the entire season.

SUMMARY.

1 . Among the smaller fish males were slightly more prevalent
than females. Males weighing more than 4,000 grams were not
obtained. Females attain a much greater length and weight than
males. The greater weight of the females was not always due to
the presence of eggs or embryos.

2. A comparatively definite relationship exists between weight
and length of fish. The relationship of length increase to weight
increase for small fish is approximately ten times as great as for
large specimens.

3. Of the mature females captured 78 per cent, carried embryos.
This species apparently breeds throughout the year. The average
number of embryos carried by the females is greater than six.

4. The shoals apparently consist of fish of all sizes and of both
sexes. The shoals probably remain in the same region throughout
the summer.

REFERENCES.

1. Quigley, J. P.

'28 BIOL. BULL. LIV., 165.

2. Lucas, C. C.

Personal communication.

3. Craigie, E. H.

'27 Contrib. to Canadian Biol. and Fisheries, N. S., Ill, No. 22, 491.

4. Ford, E.

'21 Jour. Marine Biol. Assoc. of the United Kingdom, N. S., XXII. , 468.

ALG^E OF PONDS AS DETERMINED BY AN

EXAMINATION OF THE INTESTINAL

CONTENTS OF TADPOLES.

VIVIAN FARLOWE,
UNIVERSITY OF VIRGINIA.

INTRODUCTION.

During the last few years a considerable amount of research
has centered around the food taking of small fresh-water fish.
This work has emphasized the dependence of small fish on algae
and in turn these fish as a source of food for the game fish. In
reviewing literature the writer has found comparatively little
scientific work on the feeding habits of the tadpole and frog.

The tadpole as well as the small fish is an indirect source of
food for the human race. Tiffany ('22) states: "For most of the
young fishes examined the complete story reads: 'no phyto-
plankton, no gizzard shad.' ' It may also be said, no algae, no
tadpole.

The writer wishes to express her gratitude to Dr. Bruce D.
Reynolds, who suggested this problem and who has greatly as-
sisted by his advice and criticism in the preparation of this paper;
also to Professor I. F. Lewis and Dr. E. M. Betts for helpful

criticisms.

METHODS.

During the summers of 1927 and 1928 one hundred tadpoles
and one hundred pond collections were taken from five ponds on
the campus of the University of Virginia and in the surrounding
vicinity. Two of the ponds measured approximately 250 ft. x 100
ft., one 150 ft. x 50 ft., one 100 ft. x 30 ft., and one 50 ft. x 20 ft.
The ponds which were studied did not have active outlets.

Two examinations of each of these ponds were made during the
summer of 1927 from July 15 to August 28, and two were made
during the summer of 1928 from June 20 to July 5. Each collec-
tion from a pond consisted of five tadpoles 1 which measured from

1 Of the 100 tadpoles used in these experiments, 94 were Rana clamilans and 6 R.
catesbeiana.

443

444 VIVIAN FARLOWE.

one and three-fourths inches to five inches long and five collections
of sediment taken from the edges of the ponds. The tadpoles
and pond collections were put in separate containers. Im-
mediately after returning to the laboratory the tadpoles were
killed and the intestines removed. Three slides were made of
material taken from each digestive tract, one from the anterior
and one from the middle regions of the small intestine, the third
from the anterior region of the large intestine. A study of each
of the slides was made under the high power of the microscope.
The algae from each region were identified and recorded. The
pond collections were studied in a similar way. Three slides
were made from each of the pond collections. The algae from
each slide were identified and recorded.

During the summer of '27 the tadpoles were collected from the
pond, and then the pond collections were made without any
effort to correlate the position of the tadpole and the pond collec-
tion, but in the collections made during the summer of '28 a tad-
pole was caught and from the same place a pond collection was

made.

THE PROBLEM.

The experiments presented in this paper were not undertaken
primarily for the purpose of studying the food of tadpoles, but
rather in order to ascertain if the algae found in the alimentary
tract of tadpoles can be relied upon as an index to the micro-
scopic flora of the ponds in which the tadpoles are living. In
other words, does the tadpole feed on different kinds of algae or is
it selective in its feeding habits? If not selective, is it as good a
collector of algae as the investigator interested in studying them?

EXPERIMENTAL.

In following up this problem observations were made on four
collections, made at different times, from each of five ponds.
The results obtained are shown in tabular form.

By referring to Table I. it will be seen that the number of
species of algae obtained from the intestine of the tadpoles ex-
ceeded the number obtained from the pond collections in every
case except two, and in these instances they were the same — -the
pond collections being made where the tadpoles were caught.

Attention is also called to the relative number of algae found in

ALGJE OF PONDS.

445

the intestines of tadpoles and the ponds from which they were
taken, in large and small ponds (Table I.). It is evident that,
when making collections from small ponds, the investigator is
able to find most of the algae present; whereas if the pond is a
large one there is an appreciable difference between the number
of species of algae obtained by the two methods — the ratio being
approximately 4 : 3 in favor of the tadpole.

TABLE I.

SHOWING THE TOTAL NUMBER OF SPECIES OF ALGAE TAKEN FROM THE INTESTINAL

TRACT OF FIVE TADPOLES AS COMPARED WITH THE TOTAL NUMBER

FOUND IN FIVE COLLECTIONS MADE FROM THE SAME PONDS.

Size of Pond.
250 x IOO ft.

Collections Made during
Summer of 1927.

Collections Made during
Summer of 1928.

Jun. is-Aug.n.

Aug. n-Aug. 28.

Jun. 2i-Jun. 27.

Jun. 27-July 5.

Tadpole.

Pond.

Tadpole.

Pond.

Tadpole.

Pond.

Tadpole.

Pond.

50

54
52
35
35

32

42
46
30
30

59

45
47
63
46

39

37
44
50

44

63

44
65
56
47

49

44
56
47
39

58
56
59
47

44

48

49

46

41
44

250 x IOO ft

150 x 50 ft. ...

IOO x 50 ft. . . .

50 x 20 ft.

As stated in a paragraph under Methods, three examinations
were made of each pond collection and of each tadpole — one from
the anterior region of the small intestine, one from the middle
region of the small intestine, and one from the large intestine.
Table II. shows the distribution of the species in different regions
of the intestinal tract as compared with the total number found
in the tadpole and the total number found in the pond collections.
Usually more species of algae were found in the anterior end of
the small intestine, but there is not a great variation in numbers
in the three regions. Most of the algae found in the large in-
testine show slight evidence of having been acted upon by the
digestive juices.

Even though the species of algae found in the tadpoles out-
numbered those in the pond collections, algae which did not occur
in the tadpoles' intestines were found in collections made from
the pond. There was one exception, and in this case the tadpole
and pond collection were taken from the same place. In this
entire work only five species of algae were found in pond collections

446

VIVIAN FARLOWE.

TABLE II.

THE TOTAL NUMBER OF SPECIES OF ALG.E FOUND IN DIFFERENT PONDS,
TIIK NUMBER FOUND IN TADPOLES AND THE NUMBER FOUND IN

DIFFERENT REGIONS OF THE INTESTINE.

A. S. Int., anterior end of small intestine; M. S. Int., middle region of small
intestine; A. L. Int., anterior end of large intestine.

Pond.

Tadpole.

A. S. Int.

M. S. Int.

A. L. Int.

29

36

23

22

20

30

34

22

16

19

35

44

23

24

19

33

33

24

19

19

30

45

21

24

31

28

33

28

26

21

3i

40

26

20

24

32

3i

19

M

14

27

50

26

21

36

16

34

18

13

16

25

35

29

12

16

18

35

26

14

22

35

34

21

26

16

24

36

21

21

27

32

38

21

16

, 21

27

36

2O

22

22

32

34

31

15

18

25

38

22

21

25

24

32

18

18

19

36

49

32

26

30

TABLE III.

