•V

<\

BIOLOGICAL BULLETIN

OF THE

firmrine Biological laboratory

WOODS HOLE, MASS.

lEMtorial Staff

E. G. CONKLIN — Princeton University.

JACQUES LOEB — The Rockefeller Institute for Medical Research.

GEORGE T. MOORE — The Missouri Botanical Garden.

T. H. MORGAN — Columbia University.

W. M. WHEELER — Harvard University.

E. B. WILSON — Columbia University.

Managing EMtor

FRANK R. LILLIE — The University of Chicago.

VOLUME XXX.

WOODS HOLE, MASS.
JANUARY TO JUNE 1916

PRESS OF

THE NEW ERA PRINTING COMPANY
LANCASTER, PA.

CONTENTS OF VOLUME XXX.

No. i. JANUARY, 1916.

PAGE.

WODSEDALEK, J. E. Causes of Sterility in the Mule . ... I
KANDA, SAKYO. Studies on the Geotropism of the Marine Snail,

Littorina littorea 57

KANDA, SAKYO. The Geotropism of Freshwater Snails 85

Xo. 2. FEBRUARY, 1916.

LEWIS, MARGARET R., AND ROBERTSON, \\'M. REES B. The
Mitochondria and Other Structures Observed by the Tissue Cul-
ture Method in the Male Germ Cells of Chorthippus curtipennis

Scudd 99

ROBBINS, W. J. Notes on the Physiology of Fucus spermatozoids . 125
MOORE, A. R., AND KELLOGG, F. M. Note on the Galvanotropic

Response of the Earthworm 131

WILDER, HARRIS H. Palm and Sole Studies. I to IV 135

NEWMAN, H. H. Heredity and Organic Symmetry in Armadillo

Quadruplets 173

No. 3. MARCH, 1916.
WILDER, HARRIS H. Palm and Sole Studies. \ and VI .... 211

No. 4. APRIL, 1916.

WILLIS, H. S. The Influence of the Nucleus on the Behavior of

Amceba 253

SUMNER, F. B. Notes on Superfetation and Deferred Fertilization

Among Mice 271

GOODALE, H. D. Further Developments in Ovariotomized Fowl . . 286

BACHHUBER, L. J. The Behavior of the Accessory Chromosomes
and of the Chromatoid Body in the Spermatogenesis of the

Rabbit 294

iii

iv CONTENT-

No. 5- MAY. 1016.

LlLLIE, \\.\i I'll S. The Theory of Anifsthesia 311

GLASER, R. \\".. AND CHOI>MAN, J. \\". Th<- Xnture of the Poly-
hedral liodirs round in Inserts . . 367

CHILD, C. M. A\i<il Susccptihility Gradients in the Early De-

relopment of the Sea I 'rehin 391

No. 6. JUNE, 1916.

I-'.i'^liteenth Annual Report of the Marine Biological Laboratory... 407
M. \CKLIN, C. C. A mitosis in Cells (iroicin^ in Vitro 445

Vol. XXX. January, 1916. No. i

BIOLOGICAL BULLETIN

CAUSES OF STERILITY IN THE MULE.1

J. E. WODSEDALEK.

(NINE PLATES.)

CONTENTS.

I. Introduction i

II. Comparison of the Horse and the Ass : 3

III. Characteristics of the Mule 7

IV. Material and Methods 9

V. Appearance of the Testicular Tissue 9

VI. Spermatogonia 10

VII. Primary Spermatocytes 15

1. Early Prophase 15

2. Synapsis 17

3. Abnormalities in Mitosis 23

VIII. Giant Cells 32

IX. Multinucleate Cells 33

X. Destruction of the Germ Cells 34

XI. Summary 36

INTRODUCTION.

For several years the author has been contemplating to make a
study of the germ cells of the mule with the aim of offering a
thorough explanation of the causes of sterility in this hybrid,
but it was not until the past year that material for such an
investigation was available. Cytological studies of the sex cells,
especially those of animals, in recent years have added much to
our knowledge of the mechanism of heredity; and such studies
have kept fairly good pace with the comparatively rapid advances
in our knowledge of plant and animal breeding which have been
made since the rediscovery in 1900 of the results of Mendel.
And while this study is now recognized as one of the chief methods
of confronting many problems in genetics, and must be carried

1 From the zoological laboratories, University of Idaho.

2 J. E. WODSEDALEK.

on in correlation with animal and plant breeding before any
adequate explanation of the results of hybridization can be ob-
tained, the study of the germ cells in hybrids themselves has thus
far been somewhat neglected.

Guyer ('oo) in his excellent work on the spermatogenesis of
normal and hybrid pigeons says: "It is a remarkable fact that
no attempt has been made so far to investigate carefully the
spermatogenesis or ovogenesis of hybrid forms. In all the mass
of literature discussing or touching upon hybridism, so far as I
have been able to ascertain, there has been in no instance an
approach to a thorough study of the germ cells. Yet almost
every writer states that through the study of hybrids, we have
perhaps the best opportunity for gaining a clew to many of the
most vital points in the great problem of heredity. A number
of investigators have remarked that in certain instances the
anthers, ovary, or testes as the case might be, were defective,
and have let the matter go at that."

It is even more remarkable now, since fifteen years have
elapsed and no further attempt has been made to investigate
carefully the spermatogenesis or ovogenesis of other hybrid
forms. Professor Guyer's work on hybrid pigeons which was
published in 1900, the same year in which Mendel's investigations
were rediscovered, so far remains as the only outstanding piece
of work on that particular subject. It is all the more remarkable
when we consider the fact that such a great deal of cytological
work has been done during the past fifteen years and that by far
the largest bulk of the work on hybridization has been done
during the same period.

The mule is probably one of the best known hybrids for he is
raised in practically all parts of the civilized world. He has
been known for many centuries and has been used more or less
in Europe since the days before Christ, for Varro who wrote in
the first century B.C. refers to mules in Roman agriculture.
Sterility in mules has no doubt been a subject of discussion for as
many centuries but the real nature of the cause of sterility in
these hybrids has heretofore never been carefully investigated.
The spermatogenesis of the horse which was carefully worked
out at the University of Wisconsin Zoological Laboratories a

CAUSES OF STERILITY IN THE MULE. 3

year ago (Wodsedalek, '14), has rendered several phases of this
difficult problem much more intelligible than would have been
possible otherwise. It might be added that certain phases of
this problem would have become even more perspicuous had the
spermatogenesis of the ass also been carefully worked out and
thoroughly understood; but unfortunately the jack material
has so far been unavailable.

COMPARISON OF THE HORSE AND THE Ass.

Hayes ('04) says: "Owing to the extreme want of uniformity
in the gaps left, during the process of evolution, between de-
scendants from similar ancestors, we are unable to lay down any
exact general rules for classification. The inclusion of horses,
asses, and zebras in the genus Equns admits of no controversy,
because they are the only possessors of the distinguishing char-
acteristic of having only one complete hoofed toe on each foot."

While there seems to be no doubt that the horse, ass, and zebra
have descended from common ancestors, at the present time, there
is considerable difference between the horse and the other two
Equidse. The only distinctive difference between the asses and
zebras in general seems to lie in the tiger-like stripes which are
invariably lacking in the ass. The differences between asses
and horses are marked, as can be seen from the following quota-
tion of a detailed comparison by Hayes ('04). "Some of the
following differences between asses and horses are relative, and
others absolute. Most of these differences also exist between
zebras and horses.

"i. The ass has, practically speaking, chestnuts only on the
forelegs, which is a peculiarity that is met with in certain breeds
of horses (p. 319). In some cases, the ass has vestiges of chest-
nuts on his hind legs. The chestnuts and ergots (p. 319) of the
ass are much thinner than those of the horse.

"2. The ass has a tufted tail, somewhat like that of an ox, an
erect mane, and no forelock. The horse, with the exception of
Prjevalsky wild horse (p. 640), has a bushy tail, drooping mane,
and a forelock, when they have been allowed to grow. The dif-
ference in the mane is due to the length of the hairs of the part.
In almost all breeds of horses the hairs of the tail grow long from

4 J. E. WODSEDALEK.

the root of the dock. In the ass they do so only as they approach
the end of the dock.

"3. As a rule the ass has five loin vertebrae, and the horse six
(p. 422). I have never heard of an instance, in the domestic
ass, of the number of these bones exceeding five. If, however,
we examine the skeleton of the Mountain Zebra which is in
the museum of the R. C. S. Lincoln's Inn Fields, we shall see
that it has six loin vertebrae. The number of these bones is sub-
ject to variation in all vertebrates.

"4. In the horse, the lachrymal duct, which is the canal that
conveys tears from the eye on each respective side into the nostril,
has its opening near the inferior commissure of the nostril, and
on the line of union between the dark-colored skin and the pink
mucous membrane. In the ass and mule, it is situated at the
inner face of the outer wing of the nostril. This orifice is some-
times double.

"5. In the ass, the false nostril extends higher up than in the
horse.

"6. The male ass has two rudimental teats in the form of small
tubercles. They are usually absent in the horse.

"7. The vocal sounds of the ass (braying) are produced in a
different manner from those of the horse (neighing). . . . We
may, therefore, conclude that braying can be performed only
during strong contractions of the muscles of the abdomen and
chest. It is evident that this muscular contraction is not re-
quired in the neighing of the horse.

"8. In the ass, the deep depression at the base of the epiglottis
is covered by a thin membrane, which is capable of vibrating,
and which is wanting in the horse. It may have some influence
in causing the voice of the ass to differ from that of the horse. . . .

"9. The ass hardly ever has any irregular markings on its
coat such as a 'star,' 'blaze,' 'reach,' or 'stockings,' all
of which are very frequent among horses. A small star, on one
or two occasions, is the only mark of the kind I have ever seen
in the ass, of which animal I have not had much experience.

" 10. I believe I am correct in saying that the color of the ass
is never of a bright bay, chestnut, red or blue roan, or nutmeg
gray. I have seen mules of an iron-gray color; but have not

CAUSES OF STERILITY IN THE MULE. 5

observed it in the ass. This conservatism in color and freedom
from irregular markings, shown by the ass, is very remarkable;
considering how greatly the coat of the horse varies in this respect
and that the ass has, in all probability, been longer under the
influence of domestication than the horse.

"n. The ass is higher over the croup, than at the withers
which is a peculiarity that tends to make his withers appear unduly
low (p. 241). The spines of the vertebrae at the withers are only
a little shorter in the ass than they are in the horse. As a rule
horses are higher at the withers than they are at the croup.

"12. The horse's dock is thicker, stronger, and shorter than
that of the ass.

"13. The horse, on each side of his croup and covering his
pelvis, has, underneath, and closely adhering to the skin of the
part, a thick and extremely dense layer of connective tissue, which
is so close and hard, that it looks like horn, when the skin has
been tanned and dried. These two patches of thickened skin
are separated from each other about four or five inches apart, so
that there is a strip of skin of ordinary thickness running down
the croup towards the tail. . . . The 'shell' is connected to the
skin so closely that the two form one piece; although their re-
spective consistencies are different. If a section be made through
the hide their line of union may be readily seen. In the ass, the
'shell' is not confined to the skin that covers the pelvis; but also
extends over the ribs, which are consequently not as sensitive
to the effects of blows as are those of the horse. . . .

"14. The ass has no tufts of hair at the fetlocks (p. 290).

"15. Messrs. Tegetmeier and Sutherland appear to have been
the first to note the difference between the respective periods of
gestation of asses and horses; the former period being twelve
months; the latter, eleven months.

" 1 6. The foot of the horse is more highly specialized than that
of the ass. One of the best marked evolutionary changes in the
equine foot, was the gradual curtailment of its posterior bearing
surface (frog and sole), until, in the horse (Fig. 382), the length
of this bearing surface is not much greater than its width, and is
included between the heels and the 'toe' of the hoof. In the
ass, this bearing surface is relatively longer than in the horse,

6 J. E. WODSEDALEK.

and extends some distance beyond the heels to the rear (Fig. 443).
Also, the shape of the foot of the ass, from above downwards, is
more or less cylindrical; and that of the horse is more or less in
the form of a truncated cone.

"17. The teeth of the horse are more highly specialized than
those of the ass. . . . We can see in Fig. 643, that the gradual
lengthening of the crowns of the molar teeth is a well-marked
feature in equine evolution ; and from a study of the mechanism
of equine dentition, we learn, that for purposes of food-prehension
and mastication, the growth of the incisors must be proportionate
to that of the molars. Hence, we may assume that the crown of
the incisors, like those of the molars, have gradually lengthened
during the evolution of the horse; and consequently that the
acuteness of the angle made by the upper and lower incisors of
the horse of today increases with age, to a greater extent than
it did in the case of his ancestors. On turning to the domestic
ass, we find that in old donkeys (Figs. 448, 449 and 450), the angle
in question is greater in the horses of similar ages (Figs. 445, 446,
and 447). Consequently we may infer that the teeth of the
domestic ass are of an older type than those of the horse. Also,
the incisors of the donkey are relatively narrower than those of
the horse." In addition it might be said that the ears of the ass
are much larger and longer.

All of the differences enumerated above are skin-deep, or
anatomical in nature. The present study has revealed another
difference, a cytological one, which is probably the most impor-
tant, for it lies in the finer structure of the cells of these animals
and no doubt is at the base of all of the differences perceptible
to the naked eye. It was learned that the cells of the horse con-
tain thirty-seven chromosomes (Wodsedalek, '14). The present
study shows that the cells of the mule possess fifty-one chromo-
somes. This would suggest that the cells of the ass possess about
sixty-five chromosomes, if nothing unusual happens in the de-
velopment of the hybrid, thus making the remarkable difference
of twenty-eight chromosomes between the cells of the horse and
those of the ass. This matter is considered at length in the latter
part of the paper.

CAUSES OF STERILITY IN THE MULE. 7

CHARACTERISTICS OF THE MULE.

The mule is a hybrid having for sire a jackass or male ass,
commonly termed a jack, and a mare for dam. If, however, a
stallion be bred to a she-ass which in the United States is called
a "jennet" and in England a jenny, the result is a hybrid known as
a hinny. It seems to be a well-established fact that no special
distinction with respect to appearance or conformation can be
made between mules and hinnies, because in both of these hy-
brids of both sexes the proportional resemblance to the horse and
ass is of infinite variety. Mares are generally taller than jennets,
hence the mules as a rule are of a greater height than hinnies,
which appears to be the only difference between these two hybrids.

There seems to be a great deal of difference of opinion as to
which one of its parents the mule most resembles. For example,
Hayes ('04) says: "The coat, mane, tail, feet, and voice are more
or less intermediate between those of the horse and those of the
ass. The hind chestnuts in some cases are well developed, as
in the ordinary horse; in others, they are in a comparatively
vestigial form, and one or both may be absent. These hybrids
"resemble the ass more than the horse in their placid temper,
great adaptability to work, and longevity. In carrying or draw-
ing loads they are superior to horses of the same weight, and they
can obtain the necessary energy from food which horses cannot
digest, as for instance, the reeds on which the mules of the south
of France live. I have experimentally established the fact that
their digestive power is higher than that of horses" (Sanson).

Lydekker ('12) says; "In the case of both mules and hinnies
the general build and appearance of the animal accord with the
type of the sire, although in the manner of bodily size the dam is
followed. Mules are therefore asinine in appearance, although
with a more horse-like tail, and relatively large ears; whereas,
the more horse-like hinny is small. If, however, females of the
great Poitou ass were to be utilized for hinny-breeding, the
progeny would probably be of larger stature. One exception to
the ass-like character of the mule is that it lacks the white belly
of its male parent."

Wilcox ('07) says: "It is not true, however, as sometimes
asserted, that the mule eats less than the horse. On the contrary

8 J. E. WODSEDALEK.

the mule has an excellent appetite. However, the mule can go
longer without food and can live on coarser and more unpalatable
food than could be expected to give results with the horse.
Contrary to a prevailing belief also, the mule is equally suscept-
ible to various diseases as the horse. In the Philippines our
mules suffered as much as horses from surra. Glanders has
always been dreaded as one of the scourges of the army mule.
Even colic and other digestive troubles familiar to horse raisers
also occur among mules."

Plumb ('06) says: "The characteristics of the mule partake
of both sire and dam. There is the long ear, slender body,
tufted or slightly haired tail, and small, slender foot, and braying
voice of the ass."

"The temperament of the mule is quiet and patient, while for
steadiness under the collar and hard pulling he has no equal in
the equine world. . . . Horses are more nervous and uncertain
in temperament than mules, and are more subject to fright and
consequent runaway.

"The endurance of the mule is remarkable. . . . Mules usually
live to a greater age than horses, and perform their work with
regularity and on less feed, a most important point in their favor.
Cases are recorded of mules living to seventy years of age. . . .
The resistance of the mule to disease, its activity, sureness of foot,
docility, and easiness of keep, have resulted in its finding much
favor in the army service." The sureness of foot is no doubt
inherited from the ass.

That mules and hinnies are sterile is a well-known fact. And
if degeneration of the sex cells of all of these hybrids is as pro-
nounced as it was found to be in the material studied in this in-
vestigation— and in all probability it is — the cause of sterility
will remain obvious. A few cases of fertile mules have been
recorded, but careful investigations show that such records are
not authentic and practically all of the writers on mules regard
them as unreliable.

Lydekker ('12) says: "With the possible exception of a few
instances in which the female is stated to have produced off-
spring, the mule is sterile; as, indeed, might be expected to be
the case when the difference between its parents is borne in
mind."

CAUSES OF STERILITY IN THE MULE. 9

Hayes ('04) in remarking about the matter of sterility of
mules says: "Scientific research has amply proved that mules
and hinnies are absolutely sterile, although cases are on record
of induced lactation occurring in females of this kind. . . .
Accounts not infrequently appear in the American and other
papers, of mules which are seen suckling young, and the conclusion
is at once arrived at that these young are the offspring of the
animals that are supporting them, but it may be regarded as
perfectly certain that they are merely adopted foals, which by
their endeavors to suck female mules have developed in the latter
abnormal lactation."

MATERIAL AND METHODS.

The material for this study was obtained through the courtesy
of my friend, Dr. J. W. Kalkus, professor of histology and path-
ology at the State College of Washington, Pullman. Dr.
Kalkus made every possible effort to secure the material for my
studies and generously placed at my disposal material obtained
from three animals.

The hybrids were about two years old at the time the material
was removed from them. It is probable that all three of the
mules were by the same sire. The tissue was fixed in Flemming's
fluid, and stained with Heidenhain's iron hematoxylin. The
sections were cut from five to nine microns thick.

•

APPEARANCE OF THE TESTICULAR TISSUE.

The first noticeable difference in the testicular structure of the
mule and the horse is that the seminiferous tubules of the mule
are in general much smaller than those of the horse. There is
considerable variation among the tubules of the mule. The di-
ameter of some tubules is fully twice that of others, and all
gradations between the two extremes may be found. The large
tubules are characterized by large clear areas in their lumens.
This clear space diminishes as a rule in proportion to the di-
ameter of the tubules. In most of the small tubules the cells
occupy all of the lumen. This structure, however, is not always
the case in the respective tubules, for occasionally we find large
tubules with solid lumens and on the other hand small tubules
with clear lumens.

IO J. E. WODSEDALEK.

The difference in the interstitial cells is quite obvious. They
are much larger, much more numerous, and more regularly dis-
tributed in the horse than they are in the hybrid. The nuclei in
the former are large and usually contain one or two large nucleoli,
and several small karyosomes; while those of the mule are small
and contain many small karyosomes, and the large nucleolus is
not as constant as it is in the horse. The cytoplasmic content
of these cells in the horse, too, is larger and not as variable in size
as in the mule.

SPERMATOGONIA.

The spermatogonial cells of the mule are considerably larger
than those of the horse in the corresponding stages. They gener-
ally occupy the usual position next to the tubule wall, though oc-
casionally they are crowded out of this position by the numerous
nurse cells and pushed further in toward the lumen. They occur
in relatively fewer numbers in the hybrid than they do in the
horse. In cross sections of the tubules they very seldom form a
complete ring within the tubule wall, and are never found in a
crowded condition neither along the entire tubule wall nor in
small areas, as is often the case in normal mammalian tissue. The
cells are usually far apart, and it seems safe to state that in a
large majority of the tubules only about twenty to thirty per cent,
of their inner surfaces is covered with these cells. It seems
equally as correct to say that in cases where the spermatogonial
cells are most abundant they never exceed covering more than
sixty per cent, of the tubule's inner surface; and it might be said
that such extreme cases are rarely found. Many of the tubules
contain only a comparatively few spermatogonial cells and not
infrequently tubules are found in which no cells can be identified
as such. The cells occupying the position usually occupied by the
spermatogonia greatly resemble the ordinary nurse cells.

During the prophase of the spermatogonial cell the nucleus is
round and contains a large nucleolus which is usually heart-
shaped, several small karyosomes, a large number of small
granular masses and extremely thin linen strands (Figs. I and 2).
In this respect the cells resemble those of the horse in the cor-
responding stages, except that they are much larger. No centro-
some can be detected at this stage although a dense mass of

CAUSES OF STERILITY IN THE MULE. 1 1

cytoplasm can often be seen near the nucleus. This is in all
probability the idiozome. These cells have definite walls which
mark them off clearly from the neighboring cells. They are
almost spherical and have the appearance of smear cells, a con-
dition presumably due to the fact that they are not crowded.

As growth proceeds all parts of the cell increase in size. When
the maximum size is reached, dense masses of chromatin make
their appearance which become more and more defined until the
chromosomes of various shapes and sizes are formed. Some-
times the chromosomes are so close together that an accurate
count is impossible; but more frequently they are sufficiently
separated so that many definite counts were possible.

There are fifty-one chromosomes found in the spermatogonia
of the mule. Fifty of these are the ordinary chromosomes or
autosomes, and one is the accessory which is larger than the others.
Fifty-one chromosomes is fourteen in excess of the number found
in the spermatogonial cells of the horse which has, at least in
some breeds, a total of thirty-seven, thirty-six autosomes and
one accessory (Wodsedalek, '14).

A considerable amount of time was spent in an attempt to
ascertain which of the chromosomes are of paternal and which of
maternal origin, or in other words which were contributed by the
jack and mare respectively. And while the author has obtained
some fairly definite ideas in regard to the individuality of the
chromosomes in the mule, he does not feel that any positive state-
ments in regard to the matter can be made until the sex cells of
the ass will have been thoroughly investigated.

It is quite certain, however, that the accessory chromosome
retains its individuality, for its appearance in the mule is identical
to that of the accessory of the horse. The mule no doubt re-
ceived the accessory from the mare in the same manner as it is
contributed by the dam to her male offspring in the horse, in
view of our knowledge of the behavior of this body in the sper-
matogenesis of the horse (Wodsedalek, '14), and particularly
according to our acknowledge of the behavior of the accessory
chromosomes in relation to sex determination in another mammal,
the pig (Wodsedalek, '13). In regard to the other chromosomes
it is definitely known that in the horse they vary considerably

12 J. E. WODSEDALEK.

in size and shape (Wodsedalek, '14), and the same is probably
the case among the chromosomes of the ass. And the fact that
some of the chromosomes of the ass are apparently of the same
size or about the same size as some of the chromosomes of the
horse, makes it rather difficult to discriminate all of the maternal
from the paternal chromosomes in the hybrid. It appears,
however, that most of the large chromosomes in the mule are
paternal in origin. A careful study of the spermatogenesis of
the ass should throw considerable light on this subject.

It is also definitely known, as was stated before, that the horse,
at least of some breed or breeds, possesses thirty-six chromosomes
besides the accessory, or sex-determining chromosome, in the
spermatogonial cells. The reduced number of the ordinary
chromosomes in this animal is eighteen, which number enter
into the formation of the one type of sperm, and eighteen plus the
accessory or nineteen in the other type. Since the spermato-
genesis of the horse with regard to the accessory (Wodsedalek,
'14) bears such a striking resemblance to the spermatogenesis of
the pig, in which animal the problem of sex-determination was
carefully worked out (Wodsedalek, '13), the author in his paper
on the spermatogenesis of the horse assumed that the sperm cells
possessing eighteen chromosomes are male determining and those
possessing nineteen are female determining. It was further
assumed on account of the behavior of the accessory in the horse
that the oogonia of the dam contain thirty-eight chromosomes,
or thirty-six autosomes and two accessories, while the mature
ova contain the reduced number of nineteen chromosomes, or
eighteen autosomes plus one accessory.

The above figures are in all probability correct, but the actual
count of the chromosomes in the female tissue must be made
before it can be regarded as authentic, regardless of the strong
evidence obtained in favor of the assumption through the com-
parison of the behavior of the chromosomes in both the male
and female germinal and somatic cells of the pig. Great as the
similarity is, the mere fact that the pig is the only case among
the vertebrates in which the relation of the accessory chromo-
somes to sex-determination has been clearly and completely
demonstrated, any remarks based on such a comparison must
necessarily be of a more or less reserved nature.

CAUSES OF STERILITY IN THE MULE. 13

The investigations on the relation of the accessory chromosome,
or group of chromosomes, to sex-determination in at least another
vertebrate, the man, must not be overlooked. And while there
is some difference of opinion in regard to this matter in man and
the results are not entirely conclusive, they should nevertheless
add more strength to the above assumption. Guyer ('10) found
that the spermatogonia of man contain twenty-four chromo-
somes; and that twelve chromosomes appear for division in the
primary spermatocyte of which ten are evidently bivalent and
two accessories. During division ten chromosomes pass to one
pole and ten plus the two accessories to the other, giving rise to
two different types of secondary spermatocytes which eventually
give rise to two types of spermatozoa; the one type containing
ten chromosomes and the other ten plus the two accessories.

Montgomery ('12) confirms the number of chromosomes found
by Guyer but disagrees with him in regard to the accessory.
Von Winiwarter found forty-seven chromosomes in the male, of
which forty-six unite at reduction and form twenty-three, bi-
valents and the accessory remains unpaired. Two types of
spermatozoa are produced, the one containing twenty-three
chromosomes and the other twenty-three plus the accessory, or
twenty-four chromosomes. In the female he had some difficulty
in obtaining the exact number of chromosomes, but his best
counts gave forty-eight, which number fits in with the results
obtained in the male.

The cause of the difference of opinion is probably due to the
fact that Guyer used Negro material in his investigation, while
Von Winiwarter studied tissue obtained from a white man.
In view of the difference between these two human races it is
not at all improbable that a difference in the number of chromo-
somes also exists. The Negro is fully as far removed from the
white man as is the ass from the horse, where a great difference
in the number of chromosomes apparently exists. This vast
difference in number appears to be, at least in part, responsible
for the sterility of the mule. The difference between the white
man and the Negro in regard to the number of chromosomes
according to Guyer and Von Winiwarter is equally as great, but
unfortunately the mulatto is fertile.

!4 J. E. WODSEDALEK.

The number of chromosomes in the spermatogonial metaphase
stage of the mule is, as was stated before, fifty-one, or fifty
autosomes and one accessory. The number in the corresponding
stage of the horse is thirty-seven or thirty-six autosomes and
one accessory; while the reduced number is eighteen in the one
type of sperm and nineteen, including the accessory, in the other.
According to the previous discussion it seems fairly safe to predict
that in the oogonial cells of the mare there are thirty-eight chro-
mosomes or thirty-six ordinary chromosomes and two accessories,
and that the reduced number in the matura ova is nineteen in-
cluding the one accessory. This would suggest that nineteen
of the fifty-one chromosomes including the accessory are maternal
in origin or contributed by the dam, and the remainder or thirty-
two are paternal in origin or contributed by the jack. On the
customary inference that the reduced number of chromosomes is
always one half the number in the spermatogonial and somatic
cells we should expect to find sixty-four chromosomes in the ass,
with the possible addition of one accessory in the jack and two in
the jennet; thus making a total of sixty-five and sixty-six re-
spectively in the two sexes of this animal.

These plausible figures for the jack and jennet would, of
course, be expected only in the event that no cellular abnormalities
occurred in the mule from the time of the fertilization of the
dam's ovum by the jack's sperm, through the succeeding de-
velopmental stages of the hybrid up to maturity. In such a case
the full quota of chromosomes would be handed on to the sper-
matogonial cells of the mule.

The normal mitotic figures of the spermatogonial cells seem
to indicate that probably no such abnormalities occur in the
various preceding stages. But even such an apparently normal
condition cannot be relied upon too strongly. For while the
spermatogonial cells are normal during mitosis, as near as can be
detected, that does not necessarily preclude that nothing unusual
happens in the developmental stages of the hybrid, particularly
at the time of fertilization and early cleavage stages. At any
rate it would not be at all surprising if something unusual did
happen to bring about the fifty-one chromosomes. The need of a
careful study of the chromosomes in the jack is obvious. A

CAUSES OF STERILITY IN THE MULE. 15

•

study of the sex cells in both sexes of the hinney, as well as those
of the female mule and she-ass or jennet, would also add materi-
ally to this immensely interesting problem of sterility of the horse
and ass hybrids.

Fig. 3 represents a polar view of the metaphase stage of a
spermatogonial cell. The fifty-one chromosomes are distributed
throughout the entire plane of the equator and the larger ones,
as a rule, are arranged along the edge. The heart-shaped ac-
cessory can be readily distinguished not only on account of its
shape, but also on account of its large size. It is invariably
at or near the edge of the equator. Fig. 4 is a side view of the
spindle and a representation of one of the many perfect spindles
found in this stage. There are no scattered chromosomes to be
seen; each enters the equatorial plate and divides. The fact is
further evidenced by the perfect divisions which follow. During
the anaphase stage the chromosomes move to the opposite poles.
Figs. 5 and 6 show that there are no stragglers among the chro-
mosomes. The accessory which also divides in this stage, can
practically in all cases, be detected on account of its larger size.
In the telophase stage when the chromosomes are at the opposite
poles the cytoplasm begins to constrict in the center. Soon the
chromosomes become ragged and later disintegrate while the
nuclear membrane makes its appearance. Following the last
spermatogonial division the cell enters into the period of great
growth and represents the early stage of the primary spermato-
cyte.

PRIMARY SPERMATOCYTES.

i . Early Propliase.

The most important phase of this problem lies in the primary
spermatocytes. These cells are the result of the last spermato-
gonial division as in normal forms, and while their behavior is
always interesting and more or less difficult to analyze in any
case of spermatogenesis, it is especially so in this hybrid. From
the time of the appearance of these cells until the period of their
final destruction one is confronted with numerous fascinating
categories.

During the early prophase the primary spermatocytes resemble

T6 J. E. WODSEDALEK.

the early spermatogonial stages, except that the cells are larger
and the linin strands a little more conspicuous. The large
nucleolus is again very distinct and appears as it did in the
spermatogonia. A large, irregularly shaped karyosome is also
usually present, as are a number of small ones. As the cell in-
creases in size the linin strands become more conspicuous and the
chromatin granules more numerous. The karyosomes, too,
become larger and are surrounded by dense masses of chromatin
granules. One of the karyosomes is always large and has from
eight to twelve linin strands radiating from it. Frequently a
few strands can be seen radiating from the small karyosomes.
These radiating strands together with many others, form a
continuous network of linin. Later the chromatin granules begin
to mass around the linin strands and the karyosome begin to
disappear (Fig. 7). Soon all of the chromatin material seems to
be arranged along the linin strands, which take on the appearance
of a network of chromatin threads. The threads are very granu-
lar and vary a great deal in length and somewhat in thickness

(Fig- 7)-

There is no definite arrangement of these threads in the net-
work of the various cells. Each cell seems to have its own hit-
and-miss tangle. In some cells the long threads are more nu-
merous and the short connecting threads are proportionately
fewer in number. In others there are many short components
forming the network. The nuclei at this stage vary considerably
in size.

Synizesis, which is so commonly observed in the horse tissue,
does not occur in the mule. The nearest approach to it is a
small clump of threads which is frequently seen and seems to be
formed in the position occupied by the large karyosome. This
clump persists in many cells until the threads break up into
chromosomes. The mass of chromosomes resulting from such a
clump of threads can also be frequently seen. It is probable that
the clump of threads is brought about by the tangling up of the
linin strands, which extend from the karyosome.

The spireme stage which follows synizesis and synapsis in the
horse is also lacking in the mule. The expanded network takes
its place, and only parts of the network resemble the spireme of the

CAUSES OF STERILITY IN THE MULE. I'J

horse. Even in cases where long threads occur the resemblance
is slight, for invariably the thread is made up of portions of
various thicknesses, while the spireme in the horse is of a more
uniform diameter.

2. Synopsis.

It appears that thus far there was no necessity for the paternal
and maternal chromosomes in the cells of the hybrid to cooperate
to any noticeable extent in functioning, and while some conflicting
tendencies may be in operation, each group mixed with the other,
seems to have gone on performing its functions normally.
Material from both parents undoubtedly prevails in the somatic
cells of the mule as it does in the sex cells. Yet these materials
in the somatic cells, coming from two vastly different animals,
do not in the least interfere with each other physiologically. And
while no somatic tissue of the mule has been studied, any sus-
picion of abnormalities, such as disintegration and decay for
example in the muscle cells, could hardly be substantiated by
our knowledge of the health, strength, endurance, and longevity
of these hybrids. This suggests that developmental processes are
also normal. It is highly probable, then, that since the physio-
logical and developmental processes are normal that the mechan-
ism of cell division during the process of development and growth
is also normal ; the same as it is in the spermatogonial cells where
perfect mitotic figures appear and all of the chromosomes divide.

The real conflict ensues during the various stages of the primary
spermatocyte as is so plainly evidenced by the numerous ab-
normalities which occur in these cells. In fact the conflicting
tendencies are so great that the destruction of each cell is in-
evitable, and no spermatozoa are produced, causing the hybrid
to be sterile. Up to this stage the paternal and maternal plasmas
evidently retain their individuality and would undoubtedly
continue to do so were it not for the wonderful phenomena of
reduction, which necessitates a fusion of the chromatin com-
ponents of the germ cells at this stage of maturation, as is known
in many of the normal forms in which spermatogenesis was care-
fully studied.

It is now a well-known fact that in maturation of the germ
cells a reduction of the ordinary number of chromosomes to

18 J. E. WODSEDALEK.

one-half takes place. Before this actual reduction, there is usually
a so-called pseudo-reduction, in which the chromosomes unite in
pairs so that when the cell is ready for division, although only
half of the regular number of chromosomes appear, each is really
double or bivalent and equivalent to two of the single or univalent
type.

In this hybrid, as was suggested before, it may be supposed
that in the somatic cells and in the spermatogonia, the chro-
mosomes from the paternal and maternal animals lie side by
side and carry on the customary functions of the cells; but when
it comes to an actual fusion of the chromosomes to form the
bivalent type necessary for reduction, the incompatibility of the
two plasmas renders the pairing difficult and incomplete, or
prevents it entirely.

In the horse pairing or pseudo-reduction takes place during
the period of synizesis. This period is characterized by massing
of the chromatin threads in the center of the nucleus, and later
the nuclear wall expands and the entire mass passes to one side
of the nucleus leaving a large clear area in the remaining portion
(Wodsedalek, '14). This condition is much the same as in the
pig (Wodsedalek, '13), except that in that animal the nucleoli
were invariably found within the mass of threads and in a position
nearest to the nuclear wall; while in the horse the nucleoli are
almost invariable within, or next to the clear area. Shortly
after the collapse of the chromatin material, the threads pair
and appear in apparently half the original number and twice
as thick. Later they expand and again occupy the entire contents
of the nucleus. Then follows the period of growth, and eventu-
ally the threads break up into the bivalent chromosomes which
appear in half the original number found in the spermatogonia,
plus the accessory.

The pairing of the various components, in the horse, takes
place simultaneously, and when the spireme appears it possesses
a fairly uniform diameter throughout. This fact indicates
clearly that each portion of the spireme is of a bivalent nature,
for should some parts remain unpaired or univalent in nature,
such parts could be readily discerned by the sudden and notice-
able decrease in their diameters. No pairing of parts in the horse

CAUSES OF STERILITY IN THE MULE. IQ

was ever observed after the synizesis period, and no chromosomes
ever seem to remain unpaired in the primary spermatocyte, with
the exception of the accessory which is always unpaired and
passes undivided to one pole or the other.

In the mule the period of synapsis is extremely fascinating,
and the facts presented here will no doubt stimulate many
investigators in the study of hybrid sex cells. The author con-
siders himself exceedingly fortunate in being able to obtain
hybrid material which was in such excellent condition for this
investigation, especially since the hybrid in question is the off-
spring of two vastly different parents. Thus far, it appears that
there is no case on record among the vertebrates, showing that
this important step in the process of maturation of the sex cells
of hybrids has been fully and conclusively demonstrated; and
since this important phenomenon has never been clearly demon-
strated in a hybrid of this nature, and since this material given
to me by Dr. Kalkus appeared promising from the start, the
author spared no time and energy in making a careful and pro-
longed study of this particular phase in the maturation of the
sex cells of the mule.

It might be stated at the outset that little fusion of the chro-
matin material takes place, and that there is no definite time for
such fusion. As soon as the various chromatin threads in the
network of the nucleus become well defined, there appear in-
dications of pairing of some of the threads. This process con-
tinues even after the network breaks up and the chromosomes
make their appearance (Figs. 8, 9, and 12). Fig. 8 represents an
early stage in which two or three of the threads had fused, and
a beautiful case of two long threads lying side by side almost
across the entire diameter of the large nucleus. Fig. 9 shows
a cell in which many more threads had fused and some are still
in the process of fusing. Fig. 10 shows a case where only a
small amount of fusion had taken place before the parts of the
chromatin network had become loose; but the process of pairing
seems to have continued in two or three instances regardless of
the fact that some of the chromosomes were in the stage of dis-
integration. Fig. ii represents a more advanced case, where
apparently a great deal of the chromatin material had fused

2O J- E. WODSEDALEK.

before the network broke up and still there are indications of
pairing. Cells of this nature in which a great deal of fusion had
apparently taken place, invariably show more or less pronounced
indications of decay, and the question arises as to whether this
unusual amount of fusion is due to the condition of decay or
whether the degeneration sets in because of the unusual amount
of fusion. It appears, however, that the great amount of fusion
is caused by the existing degenerate condition of the cell in
general, for invariably masses of chromatin material bearing no
resemblance to normal chromosomes or threads are present in
these cells. Fig. 12 shows a still further advanced stage, in
which conjugation continues.

There is no regularity in the amount of fusion that takes place
in the various cells. In some cells many more chromosomes
conjugate than in others. This fact was apparent during the
various stages of the comparatively long synaptic period, and
became more obvious when numerous accurate counts of chro-
mosomes were made in cells in which all indications of pairing
had ceased. The fact that the nucleus expands enormously
during the "spireme stage" was a great aid in making many
accurate counts. The chromosomes, as a rule, were well segre-
gated with only partial overlapping. Here and there were tubules
with cells in exceptionally good condition for this particular in-
vestigation, thus rendering possible many definite counts. The
number of chromosomes in the different cells at this stage vary
considerably. The smallest number found was thirty-four and
the largest was forty-nine. This seems to indicate without a
doubt that in the few cases where only thirty-four chromosomes
could be counted that thirty-two of the fifty-one univalent
autosomes had fused. In the few cases where as many as forty-
nine counts were made it appears equally as certain that only
four of the fifty-one chromosomes had fused. It must be re-
membered, however, that these two extreme counts were ob-
tained in only a very few cases and stand out as exceptions. All
other counts between thirty-four and forty-nine were obtained,
but by far the greater majority of counts lie between forty and
forty-five, with a fairly good number of thirty-eight and forty-six.

A very interesting as well as important feature about the

CAUSES OF STERILITY IN THE MULE. 21

chromosomes at this stage is the fact that the bivalent ones can
as a rule be readily distinguished from the univalents, and the
loss of univalents can be accounted for by a proportional increase
of the bivalents. If, for example, forty-one univalent chromo-
somes can be detected, it means a loss of ten others. This
loss of ten univalent chromosomes can usually be accounted for
by the presence of five large or bivalent chromosomes, making a
total of forty-six. If only thirty-one chromosomes can be
counted, which means the disappearance of twenty univalents
as such, one can usually find ten large bivalent chromosomes,
making a total of forty-one. This part of the problem was so
fascinating that the chromosomes of hundreds of cells, in which
they were well scattered, were carefully counted and an attempt
was made to account for the loss or gain, depending on the type
of chromosomes that were considered first. In some cases those
which appeared to be univalent were counted first and then the
bivalents. In other cases those which appeared to be bivalent
were counted first and then the univalents, and it is surprising in
what a large percentage of cases the necessary number of fifty-
one, in terms of univalents, was actually obtained.

Occasionally a discrepancy of one or two, and in rare cases,,
three bivalent chromosomes, was noted, or in other words, it was
sometimes impossible at first sight to distinguish all of the
bivalent chromosomes. Even in cases when the individual chro-
mosomes were carefully examined in regard to their single or
double nature, one could not always be positive whether he was
dealing with a single or a compound body. This fact, however,
should not be so very surprising because two of the smallest
type of single chromosomes may fuse to form a bivalent one of
practically the same size as some of the larger univalent chromo-
somes. The matter suggests rather strongly that it is not
always the same chromosomes that fuse, even in cases where the
same counts are obtained. On the other hand there is a possi-
bility of some of the large chromosomes being tri- instead of
bivalent, since the fusion of three threads, as well as three chro-
mosomes, was observed in several cases (Fig. 12). In cells in
which decay is noticeably under way it frequently occurs that
many chromosomes fuse together and form a large body, but this
cannot be regarded as fusion in the same sense as synapsis.

22 J- E. WODSEDALEK.

It can now be clearly seen that great variation exists in the
amount of fusion which takes place in the chromatin material
of the various cells in this stage of the maturation process. In
some cells fully eight times as many chromosomes pair as in
others. The question at once arises why should such a decided
variability exist in this respect. It is not very probable that the
various cells resulting from the last spermatogonial divisions
differ so greatly in their make-up, as no such variability was
observed in the spermatogonial cells themselves. The fate of
the different primary spermatocytes with respect to the extent
of pseudo-reduction of their chromatin material is evidently
determined at the time of the formation of the network of the
linin threads, for these linin threads seem to persist as the axes
of the later chromatin threads which eventually break up into
chromosomes.

The nature of the cause of the linin strands connecting with
each other in such confusion as they are found, must necessarily
remain a matter of conjecture at present. The same force which
in normal sex cells at this stage causes the linin strands to come
in contact at their free ends, thereby forming long threads, or
loops, or a continuous strand, is apparently brought into action
in the hybrid cells but not with the same precision. The fact
that the two plasmas are so different apparently causes the
strands to connect with each other at any point and at any angle,
instead of at their free ends only, and thereby forming a con-
tinuous network rather than long strands. The tendency of the
strands to connect with others, however, must be well pronounced
for very seldom can any free ends be seen in the earlier stages
(Figs. 8 and 9). Later on, of course, free ends make their ap-
pearance but these are caused by the breaking up of the network
in parts rather than original failure at fusion. In some cells the
end to end fusion of the original components is more pronounced,
since many long threads appear in such cases. However, not a
single case was seen in which several of the cross-connecting
pieces were not present, which condition was never observed to
occur in the corresponding stages in the sex cells of the horse.

The fact that the ends of the threads are fused on to some
other part of the chromatin network seems to be a drawback

CAUSES OF STERILITY IN THE MULE. 23

in their fusion. Frequently two threads, which have some affinity
for each other, and lie in a position parallel to each other and are
not too taut, fuse along the central portion with the ends remain-
ing in the shape of V's, because the two threads are attached at
each end some distance apart. At times when parts having an
attraction for each other are so arranged as to form an acute
angle, fusion begins at the point of the angle and continues a
short distance, leaving a V at one end. The ends remain unfused
at that point, until the network breaks up into chromosomes,
when the temporary connections of the two chromosomes in
question, too, are severed, and thereby complete their fusion.

There is also considerable evidence that actual migration for
purpose of fusion of some parts of the net-work takes place before
the connections are entirely severed, and even before there are
any noticeable indications of such connections separating, for
frequently we find a bivalent thread with both ends still attached
to other threads.

3. Abnormalities in Mitosis.

The abnormalities in mitosis bear a striking resemblance to
the abnormalities found in the sex cells of hybrid pigeons by
Guyer ('oo). In speaking of abnormalities in pigeons, Dr.
Guyer says: 'The abnormalities in mitosis are in the nature of
multipolar spindles and asymmetrical division and distribution of
the chromosomes (Figs. 28-39). These may exist independently
one of another, or both may occur together in the same cell.
They are more pronounced in sterile birds but may at times be
seen in the fertile forms. In very many of the division of primary
spermatocytes one or the other, or both of these phenomena are
seen. It is a curious fact that the multipolar spindles seem to
be confined largely to the primary spermatocytes, and one is
prompted immediately to associate the fact with the pseudo-
reduction or formation of bivalent chromosomes which occurs
normally at this stage of spermatogenesis. The irregularities
in chromatin distribution are also seen for the most part in the
primary spermatocytes."

The abnormalities in the mule are also in the nature of multi-
polar spindles and there are attempts at asymmetrical division
and distribution of the chromosomes; but in addition to these

24 J- E. WODSEDALEK.

types of abnormalities we have cases of scattered chromosomes
which do not enter into the spindle at all (Figs. 16-36). Guyer's
suggestion regarding the curious fact that the multipolar spindles
seem to be confined largely to the primary spermatocytes prompts
one immediately to associate the fact with the pseudo-reduction
or formation of bivalent chromosomes which occurs normally at
this stage of spermatogenesis, is considerably strengthened by
the results of this investigation on the mule, since a thorough
study of the entire phenomena of pseudo-reduction was possible
in this hybrid.

The diverse types of multipolar spindles and other abnormal-
ities that occur in the primary spermatocytes of the mule are
shown in Figs. 16-36. Fig. 16 represents a polar view of a
metaphase stage. Twenty-eight chromosomes, some bivalent
and some univalent, can be seen in the equatorial plate, while
sixteen all apparently univalent, together with the accessory are
scattered about in the cytoplasm. Fig. 17 shows a meager
attempt at spindle formation. A bunch of chromosomes is
found at the equatorial plate and only a few spindle threads
could be seen and those appeared to be broken up. There was
no sign of the centrosomes being present. The other chromo-
somes, both bivalent and univalent, are scattered about. The
cell was no doubt in the process of decay. A trivalent chromo-
some can be seen at the lower part of the figure. Trivalent
chromosomes are not uncommon but the three components
can seldom be seen at this advanced stage. They usually fuse
together, forming a large chromosome, frequently spherical in
shape. Especially is this true when the cell is in the process
of deterioration. The large spherical body at the upper part
of Fig. 1 8 was probably formed in this way, though it is possible
that it is composed of more chromosomes.

Fig. 1 8 represents a cell similar to that shown in Fig. 17.
These cells were found together with many other similar cells
in the same tubule, and several tubules bearing such type of
cells were observed. All of these cells were characterized by a
large clump of chromosomes, which was presumably an attempt
at plate formation, and a large number of chromosomes, uni-,
bi-, tri-, and even quadrivalent, are scattered about the entire

CAUSES OF STERILITY IN THE MULE. 25

contents of the cell. The cytoplasm of these cells was invariably
degenerated ; the remains being closely assembled about the large
clump of chromosomes, leaving the periphery of the cell clear
in appearance. The cell wall, however, retained its normal out-
line up to this stage of degeneration, which 'seems to suggest
that the clear space was occupied by a fluid. While there were
indications of spindle threads more or less pronounced in some
cells, many of the cells showed no signs of the presence of such
threads (Fig. 18). In such cells they either disappeared or had
never been formed. It is a curious fact that such type of cells
usually appeared in large numbers in the same locality and that
only several tubules bearing them were found, though occasionally
single individuals of that type were found in other tubules. The
heart-shaped accessory could usually be seen somewhere off by
itself (Fig. 1 8).

Figs. 19-21 show a common type of freak in which the spindle
is bipolar and a number of chromosomes did not find place in the
equatorial plate but remain scattered about, either somewhere on
the spindle or in the cytoplasm. Figs. 19 and 20 each show the
large accessory at one pole. In Fig. 21 the accessory could not
be easily detected.

Fig. 22 shows an uncommon type of unequal division in which
five large chromosomes have passed over to one pole, while about
thirty-six remained in the original place. This indicates that
the total number of chromosomes in that cell is forty-one and
that ten of them, in view of the previous evidence and discussion,
must necessarily be bivalent. Each of the five at one pole on
account of their size appear to be bivalent. One of them, in
which the two parts have not fused along their entire lengths,
obviously shows its bivalent nature. Further evidence that
these five are double can be based on the fact that only about
five of the chromosomes in the big group appear to be double.
This shows then that no chromosomes have divided, but that
five bivalent ones were pulled over to one pole in their entirety.
Each of these groups without further division would in all prob-
ability, if decay did not set in too early, form a separate nucleus.
And it can be safely assumed that the binucleated condition of
cells sometimes seen in this tissue, in which one nucleus is much
larger than the other, was brought about in this fashion.

26 J. E. WODSEDALEK.

Fig. 23 shows still another type of cell with a bipolar spindle
or an attempt at one. A ring of ten chromosomes can be seen near
one pole and partly pushed out of the cell through the cell wall
which had been broken at that place. Judging by the conver-
gence of the spindle threads in that half of the cell one would
conclude that these ten chromosomes never occupied a position
in the plate, and it appears as though this was an attempt on the
part of the cell to get rid of some of the chromatin material which
did not harmonize with the other parts of it. The clump of
chromosomes was apparently repelled by the others with suf-
ficient force to cause them to break through the cell wall. Only
about twenty other cases similar to the one just described were
observed in all of the tissue studied. However, a number of
these the author is inclined to regard, on account of circumstan-
tial evidence, as being due to methods of technique, though
even in the more suspicious cases he is not absolutely positive
that such in the real cause. In at least a dozen of cases there
seems to be no doubt but that the expulsion of masses of chro-
mosomes is actually due to the activity of the cell itself. Among
these twelve cases a regular series from partial to total expulsion
of a mass of chromosomes can be demonstrated. The number of
chromosomes comprising such a discarded mass varies from five
to nineteen.

The assumption that an effort is made on the part of some cells
to discard some of the chromatin material is further substantiated
by the fact that in many cells a mass of chromosomes was seen
closely applied to the inner side of the cell wall. At times the
cell wall bulges out at such a point and has the appearance of
being quite taut in the immediate vicinity of the chromosomes,
giving the impression that some repelling force is acting upon
them. The most notable evidence in favor of this assumption
lies in the morphology of the cell, represented in Fig. 24. The
unfortunate thing about this matter is, of course, that only a
single cell of its kind, the one here represented as accurately as
possible, was found even after a long and diligent search for
more was made. It may at first seem very unscientific or even
ridiculous to attach so much value to a single cell. Nevertheless,
the significance of the structure of this cell, lonely as it is, must

CAUSES OF STERILITY IN THE MULE. 2."]

not be underestimated. The cell is undoubtedly a primary
spermatocyte in the telophase stage. The cytoplasm has failed
to divide. This is in accordance wiih all other observations
made on primary spermatocytes in this hybrid, since not a single
case was noted in which the cytoplasm showed even the slightest
indications of division or construction.

Just exactly what the nature of the cell was in its metaphase
stage with regard to the arrangement of the chromosomes can
not be definitely stated. One thing seems fairly certain, and
that is that no division of individual chromosomes took place.
For the number of chromosomes can still be fairly accurately
determined since they had become only partially disintegrated.
A count of the chromosomes in the various parts of the cell
reveals a total of about forty-three, fourteen in the oblong
nucleus, ten in the spherical one, three in the cytoplasm, and
sixteen in the small spherical nucleus outside of the cell proper.
The total of forty-three shows that eight chromosomes are lacking
to make the total of fifty-one contributed by the last spermato-
gonial division. In accordance with the prevailing scheme of
synapsis in other primary spermatocytes this means that eight
of the forty-three chromosomes should be bivalent and thirty-five
univalent. In examining the various chromosomes with regard
to size, nine of them seem to be noticeably larger than the rest;
five are in the oblong nucleus and four in the large spherical one-
One of the nine is possibly the accessory, thus leaving the required
number of eight bivalent chromosomes. The three left behind
in the cytoplasm and the sixteen outcasts are all univalents.

As to the origin of the two nuclei within the cell, several
possibilities suggest themselves. The possibility which appears
most tenable is that the cell in its metaphase stage resembled
the one represented in Fig. 32. It can be seen from that figure
that the cell possessed a large spindle, at one side of which was a
plate of chromosomes with threads extending only to one pole.
In another part of the cell is a group of about twenty chromo-
somes with no signs of spindle formation. It can be assumed
that the two nuclei shown in Fig. 24 were formed from two
masses of chromosomes resembling the two plates of chromo-
somes shown in Fig. 32, and that the nuclei were reorganized

28 J. E. WODSEDALEK.

without division of the chromosomes. The latter phase of the
assumption is based on the fact that indications of the actual
division of the chromosomes in the primary spermatocytes are
extremely rare, and that occasionally it can be seen that the
nuclei actually enter the process of reorganization before the
chromosomes had divided. This is especially true in cells with
multipolar and double spindles. As to the nucleus outside of the
cell (Fig. 24), it may be suggested that the chromosomes con-
tained therein resembled in the metaphase stage of this cell the
group of chromosomes without spindle threads represented in
Fig. 32, and that later they were expelled from the cell after the
fashion shown in Fig. 23. The three chromosomes in the cyto-
plasm were probably treated with indifference by the three large
groups of chromosomes, being neither attracted nor repelled by
them, and thus remained scattered about in the cytoplasm. The
material necessary to construct a separate nuclear wall for the
three deserted chromosomes was apparently lacking.

Another possibility as to the origin of the nuclei in question
is that the cell may have had two spindles, as are shown in Fig. 29,
with an additional group of chromosomes without the spindle
threads, for such cells were also observed. Still another possi-
bility is that the cell may have possessed three spindles like those
shown in Fig. 34 and that the chromosomes belonging to one of
the spindles were eventually expelled from the cell and there
reorganized a nucleus while the other two spindles remained
within the cell.

There seems to be no doubt but that the chromatin material
found in the small nucleus on the outside of the cell shown in
Fig. 24, originally formed a part of the inner contents of the same,
for a distinct connection in the form of coarse granular threads
still persists between it and the chromosomes within the cyto-
plasm, as well as the two large nuclei themselves. The group of
sixteen chromosomes was no doubt expelled from the cell and
there formed a separate nucleus. The nuclear membrane is
thin but well defined. It can also be seen that a quantity of
cytoplasm was forced out with the chromosomes. Some of it,
very finely granular, persists around the nuclear membrane.
The rupture in the cell wall occasioned by the expulsion of such

CAUSES OF STERILITY IN THE MULE. 2Q

a mass of chromosomes had evidently regenerated in this cell.
That the cell had begun to degenerate is evidenced by the
presence of large vacuoles and other irregularities in the cyto-
plasm. The signs of decay always seem to be manifest in the
cytoplasm first and then in the chromatin.

The fact that more cells of exactly this type could not be
identified is no doubt due to decay which almost invariably sets
in before the aim of the cell is accomplished, as will be pointed
out later. It is evident that at least some of the other cells
seemed to act in the same direction but succumbed before the
reorganization of the nuclei could take place. Then, too, it
seems reasonable to believe that the isolation of a particular
group of chromosomes, as a necessary preliminary step to the
function of expulsion, depends largely on the relative arrangement
of the linin strands representing such a group of chromosomes in
the early stages of these cells. According to our knowledge of
the extreme variability in the structure of the linin strand net-
work, such a favorable arrangement of the unwelcome chromatin
material can possibly be regarded as being accidental.

Judging by the appearance of the chromosomes, the author is
inclined to think that the material expelled by the cells is largely
that which was contributed by the mother of the hybrid; and
while the number of chromosomes so expelled varies in the differ-
ent cases, nineteen chromosomes including the accessory which
is the number presumably of maternal origin were found in a
few instances. There were never more than nineteen chromo-
somes in the expelled group. This, however, is only a matter of
suggestion, for the situation can not possibly be cleared up until
a careful study has been made of the sex cells of the ass.

Fig. 25 shows a type of cell in which there is one' large spindle
and a bunch of chromosomes forming a small plate located at
right angles to the large one. It can be seen that a number of
threads radiate from the chromosomes in the small plate and
that some of them extend to the poles of the large spindle. The
large chromosome in the upper part of the spindle is the accessory.
Fig. 26 shows a peculiar quadripolar spindle. Most of the chro-
mosomes are in the plate of the large spindle and many of them
show signs of division. A small group of chromosomes is located

3O J. E. WODSEDALEK.

at right angles to the right of the large plate, and spindle threads,
extending from them and convergingat a conspicuous centrosome,
could be plainly seen. On the other side some of these chromo-
somes seem to be connected by threads to the one pole of the
large spindle and some to the other. At the upper pole of the
large spindle is a group of eleven chromosomes with spindle
threads converging at a centrosome near the cell wall. There was
a slight indication of threads on the other side passing to the
upper centrosome of the large spindle, but these could not be
definitely made out since the spindle in question was at a some-
what oblique angle to the large spindle, and the threads on the
opposite side were obstructed from view by the group of eleven
chromosomes. The centrosomes in this cell are unusually large
and conspicuous. The large chrosome at the lower portion of the
main spindle is the accessory.

Fig. 27 is apparently another attempt at a double spindle,
the small one at right angles to the large. There was no sign of
centrosomes. Fig. 28 is a polar view of a cell with two spindles.
In each case the small chromosomes are in the center of the
plate. Seventeen chromosomes form the plate in one spindle
and twenty-five in the other, making a total of forty-two. There
then should appear ten large chromosomes, including the acces-
sory in this cell. In the smaller spindle containing seventeen
chromosomes, four are undoubtedly bivalent, and in the large
spindle of twenty-five chromosomes, six should be larger than
the rest, which upon very careful examination, seems to be the
case, but some of the other chromosomes are also somewhat
larger than usual. This condition is possibly due to the fact that
decay is well under way in the cell and the chromosome may have
become distorted.

Fig. 29 shows an excellent case of two spindles, one is larger
than the other and also contains the larger chromosomes. It is
probable that the large spindle is paternal in nature, and the
small one maternal.

Fig. 30 shows an unusual freak with the upper part giving the
appearance of two closely applied spindles. At the opposite
end, however, the threads do not converge at two separate poles
nor at a single pole but remain loose. At first sight one gains the

CAUSES OF STERILITY IN THE MULE. 3!

impression that the spindle was a tripolar one, with two poles on
one side of the double plate and a single pole on the opposite
side, and that the single pole was cut off in the process of sec-
tioning. But a careful investigation shows that the cell is entire
and that no part of it was cut off. Guyer ('oo) represents a
similar case in hybrid pigeons, and describes it as a case in which
each fiber of the unusually loose spindle seems to terminate at
one end in a small centrosome-like dot or granule.

Fig. 31 shows a tripolar spindle which is really a case of two
spindles fused together. All of the chromosomes are found on
the spindle and some are apparently in the process of division.
Fig. 32 shows another type of tripolar spindle and a group of
nineteen chromosomes, including the accessory, in the cyto-
plasm, corresponding to the group of chromosomes which are
sometimes expelled from the cell. Fig. 33 shows a degenerate
cell with three ragged spindles converging at a common point.
The accessory chromosome is off by itself and from it extends a
thread toward the point of convergence of the three spindles.
The cell is well along in the stage of decadence. Some of the
spindle threads are broken up and others seem to have fused
together.

Fig. 34 shows a cell with three perfect spindles meeting at a
common conspicuous centrosome. The bivalent chromosomes
appear to be distributed among the three spindles. Fig. 35
shows an extremely rare type of quadripolar cell with five spindles
closely and symmetrically arranged, with three chromosomes
remaining loose in the cytoplasm. Fig. 36 shows a cell which is
the only one of its kind that was found and is no doubt the ana-
phase stage of a cell with a quintespindle arrangement similar to
that represented in Fig. 35. Eighty-five chromosomes, including
the accessory near the centrosome to the left, could be distin-
guished as represented. Six of the chromosomes in the central
spindle evidently had not divided. This in terms of univalence
would suggest that there is a total of ninety-one chromosomes
in the cell. The accessory remains undivided as it does in the
horse in the primary spermatocyte. Without the accessory,
then, there would be ninety chromosomes, which suggest rather
strongly that forty-five autosomes, of which five were presumably

32 J. E. WODSEDALEK.

bivalent lined up for division in the metaphase stage of this cell
and that all but the six in the center had divided. Only two or
three of these six remaining undivided could possibly be taken
for bivalents, which fact indicates that the bivalents as well as
the univalents had divided. The centrosomes were well defined
and there were only slight indications of deterioration in the cell.

The researches of Hansemann ('91), Lustig and Galeotti ('93)
and others, show that asymmetrical mitoses are of very general
occurrence in carcinoma cells and other pathological tissues;
and Schottlander ('88) and Galeotti ('93) among others have
shown that similar abnormalities in mitosis may be produced
artificially in many tissues by treatment with dilute solutions of
various drugs, such as quinine, chloral hydrate, cocaine, potassic
iodide and antipyrin.

At first thought, in view of what is known about the effects
of drugs on mitosis, it would appear that the abnormalities in
division of the primary spermatocytes are probably due to some
deleterious chemical substances which may be produced by the
degenerative processes going on in the tubules. This however,
does not account for the fact that the abnormalities in the mule
are practically entirely restricted to the primary spermatocytes.
It appears more certain that the abnormalities in this hybrid are
due to the conflicting tendencies brought into action within the
cells themselves at this stage of spermatogenesis when cooperation
between the two parent plasmas becomes necessary but is
impossible on account of their vast dissimilarity.

GIANT CELLS.

Giant cells appear occasionally but do not seem to be as
numerous as was described by Guyer (*oo) in hybrid pigeons.
Fig. 37 represents a large cell of rare occurrence. It appears to
be a primary spermatocyte with three spindles in the anaphase
stage, but the chromosomes are unusually small and resemble
more those of the nurse cells.

Fig. 38 shows an enormous cell with four large nuclei. The
cell is evidently a primary spermatocyte in the early prophase,
as the size and structure of the nuclei, and the uniform consist-
ency of the cytoplasm are identical to those of normal cells in

CAUSES OF STERILITY IN THE MULE. 33

this stage (Fig. 7). The abnormality appears to have taken
place in the last two divisions of a spermatogonial cell ; only the
nuclei have divided and therefore remained in the same mass of
cytoplasm. Primary spermatocytes with four nuclei appear
occasionally, but in practically all cases there is evidence that
the abnormality had occurred within the cells themselves and
not in the spermatogonia.