COLLECTIONS MADE DURING SUMMER OF 1927.

Total

Total

Number
Species
from Both

Percentage of
Those Found
in Tadpoles.

Percentage of
Those Found
in Pond.

Number
Species
from Both

Percentage of
Those Found
in Tadpoles.

Percentage of
Those Found
in Pond.

Sources.

Sources.

50

86.20

55-17

70

82.85

55-71

68

79-32

61.76

58

83.10

63.79

70

74.28

65-71

57

82.62

77.19

45

77-77

66.66

68

93.64

73-23

37

94-59

81.08

53

86.79

75.28

COLLECTIONS MADE DURING SUMMER OF 1928.

67

94-03

73-13

64

95-31

77-50

50

88.

88.

58

96.55

84.48

66

98.48

84.84

62

95-17

74.19

60

94-33

78.33

56

IOO.

83.91

48

97.91

81.25

47

93.61

93.61

Showing total number of species of algae taken from each pond, including the
percentage of those obtained from tadpoles and from pond collections.

OF PONDS. 447

which were not also observed in the tadpoles. Evidently these
species were very rare, for only one was encountered the second
time. The fact that these algae were not found in the tadpoles
does not indicate, therefore, that the tadpoles refuse to eat them.
The variation in percentage of algae from the two sources is
less when pond collections and tadpoles are taken from the same
place. This may be seen by referring to Table III. The pond
collections made during the summer of 1928 were taken from the
immediate vicinity in which the tadpoles were caught, while
those made during the summer of 1927 were taken without regard
to this matter.

SUMMARY.

It is a well known fact that tadpoles feed on microscopic
plants. The importance of this animal as a collector of algae is
clearly demonstrated. In comparing the intestinal contents of
one hundred tadpoles with pond collections made from the same
ponds, the number of species of algae obtained from the tadpoles
exceeded the number obtained from the collections in every case
except two; and in these instances, they were the same. It may
be stated, therefore, that an examination of the intestinal con-
tents of tadpoles affords one of the best and easiest methods of
determining the species of algse present in ponds. This is es-
pecially true in large ponds, and applies particularly to the phyto-
plankton.

In this examination one hundred and seventy species and
varieties of phytoplankton were found. Of this number, one
hundred and sixty-five were encountered in the intestines of
tadpoles.

CONCLUSION.

1. The food of green-frog tadpoles consists chiefly of algae.

2. The algae from pond collections and from the intestinal
contents of tadpoles taken from the same ponds do not differ as
much in small ponds as they do in the larger ones.

3. The anterior region of the small intestine is considered to be
the best region for making examinations for algae.

4. The species of algae taken from the intestines of tadpoles
constituted, on the average, 89.73 + per cent, of the total found.

448 VIVIAN FARLOWE.

An examination of the intestinal contents of tadpoles affords
one of the best and easiest methods of obtaining a collection of
alga; from ponds.

BIBLIOGRAPHY.

Cahn, A. R.

'27 An Ecological Study of Southern Wisconsin 'Fish. Illinois Biological

Monographs.
Coker, R. E.

'18 Principles and Problems of Fish Culture in Ponds. The Scientific Monthly.
Forbes, S. A.

'14 Fresh Water Fish and their Ecology. Illinois State Laboratory of Natural

History.
Mann, A.

'21 The Dependence of Fishes on the Diatoms. Ecology, 2: 79-83.
Tiffany, L. H.

'20 Algal Food of the Young Gizzard Shad. Ohio Journal of Science, 21: 113-

122.
Tiffany, L. H.

'22 Some Algal Statistics Gleaned from the Gizzard Shad. Science, 56: 285-

286.
Tiffany, L. H.

'26 Algal Collection of a Single Fish. Michigan Academy of Science, Arts and
Letters, Vol. VI.

FURTHER OBSERVATIONS ON THK CHEMICAL

COMPOSITION OF WOODS HOLE SEA

WATER— THE CHLORINE

CONTENT AND SALT

ANALYSIS.

IRVINE H. PAGE,

ELI LILLY RESEARCH LABORATORY, MARINE BIOLOGICAL LABORATORY, WOODS

HOLE, MASS.

From time to time we have had occasion to make further ob-
servations on the sea water at Woods Hole since the publication of
the original analysis (i). Though not in any sense complete it is
believed that the following data may prove useful and therefore
they are presented.

It should be pointed out that our aim has been always to select
methods of analysis which would adapt themselves to the use of
relatively small fluid volumes, as only in this way can they be-
come applicable to the investigation of physiological and biolog-
ical problems. From the large number of analyses of sea water
tabulated by the Hydrographic Laboratory of Copenhagen,
Knudsen, Dittmar (2) etc., further data of this kind have oceano-
graphic interest but little more. There has, therefore, been made
a conscious attempt to utilize more sensitive methods which
require small samples for analysis, albeit the absolute values
may not be quite as accurate.

DETERMINATION OF CHLORINE.

Since many physiological activities are sensitive to slight
changes in the tonicity of the surrounding medium it seemed of
interest to determine whether the chlorine content of the Woods
Hole sea water varied to a significant degree from day to day.
The method employed "was as follows: Standard AgNO.3 was
made such that I cc. was equivalent to 10 mg. chlorine. This was
standardized against pure NaCl since it has been shown by
Thompson (3) that this salt may be substituted for standard
water from the Hydrographic Laboratory. The AgNO3 was

449

450

IRVINE H. PAGE.

kept in the dark in a glass stoppered brown bottle and the
standardization repeated at the end of the series of determina-
tions. The method, thereafter, followed in detail that presented
by the Association of Official Agricultural Chemists (4). The
burette used was of 50 cc. capacity, standardized by the Bureau
of Standards, Washington. 15 cc. samples of sea water were
measured with a standardized pipette and diluted with distilled
water to 35 cc. before titration.

Samples were taken from the laboratory tank. This tank is
fed by water taken about 125 feet from shore. The other samples
were taken from surface water as follows: (i) Buzzards Bay one
half mile North of Robinson's Hole. (2) Cuttyhunk 300 feet
from shore on the "Sound" side. (3) Tarpaulin cove one half
mile out in the Sound; water 80 feet deep. (4) East of Nobska;
water 28 feet deep.

Duplicate titrations were made and it may be said that these
determinations but rarely disagreed.

The temperature was taken with not great accuracy, employing
a standard 50 degree laboratory thermometer. Such slight
changes as observed during these observations were not consid-
ered significant.

Grams of chlorine per kilogram were calculated from Thomp-
son's empirical formula—

Clw = = 0.008 -f 0.99980 CU -- 0.001228 C\v2

where C\w •- = grams of Cl per kilogram and Clt, = grams Cl per
liter at 20° C. A graph prepared by using the more common
range of Cl contents was found useful.

The salinity — defined as the weight in grams of all the salts
dissolved in a kilogram of sea water, after the carbonates have
been converted to oxides, the Br and I have been replaced by Cl
and the organic matter has been completely oxidized — was
calculated from the relation derived by Knudson—

So/oo == 0.030 + 1.8050 Clw

Of course it must be recognized that this is only an approxima-
tion, as Giral (5) has emphasized.

During these observations it should be stated that the weather
was in general extremely bad, rain alternating with fog for dis-

CHEMICAL COMPOSITION OF WOODS HOLE SEA \\ATER. 45!

TABLE I.

CHLORINE CONTENT OF \\~OODS HOLE SEA WATER I)i RIM, TIIK Si M.MER OF ig28.

Date.

Source.

Tcinprraturf.

Grams Cl
per Liter.

1 .rams Cl
IKT Ki'.oKr.im.

So'oo.

July 16. . .

Laboratory tank

2 1 degrees

17.80

17.42

.Si. 17

18. . .

*

22

17.80

17.42

31-47

21 ...