Fig. 39 is another giant cell in which all of the chromosomes
seem to be in the huge non-symmetrical quadripolar spindle.
About a hundred chromosomes could be counted in this cell,
some of which are apparently the result of division and others are
in the process of dividing. There are undoubtedly many more
chromosomes present but the exact number could not be de-
termined owing to the position of the large equatorial plate. It
is probably a spermatogonial cell which possessed two complete
nuclei and the number of chromosomes in excess of one hundred
and two may be due to the fact that many of the chromosomes
have divided. Parts of the spindle seem to be in the late meta-
phase and others in the anaphase stage.

MULTINUCLEATE CELLS.

Primary spermatocytes with nuclei ranging from two to six
in number are not uncommon (Figs. 40-43). The most common
ones are those with two and four nuclei. Such cells appear to be
in the late telophase stage. There is much irregularity in size
among the nuclei of the same cell. For example, when two
nuclei are present they are sometimes of similar size and then
again one is much larger than the other. In cases where three
nuclei are present they are sometimes of practically the same size,
or form a gradation in sizes, or two large and a small one, or two
small and a large one appear, and so on. Frequently one or
more chromosomes were left in the cytoplasm while the others
formed one or more nuclei. Fig. 41 shows a decaying cell with
two nuclei, and a large chromosome left in the cytoplasm but
connected with one of the nuclei by a coarse strand. Such
strands were not always present. It is possible that the chro-
mosome in question is the accessory, though one of the chro-
mosomes within the nucleus to the left also resembles it.

34 J- E. WODSEDALEK.

The formation of more than one nucleus is not at all strange,
when we consider the various freakish spindles of the primary
spermatocytes and the fact that the cytoplasm of these cells was
never observed to be in the process of division. A spindle like
those shown in Figs. 34-36 for example, may, if decay does not
set in beforehand, form five or six nuclei. A cell like that rep-
resented in Fig. 29 may eventually possess four nuclei, one like
that shown in Fig. 31 three nuclei, and so on. In all of the com-
paratively few cases that reach the telophase stage before decay-
ing, we must necessarily expect some such abnormalities.

DESTRUCTION OF THE GERM CELLS.

In speaking of degeneration of the germinal cells in hybrid
pigeons, Guyer (*io) says: "Degenerative processes were in
progress in the testes of all the sterile forms, but were most
pronounced in hybrids between very divergent species, or between
unlike hybrids from birds which were themselves descendants
of fertile hybrids. There were some such extreme cases of de-
generation that only the layer of cells lying along the wall re-
mained in the tubule. Where such a degree of degeneration
exists there is, of course, no approach to the formation of sper-
matozoa. There is often a strong invasion of wandering cells
into the tubules, especially where the degenerative activity has
become extensive. The interspaces between such tubules are
also usually packed with cells which have much the same ap-
pearance as white blood corpuscles. Little clumps and globules
of deeply staining cytoplasm are scattered about among the cells
within the tubules. The oval nuclei of the Sertoli cells are also
generally to be seen in varying numbers.

"In some places it really looks as if the germinal cells them-
selves lose their cell walls and characteristic appearance and
become leucocytes, but this point will require very careful study
before any definite conclusions can be reached. In other tubules
the original cells, or cells which have wandered in, settle down
and take on exactly the appearance of the large stroma cells
which are ordinarily present outside the tubules. The cytoplasm
of such cells has a peculiar alveolar-like appearance that is very
characteristic."

CAUSES OF STERILITY IN THE MULE. 35

The destruction of the sex cells in the mule appears to be re-
stricted to the primary spermatocytes, where it seems inevitable.
The incompatibility of the two widely different plasmas contri-
buted by the parents renders completion of spermatogenesis in
this hybrid impossible and the animal remains sterile. The
process of deterioration sets is as soon as cooperation of the chro-
matin material becomes necessary in the primary spermatocyte.
By far the largest amount of destruction takes place in the
"spireme stage" before the chromosomes can make their ap-
pearance. Sometimes the nucleus in this stage becomes enor-
mously distended and the thin chromatin threads break up into
numerous fragments before any noticeable pairing takes place.
Then again, the fragmentation takes place after some of the
threads fuse (Fig. 10), or after the chromosomes make their
appearance (Fig. 44).

The cytoplasm seems to decay first and large masses of naked
nuclei in various degrees of decadence can be seen in some of the
tubules. Small masses of chromatin material, pieces of threads,
cytoplasm, and cell wall, are also abundant in many of the
tubules. Occasionally small, naked, though well-defined nuclei
are seen. Whether these are formed from cast-off material like
the one shown in Fig. 24, or whether they are small nuclei
similar to those shown in Fig. 42, retaining their structure after
the cytoplasm had fallen to pieces is difficult to determine.
The latter suggestion can easily be conceived, and the former is
not unreasonable, since there is no logical reason to make one
believe that the expelled material invariably remains attached
to the cell. A much more legitimate assumption would be that,
since the material is cast off by the cell, it ordinarily leaves the
cell entirely and becomes free among the other material in the
lumen of the tubule. This assumption would tend to explain
further the extremely rare occurrence of cells similar to the one
shown in Fig. 24.

Figs. 44-52 show various types of degenerating cells. Fig. 44
is in the late prophase, and Fig. 45 in the metaphase showing a
partly expelled mass of disintegrating chromosomes. Figs. 46,
48 and 49 show the collapsed condition of the cytoplasm and the
dense nature of the nuclei. The nuclei sometimes retain their

36 J- E. WODSEDALEK.

normal shape but the chromatin material within fuses into a
continuous mass, which is often perforated by many vacuoles
(Figs. 46 and 49). Occasionally a single large vacuole appears
in the center of the nucleus leaving the chromatin material
arranged in a layer next to the nuclear membrane (Figs. 48 and

52).

Fig. 47 is a degenerate cell possibly in the late prophase stage
in which many of the chromosomes have fused into large spherical
bodies which remained scattered about in the cytoplasm. Fig. 50
shows a cell with an ill-formed nucleus and a mass of chromatin
material left out in the highly degenerated cytoplasm. Fig. 51
shows a cell in an advanced stage of destruction in all of its parts.
Fig. 52 shows a collapsed nucleus with a portion of naked cyto-
plasm.

On account of the broadcast destruction of the primary sper-
matocytes in the early stages, a comparatively few are left that
attain the metaphase and early anaphase stages (Figs. 18-45)
when they, too, meet their fate; and the remaining few that
survive the anaphase succumb soon after, and no secondary sper-
matocytes nor spermatids, and consequently no spermatozoa
are formed and the hybrid therefore remains sterile.

The invasion of tubules by wandering cells is not as pronounced
in the mule as in pigeon hybrids, as described by Guyer ('oo).
Even where the degenerative activity was most pronounced the
leucocytes were rare or absent altogether. Nor was there any
indication of the germinal cells themselves losing their cell walls
and characteristic appearance and becoming leucocytes, as some-
times appears to be the case in hybrid pigeons according to the

same author.

SUMMARY.

1 . The parents of the mule are widely different as can be seen
from the comparison of the horse and the ass.

2. The mule partakes of both the sire and the dam, but appears
to resemble the ass more than the horse, both in structure and
habits.

3. The greatest difference seems to lie in the relative number of
chromosomes in the cells of these two animals. The horse
has thirty-seven and the mule fifty-one. This suggests that the

CAUSES OF STERILITY IN THE MULE. 37

number in the ass is about sixty-five, thus making a difference of
twenty-eight chromosomes between the parents of the hybrid.

4. The seminiferous tubules of the mule contain a much
smaller amount of germ cells than do the tubules of the horse.
Some of the tubules of the mule are entirely devoid of sex cells.

5. The sex cells of the mule are larger than those of the horse
in the corresponding stages.

6. All of the fifty-one chromosomes in the spermatogonial
cells of the mule enter the spindle for division. The mitotic
figures are normal and there are no straggling chromosomes.

7. The accessory chromosome of the hybrid, which is undoubt-
edly maternal in origin, resembles entirely the accessory of the
horse, which fact shows that this sex-determining chromosome
retains its individuality.

8. The period of synizesis which is so obvious in the primary
spermatocytes of the horse, is lacking in the mule.

9. The spireme of the horse is also lacking in the mule, but is
replaced by a continuous network of chomatin threads, parts
of which sometimes resemble the spireme to a certain extent.

10. There is no definite time for the pairing of threads or
chromosomes in the hybrid; the synaptic period begins at the
time the chromatin threads make their appearance and continues
through the prophase.

11. The pairing of chromosomes, or pseudo-reduction, is
always incomplete and very inconstant. The number of chro-
mosomes in the late prophase of the primary spermatocytes
varies from thirty-four to forty-nine. The greatest majority of
counts of chromosomes lie between forty and forty-five. The
expected number, if reduction was complete, would be twenty-
five besides the unpaired accessory.

12. In the late prophase of the primary spermatocytes the
bivalent chromosomes can as a rule be readily distinguished
from the univalent ones. The number of chromosomes which
the various cells lack in order to make the original total of fifty-
one, in terms of univalence, can usually be accounted for by the
proportional increase in the presence of bivalents in such cells.

13. Up to the early prophase of the primary spermatocytes
there seems to be no necessity for the paternal and maternal

38 J. E. WODSEDALEK.

chromosomes to cooperate in functioning. Each group seems to
go on performing its functions normally. The real conflict
ensues during the various stages of the primary spermatocyte,
and is no doubt occasioned by the necessity for cooperation on
the part of the paternal and maternal chromosomes in the process
of conjugation or pseudo-reduction.

14. Abnormalities in mitosis occur invariably in primary
spermatocytes that attain the metaphase stage.

15. Giant cells are occasionally seen.

1 6. The chromatoid body which is very conspicuous and
constant in the horse is entirely lacking in the mule.

17. There is considerable evidence that primary spermatocytes
make an attempt to eliminate some of the chromatin material.

1 8. The chromosomes expelled by the cells appear to be those
which were contributed by the mother of the hybrid.

19. Destruction as well as abnormalities in mitosis seem to be
restricted to the primary spermatocytes. Most of the cells
disintegrate during the prophase, especially during the period of
synapsis. Others meet their fate in the metaphase or early ana-
phase stages. The remaining few that survive the anaphase
succumb soon after, and no secondary spermatocytes nor sper-
matids, and consequently no spermatozoa are formed and the
hybrid remains sterile.

20. There are no authentic cases on record showing that fer-
tility ever occurs in this hybrid.

LITERATURE CITED.
Galeotti, G.

'93 Ueber Experimentelle Erzengung von Unregelmassigkeiten des karyokine-
tischen Processes. Beitr. z. patholog. Anat. u. z. Allg. Pathol., XIV., 2,
Jena, Fischer, 1893.
Guyer, M. F.

'oo Spermatogenesis of Normal and of Hybrid Pigeons. University of Cin-
cinnati Bulletin No. 22.

'10 Accessory Chromosomes in Man. BIOL. BULL., Vol. XIX., No. 4.
Hansemann, D.

'91 Karyokinese und Cellularpathologie. Berl. Kin. Wochenschrift, No. 42

1891.
Hayes, M. H.

'04 Points of the Horse. New York.
Lustig and Galeotti, G.

'93 Cytologische Studien ueber pathologische menschliche Gewebe. Beitr.
Path. Anat., XIV., 1893.

CAUSES OF STERILITY IN THE MULE. 39

Lydekker, R.

'12 The Horse and its Relatives. New York.
Montgomery, T. H.

'12 Human Spermatogenesis. Jour. Acad. Nat. Sci. Phlla., XV.
Plumb, C. S.

'06 Types and Breeds of Farm Animals. New York.
Schottlander, J.

'88 Ueber Kern und Zelltheilungsvorgange in dem Endotheil der enzandeten

Hornhaut. Arch. f. mikr. Anat., XXX., 1888.
Wilcox, E. V.

'07 Farm Animals. New York.
Wodsedalek, J. E.

'13 Spermatogenesis of the Pig, with Special Reference to the Accessory

Chromosomes. BIOL. BULL., Vol. XXV., No. i, June, 1913.
'13 Accessory Chromosomes in the Pig. Science, N. S., Vol. XXXVIII., No.

966, pp. 30-31-

'14 Spermatogenesis of the Horse with Special Reference to the Accessory
Chromosome and the Chromatoid Body. BIOL. BULL., Vol. XXVII.,
No. 6, December, 1914.

4O J. E. WODSEDALEK.

EXPLANATION OF PLATES.
PLATE I.

(All of the drawings were made with the aid of a camera lucida, X 3,000.)

FIG. i. Early spermatogonial cell showing a large heart-shaped nucleolus and
two small karyosomes.

FIG. 2. Resting stage of a full grown spermatogonial cell showing the large
triangular or heart-shaped nucleolus and the two spherical karyosomes.

FIG. 3. Late prophase of spermatogonial division showing fifty ordinary chro-
mosomes and the large accessory which can easily be distinguished.

FIG. 4. Metaphase of division in a spermatogonial cell showing a perfect spindle
with the accessory dividing at the left.

FIG. 5. Anaphase of division of a spermatogonial cell showing a perfect mitotic
figure.

FIG. 6. Late anaphase of division of a spermatogonial cell with perfect mitosis
again in evidence.

BIOLOGICAL BULLETIN, VOL. XXX.

PLATE I.

1

I

A»v:*;> I

J. E. WOC8EDALEK.

42 J. E. WODSEDALEK.

PLATE II.

FIG. 7. Resting stage of a primary spermatocyte showing the characteristic
heart-shaped nucleolus and the karyosomes from which radiate fine linin strands.

FIG. 8. Primary spermatocyte showing a continuous network of chromatin
threads, some of which have paired and others are in the process of fusion, but most
of them are unpaired.

FIG. 9. Primary spermatocyte showing a continuous network of chromatin
threads, an unusual number of which have fused; some are still unpaired while
others are in the process of fusing.

FIG. 10. Primary spermatocyte in process of degeneration in the "spireme
stage."

FIG. it. Primary spermatocyte in the late "spireme stage," showing signs of
paired and pairing threads, paired and single chromosomes, and also signs of
degeneration.

FIG. 12. Late prophase of primary spermatocyte showing univalent and bi-
valent chromosomes, and one trivalent chromosome. Some of the univalent
components are still attached to each other at various angles. The accessory chro-
mosome is very distinct.

BIOLOGICAL BULLETIN, VOL. XXX.

PLATE II.

. ^<te&&&&;>^£££gj<ifr'

8

.-*"•"•".'•'•'"

--"f . ".

i :N.
•.•\v-.V-. •;•• '.\

K

\

J. E. WODSEDALEK.

44 J- E. WODSEDALEK.

PLATE III.

FIGS. 13-15. Late prophase of primary spermatocytes after the partial synapsis
had taken place.

FIG. 13 shows forty chromosomes, about eleven of which appear to be bivalent.

FIG. 14 shows thirty-eight chromosomes about thirteen of which are bivalent,
and Fig. 15 shows forty-three chromosomes about eight of which are bivalent
Note the heart-shaped accessory in each of these cells.

FIG. 16. Polar view of a metaphase stage of a primary spermatocyte, showing
twenty-eight chromosomes in the plate, and seventeen including the accessory are
scattered in the cytoplasm.

FIGS. 17 AND 1 8. Degenerating primary spermatocytes in the metaphase stage
showing many scattered chromosomes and the remains of the decayed cytoplasm
congregated about the plate or large clump of chromosomes.

BIOLOGICAL BULLETIN, VOL. XXX.

PLATE III-

13

,,... •••••;:••"' '"''v'. " ' '"\ '•'"••••••.,v

^ff ' ' V..,;v^v

/r':" —

•A«LV

1 C

15

••'•••. ••- • 4

14

.=••--!-"/; . -

/fl'SLl I

J. E. WODSEDALEK.

46 J. E. WODSEDALEK.

PLATE IV.

FIGS. 19-21. Freak types of primary spermatocytes in the metaphaae stage
showing most of the chromosomes in the equatorial plate and several scattered
about on the spindle or in the cytoplasm.

FIG. 22. Primary spermatocyte showing an uncommon type of unequal division
in which five large chromosomes have passed over to one place while about thirty-six
remained in the original place.

FIG. 23. A somewhat degenerate metaphase stage of primary spermatocyte
showing the expulsion of a bunch of chromosomes.

FIG. 24. A late telophase of a trinucleated primary spermatocyte. The nu-
cleus on the outside of the cell wall was apparently formed from a group of chro-
mosomes which were expelled by the cell.

BIOLOGICAL BULLETIN, VOL. XXX.

PLATE IV.

J. E. WCDSEDALEK.

48 J- E. WODSEDALEK.

PLATE V.

FIGS. 25-30. Abnormal mitotic figures of primary spermatocytes (fully ex-
plained in the text).

BIOLOGICAL BULLETIN, VOl . XXX.

PLATE V.

• ;

'. .!

f-

•••.;-

- -- . . •-

..

'M^^^;S^^^|^\

'••• '• -• ••' '

V

J. E. WODSEDALEK.

50 J. E. WODSEDALEK.

PLATE VI.

FIGS. 31-36. Abnormal mitotic figures of primary spennatocytes (fully ex-
plained in the text).

BIOLOGICAL BULLETIN VOL. XXX.

•:•>•'.:

31

34

'*;%&$#^W0M£-^

^i^Sii:.^^ :- ^'i.

J. E. WOD8EDALEK.

52 J. E. WODSEDALEK.

PLATE VII.

Fie. 37. A giant cell with three spindles in the early anaphase stage.

FIG. 38. A giant cell with four large nuclei, each resembling the nucleus of the
primary spermatocyte. (See Fig. 7.)

FIG. 39. A giant cell with a huge non-symmetrical quadripolar spindle con-
taining over one hundred chromosomes. It is probably a spermatogonial cell
which possessed two complete nuclei and the number of chromosomes in excess
of one hundred and two may be due to the fact that many of the chromosomes have
divided.

FIG. 40. A quadri-nucleated cell.

BIOLOGICAL BULLETIN, VOL. XXX.

.•* ;- • >

"X

••-,. %'

J. E. WOD8EDALEK.

54 J- E. WODSEDALEK.

PLATE VIII.

FIG. 41. A bi-nucleated cell in process of degeneration showing a chromosome,
possibly the accessory, which was left out in the cytoplasm.
FIG. 42. A cell with six well-defined nuclei.
FIG. 43. A degenerating cell with three nuclei.
FIGS. 44-46. Cells in process of degeneration (explained in the text).

BIOLOGICAL BULLETIN, VOL. XXX.

•-.

'

<qh

$ :^ '

$&:|

• jyjv-'-j
I fij

>;.

; S

- Vv-.

|

46

J. E. WODSEDALEK.

56 J- E. WODSEDALEK.

PLATE IX.

FIGS. 47-52. Various types of cells in process of degeneration (explained in
the text).

BIOLOGICAL BULLETIN, VOL. XXX.

X"

51

. ;-
' ' '

52

J. E. WODSEDALEK.

STUDIES ON THE GEOTROPISM OF THE MARINE
SNAIL, LITTORINA LITTOREA.1

SAKYO KANDA.

CONTENTS.

I. Introduction 57

II. Material and Methods 58

III. Experiments 61

1. Preliminary Experiments with Gravity and Light 61

2. The Relation between the Pressure of Gravity and the Precision of

Orientation 63

3. What Determines whether the Head End will be Directed Up or

Down? 66

4. The Effects of Light and Surface-Film of Seawater 71

5. The Effect of the Surface-Film of Sea- water 74

6. The Effect of Chemicals 76

IV. Discussion 76

V. Summary and Conclusions 82

VI. Bibliography 83

I. INTRODUCTION.

In 1888, Loeb pointed out that "Die Schwerkraft der Erde,
wenn sie senkrecht gegen die ventrale Seite der Schabe gerichtet
ist, wirkt als Reiz, der dieselbe zu Bewegungen veranlasst"
(8, p. 9). Since then, with modifications of his method, Daven-
port and Perkins (2) and Frandsen (3) have investigated the
same problem on a slug, Limax maximus. The former drew the
conclusions that, "the precision of orientation of the slug varies
directly with the active component of gravity" (2, p. 105) and
that "this tendency (to go either up or down) must be ascribed
to some internal condition of the individuals, for it varies in
different individuals and in the same individuals at different
times" (2, p. no). The latter reached rather different con-
clusions, thus: 'The different geotactic response, on a glass
plate, of different individuals is due mainly to two factors: (a) The
quantity and quality of the slime secreted, which is a very im-
portant factor; (6) the relative proportions of the length of the

1 From the Marine Biological Laboratory, Woods Hole, Mass., and the Physio-
logical Laboratory, University of Minnesota, Minneapolis, Minn.

57

58 SAKYO KANDA.

anterior and posterior regions of the animal's body. All the
conditions being. the same, it is this factor which 'determines
whether the head end will be directed up or down ' ' (3, p. 205).

The reactions of the marine forms of Littorina littorea, L. rudis,
etc., to light and other influences have been studied by Mitsu-
kuri (12), Bohn (i), Haseman (4), and Morse (14). Unfortun-
ately, however, none of them has taken into consideration the
response of the animals to gravity, although the effect of gravity
upon Littorina and upon gastropods in general is remarkable
and easily observed at the seashore.

Following the example of Frandsen and of Davenport and
Perkins, an attempt was made by the writer with Littorina
littorea to determine: (i) "What relation exists between a vari-
ation in the pressure of gravity and the precision of orientation?"
and (2) What "determines whether the head end will be directed
up or down?"

The experimental work was done in the physiological depart-
ment of the Matine Biological Laboratory at Woods Hole, Mass.,
during the summers of 1912 and 1913. The results obtained in
1912 used in this paper are indicated by the date, " 1912." The
others were all secured in 1913.

II. MATERIAL AND METHODS.

i. Material. — The animal used in all the experiments was a
marine snail, Littorina littorea, which is numerous about Woods
Hole. The snails were collected by the writer in the morning or
afternoon, just before a new series of experiments was carried on,
and were kept in a large glass dish in running sea-water during
experiments. The size of the animal used was about 1.5 x i.i
cm.1 It was found that this was the more convenient size for
experimental purposes, because the bigger, i. e., older, ones re-
treated into and remained for a long time in their shells when
handled. The younger animals are more active and quicker to
respond to stimuli.

The same individuals could not be used throughout any series
of experiments; since their movements became abnormal, due

1 The laboratory collector, Mr. Gray, told me that the snails which I used were
about one year old.

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA. 59

possibly to fatigue. The same individuals could be used only
for four or five trials. Ten individuals, sometimes II or 12
(with anticipation of falls), were used for each trial of a series
of experiments. As will be seen in the tables, however, very often
less than 10 individuals were left on the support for observation,
the rest having fallen down.

2. Methods. — Loeb's method (8) modified by Davenport and
Perkins (2) and by Frandsen (3), of the different angles of in-
clination of a support, on which the animals are to be placed, was
adopted with variations. The support was a plain glass plate
about 22 x 19 cm., marked off on one side into squares of one
sq. cm. each, whereby it could be readily determined how far the
animals moved from their original places.

The support, or plate, was placed in an apparatus by which it
could be held at any ten-degree angle between the horizontal and
vertical (see Fig. i).

FIG. i. G. P. = a plain glass plate. The rest of the apparatus is all wood.

The animals were placed on the unlined side of the plate,
moistened and held upside down at an inclination, say 10° to

60 SAKYO KANDA.

the horizontal, so that their heads were all directed upward.
Meamvhile sea-water was poured, two or three times, over the
animals and the plate surface to prevent their moving before the
commencement of the experiment, and at the same time to get
them to stick tightly to the surface. When the desired number
of individuals had been so placed, the support was reversed, and
placed at the desired angle in the holding apparatus, so that their
heads were, now, directed downward. The animals being
negative to gravity, this procedure was necessary in order to
determine their movements of orientation in a desired time.

Experiments were conducted either in sea-water in a glass
aquarium, or in air. To exclude the effect of light in either case,
a square box painted black inside was employed to cover the whole
arrangement described above.

After the desired time for a particular experiment, the cover-
box was removed. Then the movements which the animals
had made were recorded (with detailed notes) as nearly as
possible as follows: A movement of 180° from the original was
designated as "oriented"; 90° as "horizontal"; and o° as "ori-
ginal." If an animal was observed to have crawled downward
from the original place, it was recorded as positive to gravity.
It was noticed that some individuals did not crawl downward
quite vertically, but no discrimination as to these is made in the
tables. Quite a few individuals, that crawled horizontally, are
also arbitrarily classified under "horizontal," although such
movements are believed by the writer to have no great signifi-
cance.

A question might be raised about middle spaces between
1 80° and 90°, and also between o° and 90°. Such positions were
seldom observed; but if any were observed, they were recorded
as "oriented" at a position between 90° and 180°; as "horizontal"
between 90° and 45°; and as "original" between 45° and o°.
Moreover, since quite a number of individuals would possibly
have become oriented if they had been given longer than one
minute, it might have been better to describe them as " orienting "
rather than "horizontal." The animals whose position was
specified as "original" should not be interpreted as indifferent
to gravity. They simply failed to respond to it during the time

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA. 6l

when they were under experimentation. The time factor was
important. This will be confirmed by the following preliminary
experiment A. But for the more detailed study, the shortest
possible time was adopted, to avoid fatiguing the animals.

In all the following tables, the signs indicate positive either to
gravity or light by + and negative by — .

III. EXPERIMENTS.
i . Preliminary Experiments with Gravity and Light.

Preliminary Experiment A . — A tall beaker full of sea-water was
placed upside down in a large dish filled with sea-water. Special
care was taken to exclude air bubbles from the beaker. It was
then supported on two pieces of glass-tubing about 4 mm. in
diameter in the dish in order to let the sea-water within com-
municate freely with the outside at the bottom. By means of
glass-tubing, sea-water was run into the dish so as to have fresh
sea-water always at the bottom, but so as not to let any air
bubbles into the beaker. Thirty selected snails were placed
under the beaker, and between the two pieces of the glass-tubing,
which was not high enough to permit the snails within to crawl
out from under the beaker. The whole arrangement was then
covered with the box already described, to exclude light.

About ten minutes later it was found that all the snails had
crawled up the vertical (inside) wall of the beaker and gathered
at or near its top. The sea-water would be expected to be fresher
and better supplied with oxygen at the bottom than at the top.
The snails, however, crawled upward just the same. This was
repeated several times, but there was no exception to this rule.

Such results indicate that the upward movements of the snails
are not caused by the lack of oxygen but by gravity. This is also
fully supported by the fact that the snails crawl upward on moist
rocks, or on a moist glass plate, in air where the amount of oxygen
does not vary.

Moreover, in leaving the horizontal bottom of the dish to take
positions on the vertical wall of the beaker, the snails must have
oriented themselves against the pull of gravity. This orientation
of the snails cannot be explained by the mechanical theory of
geotropism. It is an active process, one of true response to

62

SAKYO KANDA.

stimulation. Loeb's conclusion which was already referred to,
is thus confirmed.

Preliminary Experiment B. — According to Bohn and Mitsukuri,
Littorina littorea is negatively heliotropic. To test their results,
a simple method of "righting" experiments was adopted as fol-
lows:

Ten selected individuals were placed dorsal side down on each
of two glass plates. The one was covered with the box, and at
the same time, the other was exposed to diffused daylight, both
for five minutes. The two lots were alternately covered and
exposed to daylight, after the animals had been refreshed in
sea-water and again turned "on their backs." The results
given in Table I. show that light considerably affects the "right-
ing" response of the animals.

TABLE I.

THE "RIGHTING" OF SNAILS ON A HORIZONTAL GLASS PLATE IN DARKNESS AND
DAYLIGHT FOR FIVE MINUTES (1912).

No. of
Trials.

No. of
Ani-
mals.

Darkness.

No. of
Ani-
mals.

Daylight.

Righted.

Not Righted.

Righted.

Not Righted.

I

10

4

6

10

3

7

I

10

8

2

IO

2

8

I

10

6

4

10

3

7

I

IO

5

5

IO

2

8

I

IO

6

4

10

3

7

I

IO

6

4

10

2

8

I

10

4

6

10

4

6

I

10

4

6

10

2

8

I

10

5

5

IO

3

7

I

IO

4

6

IO

2

8

Total. .10

'100

52 or 52%

48 or 48%

IOO

26 or 26%

74 or 74%

In explanation of this "righting" reaction of the snails, it may
not be out of place to refer to Loeb's analysis in the starfish.
According to him, the "righting" of the starfish is a stereotropic
phenomenon, but not geotropic (10, pp. 64-65). Recently Moore
confirms this conclusion through experiments (13, p. 237), while
Jennings seems to have been misled in this respect (5, pp. 120-
148). The "righting" of the snails may also be a stereotropic
phenomenon, though it makes no difference for the present
purpose. In fact, however, as will be seen later, contact stimuli
interfere with the snails' reaction to gravity. The main point in

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA. 63

regard to Preliminary Experiment B is that light is a factor in
the behavior of these animals.

Preliminary Experiment C. — According to the "mechanical
theory" of geotropism, an animal becomes oriented head up
because of its heavier posterior region, or it orients itself with its
head down, because of its heavier anterior region or head. An
experiment was, therefore, made to determine which region of
the snails, the anterior or the posterior, was heavier. This was
simple. When an individual was placed on a glass plate in sea-
water with its dorsal side down, it came out of the shell, and
made an "effort" to "right." It was then carefully dropped and
watched as it sank, before it retreated into the shell. Every
one of them sank with its anterior region up, as expected. The
was done also with those individuals which were positive to
gravity at a certain angle of inclination. There was, however,
no exception.

With the same point in mind, empty shells were tested. One
of them was fixed by its ventral side on a small square cover glass,
with as little glue as possible. At the three corners of the cover
glass, fine threads were fastened. The center of gravity was
found by suspension to be located somewhere in the posterior
region. This was done with a number of shells and the same
results were obtained in every case. These results, therefore,
agree with the other observations, indicating that the posterior
region of the snails is heavier than the anterior region.

From the above results, the writer is justified in concluding
that the center of gravity of the snails is located in the posterior
region; and a possible inference from this conclusion is that the
posterior region of the snails has a greater specific gravity than
the anterior region. This fact and inference may have a bearing
in deciding whether the orientation of the animals against gravity
is due to purely physical or mechanical pull. This question will
be considered throughout all the following experiments.

2. Experiments to Show the Relation between the Pressure of Gravity
and the Precision of Orientation.

Experiment A. — Davenport and Perkins asked and answered
by experiment: "What relation exists between a variation in the

64

SAKYO KANDA.

pressure of gravity and the precision of orientation of the slug?"
(2, p. 100). An attempt was made by the writer to determine
experimentally the same question for Littorina littorea. The
experiments were conducted in sea-water. Light was excluded
by means of the cover-box. The results given in Table II. show
that the higher the angle the larger is the number showing
negative geotropism, and the lower the angle the larger the number
of (so-called) positively geotropic animals.

TABLE II.

GEOTROPISM OF SNAILS AT THE DIFFERENT ANGLES OF INCLINATION OF A GLASS
PLATE IN SEA-WATER IN TOTAL DARKNESS.

At beginning of experiments each head pointing downward. Table shows re-
sults after one minute.

— Geotropism.

+ Geotropism.

Temp, of Sea-
water.

Angles.

No. of
Trials.

No. of
Ani-
mals.

Oriented.

Horizontal.

Crawled
Downward.

Did Not
Move.

No.

*

No.

*•

No.

*.

No.

Jfc

TolO OTO (^
102 ~2l V-.

Q0°

35

300

300

IOO

O

O

0

o

O

0

i8i°-i8^°C.

78f°

22

2OO

2OO

100

0

O

0

o

0

0

i8j° C.

675°

22

2OO

I9O

95

3

1.5

5

2.5

2

I

19 — 19^ ^'

S6|°

23

200

180

90

9

5-5

7

3-5

4

2

I9i°-i9i° C.