'

21

17.86

17.48

31.58

23. ••

'

21

17.77

17-39

31.42

26. ..

'

21.8

17-77

17-39

31.42

28. . .

*

21

17.86

17.48

3I.S8

August i .

*

20-5

17.80

17.42

31-47

July 17. ..

Buzzards Bay. .

2O

17-93

17-54

31.69

*

17. . .

Cuttyhunk

20

18.00

17.60

.51.7';

*

' 21 ...

Off Tarpaulin Cove

20

17-93

17-54

31.69

t

' 21 ...

East Nobska

20

17-70

17.32

3I-27

agreeably long intervals. The results, do not show any very
marked changes in the Cl content of the water but it is altogether
possible that a dry summer may increase the Cl content. Sam-
ples taken from other points along the uneven coast of Woods
Hole show more evident variations, as was to be expected.

SEA SALT ANALYSIS.

Samples of the dried sea salt taken from the laboratory tank
during the summer of 1926 have been analysed, employing the
classical methods as given in the Bulletin of the Official Agricul-
tural Chemists (4) and by Scott (6). Though not complete,
these data are presented, as they may be found useful.

SEA SALT OF WOODS HOLE.

Xo. i

PerciMitam-.

Sodium 30.68

Magnesium 3. 3 i

Calcium 1.27

Silica 0.014

Phosphate Trace

Nitrate. . . Trace

No. 2.

30.49
3.48

1. 12
0.018

Trace
Trace

The above analyses would tend to confirm the suggestion made
in our former paper that the Kramer-Gittleman direct method
for the determination of sodium, while very convenient for rela-
tive data, may give an absolute value which is low. One must
remember, however, that using the Haywood and Smith Method

IRVINE H. PAGE.

or that of Dittmar the sodium determination comes out low,
as has been the universal experience of analysts. The values are
then corrected by employing Dittmar's method (2) of "total
Milphates." The older methods for sodium determinations are so
nbersome (as reference to Dittmar's article will show) that
there is still some doubt as to the accuracy of the results.

During the Summer of 1928 we have again confirmed Atkins'
(8) and Harvey's (9) work on the nitrates and phosphates.
Samples of the Woods Hole water showed only the smallest trace
of NO3 and PO4 during July 1928, the time at which our analyses
were made this year. This change is, as they have shown, due to
seasonal variations in the plankton.

SUMMARY.

1. The chlorine content of Woods Hole sea water has been
examined over a three-week period and shown not to vary within
any large range.

2. Analyses of the sea salt are presented.

REFERENCES.

1. Page. BIOL. BULL., 52, 161 (1927).

2. Dittmar. Challenger Report I. (Phys. and Chem.) i, (1884).

3. Thompson. Jour. Am. Chem. Soc., 50, 681 (1928).

4. Methods of Analysis of the Association of Official Agricultural Chemists,

Washington, D. C., 1925.

5. Giral. Publications de Circonstaiice No. 90 (1926).

<>. Scott. Standard Methods of Chemical Analysis. 3d Ed., D. van Nostrand

Company, New York, N. Y.

7. Hay wood and Smith. Bull. 91, Bureau of Chemistry.
vS. Atkins. Jour. Marine Biol. Assn., 15, 191 (1928).
9. Harvey. Jour. Marine Biol. Assn., 15, 183 (1928).

THE PRECIPITATION OF CALCIUM AND MAGNESIUM
FROM SEA WATER BY SODIUM HYDROXIDE.

ELEANOR M. KAPP.1

In the course of an investigation into the modification of sea
water for use as a perfusion medium (Kapp, '28), it became
necessary to know something of the relative amounts of calcium
and magnesium precipitated by sodium hydroxide. Haas ('16)
suggested that the first flat portion of his titration curve for sea
water was coincident with the precipitation of Mg as hydroxide,
the second with that of Ca. That this was a reasonable assump-
tion is further suggested by the solubility product constants for
the hydroxides of Mg and Ca, which are 1.2 X io~n and 4.1 X
io~6, respectively (Johnston, '15). To obtain more exact in-
formation concerning this behavior of Mg and Ca, the following
experiments were run on sea water taken from the English Chan-
nel outside the Plymouth breakwater, and from Great Harbor,
Woods Hole, Mass.

Graded amounts of 10 normal NaOH (practically carbonate-
free 2) were added to 100 c.c portions of sea water. The flasks
were stoppered and the contents thoroughly mixed. The
supernatant fluid was filtered off as soon as the precipitate had
settled somewhat (within four hours in all cases), and Ca and Mg
were determined in separate samples of the filtrate. Ca was
precipitated as oxalate from 25 cc. samples according to McCrud-
den's ('09) method, and allowed to stand in the refrigerator for
at least 18 hours. The oxalate, after washing, was determined
with permanganate. The Mg determinations were carried out
according to the method of Willstatter and Waldschmidt-Leitz
('23) on duplicate 5 cc. samples from each filtrate. Values for
total Ca and Mg were obtained by the same techniques from
samples of untreated sea water, and show good agreement with
the figures compiled by Clarke ('24) for sea water from a wide
range of sources.

1 From the Laboratory of the Marine Biological Association, Plymouth.
- Made up from the filtrate of a 50 per cent, solution in which the carbonate had
been allowed to settle.

453

454

ELEANOR M. KAPP.

The behavior of Mg and Ca was investigated by Irving1
('26), but major emphasis was placed by him on equilibria within
the biological limits of alkalinity. An extension of these in-
vestigations and an explanation of certain discrepancies which
were encountered follow.

The data for Mg for Plymouth sea water are given in Fig. i,
and roughly agree with my results obtained on Woods Hole sea
water by a less reliable technique. The curve for the precipita-
tion of Mg as drawn by Irving is inaccurate, as owing to the
scarcity of his points he completely missed the plateau. Fig. I,
however, substantiates the points he did determine.

.050

h

0)

•*» .040

0)
Pi

43

«H

P.

030

O
g

a.020

.010

,000

Total Magnesium

o - determined immediately
• - at equilibrium
+ - C08- saturated

j_

.100 .200

Mols NaOH added per liter

FIG. i. The precipitation of Mg from Plymouth sea water in relation to the

amount of NaOH added.

1 Unknown to me when this work was undertaken.

PRECIPITATION OF CALCIUM AM) M.\(.NI-;sirM.

455

The data for Ca show that the results may be considerably
modified by a slight variation in procedure. The Ca curve -
plotted as hollow circles in Figs. 2 and 3 both differ markedly
from the one obtained by Irving. His technique was substanti-
ally the same as mine, with the exception that his original sample-
of sea water, after the NaOH had been added, were shaken for 24
hours instead of being filtered at once, so that equilibrium was
insured. Since CaCO;! tends to remain supersaturated, it was
suspected of being the cause of the discrepancy. A control
experiment was therefore set up, in which the NaOH was added
very slowly as a normal (instead of 10 normal) solution, in order
to avoid local high concentrations of hydroxide, and the stoppered
mixtures were allowed to stand with occasional shaking for one
week. At the end of this time they were filtered and analyzed.

$-.

<D

5.010 -

-d
a>
•p

05
-P

TH

p<

•H
O

I

«J
O

.005 -

determined immediately
at equilibrium
C0a- saturated

. 100 . 200

Mols NaOH added per liter

FIG. 2. The precipitation of Ca from Plymouth sea water in relation to the

amount of NaOH added .

30

456

KLKANOR M. KAPP.

The ('a curve thus obtained differs from the first ones, this time
confirming the results of Irving. Its points are shown in Fig. 2
as black circles. The difference between the two curves is
therefore due only to the slowness with which CaCOs is precipi-
tated, and can be controlled by taking the time factor into account.
The same situation does not exist in the case of Mg, as can be
seen from the black circles plotted in Fig. i, which coincide with
the original curve.

The effect of increasing the amount of carbonate was obtained
by saturating several samples of sea water with CO2 before the
addition of the alkali. Increasing quantities of normal NaOH
were then added very slowly, to allow the gelatinous precipitate
which formed to redissolve, until the third sample, to which 11.5
cc. had been added, remained cloudy. The mixtures were aerated
to drive off excess CO2, and allowed to stand in contact with the
atmosphere for one week. During this time a crystalline precipi-

.010

o
O,

•tf

<u
•p

efl

•p

*H
ft
^

o

8

p.

05
O

VI

.005

Total Calcium

•100 .200

Mols NaOH added per liter

FlG. 3. The precipitation of Ca from Woods Hole sea water in relation to the

amount of NaOH added.

PRECIPITATION OF CALCIUM AND MAGNESIUM.

457

tate had formed, and the solutions were filtered and analyzed as
before. The results are shown by the crosses in Figs. I and 2,
and are strikingly different from the other precipitations. In.
this case the addition of a small amount of alkali precipitates
only the Ca, while the Mg is affected by larger amounts.

PH
12

11

10

3

.05 .10 .15
Mols NaOH added per liter

FIG. 4. The effect of NaOH on the pH of sea water (after Haas).

With reference to the reason for the shape of the Haas titration
curve (Fig. 4), it is clear that Haas' own statement, mentioned
previously, must be modified somewhat. As he suggested, Mg
is precipitated rapidly by NaOH over the range where his titra-
tion curve shows a plateau. At a region corresponding to the
addition of o.i mols of NaOH per liter of sea water, the titration
curve begins its second rise, and the Mg curve flattens out. A
small amount of Ca, however, is precipitated throughout, owing
its first precipitation to the insolubility of the carbonate,1 which
is intermediate in this respect between Mg and Ca hydroxides.

I am deeply indebted to Dr. E. J. Allen, F. R. S., of the Marine
Biological Association, Plymouth, for facilities extended to me

1 Ks.p. = .98 X io-» (Johnston, '15).

458 ELEANOR M. KAPP.

during this investigation. 1 also wish to thank Prof. M. H.
Jacobs and Mr. H. W. Harvey for their helpful interest.

REFERENCES.
Clarke, F. W.

'24 U. S. Geol. Survey, Bull. 770, p. 127.
Haas, A. R.

'16 Jour. Biol. Chem., Vol. 26, 515.
Irving, L.

'26 Jour. Mar. Biol. Ass., Vol. 14, 441.
Johnston, J.

'15 Jour. Am. Chem. Soc., Vol. 37, 2001.
Kapp, E. M.

'28 Science, Vol. 67, 513.
McCrudden, F. H.

'09 Jour. Biol. Chem., Vol. 7, 83.
Willstatter, R., and Waldschmidt-Leitz, E.

'23 Ber. v. d. deut. chem. Ges., Vol. 56, 488.

FURTHER OBSERVATIONS ON THE EFFECT OF

HIGH FREQUENCY SOUND WAVES ON

LIVING MATTER.

E. NEWTON HARVEY, ETHEL BROWNE HARVEY AN-D
ALFRED L. LOOMIS.i

»

Interest in the biological effects of very high frequency sound
waves started with the investigations of Wood and Loomis (i)
who devised methods for producing intense "supersonic" vibra-
tions and described many of the phenomena connected with them.
The reader is referred to this paper for a description of the two
kilowatt generator and methods of working with the waves. The
apparatus was of such high power and the sound waves of such
great intensity as to produce considerable heating. It seemed
highly desirable in working with cells to reduce the heating
effects of the vibrations, and to observe the cell with the micro-
scope while being radiated. After many attempts to use the
high power oscillator as the source of the waves and to lead them
to the material on the stage of a microscope along capillary rods
and tubes, a low-powered apparatus was decided upon as the
most convenient for the purpose. This has previously been
described by Harvey and Loomis (2) together with some of the
effects of these supersonic waves on living organisms, cells and
tissues. The outfit consists of a 75 watt high frequency oscillator
and a quartz crystal whose vibrations, produced in the electric
field by reversal of the piezo-electric effect, travel through any
medium in contact with the crystal. A frequency of 400,000 per
second was used and the material mounted directly on the crystal
which served as a microscopic slide. Schmitt, Olson and
Johnson (3) have also described various biological effects using a
250 watt generator with crystal immersed in xylene. They lead
the sound waves along a rod of small diameter ending in a micro-
needle, which could be inserted into the material to be studied.

Some additional effects have been recently observed with our

1 From the Marine Biological Laboratory, Woods Hole, the Physiological
Laboratory, Princeton University, and the Loomis Laboratory, Tuxedo Park, N. Y.

459

460 E. N. HARVEY, E. B. HARVEY AND A. L. LOOMIS.

75 watt outfit in its original form and also modified to use higher
frequencies by changing the capacity, inductance and crystal.
The new quartz crystal was a spectacle lens which happened to be
cut in the proper direction, kindly loaned by Dr. Kenneth Cole.
The natural frequency of this crystal was approximately one and
one quarter million per second. Its thickness varied from I to
1.8 mm. and consequently the distance between the tin foil
electrodes, was much less than in the original 7 mm. crystal, giving
a far more intense electrical field and greater effects. A few
experiments have been made with a 2.25 million crystal which
vibrates well and gives the same effects with Elodea as the 1.25
million. A 6 million crystal, 0.45 mm. thick, does not vibrate
strongly. We are at present engaged in increasing the frequency
to the highest point possible to see how biological effects will vary
with the frequency.

A convenient means of finding the resonant frequency of the
crystal is to set it up between the two tin foil electrodes with
holes in their centers (to allow light to pass for microscopic ob-
servation) and then place a drop of water on the crystal. At
various settings of the condenser the water will be violently
agitated and broken up into fine droplets like steam. Low
melting point crystals placed in the water show that the tempera-
ture does not rise but that the "steam" is mechanically formed,
as observed in various ways by Wood and Loomis (i), and not a
condensation from vapor. The exact specifications for an oscil-
lator giving various frequencies will be found as an appendix to
this paper.

If an Elodea leaf covered with a cover slip is mounted on a
crystal whose resonant frequency is 400 kilocycles, and relatively
weak (by reducing filament current) sound waves sent through
the leaf, it can be observed under the microscope that only certain
areas in the leaf show the characteristic whirling of the chloro-
plasts described in our previous paper (2). The areas do not
correspond to any position on the crystal but to some peculiarity
in the leaf, as moving a leaf to a new position over the crystal
does not necessarily change the areas of marked whirling. These
areas of whirling are most marked where air bubbles, which vi-
brate strongly, are caught under the leaf and where the cells are
several layers in thickness, near the midrib (which also contains

EFFECT OF SOUND WAVES OX LIVING MATTER. 461

air in intercellular spaces). Part at least of the condition for rapid
whirling is the distance of the leaf from the crystal. By attaching
the coverslip to a mechanical device for adjusting its distance
from the crystal, the amount of water between coverslip and
crystal can be varied and a slight change in this layer of water
will cause whirling in a given area to start or to stop. These
effects are no doubt due to interference of two sets of sound waves
resulting in complicated interference patterns with nodes and
internodes. Fine particles like red blood corpuscles suspended
between crystal and coverslip can be observed to collect in nodes
forming such a pattern. The chloroplasts in Elodea cells cannot
do so since they are restricted in movement by the cell walls but
in a region which happens to be an internode, they will undergo
rapid whirling movements. The part played by an air bubble in
causing rapid whirling is no doubt to offer a reflecting surface
around which interference pattern and nodes appear. The
whirling itself is probably due to the radiation pressure of the
sound waves as they pass through the cells.