45°

25

20O

163

81.5

13

6..S

12

6

12

6

I95°-20° C.

33f°

32

2OO

152

76

IS

7-5

16

8

17

8-5

19^-20° C.

22^°

33

20O

130

65

22

n

23

ii-S

25

12.5

20°-2oi° C.

iij°

23

2OO

no

55

26

13

34

17

30

15

Total.

2IS 1.700

LSII or 80%

07 or <;.7%

QO or <?.2%

As already pointed out, however, it is misleading to consider
that the negative geotropism of the snails reaches zero on the
horizontal surface of the earth. The force of gravity stimulates
the snails, when it happens to pull along the line of the dorso-
ventral axis of the animals, that is, on the horizontal surface,
as the preliminary experiment A has shown. But the larger the
angle the shorter the time required for orientation. This is
shown in the "horizontal" column, when the number of the
animals, most of which were "orienting" at the end of one minute,
decreases as the angle of inclination to the vertical increases.

Experiment B. — It has been supposed that hydrostatic pressure
affects the reaction of an animal to gravity. To decide this
question, a series of experiments was made in air. Light was

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA. 65

excluded as usual. The results given in Table III. show an
interesting difference compared with those of Table II., which
must not be overlooked. The number in the negative geotropism
column in Table III. is always higher than the corresponding one
in Table II., and that in the positive column in the former is
always lower than that in the latter. A possible explanation of
these differences is found in the buoyancy effect of the water,
which decreases the weight of the animals. In consequence,
the pull of gravity is more effective on the animals in air. It
would seem therefore that the pull of gravity on the animal
as a whole must be one factor in its orientation.

TABLE III.

GEOTROPISM OF SNAILS AT THE DIFFERENT ANGLES OF INCLINATION OF A GLASS
PLATE IN AIR IN TOTAL DARKNESS.

At beginning of experiments each head pointing downward. Table shows results
after one minute.

— Geotropism.

-+- Geotrop.

Original.

Temp, of
Room.

Angles.

No. of
Trials.

No. of
Ani-

Oriented.

Horizontal.

Crawled
Downward.

Did Not
Move.

mals.

No.

H

No.

*

No.

*

No.

t

2l|°-22j°C.

90°

2O

IOO

IOO

IOO

0

o

0

o

O

o

25° C.

78f°

14

IOO

IOO

IOO

O

o

0

o

O

o

22^°-23° C.

67^°

IS

IOO

IOO

IOO

o

o

o

o

o

0

23° C.

S6|°

13

IOO

92

92

2

2

2

2

4

4

23|° C.

4.S°

13

IOO

86

86

6

6

6

6

2

2

2lJ° C.

33 1°

17

IOO

80

80

8

8

9

9

3

3

22f°-23° C.

22^°

13

IOO

67

67

13

13

18

18

2

2

22° C.

IIj°

14

IOO

60

60

13

13

27

27

O

O

Total .

119

800

727 or 00.8%

62 or 7.7%

II or 1.7%

The above fact reminds the writer of the "mechanical theory"
(6, pp. 2-13) and, also, the "resistance or weight theory" (6, pp.
14-20) of geotropism in Paramecium. If the greater weight .of
snails in air causes the quicker orientation of the animals, and
also if the posterior region of the animals is heavier than the
anterior region (as has been proven to be the case), the theories
above mentioned are favored to some extent. Nevertheless,
no mechanical theory can satisfactorily explain why "positive
geotropism" increases with decrease in the angle of inclination.
This, if a real gravity response, must be explained on the basis of
physiological conditions, as Loeb pointed out many years ago.

66 SAKYO KANDA.

The question, however, arises: Is there any maximum or
minimum of resistance or weight which makes the snails crawl
either up or down? This question will be answered later.

Although differences exist between Tables II. and III., the
results in the latter table in general show a fair uniformity with
Table II. The hydrostatic pressure,1 therefore, seems seldom,
if ever, to affect the negative reaction of the animals to gravity.
These two tables also show a fair agreement with the results
obtained by Loeb (8, pp. 6-8), Davenport and Perkins (2, pp.
100-105), and Frandsen (3, p. 198), that is, a steady increase in
negative geotropism with increase in the angle of inclination.
Davenport and Perkins's view in this respect is fully confirmed
(2, p. 105).

3. What Determines Whether the Head End Will be Directed

Up or Down?

Experiment A, — Instead of the plain glass plate used for ex-
periments hitherto, a ground-glass plate was employed for the
following experiments, for two reasons: (i) Since the animals
live along a rocky seashore, the plain glass plate may not be a
proper support for them; and (2) the animals being positively
stereotropic, it was thought well to test the effect of contact
stimuli upon their geotropism. Light was excluded in these
experiments as usual. In the results given in Table IV. it is
interesting to note, as expected, a decided decrease in the number
of negatively geotropic and an increase of "positively geotropic"
individuals with decrease in the angle of inclination. Moreover,
below the 45°, no individuals were in their original position.

The animals can stick better to the ground plate than they do
to the plain glass. In consequence, they can resist the pull of
gravity better on the former than they do on the latter. If so,

1 On preparing Mr. Kanda's Ms. for publication it strikes me that the chief
difference between the experiments shown in Tables II. and III. lies in the buoyancy
rather than the pressure factor. Remembering that the air pressure must be added
to that due to any depth of water, the difference in pressure in Table II. as contrasted
with III. is almost negligible. But the weight of the animal in water is much less
than in air. The slight difference between Tables II. and III. is to me strong evidence
against the buoyancy theory as playing any large part. Mr. Kanda is in Japan
and I cannot call his attention to these points, so I venture to add this footnote.
— E. P. LYON.

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA. 67

this serves to indicate that resistance plays a part, particularly
in negative geotropism. The resistance theory, however, does
not explain why the animals leave the horizontal surface on which
they are placed and crawl upward on the vertical wall of a beaker
(see preliminary experiment). This fact is unexplained by any
mechanical or resistance theory.

TABLE IV.

GEOTROPISM OF SNAILS AT THE DIFFERENT ANGLES OF INCLINATION OF A GROUND-
GLASS PLATE IN AIR IN TOTAL DARKNESS.

At beginning of experiments each head pointing downward. Table shows
results after one minute.

— Geotropism.

+ Geotropism.

Original.

Temp, of
Room.

Angles.

No. of
Trials.

No. of
Ani-
mals.

Oriented.

Horizontal.

Crawled
Downward.

Did Not
Move.

No.

*

No.

Jfc

No.

a.

No.

*

22i°-22f0 C.

90°

14

IOO

IOO

IOO

O

O

0

0

O

O

22° -22^° C.

78f0

16

IOO

96

96

O

O

2

2

2

2

22f°-23° C.

O7i°

13

IOO

82

82

6

6

IO

IO

2

2

23° -233° C.

565°

13

IOO

71

71

10

10

16

16

3

3

27° C.

45°

12

IOO

6.3

63

12

12

25

25

0

O

2Si°C.

33f°

II

IOO

53

53

13

13

34

34

O

0

24j°-25|° C.

225°

10

IOO

39

39

2O

20

41

41

O

O

24f°-25f° C.

nf°

10

IOO

23

23

22

22

55

55

O

0

Total . ,

OO

800

610 or 76.2%

187 or 22. 8%

7 or .8%

Experiment B. — A few experiments were made to determine
the effect of a dry surface at the angle of 90°. The animals ex-
perimented on were carefully dried on filter paper. They were
placed on a dry smooth glass plate with their heads directed
upward. Light was excluded and the animals examined after
two minutes. The movements of the animals on the plate were
readily seen by the tracks of the mucus secreted by them.

Of 60 animals used 31 dropped off, 6 or 15 per cent, of the
remainder oriented and crawled down, 31 or 79 per cent, crawled
upward and one crawled horizontally. The striking fact is the
number of individuals that oriented positively and crawled down.
This evidence goes to show that the animals tend to become
positive on a dry surface, which would evidently serve as a pro-
tective reaction when they are. left on rocks by a retreating tide.

The movements of the animals on the dry plate were regular

68

SAKYO KANDA.

in some cases and irregular in others. The figures given are,
therefore, somewhat arbitrary. To avoid this inaccuracy, a
few typical examples of movement are shown in figures as follows:

FIG. 2.

The arrows in the figure indicate the movements during the
experiment, and the dots the movements of the animals after
being placed on the plate but before the commencement of the
experiment, i. e., before the plate was covered by the dark box.

One thing is clear. Some of the animals on the dry glass
plate and also those placed on a dry wooden plate, in an experi-
ment the results of which will be seen shortly after this, oriented
themselves and crawled downward, in spite of the mechanical
difficulties due to their shells.

Experiment C. — In the next series, a dry wooden plate was used
instead of the dry glass plate, in the same manner as in the above
experiment. Of 60 animals used 24 dropped off, 19 or 52 per
cent, of the remainder oriented and crawled downward, 6 or
16 per cent, crawled up and 4 crawled horizontally. Positive
instead of negative geotropism was dominant, even though
the experiments were conducted at the angle of 90°.

Experiment D. — The following series was of the same nature,
except that the animals were placed with their heads down, and
the time of experimentation was limited to one minute. Of 60
individuals used 33 dropped, 21 or 77 per cent, of those remaining

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA. 69

on the plate crawled downward; only 2 oriented negatively and
these did not crawl upward; 4 failed to move. As the animals
were not compelled to orient themselves in order to move down,
the highest percentage of positive geotropism (21 individuals out
of 27 or 77 per cent.) was obtained in this case.

That the character of the surface played a part is evident.
But some explanation is necessary. When the cover-box was
removed, nine individuals which had crawled down as shown by
the mucus were found with the head end of their shells directed
vertically, or with slight inclination to the vertical, i. e., their
shells mechanically "oriented" upward by the "greater pull of
gravity" on the "heavier posterior region." Those individuals
probably could not resist the pull of gravity and keep their shells
from falling by their weight, while others could. Four individuals
crawled downward not quite vertically (on a diagonal line between
vertical and horizontal). It may be possible to explain this as
a movement which was resultant of the pull of gravity on their
shells and on themselves, — the resultant of a passive tendency
to orientation and an internal response to gravitation.

From the above observations, one may explain the habitat
of the snails, "which usually live on rocky surfaces which lie
between the high-tide mark and one foot below the low-tide
mark." The limitation of upward movement is due chiefly to
the exhaustion of moisture carried by their "foot" from the sea.
They seldom, if ever, crawl on dry rocks much higher than high-
tide mark, though they do crawl on rocks clear out of sea-water.
They seem to stop crawling when the moisture, which they carry
with themselves, is exhausted. This agrees with Mitsukuri's
and Haseman's observations.

It is not thinkable that either the dry glass plate, or the dry
wooden plate makes it easier for the snails to become oriented
and to crawl downward, than the ground-glass plate did. Never-
theless on the former surfaces many animals oriented themselves
and crawled downward against the resistance of the heavier
weight of their posterior region. If this is true, then the upward
and downward movements of the snails can not be explained either
by the resistance factor or the weight factor or by the both.

One thing is, of course, possible, that "the quantity and quality

7O SAKYO KANDA.

of the slime secreted" makes it easier for the animals to crawl
downward or upward, as Frandsen thinks in the case of slugs
(3, p. 205). The question, however, arises: Why, then, did not
the snails crawl upward instead of downward on the the dry glass
and wooden plates?

In this respect, Parker and Parshley offer a suggestive ex-
planation in the case of earthworms: "In the earthworm, the
response to dryness ... is apparently the selective extraction
of water from the peripheral protoplasm of the worm, a process
which is favored by capillarity of dry surfaces over which the
worm begins to creep and is probably dependent chiefly upon
evaporation from the surface of the worm itself. Under such
circumstances, the materials in the peripheral protoplasm of the
prostomium must become concentrated and probably initiates
stimulation by undergoing some such change as partial coagula-
tion" (15, p. 363).

If so, the explanation must be looked for in physiological
conditions, that is, internal, rather than external mechanical
factors. As Parker and Parshley explain, the dryness of the
surface may cause the partial coagulation of the materials in
the peripheral protoplasm of the "foot" as well as of the anterior
geo-sensitive region of the snails, which may also result in surface-
tension changes and contraction of "muscle elements," and
cause the motion of displacement or coalescence of the fluid
materials in the protoplasm of the muscle-fibers and internal
cells (as Lillie thinks; 7, pp. 252-255), which possess different
specific gravities. Under such circumstances, we may assume a
disturbance in the peripheral as well as internal cells, especially
perhaps in the cells of the geo-sensitive region or the geotropic
organs, statocysts, which disturbance may be assumed to stimu-
late and cause the response of the animals. Whether the animal
will go up or down will depend on the other stimuli besides gravity,
such as character of surface and degree of dryness which accom-
pany the stimulation of gravity itself. This view quite agrees
writh Loeb's (9) and Lyon's (n or 6, pp. 20-21).

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA.

4. The Effects of Light and Surface-film of Sea-water.

Experiment A. — Experiments were conducted in the aquarium
already described, which was closely surrounded by a four- fold
black cloth, all over the sides and bottom. The snails were
placed on the glass plate about 2.5 cm. below the surface film of
sea- water. Part of the glass plate extended above the surface.
By this arrangement, it was possible at the same time to test
Haseman's conclusion that the upward and downward movements
of the animals "are not due directly to either geotropism or pho-
totaxis, but to the action of the film of water" (4, p. 113). The
results are given in Table V.

TABLE V.

GEOTROPISM OF SNAILS IN AND OUT OF SEA-WATER IN DIFFUSE DAYLIGHT AT
DIFFERENT ANGLES OF INCLINATION OF A GLASS PLATE.

At beginning of experiments each head placed pointing downward. Table shows
results after one minute.

— Geotropism and

— Heliotropism

rt

[fl

-f- Heliotropism.

Geotropism.

Crawling

£

"ri

Hori-

u

I*-.

o

u

C

H
o

•g

Crawling
up in

Crawling
up in

At Sur-
face-

Crawling
Down-

Crawling
Diago-
nally

zontally
in Sea-
water.

0

Air.

Film.

ward.

Down-

E

fc

O

2;

ward.

H

No.

1°

No.

*

No.

*

No

*

No.

t

No.

Jf

2Ii°-I9i° C

90°

15

50

14

28

21

42

9

18

O

O

O

O

6

12

20^° C.

45°

14

50

7

14

17

34

3

6

7

14

12

24

4

8

20§° C.

ill"

10

50

o

o

2

4

o

o

15

30

27

54

6

12

Total

39

750

73 or 48.6%

61 or 40.6%

16 or

10

6%

At the angle of 90°, none were positively geotropic and only
six individuals out of 50 crawled horizontally in sea-water
(possibly due to the effect of light). At the angles of 45° and
n/4°, the number showing " positive geotropism " increased, as
would be expected. At the end of one minute, 21 individuals
out of 150 (14 at 90° and 7 at 45°), as shown in the first column
of negative geotropism, crawled upward clear through the surface-
film of sea-water into the air. 12 individuals (9 at 90° and 3 at
45°), as shown in the last column of the same, crawled upward
until they reached the surface-film of sea- water, and then

72

SAKYO KANDA.

crawled horizontally. Was this due "to the action of the film
of water," as Haseman claims? The question will be answered
after consideration of a few other experiments.

Experiment B. — The next series of experiments was of the same
nature as the above, but the animals were placed with their heads
upward instead of downward at the beginning. The results
given in Table VI. show that the starting position, that is, placing
the animal head downward or upward, at the beginning of the
experiment, makes a difference in the result.

TABLE VI.

GEOTROPISM OF SNAILS IN AND OUT OF SEA-WATER IN DIFFUSE DAYLIGHT AT
DIFFERENT ANGLES OF INCLINATION OF A GLASS PLATE.

At beginning of experiments each head placed pointing upward. Table shows
results after one minute.

i

u

— Geotropism and 4- Heliotropism.

+ Geotropism and
— Heliotropism.

W

KM

0

to

(I

bo

. of Trials.

of Animals.

Crawling
Up in
Air (No
Hesita-

Crawling
Up in
Sea-
water
(Not yet

Hesi-
tated at
Surface
then
Crawled

Hesi-
tated at
Surface
and
Crawled

Crawling
Down-
ward.

Crawling
Diag.
Down .

Crawling
Hori-
zontally
in Sea-
water.

a
E

V

0

Surface).

at

Surface).

Through
into Air.

Hori-
zontally.

No

4

No

$

No

4

No

4

No

$

No

<

No

<

igk° C.

90°

U

SO

23

46

O

O

6

12

II

22

O

o

0

o

IO

20

2l|° C.

45°

13

So

38

12

24

3

6

5

10

I

2

8

16

2

4

2l|° C.

IO

So

3

6

24

48

o

o | o

O

8

16

ir

22

4

8

Total. .

•37

I "O

rnfS nr in (

->8

or 1 8 6

16 or

10.6%

Here, again, eleven individuals at the angle of 90° and five at
45° crawled horizontally beneath the surface-film of sea-water.
Besides these, six individuals at the angle of 90° and three at 45°
"hesitated" at the surface-film when they reached it, and then
crawled upward through the film. This constitutes a puzzle for
the surface-film theory. But let us see further.

Experiment C. — Similar experiments were made in direct sun-
light. The glass plate in the aquarium was placed at 45°, which
made it parallel to the rays of sunlight as nearly as possible.
The outside and bottom of the cylindrical glass aquarium were
surrounded by black cloth as already stated; and two sections,
a and b (see Fig. 3), separated within by the glass plate, were

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA. 73

covered above so that the sunlight penetrated parallel to the glass
plate and through the open portion, c. So arranged, the intensity
and direction of the rays of the sunlight had to affect the reaction
of the snails to gravity.

Since the animals were negatively heliotropic, it was expected
that most of them would crawl downward in this arrangement.
Their heads, therefore, were placed upward at beginning, so that
definite movements of orientation could be observed. The

FIG. 3.

results after one minute were as follows: Of 50 animals in 15
trials, 5 or 10 per cent, had crawled upward, all diagonally; 37
or 74 per cent, had crawled downward, 22 of them being well
oriented and 15 diagonally; 8 or 16 per cent, crawled horizontally
and consequently across the lines of direction both of light and
of the effective component of gravity.

The predominance of downward movement was no doubt due
to the intensity and direction of the sunlight. This point be-
comes clearer in the next experiments, but it is clear that the
surface-film has nothing to do with such reactions.

Experiment D. — With the same point in mind, other experi-
ments were made in a similar way, but the animals were placed
about 1.5 cm. above the surface of the sea-water. In this case
29 individuals out of 50 oriented themselves downward, and
crawled in that direction through the surface-film of sea-water
into it. Three crawled horizontally at the surface-film instead
of going into the sea-water. 18 or 36 per cent, crawled upward.

74

SAKYO KANDA.

5. The Effect of the Surface-film of Sea-water.

The last of this series was planned in a little different way from
the above, entirely excluding light.

As a preliminary test, the snails were placed on the glass plate
at the angle of 56^° in the aquarium, which was half filled with
sea-water. When those crawling upward reached the surface-
film of sea-water and "hesitated," as Haseman calls it, beneath
the film, the whole arrangement was covered with the dark box.
Ten seconds later, the box was removed for observation. Nearly
all that had "hesitated" at the film, were found to have crawled
upward through the film.1 Certain individuals at that time had
already crawled upward as high as 3 cm. from the film, some 2
cm., and others were just above the film. This experiment was
repeated several times and these results were readily demon-
strable. Judging from the results, the animals did not seem to
have "hesitated" long, after the light was excluded. The
"hesitation" of the snails at the surface of sea- water seems to the
writer to be due chiefly to the effect of light instead of to action
of the surface-film of sea-water. This point also becomes evident
after further consideration.

Quantitative experiments were conducted as indicated above.
The results are tabulated in Table VII. and show rather compli-
cated conditions.

TABLE VII.

GEOTROPISM OF SNAILS AT THE DIFFERENT ANGLES OF INCLINATION OF A GLASS
PLATE IN AND OUT OF SEA-WATER IN TOTAL DARKNESS.

At beginning of experiment each head placed pointing downward.
results after one minute and after another half minute.

Table shows

Temperature of Sea-
water.

in

V

be

_rt

H
"o
o

No. of Animals.

— Geotropism

+ Geo-
tro-
pism.

Crawled
Horizon-
tally in
Sea-water.

A. After One Minute.

5. After Another
Half-minute.

|,
u
0

. *
> £

a

Stopped
lust Under

U

u •
<5-5

Crawling
up in
Sea-water.

" U

3 u

hi

C'S i
^ rt u

ho jj
1^ -^ rj

H

ZZ

H
U £

No.

$

No.

*

No.

«

No.

No.

No.

No.

No.

22° C.
2l\° C.
2\\° C.

90°

45°iQ

12

14
II

50
50
50

26

23

I

52
46
2

18
II
13

36

22

26

6
6
5

12
12
IO

18

14
II

4
2

8

4
i

5

O
3

21

O
5

IO

Total

37

ISO

I'OQ or 72%

67

24

15

1 They change their "minds" very rapidly!

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA. 75

"B" covers the activity of those animals which were still
below the surface at the end of one minute.

After the first minute, twenty-six individuals at the angle of
90°, twenty-three at 45°, and one at nJ/±° were found to have
crawled upward through the surface-film of sea-water. Eighteen
individuals at the angle of 90°, eleven at 45° and thirteen at 11%°
were just beneath the film of sea- water. And after another half-
minute, sixteen individuals at the angle of 90° out of the eighteen,
which were beneath the film of sea-water at the end of one minute,
nine at 45° out of the eleven, ten at n%° out of the thirteen, and
one at the same angle which was crawling horizontally at the
end of one minute, had crawled upward through the film of sea-
water. This is significant. At the time of second observation,
also, at the angle of 90° two individuals out of six, which were
crawling upward in sea-water at the end of one minute, at 45°
five out of six, and at il%° one out of ten which were crawling
horizontally at the end of one minute, had crawled upward
through the film. But at the angle of 90° two individuals out of
eighteen which wTere beneath the film at the end of the first
period, at 45° two out of eleven, and at n%° three out of thirteen,
that is, only seven individuals altogether out of 150, still remained
beneath the film even at the end of the second period. This is
decidedly contrary to the surface-film theory. And at the angle
of 90° two individuals out of six, which were crawling upward in
sea- water at the end of the first period, at u%° four out of five,
and at the same angle one which crawled downward in the first
period, were beneath the film at the end of the second period.

Referring again to the observation of Haseman that "during
a falling tide, some snails are crawling down beneath the surface,
some with the surface, and some above the surface" (4, p. 120),
and which Haseman claims is "due to the action of the film of
water," but not "to either geotropism or phototaxis," the writer
is inclined to draw the conclusion based on the results of the series
of the experiments above, that the movements of the snails in
question are not due, directly, if at all, "to the action of the film
of water," but to geotropism and heliotropism. In nature,
especially in the daytime when the sun is shining, a considerable
part would be played by heliotropism, as Mitsukuri already
observed.

76 SAKYO KANDA.

Furthermore, in the above experiment, even the seven individ-
uals, that still remained beneath the film at the end of the second
period, and whose failure to penetrate the film seems to favor the
surface-film action theory, may be considered in a different way.
The effect of light having been excluded, their failure to emerge
might be due to the susceptibility of these individuals to buoyancy
when they had to crawl out of sea-water. This must be taken
into consideration as has been shown. It might also in com-
bination with the effect of light explain the behavior of those
animals that crawled horizontally beneath the film, or that
"hesitated" there, as shown in a number of the preceding tables.
Whatever may be the explanation and even if one attributes the
slight hesitation to the film itself, it is evident that it has very
little effect in determining the total behavior of the animals.

6. The Effect of Chemicals.

Besides the above experiments, attempts were made to control
the negative geotropism of the snails by chemicals, especially by
salts and acids. But all were unsuccessful, except possibly an
alcohol experiment. Five snails were placed in a finger bowl
containing 100 or 150 c.c. sea- water, to which about 10 c.c. of
95 per cent, alcohol were added without stirring. The finger
bowl was shaded. The animals crawled upward on the vertical
wall of the bowl, but as soon as they reached the upper layers they
turned round, and crawled downward. The negative geotropism
was thus apparently reversed. But if the alcohol and sea-water
were well mixed by stirring, the reversal was not definite and the
experiment was not followed further.

IV. DISCUSSION.

That Littorina littorea is negatively heliotropic was first shown
by Bohn (i), and this fact has been confirmed by the present
writer. Morse, however, has published puzzling conclusions
from his observations on this species. He states: "During the
days of June, they were, as a rule, negatively phototactic, and as
night approached, they became positively phototactic. However
after July 18, the preponderance of positive phototataxis during
the day was very noticeable. This period of transition corre-

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA. 77

sponded to the time of change from spring to neap tide, during
which the specimens out on the rocks were exhibiting a corre-
sponding change in phototaxis, for the water did not reach them;
their behavior tallied with the description of Mitsukuri, who
showed that when desiccated, periwinkles became positively
phototactic, and when wet, turned negatively phototactic" (14,

P- H5)-1

Unfortunately, Morse has given no description of his methods
of experimentation. But, judging from his statements, he has
entirely overlooked the effect of gravity on the animal, as Mit-
sukuri did. Was not the supposed "positive phototaxis" "after
July 18" really an effect of gravity? Littorina littorea does not,
in the writer's opinion, crawl upward "on the rocks" on account
of positive heliotropism, but on account of negative geotropism
and in spite of negative heliotropism, which is the unmistakable
reaction of the animal to light. Negative heliotropism in the
animal is demonstrable even at "night," if the experiment is
conducted on a horizontal surface, and if there is any source of
light present. The writer believes that Morse's statement that
"as night approached, they became positively phototactic" is
incorrect. What really happens is this: As the light stimulus
diminishes, the gravity stimulus becomes preponderant and the
animals are controlled by their negative geotropism.

Frandsen, as already stated, proposes two factors for the de-
termination of geotropism in the slug. The second will be con-
sidered first. "All the conditions being the same,"1 he says, "it
is this factor (the relative proportions of the length of the anterior
and the posterior regions of the animal's body) which 'determines
whether the head end will be directed up or down.' If the ratio
of the length of anterior to posterior region of body is 2 : 3, or
more, and the mucus is of good quality and sufficient quantity,
the slug will be positively geotactic. If the ratio is 3 : 5, or less,
the animal will usually migrate upward, and the nearer the
ratio approaches I : 2 the more apt is the slug to respond nega-
tively. . . . All slugs have a natural tendency to move towards
the earth. This tendency is masked in the animals which are
negatively geotactic on a glass plate by the greater pull of gravity

1 Italics not in the original.

78 SAKYO KANDA.

on the disproportionately larger and heavier posterior region of
the animal" (3, p. 205).

Judging from these statements, the center of gravity must lie
somewhere in the posterior region of the animal's body. In
other words, the posterior region is heavier than the anterior.
According to Frandsen, the negative geotropism of the animal
"on an inclined glass plate," however, is not due to the heaviness
of the posterior region, but to "the relative proportions of the
length of the anterior and the posterior regions of the animal's
body." In short, Frandsen's idea may be expressed thus: The
longer (and heavier?) posterior region being pulled by the con-
stant force of gravity, the slug becomes positive to it at the
minimal resistance in favor of the ratio above mentioned; on
the other hand, it becomes negative at the maximal in disfavor
of the ratio. In the latter sense, the "resistance theory," of
course, approaches the "mechanical theory."

As pointed out before, this seems to be fairly supported by the
results obtained by the writer, which were given in Tables II.,
III. and IV. This explanation is, however, focused to a limited
group of facts; because the snail becomes negative to gravity on
the nearly horizontal surface of a glass plate, where the minimal
resistance should be expected. This fact is opposed to Frandsen's
conclusion. Furthermore, as has already been shown, many
snails on the dry wooden plate at the angle of 90° of inclination
oriented themselves downward and crawled in that direction,
even though in so doing, they met a great mechanical difficulty
on account of the heavier posterior region. If the mechanical
theory of Frandsen is true, they were under the most favorable
conditions for crawling upward instead of downward. This was
not, however, the case. Was this because of the first factor of
Frandsen, that is, "the quantity and quality of the slime se-
creted" (3, p. 205). The second factor in question is by no means
separable from the first. Both go together. To make Frand-
sen's statement clearer, therefore, it may be expressed as follows:
'The relative proportions of the length of the anterior and the
posterior regions of the animal's body" "being the same, it is
this factor," that is, "the quantity and quality of the slime se-
creted," "which determines whether the head end will be directed

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA. 79

up or down." It is fair to examine the table (V.) which is given
by Frandsen based on his results. Take the animal No. "13,"
on which the series of observations, "a" and "£" were made.
The condition of the animal was "active" in series "a" and also
in series "6," that is, it was under the "same" condition.
Being the same individual, theoretically there was no difference
in the "ratio." Nevertheless, Frandsen obtained different
results in "a" from "6." Moreover, he shows different results
in the same individual under the same condition and in the same
series of observations, "a" or "6." This is more striking in
the animal No. "23," on which the series of observations, "a,"
"&," "c" and "d," were made under the same condition, "good."
But he obtained 64.2 per cent, of negative geotropism in series
"a," 62.5 per cent, in the "b," 100 per cent, in the "c," and 100
per cent, in the "d." In other words, the same individual under
the same condition was, at least, 35.7 per cent, positive in "a,"
37.4 per cent, in "&," and o percent, in "c" and "d." This kind
of variation is also found in the animals Nos. "8," "24," "25"
and "27." If so, "all the conditions being the same," it is not
entirely the first factor, nor the second, "which 'determines
whether the head end will be directed up or down,' ' but there
must be something else besides these two factors, which makes
the response variable. Frandsen does not seem, therefore, to be
justified even by his experimental data in his conclusions. And
strictly speaking, the so-called mechanical theory of geotropism
seems to the writer to be misleading in any case, because it is
not a living response to stimulus but a purely mechanical one
which would be seen as well in the dead organism, if it could be
moved. This is no tropism at all.

In this respect, therefore, the writer is inclined to take Daven-
port and Perkins's conclusion into consideration, as, at least, one
of the factors, that is, "some internal condition of the individual."
Without this "internal," or physiological, factor, it is difficult to
explain why geotropism varies in the different individuals and also
in the same individual at the different angles of inclination of the
same support and the same angle of inclination of different
supports.

Haseman's conclusions demand close consideration on several

8o SAKYO KAXDA.

points. Haseman states: "Even more interesting is the fact
that when, with a vertical surface, a stone, upon which snails
were crawling at random, was raised out of the sea, the snails
always followed the vanishing film of water even when the vertical
surface was rotated through an angle of 180°. In this case, the
rotation of the vertical surface would reverse the direction of
motion of the film of water and the snails would at once turn
around and follow it. But if most of the water was previously
removed from the surface of the stone, in order that the film
might entirely disappear before the snails (which were crawling
downward in the direction of the vanishing film) had reached the
lower surface and if, as the film was drying up, the vertical surface
was rotated through an angle of 180°, the snails continued to crawl
for some time in the direction in which they had started. In other
words, the snails crawled upward instead of downward. They
continued to crawl thus until the rough surface, food and moisture
either deflected or stopped their movements. In the above
experiment, the mere turning of the moist but filmless surface
through an angle of 180° does not seem adequate to reverse at
once the reaction to gravity and light, if either of these have a
direct influence on the rhythmical movements of Littorina"
(4, p. 116).