Another phenomenon regularly observed is a variation in the
rate and character of the whirling as the variable condenser is
changed to vary the frequency. For instance, over a range of
10 kilocycles, there appeared maximum whirling in a given area
of the leaf at 407, 409, 410.4, 412.5, 415, and 417 kilocycles, i.e.
a maximum approximately every 2 kilocycles, with no whirling or
very slow whirling between.

In order to understand the changes in whirling motion imparted
to the biological material placed upon the quartz as the frequency
is varied, it is necessary to digress a moment and consider the
forces acting upon an oscillating quartz disk. As is well known,
a natural quartz crystal has three electric axes perpendicular to
the optic axis. (See Fig. i.)

The disk is cut as indicated by the shaded portion, i.e. so that
one of the electric axes shall be perpendicular to the plane of the
disk. If pressure is applied to the side of the disk corresponding
to A - a negative charge will accumulate there, while corre-
spondingly if a negative charge is applied there without pressure
the disk will contract as if the equivalent pressure had been ap-
plied. The same holds true with positive charges on the A + sidr.
On the other hand, when a positive charge is placed on the nega-

462 I . N- HARVEY, E. B. HARVEY AND A. L. LOOMIS.

tivc side and a negative charge on the positive side, the crystal
will expand. A rapid alternation of charges causes the crystal
to oscillate and as a first approximation the crystal can be con-
sidered to be an oscillating rigid piston. This would be rigorously
correct if the disk were perfect and infinitely large but with a
finite disk the forces are not symmetrical near the edges and a
complex wave pattern is formed in the crystal. This can easily
be seen by first considering a point 0 on the surface of the crystal
near the center (Fig. 2). If a unit negative charge is placed on
the under surface with the corresponding positive charge on the

A-

FIG. i. Quartz crystal (shaded) cut perpendicular to optic axis. Electric
axes indicated by A A, BB, CC.

FIG. 2. Vectors showing forces in point O in crystal.

upper, the crystal at "0" will tend to contract along the axis
OA + and expand along the axes OB — and OC— . The intensities
of these forces are directly proportional to the potential gradiants
along the respective axis. The forces along OB— and OC-- are
therefore only half as great as along OA - since the distances
through the crystal along these axis are twice as great as along
the axis OA- (the angles between the axis being 60°). The
vector resolution of the forces OC— and OB— along OA — shows
that they are equivalent to a force opposed to the force along
OA - • and of magnitude equal to one half that of OA — . The
vector equivalent of all three forces is therefore a single force along
OA — equal to half of what that force would be if the forces along
the axes OC- - and OB— were not present.

This symmetry does not maintain however near the edges of
the disk. Consider the point Q, Fig. 3. The axis QB is not in

EFFECT OF SOUND WAVES ON LIVING MATTER. 463

the crystal at all. The resolution of the forces along QA -- and
QC- - gives ajbrce along QX equal to the force along QA + multi-
plied by IA/S. It is clear, therefore, that the forces near the
edges are not symmetrical and tend to produce distortions which
travel in waves across the disk.

A second system of forces are also acting on the disk. As the
quartz contracts normally to the surface it expands parallel to the
surface (this effect is best seen in a rectangular plate). Thus the
series of longitudinal waves create interference patterns with the
traverse waves.

A-

FIG. 3. Vectors showing forces in point Q in crystal.

Thirdly, it has been shown that even with a perfect quartz
crystal the intensity of the piezo electric effect varies in different
parts of the crystal. Dye has photographed the distortions
produced in the interference fringes of an interferometer when
one of the plates is an oscillating quartz disk. These photo-
graphs show most beautiful and complex patterns which showly
drift across the plate when the frequency is slowly changed.

Fortunately in biological investigations under the microscope
good use can be made of these complex patterns. Thus, without
changing the position of the specimen on the crystal, one can by
merely changing slightly the frequency, cause these patterns to
shift so that any particular part of the specimen can be made to
experience forces of varying magnitude and direction. Thus in a
particular cell of Elodea the chloroplasts can, at will, be made to
rotate slowly or rapidly, clockwise, or counter clockwise, in one
vortex or in a series of vortices, while merely watching the
specimens under the microscope and observing the effects pro-
duced as the frequency is slowly varied.

By increasing the intensity, the leaf of Elodea can be agitated
so violently that the chloroplasts themselves are broken up into a

464 E. X. HARVEY, E. B. HARVEY AND A. L. LOOMIS.

fine green emulsion which completely fills the cell. This effect is
not due to heating, since crystals of ethyl stearate, melting at
30—31° C. and placed on the Elodea leaf, are not melted even after
1 5 minutes, nor does slowly heating Elodea leaves bring about this
effect. Neither is it due to possible mechanical rupture of the
cellulose wall or mixing of the vacular sap with the chloroplasts,
since unrayed cells can be crushed with a needle and their chloro-
plasts do not break up in this characteristic manner. The emulsi-
fication is caused by the tearing action of the sound waves.

Perhaps it should be emphasized at this point that these effects
are all due to high frequency sound waves and not to any influence
of the oscillating electrical field, as control experiments using
glass plates of a size similar to and replacing that of the quartz
crystal have shown.

Some of the more interesting of the effects observed with the
spectacle lens crystal and frequencies of 1,250 kilocycles are as
follows :

AmcebcB proteus or dubia,1 moving along the surface of the
crystal are not particularly affected by an intensity that causes the
inclusions in small vacuoles of the Amoeba to rotate on their axes.
Higher intensities cause a mild whirling of the more liquid regions
of the Amoeba followed by rupture of the pellicle on one side and
extrusion of the contents which join the general whirl of fluid in
the medium. There is a tendency for the Amceba to move more
rapidly during the raying as if the endoplasm became more
liquid. After this there is a sudden change in direction of move-
ment.

Both unfertilized and fertilized sea urchin (Arbacia) and star-
fish egg are violently agitated and may spin around. The jelly
is torn off and the fertilization membrane may be broken. The
eggs are thrown into rows or clumps and eventually cytolyze
either partially or completely, the cytolysis taking place on one
side and sometimes within the fertilization membrane. There is
no movement of materials inside the egg caused by raying, as can
be determined with certainty by using centrifuged eggs, the
stratified layers remaining intact until cytolysis takes place.
Cytolysis may take place from any of the stratified layers. How-
ever, if the unfertilized centrigufed eggs are placed in diluted sea
water (40 distilled water to 60 sea water) and thus made less

!.!• KELT OF SOUND WAVES ON LI VI. Ml MATTER. 465

viscous, the inside may be made to whirl. The whirling takes
place in the lighter layers, including the oil, clear and granular
layers, but only along the edge of the pigment layer, most of
which remains intact. The oil drops tend to remain together,
but the clear and granular layers become mixed and after ten or
fifteen minutes of whirling a clear zone can not be distinguished.
The direction of rotation may be reversed instantly by a slight
change in frequency. Sometimes instead of a whirling of the
protoplasm, there is a streaming of granules similar to that of an
Amoeba. No whirling of protoplasm nor movement of granules
has been observed in eggs put in dilute sea water and not centri-
fuged. This may be due to the fact that the less dense material,
where the whirling takes place is not separated out from the
heavier pigment granules.

The asters are quite unaffected by raying. Cleavage furrows
will come in normally during raying, even when the egg is vio-
lently agitated. When an egg has been slightly cytolyzed by
raying we have observed that the furrow may come in at the
proper place. Eggs in the two or four cell stage may have one or
two blastomeres cytolyzed and the others unaffected.

Arbacia plutei swimming slowly are paralyzed by a momentary
raying, presumably because the cilia are torn off. Otherwise they
look uninjured but more prolonged treatment or greater intensity
will tear them to pieces, leaving only fragments of the skeleton
behind.

The gill cilia of Mytilius do not seem to be affected by violent
agitation of the sea water about them, until the cilia and gill
filaments are actually torn to pieces.

Pigment cells well-expanded in the scales of Fundulus are not
affected, although the scales are rapidly agitated as the waves
impinge upon them.

Frog abdominal muscles mounted on the crystal show no con-
traction or movement although air bubbles and blood corpuscles
on top of the muscle tissue whirl rapidly. The waves must have
passed through the muscle tissue to reach the air bubbles and
corpuscles.

Fragments of the rays of the ctenophore, Mnemiopsis, con-
taining luminous material, mounted on the crystal in the dark
and waves passed through, are agitated and occasionally

466 I'.. N. HARYKY, K. B. HARVEY AND A. L. LOOMIS.

luminesce. There is no continual luminescence which can be
attributed to the waves but only the sporadic luminescence con-
nected with sudden movement of the fragment such as can be
( ibtuined on jarring the table containing fragments of Mnemiopsis,
even when not exposed to high frequency sound waves.

Fundulus embryos within the egg, with beating hearts, sub-
jected to waves of an intensity to agitate the eggs but not so
great an intensity as to interfere with observation of the heart
beat show no marked effect upon the character of the beat or
circulation. In fact only the effect observed was a slight increase
in rate during raying which can be accounted for by a slight in-
crease in temperature, that undoubtedly occurs when these high
frequency waves carrying considerable energy, are absorbed by
the medium. The embryos were rayed I minute and then not
rayed for one minute while the heart beats were counted. In
four experiments the rates were: Rayed — 148, 157, 140, 132;
unrayed --140, 148, 122, 122, respectively. The average in-
crease in rate was about 8 per cent., which can be accounted for
from the known effect of temperature on the heart beat of
Fundulus heteroditus ,2 by a rise of temperature from 22° C. to
about 23° C.

Perhaps it should be emphasized again from the experiments on
muscle, heart, luminous cells and chromatophores that there is no
stimulating effect of these waves similar to the stimulation by
electrical or sudden mechanical disturbance.

Fertilized Fundulus eggs mounted on the crystal can be very
violently agitated and the oil drops and granules within made to
dance. The yolk can be thoroughly stirred and the surface of the
protoplasm can be observed to move and bend. Dr. Elmer
Butler has carried these eggs to the point of hatching and finds
the development and the embryos normal. If the agitation has
continued so long as to burst the protoplasmic surface develop-
ment does not proceed. An intensity of raying which does not
destroy the surface has no effect on development while a slightly
greater intensity results in dissolution and cytolysis.

Study of a large number of cells and tissues, some of which are
recorded above, has led us to the conclusion that the effects of
these waves, apart from slight heating, are purely mechanical.
If intense enough, practically all cells can be cytolyzed. It is as

KFFECT OF SOUND \\AYHS ON I.I\I\(- MATTER. 467

if one could grasp a cell in both hands and bend it violently back
and forth at a very rapid rate. Delicate structures on the out-
side of a cell are torn off. If the cell is very small it is thrown into
nodes so quickly as to escape injury. If the cell can be held fixed
and is not too viscous, its contents can often be made to whirl
before it breaks down.

From the whirling one can gain an idea of the viscosity of the
cell contents. Perhaps the chief value of the waves for biological
investigation lies in the evidence obtained from their action
regarding the viscosity of cells. It should be emphasized, how-
ever, that comparative studies of viscosity are difficult because of
the great complexity of the sound wave patterns under the cover
slip, both horizontally and vertically. Two cells in different
portions of the same microscopic field are not necessarily exposed
to the same radiational forces and great caution must be used in
drawing conclusions regarding viscosity or resistance to tearing
by difference in behavior of cells.

High frequency sound waves offer a new means of affecting
the interior of cells without necessarily breaking down the -cell
wall. They will be of most value when a beam of given fre-
quency and controlled intensity can be sent through a cell or
tissue in a particular direction.

APPENDIX.

For those biologists who desire to construct a low-powered
oscillator, the following constructional details ought to suffice.

The following apparatus is recommended.
One No. 852 Radiotron 75 watt tube,
One tube holder

One filament transformer to give 10 volts
One plate transformer to give 2,000 volts
One 5,000 ohm resistance

Several transmitting condensers (designed to withstand 5,000
volts) with an aggregate capacity of about o.i microfarad
One rheostat

Some heavy copper strip to wind the inductance
Some fine wire to make the secondary

All of the above can be bought from any radio store carrying

468 E. X. HARVEY, E. B. HARVEY AND A. L. LOOMIS.

parts for transmitting sets, and should not cost more than $100
in the aggregate.

Fig. 4 shows the wiring diagram and a suggested arrangement
of the parts. The iron of the transformers should be on the
side of the tube away from the oscillating parts and should be at
least a foot from the tube. All the parts can conveniently be
mounted on a board 30 x 10 inches.

IIOA.C.

FIG. 4. Constructional diagram for a 75 watt oscillator. A, plate transformer;
B, filament transformer; C, rheostat; D, choke coil; E, Radiotron No. 852;
F, Blocking condenser. G, Grid leak condenser. H, Grid lead. I, Inductance coil;
J, tuning condenser; K, secondary coil; L, Variable condenser; Q, Quartz plate
between electrodes.

The rheostat should be mounted in the lead from the no-volt
A. C. house circuit and can be used to regulate the voltage. The
primaries of the transformers should be connected in parallel
across the house circuit. One side of the secondary of the plate
transformer should be connected to the center tap of the filament
transformer which point should also be grounded. The other
side of the secondary should go through a choke coil to the plate.
The choke coil can be made by winding about 100 turns of fine
wire on a bakelite tube one or two inches in diameter.

The inductance can be made by winding fifteen or twenty turns
of heavy copper wire on a bakelite tube six or eight inches in
diameter. The plate should be connected to one tap on the
inductance through a blocking condenser of about .002 microfarad
capacity. The grid should be connected to the other tap on the
inductance through a by-pass condenser of about the same
capacity and a grid leak of above 5,000 ohms resistance. The
center tap of the inductance should be grounded. The secondary
can be made by winding 100 turns of fine wire on a bakelite tube

EFFECT OF SOUND \\AVKS ON I.lVI.Mi MATTER. 469

which can be slipped inside the primary inductance. One end of
the secondary should go to one plate of the crystal holder (the
other plate of the holder being grounded). The other end of the
secondary should be connected to ground through a variable
condenser or to a rod of metal perhaps i inch diameter and ten
inches long, which is not grounded.

The quartz crystal need not be larger than one square inch.
It should be cut perpendicular to an electric axis. Its natural
frequency of oscillation will depend on its thickness.

Mm. Thick. Frequency (Approx.).

i .... 2,900,000 cycles per sec.

2. . 1,450,000

3 966,000

4- 725,000

5 580,000

etc etc.

The oscillating circuit should be tuned to approximately the
frequency of the crystal.

The crystal holder can conveniently be made out of two micro-
scope slides and two thin brass strips with holes cut in them for
use with the microscope. The microscope should be at least three
feet from the oscillator so that movements of the operators body
shall not change the frequency. The high tension lead to the
microscope should be shielded by surrounding it with a grounded
metal tube and the microscope itself should be grounded to pre-
vent small spark discharges to the observer.

REFERENCES.

1. Wood, R., and Loomis, A. L.

'27 Phil. Mag., 4, 417.

2. Harvey, E. N., and Loomis, A. L.
'28 Nature, 121, 622.

3. Schmitt, F. O., Olson, A. R., and Johnson, C. H.

'28 Proc. Soc. Exp. Biol. Med., 25, 718.

1 Kindly supplied by Dr. J. A. Dawson of Harvard University.

2 Unpublished data of Dr. Otto Glaser of Amherst College.

BIOLOGICAL BULLETIN

OF THE

flDarine Biological Xaboratorj)

WOODS HOLE, MASS.