Certainly this is not a simple matter. In the first phase,
Haseman does not state, where "the above experiment" was
conducted, what condition of sunlight or diffuse daylight, there
was, and if the experiment was in sunlight, in what direction
the rays were falling and so on. His description as well as ex-
periment is not at all accurate. Judging from the above descrip-
tion, however, he seems to have conducted the experiment "in
nature," and so in sunlight. If so and if the sun were fairly
above, it is small wonder that 'Svhen the vertical surface was
rotated through an angle of 180°," thus rotating the motion of
the film of water, " the snails would at once turnaround and follow
>ince the snails are negative to light, as has been shown. In
i In- case of Haseman where the film was reversed, in which "the
vertical surface was rotated through an angle of 180°," and "the
snails continued to < r,i\vl for some time in the direction (upward)
in \\liieli they had st. tried," another explanation is possible. The

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA. 8l

snails being negatively geotropic, the direction and intensity
of sunlight were not at the time of this experiment presumably
in favor of the reversal of negative geotropism. Therefore,
"the snails continued to crawl" "upward instead of downward."
This is, of course, reasoned upon the results which were obtained
by the writer. At any rate, Haseman seems not to have checked
the counteracting forces of gravity and light and consequently
all his experiments are unreliable.

One afternoon from four to six o'clock, Dr. Irving A. Field and
the writer made special observations of tidal (rising) influence
upon the snails at the south shore of Ram Island. The animals
were found in large numbers covering a pair of long square beams
which formed an inclined railway. These beams presented
both vertical and sloping surfaces. In this vicinity there were
also rocks and stones of various shapes, their surfaces sloping at
many different angles on which numerous snails — oriented with
their heads upward — were exposed in dim sunlight. When the
tide rose higher and higher with no waves and reached to the
areas where the animals had been exposed so long that their
outer surfaces were completely dried, nearly all of them, if not
the entire number, gradually turned head downward and crawled
in that direction; some of them moved downward several inches
from the original spots, while some others moved about "at
random" as if they were seeking food; and still others turned
downward when the surface-film of sea-water came in contact
with them, while many did so after the surface-film had passed
over them to the extent of 0.5 or i cm., more or less. If the
snails follow "the direction of motion of the film of water," and
also, if light has no influence on the rhythmical movements of
Littorina, as Haseman claims, why did not the snails "crawl
upward instead of downward"? Thus considered, it becomes
evident that Haseman's observation, or experiment, was not
accurate, while Mitsukuri's is, in this respect, confirmed by the
writer.

It is very strange to note that Haseman has no notion of the
influence of gravity on the snails, although he has observed phe-
nomena which would naturally remind one of it. "When
individuals are left high and dry on vertical surfaces during low

82 SAKYO KANDA.

tide," he says, "they come to rest 'directed upward,' i. e., with
their heads toward the sky. This is true for all sides of the stones
and is obviously due to the shape of the apertures of the shells
which makes it far easier for exposed individuals to cling thus to
vertical surfaces" (4, p. 117). How does he know that "the
shape of the aperture of the shells" makes it easier to cling to
vertical surfaces? Is it not rather true that the natural tendency
of negative geotropism of the snails, favored by the heavier
posterior region of the shells instead of the shape of the aperture,
"makes it far easier for exposed individuals to cling thus to
vertical surfaces"? This sounds more reasonable and nearer
to the fact than Haseman's supposition.

V. SUMMARY AND CONCLUSIONS.

1. Littorina littorea crawls up the vertical (inside) wall of a
beaker in a dark aquarium, though the sea-water be better sup-
plied with oxygen at the bottom than at the top. It is negatively
geotropic.

2. This snail is negatively heliotropic.

3. The posterior region of the snail has a greater specific
gravity than the anterior region.

4. In sea-water, the larger the angle of inclination (to the
horizontal) of the surface on which the animals move the larger
is the number of negatively geotropic animals; and the smaller
the angle of inclination, the larger the number of animals which
move downward and are perhaps positively geotropic.

5. In the air, the number of animals showing negative geo-
tropism is always higher than that in the sea-water.

6. On a ground glass plate, the animals are less negatively
geotropic than on a plain glass plate.

7. On a dry plain glass plate, a number of individuals oriented
positively and crawled downward even at the angle of 90° (ver-
tical) though this never happens on the moist plate. This is
more striking on a dry wooden plate.

8. When the animals are placed with their heads down on a
dry wooden plate, the highest percentage of positive geotropism
is obtained.

9. The snails "hesitate" at the surface-film of sea-water, when
light is not excluded.

GEOTROPISM OF THE MARINE SNAIL, LITTORINA LITTOREA. 83

10. In the direct sunlight, the snails are apparently more
positively geotropic than in darkness, due to the fact that they
are negatively heliotropic.

11. In the direct sunlight, 58 per cent, of the animals orient
themselves head downward and crawl in that direction through
the surface-film of sea-water into it.

12. In darkness, the snails, which "hesitated" at the surface-
film of sea-water in the daylight, crawl upward through the film.

13. From the experimental results which the writer has ob-
tained, he concludes that neither the mechanical theory, nor the
pressure theory, nor the resistance theory is adequate to explain
the phenomenon of the negative geotropism of Littorina littorea
but a physiological one, that is, the statocyst or statolith theory.
This theory is the more likely since these snails have statoliths
(17, pp. 119-120). The writer, however, has no direct evidence,
at present, in favor of the statolith theory. He is led to accept
it largely by the method of exclusion. Furthermore evidence
from many sides based on the experiments of the writer causes
him to conclude that the surface-film theory is also not correct.

In conclusion, the writer wishes here to acknowledge his
indebtedness to Professors Walter E. Garrey, Ralph S. Lillie,
and Elias P. Lyon, for their valuable suggestions and criticism
on his experiments at the Marine Biological Laboratory at Woods
Hole, Mass., during the summers of 1912 and 1913. His thanks
are also due to Professor Frank R. Lillie for the privileges of the
Laboratory and to Professor Lyon for criticism and suggestions
in the preparation of manuscript.

VI. BIBLIOGRAPHY.

1. Bohn, Georges.

'05 Attractions et Oscillations des animaux marins sous 1'influence de la
lumere. Bull. d. FInst. gener. d. Psychol., pp. 171-181.

2. Davenport, C. B. and Perkins, Helen.

'97 A Contribution to the Study of Geotaxis in the Higher Animals. Jour,
of Physiol., Vol. 22, pp. 99-110.

3. Frandsen, Peter.

'01 Studies on the Reactions of Limax maximus to Directive Stimuli.
Proc. Am. Acad. Arts and Sci., Vol. 37, pp. 185-227.

4. Haseman, J. D.

'n The Rhythmical Movements of Littorina littorea Synchronous with
Ocean Tides. BIOL. BULL., Vol. 21, pp. 113-121.

84 SAKYO KANDA.

5. Jennings, H. S.

'07 Behavior of the Starfish Asterias forreri de Lorriol. Physiol. Publ. of
the University of Calif., Vol. 4, 1907, pp. 53-185.

6. Kanda, Sakyo.

'14 On the Geotropism of Paramoecium and Spirostomum. BIOL. BULL.,
Vol. 26, pp. 1-24.

7. Lillie, Ralph S.

'12 The Physiological Significance of the Segmented Structure of the
Striated Muscle Fiber. Science, N. S., Vol. 36, pp. 247-255.

8. Loeb, Jacques.

'88 Die Orientirung der Thiere gegen die Schwerkraft der Erde. (Thi-
erischer Geotropismus.) Sitz.-ber. Wiirzburg Physiol. -med. Gesell.
s. 5-10.

9. Loeb, Jacques.

'97 Zur Theorie der physiologischen Licht und Schwerkraftwirkungen.
Archiv. f. d. ges. Physiol., Bd. 66, s. 439-466.

10. Loeb, Jacques.

'oo Comparative Physiology of the Brain and Comparative Psychology.

N. Y., G. P. Putnam's Sons, x +309 p.
it. Lyon, E. P.

'05 On the Theory of Geotropism in Paramoecium. Am. Jour. Physiol.,

Vol. 14, pp. 421-432.

12. Mitsukuri, K.

'01 Negative Phototaxis and Other Properties of Littorina as Factors in
Determining its Habitat. Annotationes Zoologicse Japonensis, Vol. 4,
Pt. I., pp. 1-19.

13. Moore, A. R.

'10 On the Righting Movements of the Starfish. BIOL. BULL., Vol. 19,
pp. 235-239.

14. Morse, Max Withrow.

'10 Alleged Rhythm in Phototaxis with Ocean Tides. Proc. Soc. Exp.
Biol. and Med., Vol. 7, pp. 145-146.

15. Parker, George H.

'n The Mechanism of Locomotion in Gastropods. Jour. Morphology,
Vol. 22, pp. 155-170.

1 6. Parker, G. H. and Parshley, H. M.

'n The Reactions of Earthworms to Dry and to Moist Surfaces. Jour,
of Ex. Zool., Vol. II, pp. 361—363.

17. Pelseneer, Paul.

'06 A Treatise on Zoology. Pt. 5. London, Adam and Charles Black,

355 P-

18. Prentiss, C. W.

'01 The Otocyst of Decapod Crustacea: its Structure, Development, and
Functions. Bull, of Mus. Comp. Zool., Vol. 36, pp. 165-251.

THE GEOTROPISM OF FRESHWATER SNAILS.1

SAKYO KANDA.

CONTENTS.

I. Introductory 85

II. Materials 86

III. Experimental 87

1. Heliotropism of Physa gyrina Say 87

2. Geotropism of Physa and Other Species, with the Lung Empty and

with the Lung Filled with Air 88

(a) Observations on Physa 88

(6) Observations on Planorbis and Limnaa 89

3. Geotropism of Physa with Lung Empty and Filled with Air, in Pres-

ence of Food 92

4. Geotropism of Physa at the Different Angles of Inclination of the

Supports in the Air and in Total Darkness 92"

(a) Experiments with a Plain Glass Plate 92

(6) Experiments with a Ground Glass Plate 94

5. Summation of Gravity and Light Stimuli 95

IV. Summary and Conclusion 96

V. Bibliography 97

I. INTRODUCTORY.

Walter (10) and Dawson (2) have investigated the geotropic
reactions of Physa and other freshwater snails in connection
with their respiratory phenomena. However, their observations
do not agree on certain points. Walter (10, p. 26) and Dawson
(2, p. 93) agree with each other that freshwater snails are nega-
tively geotropic, "when their lungs are empty." The snails
being air-breathing forms, it is, of course, necessary for them to
crawl up to the surface of water for their air supply, "although
their specific gravity is meanwhile gradually increasing through
exhaustion of the air," as Walter expresses it. For this upward
crawling, the pull of gravity would be expected to act as a
"directive force."

Walter and Dawson, however, depart from each other when
they come to consider positive geotropism in these snails. The
former states that, "after filling the lung with air, they are

1 From the Physiological Laboratory of the University of Minnesota, Minnea-
polis.

8.5

86 SAKYO KANDA.

positively geotropic," so that "they tend to climb down." The
latter, on the other hand, states that when the snails "have suf-
ficient air they become indifferent to gravity and crawl in all
directions."

Moreover, they do not agree concerning the behavior of the
snails from which "the air supply is cut off." According to
Walter, "after reaching the highest point in the flask," which
was in an inverted position in water, "and finding themselves
unable to renew their supply, their ordinary behavior, to which
there were some exceptions, was to let go and drop like dead
weights " (10, p. 27). Dawson denies this statement of Walter
as follows: "Physa, after they have been denied atmospheric
air for some time, manifest indifference to the influence of gravity,
and scatter over the sides and bottom of the bottle. They have
never been observed to let go and drop like dead weights upon
being denied atmospheric air" (2, pp. 104, 105).

In this paper an attempt has been made to compare certain
experimental results obtained by the writer with those of his
predecessors. A comparison is also made with other results
obtained by the writer with marine snails.

The experimental work was done in the physiological labora-
tory of the University of Minnesota, under the direction of
Professor E. P. Lyon, during the academic year of 1913-1914,
while the writer was holding a Shevlin Fellowship. The writer
•expresses his appreciation of the interest and suggestions of
Professor Lyon throughout the course of the work. To Professor
John M. Holzinger, the principal of the State Normal School of
Winona, Minn., the writer acknowledges indebtedness for the
identification of the forms experimented upon.

II. MATERIALS.

Common freshwater snails, Physa gyrina Say, Planorbis
trivolvis, Limn&a stagnalis, and Limncea Columella, were used for
the work. The snails were kept in glass aquaria together with
green algae. They seemed to be perfectly healthy, and were
observed to grow.

THE GEOTROPISM OF FRESHWATER SNAILS.

III. EXPERIMENTAL.

To study the behavior of an animal it is always necessary to
discriminate the* force under consideration as much as possible
from other forces which may act simultaneously with it, favorably
or antagonistically. Oxygen for pulmonate animals such as
Physa is, of course, very important. That food is another factor
in determining behavior of living animals need hardly be men-
tioned. Contact stimuli must also be considered. Light is often
important. These with gravity are the chief forces which should
be borne in mind.

The effect of the force of gravity on Physa and others is the
problem with which this paper is chiefly concerned. But in
considering this the other forces just mentioned must also be
considered. Light especially must be taken into account.

/. Heliotropism of Physa gyrina Say.

Walter (10, pp. 23-24) has experimentally shown that Physa
primeana Tyron and others are generally negatively heliotropic.
A series of experiments was made with Physa gyrina Say to
compare results with the results obtained by Walter. The
experiments were conducted as follows: Five selected individuals
were placed on a smooth glass plate with their anterior ends
facing direct sunlight. The surface of the plate was carefully
moistened. During experiments the angle of the rays of sunlight
was about 22.5°. The glass plate was horizontally placed in air.

TABLE I.

HELIOTROPISM OF Physa ON A HORIZONTAL MOIST GLASS PLATE IN AIR AT
THE ANGLE OF 22.5° OF THE RAYS OF SUNLIGHT.

Table shows results after one minute. Feb. 14, 1914, 9:30 A. M. Tempera-
ture, 22° C.

No. of
Animals.

No. of
Trials.

— Heliotropism. + Heliotropism.

Horizontally Crawled.

No.

1°-

No.

$.

No.

j6.

I

20

20

IOO O

O

O

o

2

20

2O

IOO 0

O

0

o

3

2O

20

IOO

O

O

0

o

4
5

20
20

10

4

50

20

4
16

20

80

6
o

30

0

Total 5

IOO

74 or 74%

20 or 20%

6 or 6%

SAKYO KANDA.

The results, given in Table I. confirm Walter's conclusion.
However, there are marked individual differences. Number 5,
for instance, was exceptionally positive to light, while Numbers I ,
2, and 3 were always negative. Nevertheless, 74 per cent, were
negative to light, and only 20 per cent, positive. From these
results, the conclusion may be drawn that Physa gyrina Say is
generally negatively heliotropic.

Dawson (2, pp. 60-61), on the other hand, has observed that
darkness interferes with the activity of Pliysa. The writer's
observations seem to be in accord with Dawson's. Five of the
individuals above mentioned were kept in water under frequent
observation one afternoon (from 2.40-4 P. M.) in darkness.
Three of them were observed to crawl to the surface and then
down again only two or three times; and two of them only once
during the period, one 55 minutes and the other 70 minutes after
being covered. The rest of the time they did not move at all.

2. Geotropism of Physa and Other Species, with the Lung Empty
and with the Lung Filled with Air.

As already mentioned, Walter and Dawson disagree on their
observations concerning positive geotropism in Physa and other
species after the air supply is cut off. Neither has, however,
furnished the quantitative evidence from which the conclusion
was drawn. The writer, therefore, has made some quantitative
observations on this point.

(a) Observations on Physa. — Five selected individuals of Physa
were placed in a beaker containing about 300 c.c. of water. The
beaker, which had vertical sides and a horizontal bottom, was
placed as near as possible in optimum daylight. The water which
was used for this purpose was taken from the dish in which
Physa were kept, and was filtered when necessary. The following
observations have two aspects, (i) response of Physa to gravity,
when the lung is empty, and (2) when the lung is full of air. The
results are given in Table II.

The negative geotropism of Physa, when its lung is empty, is
precise and marked. It is impossible to mistake it. It is also
evident, on the other hand, that the positive geotropism of the
animal after filling the lung, though not as precise as the negative

THE GEOTROPISM OF FRESHWATER SNAILS.

89

geotropism observed when the lung is empty, is predominant in
the majority of cases, that is, in over 90 per cent. Cases of
"indifference" to gravity numbered less than 10 per cent. That
Physa should become positive to gravity after taking in air
supply seems to the writer to be quite natural, since the animal
is primarily positively geotropic as will be shown later. More-
over, it is worth mention that each individual occasionally showed
a peculiar "habit." It crawled up as usual, turned downward
when it had reached the surface, arid crawled down, making no
effort to get air. The number of observations of this phenomenon
for each individual is given for reference in Table II. in the second
horizontal column and indicated by a star.

TABLE II.

GEOTROPISM OF Physa AT THE ANGLE OF 90° OF INCLINATION OF A SUPPORT IN
WATER, AFTER ANIMALS HAVE TAKEN AIR IN THE "LUNG-SAC."

£

-r Geotropism.

" Indifferent to Gravity" (?)

"S

*c5

o

Dates.

"s

P.

H

o

I

Vertically or
almost Ver-
tically Ori-
ented Down-

Diagonally
Oriented
Downward

Horizontally
Oriented
and

Irregularly
Crawled.

Crawled on
the Surface-
Film of

V

H

o

<

ward and
Crawled.

and
Crawled.

Crawled.

Water.

18

i

7

3 2

I

I

. .§

i*

3

I 0

O

O

<5 § <

2

6

5

O

O

I

10 o

15

2*

3

0

O

O

O

O 1O *"*

1 1 10

U

3

5

i

o

I

0

O IO CO

1 N. 6

o

ON

1 1

3*

4

o

a

o

o

ro CO H

t^ OO H

4

8

o

o

o

o

N N CO

1 1

4*

I

2

0

o

o

d d d

A) OJ QJ

PPP

5

13

0

o

o

I

19

5*

5

O

o

o

o

74

5

67 or 90.5%

7 or 9.4%

Observations on Planorbis and Limnaea. — With single
individuals of these genera, the same tests as above were made
and still more conclusive results were obtained. Since these
snails were comparatively large forms, they were favorable for
observation. If they were dislodged they fell to the bottom,
provided their lungs were "relatively empty." If the lungs on
the contrary were full of air, dislodged snails floated on the surface

SAKYO KANDA.

of water. The same was true in the case of Physa. But with
this species, as has been already pointed out, it was impossible
to exclude light in the experiments because it was inactive
in darkness. Consequently the negative heliotropism tended
to blur the geotropism. This disadvantage was entirely removed
in Planorbis and Limncea, because these forms wTere active even
in total darkness.

Observation I : Planorbis trivolvis was put in the 6oo-c.c. beaker
full of water. It either floated or sank, depending on the con-
dition of its lung as above mentioned. If it floated, i. e., was
lighter than water, observation on positive geotropism was made;
if it sank, i. e., was heavier than water, observation on negative
geotropism was made. Either event, therefore, was useful. At
about two-minute intervals, the snail was dislodged by a test-
tube cleaner, and then was covered by a dark-box. The results
are given in Table III.

TABLE III.

Geotropism of Planorbis trh'oh'is on a Vertical
Side of a Beakei Full of Water in Total Dark-
ness When it Was Lighter than the Water.
Table Shows Results Within Two Minutes.

Geotropism of Planorbis trh'oh'is on a Vertical
Side of a Beaker Full of Water in Total Dark-
ness When it Was Heavier than the Water.
Table J-hows Results After Two Minutes.

Temp, of
Water.

No. of
Trials.

-f Geotropism.

Temp, of
Water.

No. of
Trials.

— Geotropism.

No.

*•

No.

%.

19° C.

30

30

IOO

IQ° C.

IO ' IO

IOO

Discussion is hardly needed. A hundred per cent, in both
positive and negative geotropisms was obtained. The orientation
of positive geotropism in this snail was just as precise as that of
negative geotropism.

Observation 2 : In the same manner as above, Limncea stagnalis
was observed in total darkness. The results are given in
Table IV.

TABLE IV.

Geotropism of Limniea stagnalis on a Vertical
Side of a Beaker of Water in Total Darkness

When it Was Lighter than Water.
Table Shows Results After Two Minutes.

Geotropism of Limna-a siagnalis on a Vertical
Side of a Beaker of Water in Total Darkness

When it Was Heavier than Water.
Table Shows Results After Two Minutes.

Temp, of
Water.

No. of
Trials.

+ Geotropism.

Did Not
Move.

Temp, of
Water.

No. of
Trials.

— Geotropism.

No.

i,.

No.

• i.

19° C.

35

24

68.5

II or
31-4%

19° C.

25

25

IOO

THE GEOTROPISM OF FRESHWATER SNAILS. QI

As may be seen from the table, the snail did not respond eleven
times out of thirty-five trials, when it was lighter than water.
This means that it was still floating when the complete two-
minute interval was over. The failure of response to gravity
in these cases need not be interpreted as "indifference " to gravity,
because it was often observed that the snail had opened its air
cavity at the time of observation. Evidently it was taking in
more air. The geotropic response failed, therefore, because the
lung was not full of air. All internal conditions being equal,
the snail tends to crawl down, if its lung is full of air. If it
crawled downward, it crawled vertically, orienting itself with its
anterior end accurately in that direction.

Observation 3: Limncea columella, again, when lighter than
water, failed to respond to gravity nearly half the time, as Table
V. has shown.

TABLE V.

Geotropism of Limncea columella on a Vertical Side of a Beaker Full of Water in Total Darkness.
When it was Lighter than the Water. Table Shows Results After Two Minutes.

Temp, of Water.

No. of Trials.

+ Geotropism.

Did Not Move.

No.

t

19° C

35

18

51-4

17 or 48.5%

Only a limited number of observations could be made at a
sitting, as the animals ceased to respond at all. This was
probably a fatigue effect.

It was observed three or four times that the snail crawled down
with its shell in a horizontal position ; and two or three times with
the anterior end of its shell pointed up. It thus crawled down
even against mechanical disadvantage.

The writer was unable to obtain satisfactory observations on
this snail's negative geotropism. It crawled up to the surface
for air. But when it was subjected to experimentation, it
retreated into its shell and did not readily come out.

The writer by the above observations confirms Walter's con-
clusion regarding the effect of taking air on the geotropism of
Physa and other species.

92 SAKYO KANDA.

3. Geotropism of Physa with Lung Empty and Filled with Air, in

Presence of Food.

Food being one of the strongest of forces in determining
behavior, it was thought advisable to observe its effect on
Physa. Green algae were carefully placed on the bottom of the
beaker in which there were five selected individuals of Physa.
It was surprising to find that with food present Physa seldom
crawled up to the surface for the air supply. After obtaining
air moreover most of them crawled down just as precisely as
they crawled up. It must also be added that they were not in a
starving condition previous to these experiments.

It should be remembered, however, that in this case three
forces, that is to say, (i) light, to which Physa is negative,
(2) gravity, to which it is positive after taking in air, and (3)
food, to which it is presumably positive, were here combined.
The result of the combination of these three forces is the acceler-
ation of positive geotropism. Negative geotropism, on the
other hand, is retarded, even though it is very important for the
air supply.

4. Geotropism of Physa at the Different Angles of Inclination of the

Supports in the Air and in Total Darkness.

Imagining that negative and positive geotropisms of Physa
and others are due only to respiratory phenomena, Walter claims
that "so far as gravity alone is concerned, they should show no
response at all" (10, p. 26). According to Dawson, geotropism
(negative) is only possible "when their lungs are empty, but when
they have sufficient air they become indifferent to gravity and
crawl in all directions" (2, p. 93). In a certain sense, therefore,
Walter and Dawson agree on this point. This contention will
be experimentally examined in this section.

(a) Experiments with a Plain Glass Plate. — After a few trials
it was found that Physa was positively geotropic even at a slight
inclination of a plain glass plate in the air and in total darkness.
The following method of experimentation was therefore adopted.
The well-moistened glass plate being held slightly inclined, the
animal was placed upon it with its head down. When five se-
lected individuals had been so placed, the plate was reversed,

THE GEOTROPISM OF FRESHWATER SNAILS. 93

and was placed on a "rack" specially made for angle determina-
tion. The whole arrangement was covered as soon as possible
with a dark box. Curiously enough, darkness seemed not to
interfere very much with Physa s activity, if the experiment was
conducted in the air, though it did interfere in water, as already
shown. Physa oriented itself in the line of the force of gravity
and crawled in that direction. The relative weight of Physa
is about a thousand times as great in the air as in the water. This
probably made a difference in its activity in the air even in total
darkness.

The results given in Table VI. show that from the angle of
io%° of inclination to that of 563/4°, positive geotropism in-
creases as the degree of the angle increases. It should be added
that this was not because the lung was full of air. On the con-
trary, about two thirds of these positive animals sank, when they
were tested in water. This means that their lungs were empty.
Negative geotropism and horizontal crawling, on the other hand,
decrease in reverse proportion as the angle of inclination is
increased. This was not because the lung was empty. On the
contrary, about half of these negative animals floated, when they
were tested in water.

But there was a limit to the degree of inclination of the support,
beyond which Physa could not actively move on the plain glass
plate on account of the force of gravity. This is significant. At
the angle of 67^° positive geotropism suddenly decreases, and
negative geotropism and horizontal crawling increase. Gravity,
of course, is constant and always exerted vertically. But the
effective force exerted on the animals depends upon the inclina-
tion of the surface on which the animals crawl. The exertion
required to enable the animals to move on a horizontal surface is
least; that required on a vertical surface is greatest. At the
angle of 67^°, therefore, the effective force of gravity was so
great that some individuals of Physa could not actively move
against it (in air). An apparent increase of negative geotropism,
therefore, was the result. This becomes clear, when one con-
siders the failure of the experiments at the angle of 78^°, which
was due to the fact that nearly all the animals passively slid down
the plate, mostly with their heads up. The optimum inclination

94

SAKYO KANDA.

seems to be about 56°. The inference may be made that positive
geotropism in Physa is an active process, and that negative geo-
tropism, on the other hand, is due largely, though not entirely,
to "passive orientation."

The so called mechanical theory of geotropism, however, can
not be applied even in the case of negative geotropism. This is
obvious when one remembers that Physa becomes negative to
gravity when "its lung is empty," even though its specific
gravity is less than that of water; and positive when its lung is
full of air, even though its specific gravity is greater than that of
water. The essential factors, therefore, which determine the
geotropic orientation, either positive or negative, of Physa seem
to be internal, that is, physiological ones (cf. 3, 4, 5, 7, 8 and
9 in the bibliography).

TABLE VI.

GEOTROPISM OF Physa AT THE DIFFERENT ANGLES OF INCLINATION OF A SMOOTH
GLASS PLATE IN THE AIR IN TOTAL DARKNESS.

At beginning of experiments, each head placed upward. Table shows results
after one minute.

-+- Geotropism.

— Geotropism.

Temp,
of
Room.

Angles.

No. of
Trials.

No. of
Ani-
mals.

Oriented and
Crawled Down-
ward.

Crawled Up-
ward.

Horizontally
Crawled.

No.

*

No.

f

No.

4

20.5° C

67-5°

5

SO

2Q

58

II

22

10

20

20.5° C

56 M°

5

50

47

94

I

2

2

4

20.5° C

45°

5

50

4i

82

6

12

3

6

20.5° C

33X°

5

50

35

70

ii

22

4

8

20.5° C

22^°

5

50

3i

62

12

24

7

14

20.5° C

nM°

5

50

30

60

II

22

9

18

Total

30

300

213 or 71%

52 or 19.3%

35 or n.6%

From the above results the writer thinks that both Walter
and Dawson overlooked the fact that Physa is naturally positively
geotropic.

(b) Experiments with a Ground-Glass Plate. — Contact stimuli,
as stated, affect the behavior of animals. Supposedly it might
affect the geotropism of Physa. The same methods were used
here as in the above. The results given in Table VII. show a
fair agreement with the above supposition.

THE GEOTROPISM OF FRESHWATER SNAILS.

95

TABLE VII.

GEOTROPISM OF Physa AT THE DIFFERENT ANGLES OF INCLINATION OF A GROUND

GLASS PLATE IN AIR IN TOTAL DARKNESS.

At beginning of experiment each head placed upward. Table shows results
after one minute.

+ Geotropism.

— Geotropism.

Not

c/1

<*- —

Oriented and

Crawled Up Not

Crawled.

Moved.

S 5

bJQ

C

"o

o.S
£ c

Crawled Down.

Quite Vertical.

H '

0

fc

No.

t.

No.

t.

No.

Jfc

No.

jf.

go0

10

SO

38

76

O

0

5

IO

7

14

78^°

10

50

41

82

I

2

5

10

3

6

67 M°

IO

50

41

82

3

6

i

2

5

IO

U

56^°

10

50

37

74

5

IO

2

4

6

12

o
O

45°

5

50

35

70

2

4

5

IO

8

16

cs

33 H°

5

50

32

64

3

6

8

16

7

14

5

50

28

56

5

10

4

8

13

26

nM°

5

50

24

48

ii

22

ii

22

4

8

Total

60

400

276 or 69%

30 or 7.5%

41 or 10%

53 or 13.5%

Physa evidently could stick on the ground-glass plate better
than on the plain glass plate, as would be expected. Experi-
ments, therefore, could be carried on even at an angle of 90°.
Here again, it is noticeable that there is a decrease of positive geo-
tropism at this angle. The optimum inclination in this case is
between the angles of 67^° and

5. Summation of Gravity and Light Stimuli.

Physa being positive to gravity and negative to light, as
already seen, it wrould be expected to crawl downward even at a
small angle of inclination, if it were placed in a strong light. This
is just what happened. One morning the rays of sunlight were
falling at an angle of about n%°. The angle of the rays of
sunlight was nearly constant during the experiments. Ten
selected individuals were carefully placed with their heads down
on a moist plain glass plate. The plate was then reversed and
put on the rack whose angle of inclination was 11%°- They
all oriented themselves away from the rays of sunlight, that is,
downward, and crawled in that direction. Ten trials were made
and there was no exception.

Besides the above, observations on exclusion of the air were
attempted, but the results were not satisfactory. Generally
speaking however, Dawson's observations seem to be right,
although the writer observed one individual "drop" once.

96 SAKYO KANDA.

IV. SUMMARY AND CONCLUSION.

1. Physa gyrina Say is negatively heliotropic. It becomes
sluggish in its activity in darkness, particularly in water.

2. Physa gyrina Say, Planorbis trivolvis, Limncea stagnates,
and Limncea columella, are negatively geotropic when their lungs
are empty; and positively geotropic when their lungs are full of
air. Physa often comes near the surface and crawls down again
without filling its lung with air.

3. Physa, when put wTith green algae, does not often crawl to
the top.

4. At different angles of inclination of a plain glass plate in the
air and in total darkness, Physa is positively geotropic. There
is a certain limit of inclination beyond which the animal can not
actively move on account of the force of gravity.

(a) Many of the individuals in question are positive to gravity,
even though their lungs are empty.

(&) The optimum inclination of the plain glass plate on which
Physa may crawl is an angle of 56}/£0.

(c) At the angle of 67^/2°, positive geotropism decreases as
negative geotropism and horizontal crawling increase. The
negative geotropism is not necessarily the result of lack of oxygen.

(d) At the angle of 78%° no experiments are successful.

5. At different angles of inclination of a ground-glass plate in
the air and in total darkness, Physa reacts to gravity in a similar
manner as though on the plain glass plate, although the limit of
inclination is a little higher in the former than in the latter. The
optimum inclination of the ground-glass plate is between the
angles of 67^° and 78%°.