VOL. LV

JULY, 1928

No. i

CONTENTS

Thirtieth Annual Report of the Marine Biological Laboratory

PUBLISHED MONTHLY BY THE

MARINE BIOLOGICAL LABORATORY

PRINTED AND ISSUED BY

LANCASTER PRESS, INC.
LANCASTER, PA.

AGENT FOR GREAT BRITAIN

WHELDON & WESLEY, LIMITED

2, 3 and 4 Arthur Street, New Oxford Street, London, W. C. 2

Single Numbers, S1.OO. Per Volume <6 numbers), S4.5O

Entered October 10,1902. at Lancaster. Pa., as second-class matter under Act of Congress of July 16. 1894

EOttortal Statf

GARY N. CALKINS — Columbia University.

E. G. CONKLIN — Princeton University.

M. H. JACOBS — University of Pennsylvania.

FRANK R. LILLIE — University of Chicago.

GEORGE T. MOORE — The Missouri Botanic Garden.

T. H. MORGAN — Columbia University.

W. M. WHEELER — Harvard University.

E. B. WILSON — Columbia University.

Bbitor

C R. MOORE — The University of Chicago.

All communications and manuscripts should be sent to the Man-
aging Editor, the University of Chicago. Subscriptions and other
matter should be addressed to the Biological Bulletin, Prince and
Lemon Streets, Lancaster, Pa.

Notice to contributors. Every paper to appear in the Biological
Bulletin should be accompanied by an author's abstract presenting
the chief results of the investigation. The abstract should not exceed
225 words in length.

For indexing purposes there should be, in addition to the title,
one or more subject headings indicating in a word or two the divi-
sions of the subject discussed in the paper. The entire name of the
author (of each author if a joint paper) and the year of birth is also
desired.

BIOLOGICAL BULLETIN

OF THE

fiDarine Biological laboratory

VOL. LV

WOODS HOLE, MASS.

AUGUST, 1928

No. 2

KING, ROBERT L.

MAIN, ROLLAND J.

AMBERSON, WILLIAM R.

BOYD, MARJORIE

CALKINS, GARY N., AND
BOWLING, RACHEL

PARPART, ARTHUR K.

HUESTIS, R. R.

MELVIN, ROY

CONTENTS

The Contract-He Vacuole in Paramecium
trichium ..........................

Observations of the Feeding Mechanism of
a Ctenophore, Mnemiopsis leidyi ...... ''','

The Influence of Oxygen Tension upon the
Respiration of Unicellular Organisms . . ~n

A Comparison of the Oxygen Consumption
of Unfertilized and Fertilized liggs of
Fund nl us heteroclitus ............... (>-

Sti/dics on Dallasia frontata Stokes ....... i«'i

The Bacteriological Sterilization of Parame-

i i .1

The Effect of Maternal Age and of Temper-
ature Change in Secondary Xou-Disj unc-
tion ............................. TJI

O\v«en Consumption of Insect I''.ggs ...... i;-,5

PUBLISHED MONTHLY BY THE

MARINE BIOLOGICAL LABORATORY

PRINTED AND ISSUED BY

LANCASTER PRESS, INC.
LANCASTER, PA.

AGENT FOR GREAT BRITAIN

WHELDON & WESLEY, LIMITED

2, j and 4 Arthur Street, New Oxford Street, London, W. C. 2

Single Numbers, S1.0O. Per Volume '6 numbers), $4.50

Entered October 10,1902. at Lancaster, Pa., as second-class matter under Act of Congress of July 16, 1894

Eotforial Staff

GARY N. CALKINS — Columbia University.

E. G. CONKLIN — Princeton University.
M. H. JACOBS — University of Pennsylvania.

FRANK R. LILLIE — University of Chicago.

GEORGE T. MOORE — The Missouri Botanic Garden.

T. H. MORGAN — Columbia University.

W. M. WHEELER — Harvard University.

E. B. WILSON — Columbia University.

lEMtor

C. R. MOORE — The University of Chicago.

All communications and manuscripts should be sent to the Man-
aging Editor, the University of Chicago. Subscriptions and other
matter should be addressed to the Biological Bulletin, Prince and
Lemon Streets, Lancaster, Pa.

Notice to contributors. Every paper to appear in the Biological
Bulletin should be accompanied by an author's abstract presenting
the chief results of the investigation. The abstract should not exceed
225 words in length.

For indexing purposes there should be, in addition to the title,
one or more subject headings indicating in a word or two the divi-
sions of the subject discussed in the paper. The entire name of the
author (of each author if a joint paper) and the year of birth is also
desired.

BIOLOGICAL BULLETIN

OF THE

flDarine Biological laboratory

WOODS HOLE, MASS.

VOL. LV SEPTEMBER, 1928 No. 3

CONTENTS

HILL, SAMUEL K. The Influence of Molds on the Growth of Lumi-

nous Bacteria in Relation to the Hydrogen
Ion Concentration, Together with the Devel-
opment of a Satisfactory Culture Method. . 143

KAROL, JOHN J. The Sex Ratio in Peromyscus 151

PAYNE, NELLIE M. Cold Hardiness in the Japanese Beetle, Po-

pillia japonica Newman 163

NELSON, THURLOW C. Pelagic Dissoconchs of the Common Mussel,

Mytilus edulis, with Observations on the
Behavior of the Larva of Allied Genera .... 1 80

TURNER, C. L. Studies on the Secondary Sexual Characters of

Crayfishes. — VI. A Female of Cambarus
immunis with Oviducts Attached to Openings
of Sperm Ducts 193

TURNER, C. L. Studies on the Secondary Sexual Characters

of Crayfishes. — VII. Regeneration of Aber-
rant Secondary Sexual Characters 197

SAYLES, LEONARD P. Regeneration of Lumbricul 'us in Various Ringer

Fluids 202

ALPATOV, W. W. Variation of Hooks on the Hind Wing of the

Honey Bee (Apis meUifera L.} 209

PUBLISHED MONTHLY BY THE

MARINE BIOLOGICAL LABORATORY

PRINTED AND ISSUED BY

LANCASTER PRESS, INC.
LANCASTER, PA.

AGENT FOR GREAT BRITAIN

WHELDON & WESLEY, LIMITED

2, 3 and 4 Arthur Street, New Oxford Street, London, W. C. 2

Single Numbers, $1.00. Per Volume (6 numbers), S4.5O

Entered October 10.1902, at Lancaster, Pa., as second-class matter under Act of Congress of July 16, 1894

jEDttorial Statf

GARY N. CALKINS — Columbia University.

E. G. CONKLIN — Princeton University.

M. H. JACOBS — University of Pennsylvania.

FRANK R. LILLIE — University of Chicago.

GEORGE T. MOORE — The Missouri Botanic Garden.

T. H. MORGAN — Columbia University.

W. M. WHEELER — Harvard University.

E. B. WILSON — Columbia University.

C. R. MOORE — The University of Chicago.

All communications and manuscripts should be sent to the Man-
aging Editor, the University of Chicago. Subscriptions and other
matter should be addressed to the Biological Bulletin, Prince and
Lemon Streets, Lancaster, Pa.

Notice to contributors. Every paper to appear in the Biological
Bulletin should be accompanied by an author's abstract presenting
the chief results of the investigation. The abstract should not exceed
225 words in length.

For indexing purposes there should be, in addition to the title,
one or more subject headings indicating in a word or two the divi-
sions of the subject discussed in the paper. The entire name of the
author (of each author if a joint paper) and the year of birth is also
desired.

BIOLOGICAL BULLETIN

OF THE

flDarine Biological laboratory

WOODS HOLE, MASS.