6. Contact stimuli seem to interfere slightly with the geo-
tropism of Physa.

7. The combination of gravity and light (both the glass support
and the rays of light being inclined 11%° to the horizontal)
accelerates positive geotropism of Physa.

From the data here given the writer is inclined to draw the
conclusion that Physa is naturally positively geotropic. It is
little wonder, therefore, that Physa becomes positive to gravity
when its lung is filled with air.

As to the organ of geotropic orientation of Physa and other
snails, no direct experimental evidence is yet furnished. But

THE GEOTROPISM OF FRESHWATER SNAILS. 97

according to Cooke (i, p. 196), Limncea, Planorbis and Physa have
statocysts, which are situated near the pedal ganglion, and are
probably connected with the cerebral. The statocyst also con-
tains statoliths. The number of the statoliths "varies in dif-
ferent genera and species." There are a hundred in Limncea
stagnalis on which the writer has experimented, but about fifty
in Planorbis contratus and Physa fontinalis. Therefore, Planorbis
trivolvis and Physa gyrina Say very probably have statoliths.
These statocysts with statoliths maybe the organs for geotropicori-
entation. At any rate, the most probable factors of geotropic ori-
entation, positive or negative, seem to be internal, that is, physi-
ological, and not external.

BIBLIOGRAPHY.

1. Cooke, A. H.

'95 The Cambridge Natural History, Vol. 3, xi-535 p. New York,
Macmillan & Co.

2. Dawson, Jean.

'n The Biology of Physa. Behav. Monog., Vol. i, No. 4, 111-120 p.

3. Kanda, Sakyo.

'14 On the Geotropism of Paramecium and Spirostomum. BIOL. BULL.,
Vol. 26, pp. 1-24.

4. Kanda, Sakyo.

'15 Geotropism in Animals. Am. Jour. Psychol., Vol. 26, pp. 417-427.

5. Loeb, Jacques.

'97 Zur Theorie der physiologischen Licht und Schwerkraftwirkungen.
Archiv. f. d. ges. Physiol. Bd. 66, S. 439-466.

6. Loeb, Jacques.

'n Die Tropismen. Handbuch der Vergleichenden Physiologic, Bd. IV.,
Erste Hafte, S. 451. Jena, Gustav Fischer.

7. Lyon, E. P.

'05 On the Theory of Geotropism in Paramecium. Am. Jour. Physiol.,
Vol. 14, pp. 421-432.

8. Prentiss, C. W.

'01 The Otocyst of Decapod Crustacea — Its Structure, Development,
and Functions. Bull, of Mus. Comp. Zool., Vol. 36, pp. 165-251.

9. Verworn, Max.

'91 Gleichgewicht und Ostolithenorgan. Archiv. g. d. ges. Physiol.,
Bd. 50, S. 423-472.

10. Walter, Herbert E.

'06 The Behavior of the Pond Snail Lymnaeus elodes Say. Cold Spr.
Harb. Monographs, VI, pp. 26-28. The Brooklyn Institute of Arts
and Sciences, Brooklyn, N. Y.

After writing this paper the author found three important articles by W. Baun-
acke, but was not able to take these into consideration in this present paper.

11. Baunacke, W.

'i3-'i4 Studien zur Frage nach der Statocystenfunktion. Biol. Centralb.,
Bd. 33, S. 427-452, Bd. 34. S. 371-385. and S. 497~523-

Vol. XXX, February, 1916. No. 2

BIOLOGICAL BULLETIN

THE MITOCHONDRIA AND OTHER STRUCTURES
OBSERVED BY THE TISSUE CULTURE METHOD IN
THE MALE GERM CELLS OF CHORTHIPPUS CUR-
TIPENNIS SCUDD.1

MARGARET REED LEWIS,

CARNEGIE INSTITUTION.

WM. REES B. ROBERTSON,

UNIVERSITY OF KANSAS.

CONTENTS.

Introduction 99

Literature 100

Living Material 100

Fixed Preparations 101

Method 101

Observations 103

General 103

Vital Stains 104

Nucleus 105

Apical Cell 106

Primary Spermatogonium 107

Secondary Spermatogonium (Multiplication Stages) 107

First Spermatocyte (Growth Stages) 108

First Spermactocyte Division 108

Second Spermatocyte (Interkinesis) 109

Second Spermatocyte Division 109

Spermatid and Spermatozoon no

Discussion 112

Conclusion 113

INTRODUCTION.

A study of the chromosomes of Chorthippus curtipennis by
Robertson (in press) led to the desire to study the mitochondria
and other structures of the cytoplasm in order to determine if
possible the bearing of the cytoplasmic structures upon the later
development.

1 From the Marine Biological Laboratory, Woods Hole, Mass.

99

IOO MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

It was observed that this material lent itself to the study of the
living cell by means of the tissue culture method as described for
the chick embryo cell by Lewis, M. R., and Lewis, W. H. ('15)
and that not only could the most minute structures of the cell
be observed from day to day, but also these structures could be
experimented upon as readily as those of the chick embryo. It
was decided to study the mitochondria and other cytoplasmic
structures of the germ cells of Chorthippus curtipennis by means
of the tissue culture method.

LITERATURE.

Living Material. — The earliest observations upon the living
germ cells of the Arthropods were those of von La Valette St.
George ('86) in which he made a careful study of the Nebenkern
of the spermatid and described the structure and behavior of
that body more completely than many of the later investigators.

Chambers, R. ('15), in his microdissection studies on the germ
cell, for which he used the male germ cells of Disosteira Carolina
(grasshopper) and of Periplaneta americana (cockroach), gives
many interesting observations as to the behavior of the mito-
chondria during the spermatocyte divisions and he also describes
in detail the development of the axial filament of the spermatid
and spermatozoon, but apparently Chambers made no effort
to trace the cytoplasmic structures of the germ cells throughout
their development. The description of the mitochondria during
the spermatocyte divisions and the formation of the Nebenkern
of the spermatid agrees in general with that found for preparations
of Chorthippus curtipennis when stained with Janus green. The
development of the spermatozoon of Chorthippus, however,
takes place in quite a different manner from that described by
Chambers for the cockroach.

Goldschmidt, R. ('15) states that it was possible to keep the
sperm cells of the moth Samia cecropia L. alive for three weeks in
cultures of haemolymph and that during this time many follicles
finished the process of spermatogenesis. Goldschmidt does not
describe the process of spermatogenesis, but merely states that
it corresponds with that described for fixed preparations. From
these studies upon Chorthippus it is quite evident that neither the

THE MITOCHONDRIA AND OTHER STRUCTURES. IOI

details of the cellular structure nor their behavior could be
carefully studied while the cells remained in the follicle as de-
scribed by Goldschmidt, but that for this purpose it is necessary
to observe isolated cells which lie close to the cover slip.

Fixed Preparation. — The mitochondria and other structures
present in various types of fixed material have been studied
more or less in detail by numerous observers, with results which
depend very largely upon the state of preservation of the material
studied. Duesberg, J. ('n) has reviewed this literature so it
is unnecessary to repeat it here. In an earlier paper Duesberg, J.
('10) gives a clear and complete description of the behavior of
the mitochondria in Blatta germanica, which corresponds in all
but a few details with that given below for Chorthippus curtipennis.

METHOD.

The cultures were prepared in the usual manner (Lewis, M. R.,
and Lewis, W. H., '15) and all precautions were observed in order
to keep them not only chemically clean but also aseptic, except
in cases where the period of observation was to extend over only
a short time, as for instance when a vital stain was used.

Various different culture media were tried and the one which
appeared to be most nearly isotonic with the body fluid of the
grasshopper and which also was most favorable for growth was
practically Locke's solution i. e., NaCl 0.9 per cent., CaCU 0.025,
KC1 0.042 per cent., NaHCOs 0.02 per cent., dextrose 0.25 per
cent., peptone 0.2 per cent., but since the observations were made
at Woods Hole where running sea water is supplied, the same
concentration of salts was obtained by a dilution of the sea water
as follows: sea water 30 c.c.+distilled water 50 c.c. + bouillon
20 c.c.+dextrose 0.25 gram + NaHCOs 0.02 gram. The bouil-
lon was prepared in the same manner as that used for bac-
teriology, except in this case grasshopper muscle was used in
place of beef. A solution of peptone alone can be substituted
for the bouillon with rather good results. The culture medium,
which is successful, depends largely upon the amount of evapora-
tion which takes place in the technic of the individual observer.
This can be determined from the appearance of the preparation
itself, for it was found that when the medium was too concen-

IO2 MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

trated the cells formed numerous delicate pseudopodia or flagella
and when the medium was too dilute the cells became swollen.
The solution which is most nearly isotonic is one in which the
cells remain round and flatten out close to the cover slip, or
crawl along the cover slip by means of broad, flat pseudopodia.

The grasshopper was opened on the ventral side by means of
sterile scissors and the walls pinned down with sterile pins.
The testis follicles were then removed aseptically. Each follicle
of the testis is made up of a number of cysts, each of which con-
tains a number of cells all in the same stage of development.
The apical cell and the primary spermatogonia are at the blind
end of the follicle. The cysts which contain the secondary
spermatogonia, the first spermatocytes, the synapsis stages, the
growth stages, the first spermatocyte division stages, the second
spermatocytes, the second spermatocyte divisions, the spermatids,
and the spermatozoa are arranged in order back of this towards
the open end of the follicle. In order to obtain the cells in the
stage desired for observation, a follicle of the testis was placed
in a thin drop of the sterile culture medium on a sterile cover slip
and, with the aid of a binocular microscope, the wall of the cyst,
which contained the cells to be studied, was punctured with a
sharp, sterile needle so that the cells of the cyst flowed out into
the medium. The excess of the medium was drawn off by means
of a capillary pipette and the preparation was then sealed onto
a hollow ground slide by means of a vaseline ring. In case
stained preparations were to be observed, the vital stain, Janus
green or neutral red, was dissolved in the drop of the culture
medium in which the follicle was punctured. The cells released
from the cyst wall spread out in a thin layer along the cover slip
and were then studied by means of the No. 6 ocular and 2 mm.
oil immersion lens. A 4O-watt Mazda electric light was used for
illumination.

Since any stage in the development of the germ cell can be
obtained in the above manner, it was not found necessary to
watch any one cell over a long period of time, although the cul-
tures remained healthy and dividing cells were found as late as the
fourth day.

Tissue cultures of the germ cells in body fluid medium were

THE MITOCHONDRIA AND OTHER STRUCTURES. I 03

made as controls, but they were not so useful for the study of the
cell structure owing to the fact that the medium is more opaque
and that the cells do not spread out in a thin layer close to the
cover slip. Also the plasma medium is more difficult to use for
experimental purposes owing to the fact that it is easily coagu-
lated.

OBSERVATIONS.

General. — When the cyst wall is broken the cells flow out and
become attached to the cover slip. The cells appear to be formed
of a clear homogeneous cytoplasm which contains a nucleus and
granules. Numerous cells, which contain two or four nuclei
and also a correspondingly increased amount of mitochondria,
were observed in all stages of development as, for instance, a
first spermatocyte with two nuclei and a double amount of
mitochondria or a young spermatid with two or four nuclei and
also two or four nebenkern. During observations upon one
unstained preparation two second spermatocytes whose cyto-
plasm touched at one point were observed to fuse into a single
cell (Figs. 27, 28, 29). The fused cell, which resulted from the
two single cells, contained two groups of chromosomes and two
groups of mitochondria. In certain stages in the development
of the germ cell the granules are scattered throughout the greater
part of the cytoplasm (spermatogonium), while in other stages
the granules are limited to a definite area (division of spermato-
cytes). There is no indication of any network structure either
of the cytoplasm or of the nucleus. The cells of a cyst remain
attached to each other by a long thread-like process, or in some
cases by a short thick process, which appears as though the cells
had not been completely separated at division, but had remained
attached by a band of cytoplasm. Groups of spermatozoa are
attached by one end to a crescent-shaped body, while the other
end is free and lashes about continuously. Several of these
crescent-shaped bodies, each with numerous spermatozoa at-
tached to it, may be seen in one field. The cells may send out
broad, flat pseudopodia and crawl along the cover slip, or in
media which are too concentrated or to which Janus green has
been added, the cells may send out numerous delicate pseudo-
podia, which appear more like flagella. However, other factors

IO4 MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

besides the concentration of the medium may influence the size
of the pseusopodia, for Goldschmidt (1915) states that the germ
cells form flagella in the cultures in hsemolymph and that these
flagella can be caused to appear and disappear by a change of
temperature.

During mitosis the mitochondria become long, delicate threads
and lie around the spindle in such a way as to be easily mistaken
for the spindle (Figs. 14 and 22). In none of our preparations
were the spindle threads seen and the spindle itself did not show
as a cone of material, which had a different light refraction, as
it did in the chick. In one preparation the position of the spindle
at one pole was outlined (Fig. 16), but even in this case no
spindle threads were seen.

The spindle is present however and can be readily shown by
means of acetic acid vapor, which destroys the mitochondria and
coagulates the cytoplasm sufficiently to show the spindle prac-
tically the same as it is shown in figures drawn from fixed material
(Figs. 21-26).

Vital Stains. — All stages in the development of the germ cell
were studied, not only by means of the living unstained cell, but
also by means of preparations stained with Janus green and
others stained with neutral red. A few preparations were
stained with both Janus green and neutral red. The Janus
green stain and also the neutral red stain were dissolved in the
culture medium in exceedingly dilute solutions, never more than
1-50,000 parts and frequently as dilute as 1-100,000 parts.

The neutral red stains a large round granule, which is quite
different from the mitochondria, not only in size and appearance,
but also in behavior (Figs. I, 22, 36, 37, 39, 41, 46 and 49). This
granule has the same reaction to Brilliant cresylblue 2 b. as that
of the granule described by Lewis and Lewis ('15) in connection
with the "vacuole" and agrees in a few details with the "beta
globule" described by Coghill ('15). In a few cases the granule
reacts with the neutral red stain in the same manner as does the
neutral red granule, which Renaut, J. ('04) and Dubreuil, G. ('13)
describe in connection with the connective tissue development.
Owing to the fact that the literature does not furnish a satisfactory
term for this granule, it will be called simply the neutral red
granule in the following observations. Duesberg ('10) in certain

THE MITOCHONDRIA AND OTHER STRUCTURES. 1 05

figures, as for instance those of the spermatid, shows a granule in
the same position as that of the neutral red granule in our prepara-
tions, and, although the granule is stained like the mitochondria
with Benda's stain, Duesberg states that it is in all probability
not mitochondria. In preparations stained with Janus green
this granule remains unstained. The somatic cells, which form
the wall of the follicle and also the apical cell (Figs. I and 2),
are full of these bright red granules when the preparation is
stained with neutral red, but the germ cells contain only a very
few neutral red granules.

With Janus green stain however, the germ cells are shown to
contain abundant mitochondria in the form of granular threads
or of small rod shaped granules. After the preparation has been
stained for a short time the granules coalesce into larger globules
(Figs. 4 and 5) and finally they disappear in the cytoplasm.
Long, thread-like mitochondria rapidly break into granules when
stained with Janus green (Figs. 13-20).

Neither of the above stains showed the spindle threads during
mitosis. There was no appearance seen at any time during ob-
servation upon either the unstained cell or the cell stained by
means of the above vital stains, which could lead one to conclude
that the mitochondria are formed from any material at the ex-
pense of the nucleus as Wassilieff ('07) and his followers contend.

Nucleus. — -The various changes which the nucleus undergoes
during the so-called resting stage and during division can be
clearly observed throughout the development of the germ cells
from spermatogonia to spermatozoa. In the resting nucleus of
any stage chromatin threads were observed. In the nucleus
of the spermatogonium these chromatin threads or spiremes seem
to fill the nuclear space like so many sacs (Fig. i). When the
preparation was stained with neutral red, the walls of these
sacs became faintly pink and so revealed the boundaries of the
chromosomes. During the telophase, and in a few cases, in the
late anaphase of the spermatogonium, the spermatocyte and
the spermatid, the chromosomes have a granular structure
(Figs. 4, 10, 20). These granules appear to be uniform in size
and it might prove possible to count the number of granules which
compose a given chromosome. The importance of these granules
in "crossing over" phenomena may appear later.

IO6 MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

The number and also the characteristics of the chromosomes
observed in the living cells correspond with what was found in
the fixed preparations. In the first spermatocyte the chromo-
somes were as follows: Five "short rod" tetrads, three large
"compound ring" tetrads, derived from the three pair of com-
pound V chromosomes of the spermatogonium and the rod-like
sex chromosomes. In this genus, Chorthippus formerly known
as stenobothrus, there is a peculiar compounding of six pairs of
rod chromosomes to form the three pairs of V's that are charac-
teristic so far as is known of all the species of the genus (Meek,
'n, '12; Gerard, '09; Davis, '08).

This peculiar process by which the three pairs of compound V
chromosomes are formed was first observed by Robertson (in
press). The subsequent behavior of these compound chromo-
somes was the same in the living cell as has been described from
the fixed preparations and the five rods, three V's and the sex
chromosomes (present only in one of the two daughter cells,
which result from the first spermatocyte division) were easily
identified in the second spermatocytes and in the spermatids.

The Apical Cell. — The apical cell lies in the blind tip of the
follicle surrounded by primary spermatogonial cells (Figs. 1,3).
It is a round cell, which contains a more or less oval nucleus, and
is attached to the walls of the follicle by several thick cell proc-
esses. The apical cell contains both mitochondria and neutral
red granules (Figs. I and 3). The former are fewer in number
than the latter and uniformly small in size (Fig. 3). They are
arranged as granular threads mostly in a layer around the nucleus.
The neutral red granules are considerably larger than the mito-
chondria and are scattered throughout the cytoplasm and in the
cell processes. When a preparation is stained with neutral red,
these granules in the apical cell and also in all of its processes
rapidly take up the stain, so that the apical cell becomes quite
red in appearance. While the few granules in the spermatogonia
which surround the apical cell take up the stain only after a
long time and then only in a few scattered granules so that the
spermatogonia appear practically colorless. The somatic cells
(Fig. i), which form the wall of the follicle, have abundant neutral
red granules and these stain with neutral red in much the same

THE MITOCHONDRIA AND OTHER STRUCTURES. IO7

manner as did those of the apical cell. This striking resemblance
of the apical cell to the somatic cells in contrast to the germ cells
suggests the possibility that the apical cell may be more closely
related to the somatic cells than to the germ cells.

Primary Spermatogonia. — In the primray spermatogonia both
the mitochondria and the neutral red granules can be identified
in the unstained cell. In the resting cell the mitochondria appear
as delicate granular threads and at this time these threads seem
to radiate from the distal pole of the cell (i. e., the region of the
last connection with its sister cell at mitosis). The mitochondria
of the primary spermatogonia stain less intensely with Janus
green than does that of cells in a later stage of development and
when stained with Janus green the delicate threads become rapidly
distorted and appear as granules. The neutral red granules,
from 4 to i o or 12 in number, are larger, more or less round and
much more refractive than the mitochondria granules. These
neutral red granules, so far as was seen, did not seem to be located
in any definite region of the cell, but were scattered through the
cytoplasm.

Secondary Spermatogonia (Period of Multiplication). — The
secondary spermatogonial cells are smaller than the primary
spermatogonia and the mitochondria are usually in the form
of fine, granular threads scattered from the distal end of the
cell towards the nucleus. As the cell approaches the resting
condition the mitochondria become more uniformly scattered
throughout the cytoplasm. In a few observations the mito-
chondria appeared to be absent from a region at the extreme
distal pole of the cell, possibly the mitosome (i. e., the remains of
the spindle, Figs. 2, 4, 6).

During mitosis the mitochondria arrange themselves as long
threads close to the spindle and frequently they have the ap-
pearance of abnormally thick spindle threads. During the con-
striction of the cytoplasm at late anaphase the mitochondria
threads again become granular and at telophase the mitochondria
are separated into two practically equal amounts, one of which
passes into each daughter cell and from this mass the mitochon-
dria migrate towards and partly around the nucleus (Fig. 4),
as can be seen in the first spermatocyte.

IO8 MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

First Spermatocyte (Growth Period}. — As the cell grows in size
the amount of mitochondria appears to increase correspondingly
and certainly the mitochondria of the spermatocyte stain much
more intensely with Janus green than do those of the spermato-
gonium. The neutral red granules are few in number and scat-
tered throughout the cytoplasm.

In the later growth period the mitochondria become grouped
into two, or in a few cases, possibly more masses (Figs. 7 and 8
and n). Unfortunately the significance of this was not compre-
hended, but it is without doubt closely allied with some change
in the cell itself, as the massed arrangement of the mitochondria
is quite characteristic of the synapsis stages. There was no
evidence that the mitochondria granules paired during synapsis,
although the pairing of the chromosomes wras clearly seen.

First Spermatocyte Division. — During the prophase and early
metaphase of the first spermatocyte, the mitochondria migrate
away from the two masses of mitochondria granules and elongate
towards the two poles of the spindle (Fig. 12), so that when the
chromosomes are arranged on the spindle plate, the mitochondria
appear as threads which more or less closely surround the
spindle. As the chromosomes move apart, the mitochondria
become drawn out into straight, even threads closely attached to
the spindle between the two groups of chromosomes (Figs. 13,
14 and 15). Since they are much longer and more refractive
at this stage, they may be easily mistaken for the spindle fibers
and some cytologists have stated that the mitochondria in the
germ cell are but the remains of the spindle fibers. A simple
experiment shows that this is not the case in Chorthippus curti-
pennis. Figs. 21, 22 and 23 were made from the living cells in
the cultures and then an opening was made in the vaseline ring,
which supported the cover slip and a drop of glacial acetic acid
was placed on the slide near to the opening so that the fumes from
the acid passed through the opening and acted upon the prepara-
tion. The mitochondria were destroyed at once and became lost
within the coagulum of the cytoplasm, and, where previously
no spindle could be seen, there now appears a typical spindle
(Figs. 24 and 25). Fig. 26 shows the remains of the spindle fibers
near the periphery of each daughter cell, but the mitochondria,

THE MITOCHONDRIA AND OTHER STRUCTURES. 1 09

which previously lay along these threads (Fig. 23) and scattered
towards the nucleus, have now disappeared.

During the anaphase of the living cell the daughter chromo-
somes move to the extreme proximal pole of the cell and when the
cytoplasm constricts, the continuous mitochondria threads again
become granular and from the ends of these threads the mito-
chondria granules migrate away towards the two groups of
chromosomes (Fig. 22). In some instances the entire mitochon-
dria thread appeared to pass over to one daughter cell, but usually
the division of the mitochondria took place by means of the
migration of the granules into each daughter cell, so that each
daughter cell received about the same amount of mitochondria.

When the preparation was stained with Janus green the mitotic
division continues, provided it was started before the stain acted
upon the cell (Figs. 17-20), but after the division was finished,
the cell did not pass through the second spermatocyte division
as the unstained cells did. The mitochondria threads become
granular and do not usually extend from chromosome group to
chromosome group, but more frequently lie in a broad band
around the equator of the spindle (Figs. 16, 17, 18 and 19). The
ends of the granular threads swell up and become varicose before
the granules migrate away (Figs. 19 and 20).

Second Spermatocyte (Inter kinesis}. — After the mitochrondria
have migrated into the daughter cells the cytoplasmic bridge
between the two cells remains and elongates. The cells may
sometimes remain thus attached by a long process during the
semi-resting condition (interkinesis), which may continue for an
hour or even for several hours before the cells prepare for the
second spermatocyte division. The second spermatocyte has a
characteristic appearance and is easily recognized. The chro-
mosomes remain and it can be seen that they are no longer
tetrads as they were in the first spermatocyte (Fig. 30). The
mitochondria are gathered into an irregular body at one side of
the cell and from this body the mitochondria granules spread
out somewhat toward the nucleus (Fig. 30).

Second Spermatocyte Division. — The behavior of the mitochon-
dria during the second spermatocyte division is practically the
same as that during the first spermatocyte division. In this

IIO MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

case however there is usually only one mitochondrial mass
(although it may divide into two in some cases) and this mass
lies at one side midway between the two poles of the spindle
(Fig. 30). The mitochondria migrate away from this granular
mass and elongate towards the two poles of the spindle. Usually
the elongated threads spread out around the spindle, but in a
few cases they have been seen to remain mostly on the side of the
spindle where the mitochondrial body was. The threads become
homogeneous and appear to be continuous threads stretched
between the two groups of chromosomes. When the cytoplasm
constricts during anaphase the granules migrate into the daughter
cells, but they do not usually scatter around the nucleus. A few
neutral red granules are present in the cytoplasm throughout
the division and can be seen in the cytoplasm on each side of the
mitochondrial body in the spermatid (Figs. 36 and 37). When
division is completed the mitochondria granules become con-
tracted into a compact, spherical granular body, the nebenkern,
and a few neutral red granules are irregularly scattered on each
side of the nebenkern. It is interesting to note that after
acetic acid vapor, a round body can still be distinguishable in
the place of the nebenkern, so that either the mitochondria are
at this stage more resistant to acetic acid, or else there is some
other body present in the nebenkern.

In a few cases the daughter cells seemed to be unequal in size.
It has not been determined whether the inequality in size of the
daughter cells is due to the presence of the sex chromosome in
the larger cell. Davis ('08) and Gerard ('09) in their figures for
this genus each show one of the daughter cells, which result
from the second spermatocyte division, larger than the other,
but they do not call attention to this point.

The Spermatid and Spermatozoon. — The development of the
spermatid into the spermatozoon is by no means a simple process
and, even after prolonged observation, the significance of the
various steps is not clear (Figs. 38-51).

The neutral red granules do not play an active part, but remain
about the same in the cytoplasm until with the growth of the
tail the cytoplasm becomes small in quantity and they disappear.
They were not observed in the more adult spermatozoon except

THE MITOCHONDRIA AND OTHER STRUCTURES. Ill

in cases where some abnormal factor caused the cytoplasm along
the tail of the spermatozoon to become clumped into nodes, and,
in this case, one or more neutral red granules were present in
each node (Fig. 49). In the young spermatid the mitochondria
are in the form of the granular nebenkern (Fig. 38), which later
becomes a clear, homogeneous, spherical body close to the nucleus.
No trace of granules can be seen either in the unstained cell or
in those stained with Janus green. Within a short period of time
(}/2 hour) certain delicate threads appear within the clear
nebenkern (Figs. 39 and 40). These threads maybe concentric
or otherwise coiled and may represent either the rows of granules
under pressure, or they may be the edges of the membrane, which
separates the nebenkern into two half spheres, seen at different
levels, for very shortly after the threads appear, the nebenkern
slides apart into two half spheres and at once becomes granular
again (Figs. 41, 42, 43).

The behavior of the centrosome and the formation of the axial
filament, which has been described from fixed material, was seen
only in one case, and in that case it was not possible to ascertain
just what connection there is between the division of the neben-
kern and the formation of the axial filament. The centrosome
was seen as a clear paired body near the nucleus, but at a distance
from the mitochondrial body (Fig. 38). Later the centrosome
occupies a position at the posterior pole of the nucleus near the
Nebenkern. After the division of the Nebenkern it was seen as a
small opaque body close to the nucleus (Fig. 44). The body is
double and, although small, it increases in size with the growth
of the tail and later becomes the middle piece (Figs. 46, 49, 50
and 51). From this body the axial filament probably is given
off and later grows out into an extension of the cytoplasm.

The mitochondrial bodies, or the two half spheres of the ne-
benkern, now elongate as granular sacs (Figs. 45 and 46). The
wider end is towards the nucleus and the sacs gradually taper off
until they become only a few granules at the other end (Fig. 46).
As the tail grows out these bodies grow out one on each side of
the axial filament. The sacs do not increase in size, but simply
spread out along the axial filament and become crowded into the
narrow layer of cytoplasm along the tail, so that as the tail

112 MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

elongates the granules form two irregular granular strands, which
later fuse into two continuous threads of even width extending
from the centrosome body or middle piece to almost the end of
the tail (Figs. 49, 50 and 51).

Duesberg ('10) describes a small body composed of mito-
chondria at the tip of the head, but in this form such a body was
not seen either in the living spermatozoon or in those stained with
Janus green.

Unfortunately it was impossible to study this material in all
its details during the short time at Woods Hole and there are
many interesting points to be followed out, which it is hoped may
be continued in later studies upon another species of grasshopper
by Robertson.

DISCUSSION.

The above observations show that there are present in the male
germ cell from its earliest appearance throughout its develop-
ment certain granules which correspond to the mitochondria of
somatic cells. The mitochondria behave in a definite and char-
acteristic manner during the development of the male germ cell
and assume a shape and position characteristic for each stage in
the development. The mitochondria of the germ cell become
slightly blackened with osmic acid, are destroyed by acetic acid
and stained by Janus green and the usual stains for mitochondria
in fixed material. That they are not the remains of spindle
fibers has been clearly shown by means of acetic acid, which
destroys the mitochondria and causes the spindle fibers to appear.
On the other hand it does not seem possible that bodies, which
have to do only with the metabolic activities of the cell should
necessarily undergo such an exact behavior as is shown for in-
stance by the division of the nebenkern into equal parts and the
development of these two sacs of mitochondria into the two long
threads of mitochondria in the spermatozoon. Neither does the
behavior of the mitochondria seem to be entirely dependent on
changes within the cell, as for instance during the division of the
cell or the growth of the tail, where it might appear that the elon-
gation, and at the same time narrowing of the tail, might force
the mitochondria granules to assume the position of two contin-
uous threads, for in some cases where the tail does not develop

THE MITOCHONDRIA AND OTHER STRUCTURES. 1 13

as normally but the axial filament grows out within the cell, the
mitochondria develop as normally in connection with the axial
filament and form the same long threads, although in this case
these threads are wound round and round within the cell (Figs.
47 and 48).

The origin of the mitochondria is still unsolved. They cer-
tainly do not arise in the male germ cells, since they are already
present in the earliest germ cell. There is no evidence in the
above observations that the mitochondria are formed at the
expense of any nuclear material. Also there is no evidence that
the mitochondria of the male germ cell of Chorthippus curtipennis
can have any influence upon inheritance, as was suggested by
Meves ('13) in his work on Ascaris, unless it can be shown that
the tail as well as the nucleus of the spermatozoon enters the egg.

CONCLUSION.

The mitochondria as well as the neutral red granules are present
in the primary spermatogonium of Chorthippus curtipennis and,
while the neutral red granules appear to have no definite behavior,
the mitochondria do behave in a characteristic manner through-
out the development of the male germ cells. By means of the
tissue culture method the mitochondria can be seen to be present
as small, delicate granules in the primary spermatogonium. They
increase in amount during the growth stage and arrange them-
selves along the spindle in a definite manner during the spermato-
cyte division. They form the nebenkern of the spermatid and
from this develop into two equal homogeneous threads in the tail
of the spermatozoon.

WOODS HOLE, MASS.,
September, 1915.

LITERATURE LIST.
Chambers, R.

'15 Microdissection Studies on the Germ Cell. Science, N. S., Vol. XLL,

No. 1051.
Coghill, G. E.

'15 Intracellular Digestion and Assimilation in Amphibian Embryos. Science,

N. S., Vol. XLII., No. 1080.
Davis, H. S.

'08 Spermatogenesis in Acridida? and Locustida?. Bull. Mus. Comp. Zool.
Harvard College, Vol. LIII, No. 2, p. 59.

114 MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

Dubreuil, G.

'13 Le Chondriome, etc. Arch. d. Anat. Micr., Tom. 15, p. 53.
Duesberg, F.
'10 Nouvelles recherches sur 1'appareil mitochondrial des Cellules seminales.

Arch. f. Zellf., Bd. 6.
'n. Plastosomen " apparato reticolare interne " und Chromidial apparat.

Ergeb. d. Anat. u. Entwickgeschichte, Bd. XX.
Gerard, P.

'09 Recherches sur la spermatogenese chez Stenobothrus biguttulus. Arch.

Biol., torn. 24, p. 543.
Goldschmidt, R.

'15 Some Experiments on Spermatogenesis in Vitro. Proc. Nat. Acad. of

Science, Vol. I., p. 220.
La Valette, St. George, von.

'86 Spermatologische Beitrage Zweite Mitt. Arch. f. mikr. Anat., Bd. XXVII.
Lewis, M. R., and Lewis, W. H.

'15 Mitochondria and Other Cytoplasmic Structures in Tissue Cultures.

Amer. Jour. Anat., Vol. XVII., p. 339.
Meek, C. F. U.

'n The Spermatogenesis of Stenobothrus viridulus with Special Reference to
the Heterotropic Chromosome as a Sex-determinant in Grasshoppers.
Jour. Linnean Soc. (London), Zool., Vol. XXXII. , No. 208, p. 20.
'iaa A Metrical Analysis of Chromosome Complexes, etc. Philos. Trans. Roy.

Soc. London, B, Vol. 203, p. 84.
'i2b The Correlation of Somatic Characters and Chromatin Rod, etc. Jour.

Linnean Soc. (London), Zool., Vol. XXXII., p. 107.
Meves, F.

'13 IJber die Beteiligung der Plastochondrian an der Befruchtung des Eies von

Ascaris megalocephala. Arch. f. mikr. Anat., Bd. 76.
Renaut, J.

'04 Les cellules Fixes des Tendons de la queque du jeune Rat sont Toutes des
Cellules Connectives Rhagiocrines. C. R. de la Soc. de Biol., 1904, p. 1067.
Robertson, W. R. B.

Chromosome Studies. I. Taxonomic Relationships shown in the Chromo-
somes of Tettigidae and Acrididae: V-shaped Chromosomes and their
Significance in Acrididae, Locustidae, and Gryllidae: Chromosomes and
Variation. (In press, Jour. Morph.)
Wassilieff, A.

'07 Die Spermatogenese von Blatta germanica. Arch. f. mikr. Anat., Bd. LXX.

Il6 MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

EXPLANATION OF PLATES.

J. G. = Janus green stain.

N. R. = neutral red stain.

Mitochondria = black-inked granules and threads.

Neutral red granules = gray granules outlined with black ink.

Centrosome = clear round granule about same size as neutral red granule.

PLATE I.

FIG. i. N. R. Blind end of follicle. Apical cell surrounded by primary
spermatogonia. The somatic cells form the boundary 'of the follicle. Round
neutral red granules in the apical cell and also in the somatic cells. Leitz Binoc.
4b X i /i 2 lens.

FIG. 2. Unstained cell. Primary spermatogonium. Mitochondria are granu-
lar threads. Clear region at upper side of the cell is probably the mitosome (spindle
remains). Leitz Binoc. 4b X 1/12 lens.

FIG. 3. J. G. and N. R. Apical cell. Mitochondria are small black granules
and the neutral red are the larger gray granules. Zeiss 6 oc. 2 mm. lens.

FIG. 4. J. G. Two secondary spermatogonia still attached together. Chro-
mosomes granular. Telophase. Mitochondria are not in nucleus but around it.
Leitz Binoc. 4b X 1/12 lens.

FIG. 5. Same cell as Fig. 4. Globules of blue which form after the mitochon-
dria have lost their stain.

FIG. 6. J. G. Secondary spermatogonium.

FIG. 7. J. G. First spermatocyte, centrosome divided. Sex chromosome.
Early prophase.

FIG. 8. J. G. First spermatocyte. Diplotene nucleus stage. Centrosomes.
Two masses of mitochondria, one on upper surface of cell and one on the lower.
Leitz Binoc. 4b X 1/12 lens.

FIG. 9. Unstained. First spermatocyte 'bouquet-stage' synapsis. Zeiss 6 oc.
2 mm. lens.

FIG. 10. J. G. Spermatogonium, telophase. Each chromosome is in a
separate vesicle and is granular. Nuclear wall reforming from chromosome
vesicles. Leitz Binoc. 4b X 1/12 lens.

BIOLOGICAL BULLETIN, VOL. XXX.

PLATE

.M

-' ' -I 1 •

II

\

~)\

/

/ f

\{

\ /r -

•4 4&*1 • \
'"•I*00. °X ^

'X

if(

s^/f/ O

^r^;r 2

•-'«& \V "£•

9 • » «.

..-^

0 <• •-•? '

« ««,'"•""« V

o oo ' . * ~

?. * e '

a o O o « " ?-"•"

i."11 "* » *

% °B 0c00 ,^V
o / o x. 'V

~--;;>*

•A,.. .1

&•> ° « . * * •'

,1' .

^Oe o° °».»"^'

^*>_ O s\ 4k «« .

;;/ ' «

'»

• . %^

•'- C:. - v- -" - *

T-V- '*''- '""-^: -.-

.

/r

^y^,;^»

»• - - . a1* o

r-

'

j*a;

8

4

* >•

iV
f

1

m!^

ft ' iV

-x

•

*

10

M. R. LEWIS AND W. R. B. ROBERTSON.

Il8 MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

PLATE II.

FIG. u. J. G. First spermatocyte, late prophase. Nuclear wall still intact,
"two masses of mitochondria.

FIG. 12. Unstained. First spermatocyte metaphase. Mitochondria threads
have migrated towards the poles of the spindle. Zeiss 8 oc. 2 mm. lens.

FIGS. 13, 14, 15. Unstained. First spermatocyte division. Same cell at
9:30 A.M. (Fig. 13), 9:45 A.M. (Fig. 14), and 10:10 A.M. (Fig. 15). Mitochondria
are long threads close to the spindle fibers. Zeiss 4 oc. 2 mm. lens.

FIG. 16. Unstained. Mitochondria are granular threads around the spindle.
Those in the middle of the spindle occur underneath. The sex chromosome at one
pole. Spindle outlined at one pole. Leitz Binoc. 40 X 1/12 lens.

FIGS. 17, 1 8, 19. J. G. First spermatocyte division in the same cell. Fig. 17
at 9:40 A.M., Fig. 18 at 9:50 A.M., and Fig. 19 at 10:45 A.M. Zeiss 6 oc. 2 mm.
lens.

FIG. 20. J. G. End of first spermatocyte division. Mitochondria migrated
.into two daughter cells. Chromosomes granular. Leitz Binoc. 40 X 1/12 lens.

BIOLOGICAL BULLETIN, VOL. xxx.

PLATE II.

w

•I*

11

12

15

v\' Hil/i/'
W

M ,11/10

III

X

1 A

s» iffi

^!i if

/tv> i

'{ p\

\

\

18

M. R. LEWIS AND W. R. B. ROBERTSON.

I2O MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

PLATE III.

FIGS. 21, 22, 23. Unstained. Division stages drawn to show mitochondria
before the action of acetic acid. Zeiss 8 oc. 2 mm. lens.

FIGS. 24, 25, 26. Same cells after acetic acid fumes. Mitochondria have disap-
peared and the cytoplasm is coagulated. Spindle fibers can now be seen.

FIGS. 27, 28, 29. Unstained. The fusion of two cells to form a single double
cell with two groups of chromosomes and two masses of mitochondria. Leitz
Binoc. 4b X 1/12 lens. Chromosomes not all shown.

BIOLOGICAL BULLfcTIN, VOL. XXX.

PLATE III.

...- . i
h

*H * 1 » ' • - -

•

x * v* * ' », '

V

.) :

&•

^ \c

' 1AJV

m

•
i

f- ' ^ ».' •* 1

• S»yiSj;-V

27

28

M. R. LEWIS AND W. R. B. ROBERTSON.

122 MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

PLATE IV.

FIG. 30. Unstained. Second spermatocyte. Single mass of mitochondria
from which granules are elongating toward each pole of the spindle. Zeiss 6 oc.
2 mm. lens.

FIGS. 31-35. Unstained. Different stages of a cell in the second spermatocyte
division. Fig. 31 at 10:40 A.M.; Fig. 32 at 11:20 A.M.; Fig. 33 at 12:30 P.M.;
Fig. 34 at i P.M.; and Fig. 35 at 2 P.M. The mitochondria form the nebenkern.

FIGS. 36, 37. N. R. Neutral red granules present during the second spermato-
cyte division. Zeiss 6 oc. 2 mm. lens.

BIOLOGIOL BULLETIN VOL. XXX.

PLATE IV.

M •S IEWIS AND W R- B. ROBERTSON.

124 MARGARET REED LEWIS AND WM. REES B. ROBERTSON.

PLATE V.

FIG. 38. J. G. Young spermatid, centrosome divided. Nebenkern granules.
Leitz Binoc. 4b X 1/12 lens.

FIGS. 39, 40. Unstained. Young spermatid. Neutral red granules. Thread-
like appearance in the nebenkern. Zeiss 6 oc. 2 mm. lens.

FIGS. 41, 42. J. G. Nebenkern dividing is now granular. Neutral red
granules. Leitz 4b X 1/12 lens.

FIG. 43. Unstained. Nebenkern divided. Zeiss 6 oc. 2 mm. lens.

FIG. 44. Unstained. Nebenkern sacs with the centrosome close to the nucleus.
Neutral red granules. Zeiss 6 oc. 2 mm. lens.

FIG. 45. Unstained nebenkern sacs are slightly elongated. Zeiss 6 oc. 2 mm.
lens.

FIG. 46. J. G. Nebenkern sacs elongated. Axial filament. Double middle
piece body. Neutral red granules. Zeiss 6 oc. 2 mm. lens.

FIGS. 47, 48. J. G. The tail failed to develop but the mitochondria elongated
and formed the double thread as usual, although the double thread is wound round
and round within the cell. Leitz Binoc. 4bXi/i2 lens.

FIG. 49. N. R. The cytoplasm of the tail is gathered into nodes and the
neutral red granules appear. The mitochondria is now in the form of two homo-
genous long threads. Zeiss 6 oc. 2 mm. lens.

FIGS. 50, 51. J. G. Later stages of the spermatozoon.

BIOLOGICAL BULLETIN, VOL. XXX.

42

o

o°o o
0°o

39

43

vm

*V;;y-'

4T

'•'.>!'
'Wjii

48

°0<

40

\

41

51

M. R. LEWIS AND W. R. B. ROBER1SON.

NOTES ON THE PHYSIOLOGY OF FUCUS
SPERM ATOZOIDS.

W. J. ROBBINS
CORNELL UNIVERSITY

The notes here presented are of work conducted at the Marine
Biological Laboratory, Woods Hole, Mass., during the summers
of 1912 and 1913 at the suggestion of Dr. B. M. Duggar. to whose
kindness I also owe the opportunity and pleasure of working
there.

Strasburger (i) states that spermatozoids of Fucus, passing
at a distance of one or even two diameters from the egg, turn
from their path and rush toward the egg. This attraction, he
says, is a chemical one and is due to some substance secreted by
the egg which conditions the direction of motion of the spermato-
zoids. Strasburger also states that healthy spermatozoids are
strongly negatively phototactic. These two phenomena of
phototaxy and chemotaxy are used by Strasburger as a basis
upon which to account for the meeting of the spermatozoids and
the ova as follows:

The ova have a density greater than water and sink. The
spermatozoids, as they are negatively phototactic, swim down-
ward bringing them into the region where the chemotactic
influence of the eggs is sufficient to complete the union of the
spermatozoids with the. ova. Strasburger made no attempt to
discover the chemotactic agent.

Bordet (2) also attempted an explanation of the same problem.
He filled a capillary tube with crushed ova, immersed the tube in
a drop of sea water containing Fucus spermatozoids, and watched
for evidences of attraction, but found none. From this he
concludes that the ova exert no chemotactic influence on the
spermatozoids. He observed, however, that the spermatozoids
were very sensitive to contact, clinging with one cilium to the
glass slide or any solid object as, for example, a capillary tube.

Bordet also states that Fucus spermatozoids are not phototactic,

125

126 W. J. ROBBINS.

being neither attracted nor repelled by light. He believes the
meeting of the eggs and spermatozoids is due to chance.

In brief, Strasburger states that Fucus spermatozoids are
negatively phototactic and are chemotactic to the ova. Bordet
asserts that they are not phototactic and that there is no che-
motaxy.

MATERIAL AND METHODS.

The dicecious form, Fucus vesiculosus, was used in the experi-
ments.

To secure spermatozoids or ova for the investigation the
methods described by Strasburger (i) were followed.

Inasmuch as fresh material could not be secured daily, some
method of keeping the Fucns in good condition was desired.
Sinking the material in the sea was found to yield active sperma-
tozoids for only a day or two after bringing it in from its habitat.
Fucus plants wrapped in towels wetted with sea water and kept
in an ice chest yield active spermatozoids for at least five days
after being brought in. The antheridia from fruiting tips kept
in sea water in the ice chest for the same length of time were
exuded on drying, but the spermatozoids were inactive and lay
inert in the antheridia. This difference between the two treat-
ments is probably due to differences in oxygen supply. Stras-
burger advises shipping fruiting Fucus, desired for active sper-
matozoids and living oogonia, in large amounts of sea water, but
it appears that wrapping the Fucus in wet cloths and icing it
would be preferable.

EXPERIMENT ON CHEMOTAXY.

In investigating chemotaxy Pfeffer's capillary tube method
was used. The capillary tubes mounted in a drop of sea water
containing the spermatozoids were examined under the microscope
for about three quarters of an hour for change in direction of
motion of the individuals as they passed by the mouth of the
capillary tube and for collection of the spermatozoids in the tube
as described by Pfeffer and others for various Pteridophytes.
The majority of the substances, on account of the strong negative
phototaxy of the spermatozoids, were also tested by placing the
mounts in the dark and examining them after ten minutes, and

NOTES ON THE PHYSIOLOGY OF FUCUS SPERM ATOZOIDS. 12J

again in from twenty-five to forty minutes, for collection in or
about the tube.

The following substances made up in sea water were used in
the concentrations noted:

Substance. Concentrations.

Lactic acid i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol

Malic acid i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol

Butyric acid i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol

Citric acid i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol

Potassium oxalate (K.»C->On) . . . i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol

Ethyl butyrate 0.08%, 0.008%

Urea i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol

Sodium potassium tartrate i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol

Ethyl alcohol 95%, 9-5%. 0.95%, 0.095%

Cane sugar i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol

Peptone 2%, 0.2%, 0.02%, 0.002%

Potassium hydroxide i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol, o.oooi Mol

Hydrochloric acid i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol, o.oooi Mol

Potassium nitrate i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol

Diabasic potassium phosphate., i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol

Potassium iodide o.i Mol, o.oi Mol, o.ooi Mol

Ammonium sulphate 0.6 Mol, 0.06 Mol, 0.006 Mol, 0.0006 Mol

Hydrogen peroxide 3%, 0.3%, 0.03%

Sodium sulphate i.o Mol, o.i Mol, o.oi Mol, o.ooi Mol

No reactions which could be attributed to chemotaxy were
observed with any of the above substances except the acids.

It was found that with a molecular solution of the organic
acids in the capillary tube the spermatozoids were killed in the
course of a few minutes. Where o.i Mol acid was used there
was in some cases a collection of spermatozoids around the
mouth of the tube apparently attached to the cover glass or slide
by one cilium and these vibrated slowly. This collection in a
few cases was visible to the naked eye after an hour as a yellow
halo about the mouth of the tube. With o.oi Mol solution in
some of the trials there was a collection of the spermatozoids in
the tube after 45 minutes or more; o.ooi Mol had no effect.
The accumulation around the mouth of the tube containing o.i
Mol acid and the collection in the tube containing o.oi Mol
acid only occurred when sea water was used in which the sper-
matozoids were very active and numerous.

Such results might be interpreted as cases of chemotaxy.

128 W. J. ROBBINS.

Kusano (3) has described similar reactions of the swarmspores of
Myxomycetes to acids. He attributes them to chemotaxy.
An examination of individual spermatozoids, however, showed
no change in the direction of their movements passing the mouth
of the tube, the effects could by no means always be obtained,
and much the same result was observed in one case with HC1.
I believe that the results could be interpreted as a toxic phenom-
enon as explained below.

In the case of the. molecular concentration, the diffusing acid
is sufficiently strong to kill or paralyze the spermatozoids through-
out the whole mount. With acid of o.i Mol concentration the
diffusing zone of the acid which is toxic is not so wide as in the
case of acid of the molecular concentration and only those sper-
matozoids are rendered non-motile which pass comparatively
close to the mouth of the tube; hence the collection observed
after a time around the mouth of the tube. At the concentration
of o.oi Mol the diffusing acid is not toxic, and only those sper-
matozoids are rendered non-motile which pass comparatively
close to the mouth of the tube; hence the collection observed
after a time around the mouth of the tube. At the concentration
of o.oi Mol the diffusing acid is not toxic, and only those sper-
matozoids which enter the tube, as some must from several
hundred of individuals in a mount, are paralyzed and remain
there. Acid of o.ooi Mol concentration is below the toxic con-
centration and we have no effect. We do not find the effects
noted above in every mount where the acids are used because
there may be an accumulation of all active spermatozoids due to
negative phototaxy on the side away from the source of light,
in which case they do not meet the acid diffusing from the tube;
or there may be too few individuals in the mount to show distinct
collection; or they may not be actively motile due to the high
temperature or old material.

If the results noted are due to a toxic effect of the acids on the
spermatozoids, it would be expected that other substances, such
as potassium hydroxide and ethyl alcohol should also produce
the reaction. While no response to those substances was noticed,
this may have been due to poor material or to too high a tempera-
ture at the time of the experiment. In either case the spermato-

NOTES ON THE PHYSIOLOGY OF FUCUS- SPERMATOZOIDS. 1 29

zoids would be less active and no accumulation would occur. At
the time the work was done the relation between toxicity and
chemotaxy was not considered and further work on this phase
was not attempted. It would appear that the subject should be
investigated.

EXPERIMENTS ON PHOTOTAXY.

That Fucus spermatozoids are negatively phototactic is very
evident. Under the microscope the large majority will be
observed to swim away from the light and in a short time prac-
tically all the active spermatozoids will be found on the side
away from the light. If a capillary tube lies parallel to the window
the spermatozoids in the drop will collect against the tube on the
window side, blocked in their movement away from the light.
If mounted on a slide containing a bright spot of light in a dark
field, there is no collection in the bright area as Englemann (4)
found for Euglena under the same conditions. The spermatozoids
pass in and out of the bright area with no apparent response.

A slide holding a drop of sea water one fourth inch in di-
ameter was placed about two feet from a north window. In less
than five minutes practically all the active spermatozoids were
found crowded on that side of the drop away from the window.
It seemed possible this might be due to gravity.

Two slides were arranged as above, one inclined toward the
window and one tilted away from it. In both cases the sper-
matozoids were found crowded on the side away from the
window. In both cases the spermatozoids swam away from the
light, but in one case they swam with gravity and in the other
against gravity. A liter beaker half filled with sea water yello\v
with spematozoids was set on a shelf ten feet from a north win-
dow. In ten minutes far more active spermatozoids were found
in a drop taken from the side away from the window than were
found on the lighted side.

INFLUENCE OF TEMPERATURE ON ACTIVITY.

From indications during the experiments on chemotaxy it was

observed that temperature had a very considerable effect on the

activity of the Fucus sperm and that the optimum temperature

is probably relatively low. The experiment below is suggestive.

130

W. J. ROBBINS.

Sea water yellow with sperm was placed in two vials. One
vial was placed in an ice- water bath at approximately 4° C. and
the other was left at room temperature of 25°-27° C. Portions
were removed from time to time and examined for activity.

Begun at n:oo A.M.

Hours.

Time.

4°C.

25°-27°C.

H

11:30

More active than at 27°.

Active.

i

12:00

Many active.

Few active.

iH

12:30

Many active.

Few active.

2M

i:i5

Many active.

One or more in

microscope field

moving slowly.

3

2:OO

Many active.

One or more in

microscope field

moving slowly.

4

3:00

Many active.

One or more in

microscope field

moving slowly.

SUMMARY.

Using the Pfeffer capillary tube method of determining
chemotaxy, it was found that certain acids cause collection of
Fucns spermatozoids. It is suggested that this may be explained
as due to toxicity and not chemotaxy. Fucns spermatozoids are
negatively phototactic. Low temperature is favorable to the
activity of Fucus spermatozoids.

LITERATURE CITED.

1. Strasburger, E.

'84 Das Botanische Practicum, Erste Auflage. P. 392.

2. Bordet, Jules.

'94 Contribution a 1'etude de 1'irritabilite des spermatozoides chez les
Fucacees. Bull, de 1'Acad. Roy. de Belgique, 3 serie 27: 888-896.

3. Kusano, S.

'09 Studies on the Chemotactic and other Related Reactions of the
Swarm spores of Myxomycetes. Tokyo Imp. Univ., Agr. Jour. 2: 1-83.

4. Engelman, T. W.

'82 Uber lichtund Farben perception niederster Organismen. Arch. f. d.
ges. Physiol. 29: 396.

NOTE ON THE GALVANOTROPIC RESPONSE OF THE

EARTHWORM1

A. R. MOORE AND F. M. KELLOGG,
BIOLOGICAL LABORATORY OF BRYNT MAWR COLLEGE, BRYN MAWR, PA.

Since Blasius and Schweizer2 first found that many animals,
both aquatic vertebrates and invertebrates, orient themselves
characteristically to the electric current, the galvanotropic
response has been studied carefully by a number of investigators
in paramecia, motile algae, medusa strips and tentacles, certain
Crustacea and salamanders. In general it may be stated that
under normal conditions, paramecia, motile algse, and pieces of
medusa tend to move toward the kathode, while vertebrates and
Crustacea which are galvanosensitive, progress toward the anode
and avoid the kathode either by turning away from it or by
walking backward. Loeb3 and his collaborators account for the
coordinated movements of vertebrates and Crustacea under the
influence of the electric current by suggesting that they are due
to changes in the tension of associated muscle groups, viz:
flexors and extensors.

It seemed that the earthworm, Lumbricus terrestris, with its
simple locomotor system of circular and longitudinal muscles
offered favorable material for testing this idea. The following
is an account of the response of this animal to the constant elec-
tric current, based on a large number of observations made on
different individuals at different seasons of the year. It is true
that Blasius and Schweizer noted the galvanotropic response of
Lumbricus but unfortunately limited their description to the
stimulating effects of the "make" shock and the final crawling
of the animal to the kathode, consequently omitting mention
of the significant intermediary stages due alone to the flow of the
constant current.

In our own experiments, an animal to be tested was put into a

1 Drawings by Mary Cline.

2 Blasius and Schweizer, Pflueger's Archlv, Bd. 53, S. 493.

3 Loeb, J., Pflueger's Archiv, Bd. 66, S. 439.

A. R. MOORE AND F. M. KELLOGG.

hard rubber trough, 5 cm. X 5 cm. X 20 cm. and containing I cm.
depth of tap water. The current, from the ordinary lighting
circuit, passed through a water rheostat, pole changer, milliam-
meter and key, and was received at either end of the trough by

FIG. i.

non-polarizable electrodes made of wet filter paper. The strength
of current used varied between 20 and 100 milliamperes per
square decimeter, voltage no. As soon as the current was made

THE GALVANOTROPIC RESPONSE OF THE EARTHWORM. 133

the worm began to orient itself and in a few seconds the an-
terior and posterior ends were directed toward the kathode, so
that the animal took the form of a U with the concave side
toward the kathode (Fig. i). In this situation it continued to
make writhing and progressive movements, and therefore the
figure varied somewhat from time to time. If, by reason of
excessive movements, any part of the worm showed anodal curva-
ture, the latter was instantly corrected by contraction of the
muscles on the kathodal side. Because of the anterior end of the
worm being always the more vigorous in its movements, the ani-
mal ultimately succeeded, as a rule, in crawling to the kathode.

FIG. 2.

If it were cut into pieces 3 to 4 cm. long, these pieces when placed
transversely in the trough and subjected to the action of the
constant current, continued to orient themselves in the same
fashion as an entire worm (Fig. 2) but progressive movements
were absent. When the current was reversed, the specimen,
either entire or sectioned, immediately responded by turning
its ends towards the new kathode. The delicacy of the response
was lost in fatigued animals.

Since the earthworm has but two muscle systems with which
to accomplish its locomotion, viz: circular and longitudinal, it is
obvious that one-sided movement or bending can be brought
about only by unequal contraction of the longitudinal muscles
on the two sides. It would seem that this is a case in which an
entire animal responds to the electric current in the same fashion
as did Bancroft's1 medusa tentacles and bell strips. Since no

1 Bancroft, F. W., Journ. Exp. Zool., Vol. I., p 289.

134 A- R- M°ORE AND F- M- KELLOGG.

locomotor limbs are involved in the earhtworm's reaction its
response accords in as simple a way as possible with Loeb's
theory of galvanotropism. We may therefore conclude that the
constant current produces the effects described by increasing
the tension of the longitudinal muscles on the kathodal side of
the worm, wrhich results in this part being more strongly con-
tracted than the anoclal region.

PALM AND SOLE STUDIES.

HARRIS HAWTHORNE WILDER.

TABLE OF CONTENTS.

I. Introduction

II. A Primitive Palm Print 153

III. The Border Region of the Plantar Friction-Skin 157

IV. The Reduction of Line C 171

V. Friction-Skin Correspondence in Pygopagi 211

VI. The Heritability of Friction-Skin Characters 227

VII. Bibliography of Friction-Skin Configuration 245

I. INTRODUCTION.

In the study of the details of the configuration of the friction-
ridges found covering the surfaces of human palms and soles
there opens up a field of the greatest value for the biologist.
Varying greatly individually, though constant throughout the
life of a given individual; still following the lines laid down for
them in more primitive mammals, yet modified and varied as the
result of mechanical causes; showing markedly and with certainty
a direct inheritance from the immediate parents as well as from
generations more remote; they may be used with profit by the
morphologist, the ethnologist, or the student of genetics, while,
as the surest and most positive characters of an individual, they
may serve the authorities in the identification of a human body,
living or dead.

A great advantage in the study of these parts lies in the ease
with which a print of the surfaces may be taken, thus furnishing
a permanent record, accurate in even the minuter details, and
easily filed away. These prints, with the ridges marked in
black upon a white background, and reduced to a perfectly
plane surface, are much easier to study than are the real objects;
laid out upon a large table they may be compared with ease;
they are always ready for reproduction where desirable. Un-
deniably the patterns are complicated, and many new concep-
tions, and the new terminology which expresses them, confront
the beginner, as in any new field ; but this much once accomplished

135

HARRIS HAWTHORNE WILDER.

there opens up to the investigator an almost endless series of
new phenomena the study of which, in the few years during which
the subject has received special attention, has been no more than
begun.

As this paper is intended in part as an invitation, or perhaps
a propaganda, for more work in this subject, a bibliography of
the entire subject is appended, with some suggestions as to the
character of the separate titles. It may not come amiss, also,
to give in this place a key to the method of formulation now in
use, which, although somewhat artificial in the sense that it
does not have much morphological significance, is a convenient
way in which to express the fundamental conditions found in a
given palm or sole, and as such, has already been extensively
used.1

In the palm the starting-points of this system are four triradii
(Galton's deltas} which lie at the bases of the four fingers, and
may be designated as A, B, C, and D, the first situated beneath
the index finger, and so on. As in all triradii, these points form
each the meeting place of three radiating lines, at approximately
120° from each other, and of these three radiants two short ones
pass obliquely upwards (distally), defining a small digital area,
while the third follows a longer and quite variable course across
the palm. These latter are the four Main Lines, designated by
the same letters as are used for their triradii of origin, and as
their position indicates the configuration of the entire palm, a
simple method of describing their course is of first importance.

To "interpret" a given palm it is first necessary to locate the
four digital triradii, A, B, C, and D, and then to trace from these
centers their three radiants, and more especially the main lines,
following them across the palm wherever they may lead, never
crossing a ridge. Often, in this pursuit a single ridge may be
followed almost the entire distance; again, the ridge that is being
followed may come to an end, yet the direction be immediately
taken up with a new one, upon which the line may be continued.
Where a ridge forks, and thus allows two or more possible courses,
the most distal one should be taken. If, now, these courses be

1 Cf. the recent papers of Schlaginhaufen, 1906; Loth, 1906, and the two papers
of Wilder on "Racial Differences," 1904, 1913; cf. also the exposition of the method
in Martin's new "Lehrbuch der Anthropologie," Jena, 1914, pp. 360-367.

II

FIG. i. Print of a left hand, showing method of " interpretation," as ex-
plained in the text. The print shows the four main lines, and the numbeis arbi-
trarily assigned to the different parts of the margin. Formula: 9'7'S-4. Line A
becomes involved in a hypothenar pattern, giving it the value of 4.

138 HARRIS HAWTHORNE WILDER.

marked, while being followed, by a colored pencil, of red, or some
other conspicuous color, or by ink, the palm will appear like
those shown in Figs, i and 2, except that the result will be
naturally more or less unlike either model.

As the triradii of origin never vary much in position, the general
course of the main lines may be given by determining with some
precision their termini, that is, the points at which they issue
from the margin of the friction-skin area. This is easily accom-
plished by designating the several regions and points along the
margin by an arbitrary system of numbers, as here shown, using
the numbers I to 13. In this the more definite points, like
triradii or pattern-cores, are designated by even numbers, and
the intervals between these by odd. Thus 2 indicates the carpal
.triradius, lying on the proximal margin at the middle of the wrist.
When not actually present, the location of this point is equally
well determined by a parting of the lines towards the radial and
ulnar sides. 4 indicates the hypothenar pattern, a conspicuous
feature present in about 20 per cent, of white hands; when the
pattern is wanting, this number is not used. The numbers
6, 8, 10, and 12 indicate the four digital triradii, but in the reverse
order, beginning with triradius D. The odd numbers are not so
precise, and designate the entire lengths of margin between the
points just mentioned. I means any termination upon the
radial side (thumb-side) of the carpal triradius; 3 begins at this
latter point and runs up along the ulnar side as far as the hypo-
thenar pattern, and 5 lies between this pattern and triradius D
at the base of the little finger. When an hypothenar pattern is
not indicated the distinction between 3 and 5 is somewThat
uncertain, but in general, if the entire outer margin of the palm
between the lower outer corner (proximal ulnar) and triradius D
be divided into thirds, the lower, or proximal third is 3, while
the distal two thirds are 5. The boundary between these two
numbers thus corresponds to the point of location of 4 (hypothe-
nar pattern) when present. The numbers 7, 9, n and 13 desig-
nate the spaces between the fingers, 7 being that between the
little- and ring-fingers, and so on.

Thus, given these arbitrary values' for the parts of the margin,
it will be seen that in Fig. I, line D crosses the margin at 9, C
at 7, and B very high up along 5. Line A becomes involved in

PALM AND SOLE STUDIES.

139

the'hypothenar pattern, indicated by the digit 4. The entire form-
ula is thus, in the natural order, 4 • 5 • 7 • 9. In Fig. 2 lines B and C
become confluent, so that the termination of each is designated
by the number indicating the origin of the other, and the formula

FIG. 2. Print of a right hand, marked as in Fig. i. Formula: u • 10-8- 5. Here
the main lines B and C are confluent, and may each be considered to end in the
triradius of origin of the other; that is, line B ends in triradius 8, and line C in
triradius 10. The third and fourth interdigital patterns are present, the latter
with a triradius belonging to it. These features may be added to the formula in
giving a complete description.