VOL. LV OCTOBER, 1928 No. 4

CONTENTS

HARMAN, MARYT., AND

ROOT, FRANK P. The Development of the Spermatozoon in Cavia

cobaya 235

TURNER, C. L. Studies on the Secondary Sex Characters of

Cray fishes, VIII. Modified Third Abdom-
inal Appendages in Males of Cambarus
virilis 255

GRAVE, B. H. Natural History of Shipu'orm, Teredo n aval-is,

at Woods Hole, Massachusetts 260

NEWMAN, H. H. Studies of Human Twins, I. Methods of Di-

agnosing Monozygotic and Dizygotic Tuins 283

NEWMAN, H. H. Studies of Human Twins, II. Asymmetry

Reversal, of Mirror Imaging in Identical
Tirins 298

PUBLISHED MONTHLY BY THE

MARINE BIOLOGICAL LABORATORY

PRINTED AND ISSUED BY

LANCASTER PRESS, INC.
LANCASTER, PA.

AGENT FOR GREAT BRITAIN

WHELDON & WESLEY, LIMITED

2, 3 and 4 Arthur Street, New Oxford Street, London, W. C. 2

Single Numbers, $1.00. Per Volume (6 numbers), S4.50
Entered October ro, 1902, at Lancaster, Pa., as second-class matter under Act of Congress of July 16, 1804

Staff

GARY N. CALKINS — Columbia University.

E. G. CONKLIN — Princeton University.

M. H. JACOBS — University of Pennsylvania.

FRANK R. LILLIE — University of Chicago.

GEORGE T. MOORE — The~Missouri Botanic Garden.

T. H. MORGAN — Columbia University.

W. M. WHEELER — Harvard University.

E. B. WILSON — Columbia University.

BDitor

C. R. MOORE — The University of Chicago.

All communications and manuscripts should be sent to the Man-
aging Editor, the University of Chicago. Subscriptions and other
matter should be addressed to the Biological Bulletin, Prince and
Lemon Streets, Lancaster, Pa.

Notice to contributors. Every paper to appear in the Biological
Bulletin should be accompanied by an author's abstract presenting
the chief results of the investigation. The abstract should not exceed
225 words in length.

For indexing purposes there should be, in addition to the title,
one or more subject headings indicating in a word or two the divi-
sions of the subject discussed in the paper. The entire name of the
author (of each author if a joint paper) and the year of birth is also
desired.

BIOLOGICAL BULLETIN

OF THE

flDarine Biological laboratory

WOODS HOLE, MASS.

VOL. LV

NOVEMBER, 1928

No. 5

CONTENTS

HUMPHREY, R. R. Sex Differentiation in Gonads Developed from

Transplants of the Intermediate Mesoderm
of Amblystoma 317

MOORE, CARL R. On the Properties of the Gonads as Controllers

of Somatic and Psychical Characteristics, XI. 3 39

\

JUST, E. E. Initiation of Development in Arbacia, VI.

The Effect of Slowly Evaporating Sea- Water
and its Significance for the Theory of Auto-
Parthenogenesis 358

CHAMBERS, ROBERT. Intracellular Hydrion Concentration Studies,

I. The Relation of the Environment to the
pH of Protoplasm and of Its Inclusion Bodies. 369

REZNIKOFF, PAUL, AND

POLLACK, HERBERT. Intracellular Hydrion Concentration Studies,

II. The Effect of Injection of Acids and
Salts on the Cytoplasmic pH of Amoeba
dubia 377

POLLACK, HERBERT. Intracellular Hydrion Concentration Studies,

III. The Buffer Action of the Cytoplasm
of Amceba dubia and Its Use in Measuring
thepH 383

GREGORY, LOUISE H. Th? Effects of Changes in Medium during

Different Periods in the Life History of
Uroleptus mobilis and Other Protozoa 386

PUBLISHED MONTHLY BY THE

MARINE BIOLOGICAL LABORATORY

PRINTED AND ISSUED BY

LANCASTER PRESS, INC.
LANCASTER, PA.

AGENT FOR GREAT BRITAIN

WHELDON & WESLEY, LIMITED

2, 3 and 4 Arthur Street, New Oxford Street, London, W. C. 2

Single Numbers, $1.00. Per Volume (6 numbers), S4.5O

Entered October 10, 1902, at Lancaster, Pa., as second-class matter under Act of Congress of July 16, 1894

EDttorial Staft

GARY N. CALKINS — Columbia University.

E. G. CONKLIN — Princeton University.

M. H. JACOBS — University of Pennsylvania.

FRANK R. LILLIE — University of Chicago.

GEORGE T. MOORE — The Missouri Botanic Garden.

T. H. MORGAN — Columbia University.

W. M. WHEELER — Harvard University.

E. B. WILSON — Columbia University.

Managing BMtor

C. R. MOORE — The University of Chicago.

All communications and manuscripts should be sent to the Man-
aging Editor, the University of Chicago. Subscriptions and other
matter should be addressed to the Biological Bulletin, Prince and
Lemon Streets, Lancaster, Pa.

Notice to contributors. Every paper to appear in the Biological
Bulletin should be accompanied by an author's abstract presenting
the chief results of the investigation. The abstract should not exceed
225 words in length.

For indexing purposes there should be, in addition to the title,
one or more subject headings indicating in a word or two the divi-
sions of the subject discussed in the paper. The entire name of the
author (of each author if a joint paper) and the year of birth is also
desired.

\

BIOLOGICAL BULLETIN

OF THE

fiDarine Biological Xaboratorp

WOODS HOLE, MASS.

VOL. LV

DECEMBER, 1928

No. 6

CONTENTS

BODINE, JOSEPH HALL. Insect Metabolism 395

LLOYD, FRANCIS E., AND The Pulsatory Rhythm of the Contractile
BEATTIE, J. Vesicle in Paramecium 404

THRELKELD, W. L., AND Observations on Hydra and Pelmatohydra
HALL, S. R. under Determined Hydrogen Ion Con-
centration 419

MANWELL, REGINALD D. The Occurrence of Nuclear Variations in

Pleurotricha lanceolata (Stein) 433

QUIGLEY, J. P. Observations on the Life History and

Physiological Condition of the Pacific
Dog Fish (Squahis sucklii] 439

FARLOWE, VIVIAN. Algae of Ponds as Determined by an Ex-
amination of the Intestinal Contents of
Tadpoles 443

PAGE, IRVINE H. Further Observations on the Chemical

Composition of Woods Hole Sea Water
— The Chlorine Content and Salt Analy-
sis 449

KAPP, ELEANOR M. The Precipitation of Calcium and Mag-
nesium from Sea Water by Sodium
Hydroxide 453

HARVEY, E. NEWTON, Further Observations on the Effect of High
HARVEY, ETHEL B., AND Frequency Sound Waves on Living
LOOMIS, ALFRED L. Matter 459

PUBLISHED MONTHLY BY THE

MARINE BIOLOGICAL LABORATORY

PRINTED AND ISSUED BY

LANCASTER PRESS, INC.
LANCASTER, PA.

AGENT FOR GREAT BRITAIN

WHELDON & WESLEY, LIMITED

2, j and 4 Arthur Street, New Oxford Street, London, W. C. 2

Single Numbers, S1-OO. Per Volume (6 numbers), S4.50
Entered October 10, 1902, at Lancaster, Pa., as second-class matter under Act of Congress of July 16, 1894

EDttorial Staff

GARY N. CALKINS — Columbia University.

E. G. CONKLIN — Princeton University.

M. H. JACOBS — University of Pennsylvania.

FRANK R. LILLIE — University of Chicago.

GEORGE T. MOORE — The Missouri Botanic Garden.

T. H. MORGAN — Columbia University.

W. M. WHEELER — Harvard University.

E. B. WILSON — Columbia University.

C. R. MOORE — The University of Chicago.

All communications and manuscripts should be sent to the Man-
aging Editor, the University of Chicago. Subscriptions and other
matter should be addressed to the Biological Bulletin, Prince and
Lemon Streets, Lancaster, Pa.

MBL WH01 LIBRARY

b)H 1717