140 HARRIS HAWTHORNE WILDER.

will read 5-8-io-n. In this hand, also, is shown a case of
divergent ridges along the course of line A, a common phenome-
non, so that by following different ridges the line might be made
to terminate almost anywhere along a considerable extent of
margin. Following the rule above stated, however, that of
carrying a line as far distally (towards the finger-tips) as possible,
the line is made to terminate at 5.

In this way any human palm may be formulated by the use
of four numbers, which indicate the conditions so nearly that
one cannot be far out of the way if he should try and draw a
given palm from the formula alone. This can be easily proven
by anyone who knows the system and makes the attempt, using
the formula only, and withholding the original print until the
drawing is completed. This formula makes no attempt to
indicate any features other than the course of the four main
lines, yet, in point of fact, the presence of a pattern is sometimes
indicated, as in the case of Fig. 2, where the fusion of lines B
and C encloses a small area between the middle- and ring-fingers,
and makes a looped pattern there a necessity.

When a large number of formulae are accumulated they may
be readily arranged in numerical order, subdividing first by the
first number, and so on. For practical purposes, however, it is
found better to reverse the entire formula, beginning with the
number representing the course of line D, since with this latter
there is not only more precision in termination than in the case
of line A, but also line D is more variable, and thus furnishes
more classes for the first subdivision. Thus the formulae for the
two cases here illustrated should read respectively, 4-5 -7 -9 and
1 1 • 10 • 8 • 5, the first coming before the second in a numerical list.

Corresponding to the results of a morphological study of the
patterns, which shows them to have first developed on the raised
surfaces of the eleven typical pads (Whipple, 1904), traces of
this number are to be expected in their proper places. The five
apical patterns appear on the balls of the five digits, the center
or core of each coinciding with the rounded apex of the raised
area. In their most primitive form they appear as concentric
circles or ovals, or some modification of this form, limited by two
triradii (deltas), the "whorls" of the finger-print system; the
reduction of either one of the original triradii transforms the

PALM AND SOLE STUDIES.

whorl into a loop, ulnar or radial in accordance with the triradius
that is suppressed; and this figure in turn becomes an arch by the
suppression of the remaining triradius. The low arch, without
appreciable core, is thus the final stage in the reduction process
of a finger pattern. The fact that all possible stages along several
lines of degeneration are found in man indicates the reduction in
functional value as friction organs of these once so vital parts.

The six patterns of the palm proper consist of the four inter-
digitals, lying just proximal to the intervals between the fingers;
the thenar, and the hypothenar. Of the four interdigitals the
first (at the radial end of the series) lies below the wide interval
between thumb and index finger, and is found usually in close
connection with the thenar, the twTo forming loops opening in
opposite directions, and with two triradii between them. This
pattern-complex is rare in the white race; but much more frequent
in some others (e. g., the Maya-Quiches of Yucatan; Wilder,
1904). The three remaining interdigitals appear below the finger
intervals, the second lying between triradii A and B, the next
(third) between B and C, and the fourth between C and D. The
one between triradii A and B (second) is the rarest of the three
and the fourth, between C and D, is the commonest. Further-
more, this last is frequently provided with a triradius of its own,
aside from C and D, a peculiarity occasionally, but not so fre-
quently, met with in the other cases. The third pattern, between
B and C, is liable to be confused with a "false pattern" caused by
certain configurations of the ridges of the palm, but easily dis-
tinguished from the real one by its position in a depression rather
than on an elevation (Whipple, 1904). The three mounds of
this region are easily seen in most hands by bending the fingers
back and looking across the palm in various directions. They
are the plainest in children and in the fetus they are often ex-
tremely conspicuous (Retzius, 1904).

As might be expected, a human hand presenting all eleven
patterns is quite rare, yet does occur. Such a case, found in the
right hands in twin boys, has been already figured and described
by the writer.1

As each of the four interdigital pads (or patterns) possesses
typically three triradii of its own, with four for the third, it is

1 Wilder, A-nat. Anz., Bd. XXXII. , 1908, pp. 194, 195.

142

HARRIS HAWTHORNE WILDER.

FIG. 3

FIGS. 3-5. Diagrams giving the morphological explanation of the friction-ridge
configuration as found in the Primates, and especially in man. Fig. 3 represents
the condition in primitive walking mammals, and follows closely the condition

PALM AND SOLE STUDIES.

143

FIG. 4

found in Microtus, a field mouse and Crocidura, a shrew. The contact with the
ground comes upon eleven walking pads; five apical or terminal, 1-5, four inter-
digital, I.-IV., a thenar, Th, and a hypothenar, H ; eleven in all. These are surrounded
by folds of skin, two for the apical pads, four for the third interdigital, and three

144

HARRIS HAWTHORNE WILDER.

FIG. 5

for each of the others, and where these folds come together there are formed Iriradii,
or centers from which folds radiate in three directions.

Upon this original surface certain epidermic units (scales?) which are un-
doubtedly very primitive, and were there before the formation of pads and folds,
become definitely arranged in concentric circles over the pads, and in lines along
the folds. Nearly every stage in this rearrangement may be found by looking
over the more primitive mammals, especially marsupials and lemurs. The separate
rows of units tend to unite to form ridges, which possess an important function in
increasing the friction, and thus preventing slipping.

In monkeys, Fig. 4, the pads and folds are flattened down to an approximately
level surface, but the arrangement of the ridges still indicates the former condition.
The relief has become a picture.

In man, Fig. 5, the reduction in functional importance of these friction-ridges
allows all forms of degeneration, both in the individual patterns and in the ridges

PALM AND SOLE STUDIES.

FIG. 50.

145

as a whole. There is also shown a marked tendency for all the ridges to shift from
the longitudinal position of the apes to one more nearly transverse, and this ten-
dency is much more marked in right than in left hands, evidently corresponding to
use, and recording the change from the grasping of tree boughs to the holding of
tools and other objects.

While all degrees of the loss of the original configuration and the assumption of
transverse ridges may be found in different human hands, it is seldom that so
many of the mound patterns are retained as in Fig. 5 (Coll. No. 90). The formula,
however, 11-9- 7 -5, indicated the establishment of the transverse position in the
interdigital region. Fig. 5a shows a very primitive Thenar region in a Liberian
negro. (Coll. No. 571, taken by F. Starr.)

evident that in the human species a very considerable reduction
of these points has taken place. There are left in all cases,
however, the four main triradii, which are so constant as to
allow their use as the starting points in palm formulation although
it is by no means certain that they are in all cases strictly homol-
ogous. Thus, from Figs. 3 and 4 it is seen that the second inter-
digital pattern has originally two distal triradii, either of which
might persist as triradius A, and it is likely that the one that
appears is sometimes one and sometimes the other of the original
two. Other of the missing triradii appear occasionally somewhat
lower down on the palm (proximal to the pattern) especially in
connection with the fourth pattern.

A definite hypothenar pattern appears on about 20 per cent,
of hands of the white race, but the occurrence differs considerably
racially and in some may be either more or less frequent. It is
occasionally found in its more primitive form, as a whorl with

146

HARRIS HAWTHORNE WILDER.

the three original triradii, but it displays a tendency to become
drawn out along one axis and form an extensive spiral or S-formed
figure.

As is well known, the exact and detailed configuration of the
five apical patterns, the "finger-prints" of identification bureaus,
was employed by the late Sir Francis Gal ton (1892, 1893, 1895,

FIG. 6. Diagram of a human sole, with the margins numbered to correspond
to the palms in Figs, i and 2, and marked with the main lines. From an article by
the author in Pop. Sci. Monthly for September, 1903. By permission.

etc.) as the basis for his system of personal identification, and,
within the last twenty years, has become scientifically developed
by certain police investigators, notably Sir Edward Richard
Henry. At the present time the "finger-print system" has been
introduced into all civilized countries, and its scope is steadily
increasing. As the friction-skin configuration offers an abso-

PALM AND SOLE STUDIES. 147

lutely sure test, and the only one, for the identification of a
human body, alive or dead, it is certain that this use of the apical
patterns will be followed by a similar employment of the entire
palmar and plantar surfaces, as advocated by the present author
since 1902, and at this writing it is a pleasure to note that sole-
prints have recently been put in use in a Chicago maternity hos-
pital for the identification of the babies.

Something on the frequency of occurrence of the six patterns
upon the palm has been done in connection with racial differences
in friction-skin configuration1 and their progressive degeneracy
from the simian type has been followed, in some cases along
several lines, by Miss Whipple (1904, pp. 341-350).

The friction-skin configuration of the sole shows a much greater
range of variability than is seen in the palm, and correspondingly
its formulation presents a more serious problem. When the
print indicates the four main triradii it is possible, of course, to
draw the same four main lines, A, B, C, and D; using the same
arbitrary numbers to designate marginal features (Fig. 6), yet
this method is unpractical from a number of causes. In the
first place the four essential triradii lie up in the hollow between
the ball and the toes, and quite often come beyond the usual con-
tact area, and thus are not seen in a print. Again, they are by
no means as constant as in the hand, being occasionally replaced
by other triradii which have survived instead of these; and still
again, when the main lines are traceable, they quite often assume
near their free ends a parallel course and terminate close together
along the margin designated by the digit I, thus giving little or
no variety to the formula (i • I • I • i).

It has thus seemed more expedient to employ a distinctly dif-
ferent scheme for sole formulation, as illustrated in Figs. 7 and 8.
The print is interpreted, that is, is marked off into areas, by the
use of the main lines as in the palm, but as the lines themselves
are not used otherwise than as boundaries, the tracing of their
course is not so critical. Where all four main triradii are present
as in Fig. 7, this interpreting is a simple matter; yet there are
numerous cases in which one or more of the triradii lie above
(distal to) the contact area, as in the case of triradius C of Fig. 8.
Here there is, however, some divergence of the ridges at the upper

1 Wilder, Amer. Anthrop., 1904, 1913.

148

HARRIS HAWTHORNE WILDER.

margin, sufficient to indicate the position of the line approxi-
mately, and the case is assisted by the presence of a "lower"
or proximal triradius, one radient of which practically coincides
with the supposititious C line. Thus between the two a boundary
is fixed between the areas of the fourth and the third interdigital
patterns, while the second, and the boundary between it and the
third are definitely indicated by means of the main triradii.

FIG. 7. Print of a right sole, showing the method of dividing the "ball" into
its four interdigital areas, each of which may be separately described by formula.
This method is fundamentally different from that used for the palm, as shown in
Figs, i and 2. Formula: W-L-CI-0-H-H2.

Now in all cases, whether the limits are precise, as in Fig. 7,
or approximate, as in Fig. 8, the "ball" of the foot is easily
divided up into areas corresponding to the four interdigital
patterns. Of these the first, beneath the great toe, or rather

PALM AND SOLE STUDIES.

149

beneath the interval between this digit and the next, is the largest,
and is usually covered with a definite arrangement of ridges, the
hallucal pattern. It will be noted that this, the most important
of the four in the foot, is the same one that, in the hand, appears
only as an adjunct of the rare thenar pattern, and is thus of little
importance. This difference is plainly the result of the difference
in function and use of foot and hand, as in the former the main
work is performed by the heel, and by the four interdigital areas
placed in a row close together, while of this row the inner end,
below the great toe, is used the most.

The descriptive formulation of the sole configuration is thus

FIG. 8. Print of a right sole, showing a second case of the method of inter-
pretation used in the preceding figure. This print exhibits the unusual case of the
effacement of a definite hallucal pattern, similar to the sixth type 'shown in Fig. 9.
Formula: BC-L-Cl-Cl-x.

seen to be most advantageously accomplished by using the areas
rather than the lines, and describing the condition of the patterns
found upon them. To begin with, the hallucal pattern, which is
used for the classification of soles into primary groups, is found
in its most primitive form as the whorl, equipped with its three
typical triradii, A, B, and C (Fig. 9). Either triradius may be
wanting, thus allowing the ridges to open, or gush out, to use a

150

HARRIS HAWTHORNE WILDER.

figure which has no morphological significance, in the direction of
A, or B, or C. The whorl, in formula writing, may be designated
as W, and the three derived types just given, may be distinguished
by the triradius that is wanting, as A, B, or C, respectively.
Aside from these, there are compound types, where two of the
triradii fail, and where the ridges flow out in two directions. Of

W

B

C AC

FlG. 9. Diagram showing the types of hallucal pattern. W is a whorl, with
the three triradii all present, A, B, and C. In type A triradius A, the distal one.
is lost, and the ridges flow out between the first and second digits. In type B the
triradius of the name, the tibial one, is lost while A and C remain, and in type C,
the fibular triradius (C) is lost, or nearly so, giving an outlet for the ridges in that
direction. Types AC and BC may be interpreted as compound types, explicable
from the others, especially C, where the two triradii A and C lie near together. In
AC the remaining triradius is assumed to be B, and in BC it is A. In AC the rem-
nant of the pattern lies to the left of the triradius; in BC below it.

PALM AND SOLE STUDIES. 151

these the possible types are AC, BC, and AB, the distinctions
between which are obvious. The type shown in Fig. 7 is virtually
a whorl (W), although it is not typical; while the type of Fig. 8 is
extremely rare, and is most probably interpreted as an AB, with
some possibility that it is rather a BC. The three remaining
areas are conveniently designated as I., II., and III., respectively,
and are characterized as either open (0), or dosed (Cl) in accord-
ance with the condition of their proximal ends. Thus, in Fig. 7,

I. and III. are open, that is, they reach the inner margin, while

II. is clearly closed; in Fig. 8 all three are closed, but I. is open for
some distance, curves around the base of II., and abuts finally
against the side of III. Occasionally, too, an area may be closed
at the top, also, as area I. in both figures given, and sometimes
the ridges of two areas flow into each other, that is, they are con-
fluent. A confluence is easily marked by the designations of the
areas involved, united by a + sign, as, I. + III.; I. + 11.; and a
closure of the distal end may be designated by an / (loop).

This method of formulation of the sole configuration was first
attempted in 1904* in making comparisons of the sole prints of
different human races. In this article, in addition to the for-
mulas, as just described, certain exponent letters with an arbitrary
significance were employed to qualify the more general terms.
The presence of a hypothenar pattern (usually represented by a
single loop) upon the outer edge of the foot proximal to the
interdigital row, was also indicated by a capital H, added to the
rest.

Representative sole formulas, taken from this article, are here
given. The main symbols will be readily understood from the
foregoing; the meaning of the descriptive exponents, which are
of less value, may be found by referring to the article, pp. 253-254.

0
0
+ 20 H

+ i

+ 1+2

In the first of these the hallucal pattern is a whorl, while the
three succeeding areas are open fields, without patterns. The

1 Amer. Antkropol.

w

0

0

A

OCl

Cl

A.

0

+

Wsp

+ 3

Cl

B

+ 3

Cl

152 HARRIS HAWTHORNE WILDER.

second has an "A -loop," the commonest form of hallucal pattern,
where the pattern is closed, except at the interval between the
great toe and the next, where the ridges run up and attain the
margin; the second area, II, has a proximal loop, probably with
triradius, which confines a part of the ridges. In the third formula
some of the ridges run in the form of a U between areas 2 and 3,
and in the fourth a similar union involves all the ridges of areas
I and 3. .The fifth notes an unusual form of hallucal pattern,
together with a strange mix-up in the union of areas, which will
become plain from the explanation of the rest.

This method of formulating sole configuration is by no means
an ideal one, but it serves fairly well the general purpose of pic-
turing the configuration of a given sole in a few terms. A
fundamentally different method has been introduced by Schlagin-
haufen (1905), based upon the identity of the various triradii
occurring on the sole. Each one of these is designated by an
abbreviation, like "/Q" or "/I3," and the description is based
upon the position and relations of these. This method is thus a
method of description rather than a formulation, and although
in many ways convenient, it depends upon absolutely exact
homologies, which, in our present state of knowledge, is not pos-
sible. It thus seems better to rely upon some artificial method,
which is capable of picturing the essential details of a given
sole, rather than to attempt a series of homologies based upon
our present incomplete knowledge.

The two proximal patterns, thenar and hypothenar, have
naturally suffered much displacement from the peculiar lengthen-
ing of the proximal part of the foot to form the human heel, and
a discussion of their position and homologies will be found farther
on in this article. Here it may be said that the transversely
placed loop, sometimes seen on the outer edge of the sole, just
proximal to the row of interdigital patterns across the ball, may
safely be considered the hypothenar pattern, or more probably
a portion of it (Fig. 7). The thenar, rare and more or less rudi-
mentary, as in the hand, is indicated by traces of irregularity in
the lines, or an occasional loop and triradius, found upon a slightly
raised area upon the inner edge of the sole, not much distal to the
heel (Figs. 21 and 23 below).

PALM AND SOLE STUDIES.

153

II. A PRIMITIVE PALM PRINT.

Among the prints collected from college students the past
year occurred an isolated case, unique in the history of human
palms thus far known (Fig. 10).

FIG. i o. Tracings taken directly from prints of the two hands of Miss -

(Coll.

No. 647), showing the simian type to a degree not even approached by any other
human individual of the many hundreds thus far placed under observation by the
various investigators. Formula: i • 7 • i • i or i • i • i • i.

154

HARRIS HAWTHORNE WILDER.

In the left hand of this individual, the one first noticed, all
four main lines, A, B, C, and D, which proceed from the triradii
at the bases of the four fingers, instead of running transversely
or obliquely across the palm, as in all cases hitherto observed,

i\\ii\\\w///7////

hU\\\\;Uif^/^

FIG. ii. Hand of a chimpanzee (Anthropopithecus), taken from a drawing by
W. Kidd, and treated in the same way as Fig. 10, for better comparison with it.
Formula: 7-6-2 -2.

pass downward in a longitudinal course, and terminate near
together in the middle of the wrist. This gives the unheard-of
formula 2-2-2-2, if they be considered as terminating together

PALM AND SOLE STUDIES.

155

in a carpal triradius, or, what is still more remarkable I • I • I • I,
if they do not.

It is rather unusual, but by no means rare, to find line A as-
suming a low position along the lower third of the free outer
border of the palm (position 3) ; several cases have been met with

FIG. 12. Hand of a gorilla, taken from a drawing by W. Kidd and treated in the
same way as Fig. 10, for better comparison with it. Formula: 3-3'2- 1- (?).

in which this line reaches the carpal triradius (2) ; and one
instance is known of line A terminating within (on the
thumb-side of) this point (position i); but no such course has
ever been recorded for any of the other main lines, not even B,

156 HARRIS HAWTHORNE WILDER.

which, from its propinquity to A, might be thought capable of it.
Line B, at the lowest position recorded, still keeps within the upper
two thirds of the outer (ulnar) margin of the palm, and thus
never attains a number less than 5. The known range of line C
is between 5 and n, and D always curves up, from position 7 on,
save in one rather doubtful case in which it appears to curve
downwards and outwards, to reach the outer margin at 5.

The right hand of the same individual, while closely similar to
the left in general appearance, introduces a slight variation in
the form of a lower triradius, which confines a number of the
ridges of the third interdigital area, allowing line C if one wish to
so interpret it, to curve upwards and terminate at position 7,
between the ring- and little-fingers. By this interpretation the
course of Line C becomes quite normal, and suggests the possi-
bility, with a slight rearrangement of lines, of fusing lines B and
D, giving the total formula 10-7 -6- 1, which, except for the posi-
tion of line A , is not at all unusual.

This right hand, much like the left but with the profound
change produced by the presence of the lower triradius in the
third interdigital area, suggests a partial explanation for the
wholly anomalous condition of the other side, for in the left palm,
at the point corresponding to the lower triradius of the right there
seem to be one or two ridges that curve and suggest the last
vestiges of a vanished triradius like that of the right. The
presence of such a formation in this place might bring about a
fusion of lines C and D, making the first two terms of the formula
6-8, yet even thus lines A and B are not explained, and remain
wholly abnormal.

Whatever the explanation, the picture presented by these two
palm prints, especially the left, with the ridges crossing the hand
lengthwise, is strikingly like that exhibited by the large Anthro-
poids. This is shown in Figs. II and 12, put into the form of
diagrams after the sketches of Kidd (1907; Figs. 48 and 46).
Since, however, the figures of this author were drawings from the
objects, and not print impressions, and as they were designed
to show the relief, including certain of the deepest rugse, it was
often difficult to determine the exact course of the ridges, and
they are not wholly trustworthy on that account. The similarity

PALM AND SOLE STUDIES. 157

of these to the case here presented is obvious; they may even be
formulated by the method devised for human palms, the formula
being, respectively 7 • 6 • 2 • 2 and 3 • 3 • 2 • I .

The family to which the subject of this sketch belongs, No. 647
of my collection, is of good old English ancestry, and numbers
among her near relatives several persons of more than ordinary
distinction. The subject herself is an attractive young person
of excellent mind, graduating from college with distinction, and
having nothing at all abnormal about her, save this singular ar-
rangement of the palmar friction-ridges:

An investigation of the hand prints of both parents, revealed
typical European racial characters, such as the formula n -9 -7 -5
in the right hand of the father and the left of the mother; the
father's left hand bears the formula 11-7 -7 -4, and the mother's
right is ii -II -9 -5, a little unusual but quite normal for a white
person.

Unfortunately, owing to the prejudice of the family, I was not
permitted to obtain sole-prints of either the subject herself or
of her parents, and consequently can make no further report.

III. THE BORDER REGION OF THE PLANTAR FRICTION-
SKIN.

Unlike the condition in the hand, where the friction-skin of the
palm terminates along the borders of the contact surface, the
friction-skin of the foot extends up along its sides considerably
beyond the physiological sole. Since friction-skin has little or
no pigment, this extension upwards gives the feet of dark-skinned
races, when barefooted, the well-known appearance of pink slip-
pers worn over black stockings.

It thus happens that, while in the -hand a simple contact print
includes practically the entire friction-skin area, a similar print
of the foot includes only the inner portion of the ridged skin,
leaving an appreciable border entirely around the part printed,
concerning which no record is made. The relation of this tread-
area, the usual extent of a print, to the entire surface covered with
friction-skin is shown in a few examples here, and convinces one
immediately of the insufficiency of a tread-area print (Figs. 13
and 14).

158

HARRIS HAWTHORNE WILDER.

The most common loss is that of the loop upon the outer edge
of the sole, the probable equivalent of the hypothenar, or of a part
of it. This loop is extremely frequent, but as its center occurs

FIG. 13. Print of right foot of Coll. No. 22, showing hypothenar loop, and thenar
rudiment, both outside of the tread area.

near the hypothenar edge, it is more than likely to fall just beyond
the edge of a tread-area print. Thus, when completely printed,
Figs. 13 and 15 are alike in respect to the loop in question, but
with ordinary tread area prints the loop in the former would
escape observation. In Fig. 16 a single hypothenar loop would
be evident in a tread-area print, but it would never be mistrusted

PALM AND SOLE STUDIES.

159

that there was a second loop proximal to the other, or that the
two were separated by a triradius; in Fig. 17, closely similar to
16, a tread-area print would show simply an uninteresting sole,

FIG. 14. FIG. 15.

FIG. 14. Print of right foot of Coll. No. 202, showing extensive hypothenar
loop, not shown by an ordinary tread area print.

FIG. 15. Print of right foot of Coll. No. 609, showing hypothenar loop, practic-
ally identical with that of Fig. 13, but so placed that it appears in a tread area print.

devoid of all special features, and would fail to show either
hypothenar loop or the triradius between them.

Thus far, in the history of the investigation of human soles,
there have been on record but three cases in which the general
surface of the sole, proximal to the ball of the foot, shows a

i6o

HARRIS HAWTHORNE WILDER.

pattern — a West African negro,1 a Pole,2 and a Liberian negro.3
For the sake of comparison they are all reproduced here (Figs.
18, 19, 20), that of the Liberian being completed in theory

FIG. 16.

FIG. 17.

FIG. 16. Print of right foot of Coll. No. 236, showing two hypothenar loops and
• a large triradius. An ordinary tread area print would show only the more distal
of the two loops, and present a sole, in this particular like Fig. 15.

FIG. 17. Print of the right foot of Coll. No. 183, like Fig. 16, with the addition
of a thenar rudiment, the most of which would remain unrevealed by an ordinary
print.

1 Schlaginhaufen, 1905, p. 102.
- Mme. Loth, 1912, p. 603.
8 Wilder, 1913, p. 205.

PALM AND SOLE STUDIES.

161

beyond the limit of the print, to show the probable course of the
ridges. Now in all three cases we are dealing with loops that
run across the sole, and in any one of them, if the core of the loop

FIG. 18. FIG. 19.

FIG. 18. Print of Schlaginhaufen's case; a West African negro, with a loop in

the hollow in the foot. Both feet were practically the same. Compare with Fig. 14.

FIG. 19. Mme Loth's case; a negro from North America. Compare with Fig. 14.

•

had chanced to have been placed a little more laterally there
would have been shown simply a sole crossed by the usual
transverse lines. Even as it is, in two of the three cases, Mme.

1 62

HARRIS HAWTHORNE WILDER.

FIG. 20. Print of right foot of Liberian negro; H. H. W. Coll. No. 585, collected
by F. Starr. Here the loop which corresponds with the two preceding is brought
into association with a more distal one turned the other way, and perhaps, if we
may believe in the theoretical completion of the figure as indicated, there is involved
a third loop, more proximal than the others. Careful notice must be taken that
the heavy lines of the margin, which indicate the tread area, are here all that we
have, as the subject is somewhere in Liberia, and not available, and therefore, that
the completion of the lines outside of this margin are wholly theoretical. In all of
the other cases the lines outside of the tread area rest upon actual prints ob-
tained by rolling the foot. This print has been previously printed (Amer.Anlhropol.,
1913, p. 205).

PALM AND SOLE STUDIES. 163

Loth's and mine, the individuals were flat-footed and the prints
were broader than usual, so that if the feet had had the customary
height of arch these figures would have lacked in extent, and in
the case of the Liberian the loop upon the inner side might have
been entirely lost.

It thus forcibly suggests itself that such patterns as the three
just considered may not be as rare as has been thought, but that
many of the prints of the usual monotonous type, consisting, through
the middle of the foot, of transverse parallel lines without special
features, might yield interesting results if the entire friction-skin
area were included.

Some years ago I attempted to remedy this defect by rolling
the foot while being printed, first to the outer, and then to the
inner, side, but while this proved fairly satisfactory for the outer
portion, the hollow part of the sole, upon the inner side, still
remained unprinted, and left a large oval area without record.
Schlaginhaufen also, who devoted some study to this hollow
region in his extensive work on the Planta (1905) has later sought
to include in his records these neglected parts, and has places
in his printed record sheet for (i) the tread area, (2) the rolled
outer edge, (3) the inner hollow of the foot, and (4) the back of
the heel, thus making the record a complete one (1912).

In my own attempts to investigate this border area of the soles
I find it expedient, first, to make a careful study of the foot itself
with a lens, and then to prepare prints corresponding in general
to those recommended by Schlaginhaufen; but in some cases I
supplement these by strips, applied to the inked foot, and extend-
ing from one triradius, or other feature, to another, in order that
they may be correctly oriented. Naturally, the three-dimen-
sional object which the friction-skin covers makes it impossible
to reduce the whole to an exact plane, but by overlapping the
strips, superposing them carefully at the triradii, pattern-cores,
and other important places, and then tracing the whole by a
thin piece of paper which covers it, such a figure as that shown
here (Fig. 21) may be obtained, which is approximately correct.
For the details of pattern-cores, etc., the separate slips are suf-
ficient, but the exact orientation of these must naturally be
preserved.

64

HARRIS HAWTHORNE WILDER.

Prints of the tread-area and of the rolled outer edge are ob-
tained in the usual way, by the use of an inked surface, upon which
the part to be printed is first placed, and then placed, in the same
way, upon a piece of white paper; but the strips used for details,
or those covering curved surfaces, are best printed by first inking

FIG. 21. Print of right foot of Coll. No. 87, reconstructed from impressions
taken of various areas. Here the arch is very high and the tread area, indicated by
the entire line, is very inadequate. This is one of the few cases known which show a
calcar pattern, here, as elsewhere, consisting of a single loop on the heel, open to the
tibial side, and slanted somewhat distally, thus coming into close relation with the
thenar pattern, covering the small thenar eminence. This latter pattern is also in
the form of a loop, with its opening directed fibulo-distally, and is supplied with a
triradius.

PALM AND SOLE STUDIES.

165

the foot and then putting on the slips, applied something like
bandages, and pressed with the hand. To ink the foot the roller
used in inking the paper surface may be employed, care being
taken to spread the ink evenly and not too thickly. The applied
strip may be pressed or gently rubbed with the hand.

One of the chief gains resulting from the study of the completed

FIG. 22. Drawing of the plantar aspect of the right foot of a chimpanzee, to
show the normal position and relations of the four interdigital and the two proximal
pads. For comparison with Fig. 23.

friction-skin of the foot lies in the added material furnished for
the study of the proximal row of patterns, the thenar and the

1 66

HARRIS HAWTHORNE WILDER.

hypothenar. The long, extended heel of the human foot is so
unlike anything found in typical mammals that even the location
in the human subject of these once important eminences, with
their associated patterns, is by no means a simple matter. The

CaLc

ar

FIG. 23. Diagrammatic drawing of a right human foot, resting upon the heel,
as seen in a recumbent figure; inner (tibial) aspect. The thenar pad is exaggerated,
but accurately located. For comparison with Fig. 22.

best aid here comes from the comparison of the human foot
with that of some one of the large Anthropoids, preferably a
Chimpanzee (Fig. 22), where both the thenar and hypothenar
are still evident, although the drawing-out process has already
commenced. The backward extension seems here to involve
mainly the hypothenar region, and as this surmise is supported
by the relation of the underlying skeletal parts, we may look upon
the entire human sole, back of the " ball of the foot," i. e., the inter-
digital pads, as an extended hypothenar. The thenar seems to
take no part in the extension, but lies passively upon the medial

PALM AND SOLE STUDIES. 1 67

side of the foot, quite above the tread area, where in man it may
still be found greatly reduced in size and importance, but occa-
sionally quite evident (Fig. 23).

Difficult to even locate at first, one soon becomes accustomed
to picking out the thenar pad on almost any foot, and in cases
where it is really prominent, and when seen in the right light, it
forms a conspicuous object. The friction-ridges crossing this
eminence almost always exhibit some disturbance of their other-
wise even course across the foot, and occasionally show a distant
loop with a triradius (Fig. 21). As is to be expected, too, the
best developed patterns are likely to occur on the most conspicu-
ous pads, the atavistic tendency manifesting itself in these two
ways at the same time. The last vestige of the thenar pattern
is indicated upon a flat surface, without trace of a pad, by a
slight disturbance of the friction-ridges in this place (Figs. 13,
17, 24, and 25).

Miss Whipple, in her fundamental work upon the whole subject
of epidermic ridges, has considered the causes of the degeneration
of the thenar pad of man and finds the principal one in the estab-
lishment of the long arch, spanning the distance between the
ball and the heel. The thenar pad, and, as she states, the
hypothenar also, are both situated directly beneath the center
of the long arch, and their reduction thus becomes a matter of
necessity.1 Schlaginhaufen2 gives several prints of the thenar
pattern, without definitely designating it as such, and was quite
right in stating that he was the first to carefully investigate this
pattern. In its best developed form this thenar pattern, as
thus far recorded, seems to consist of a simple loop with a single
triradius (Fig. 21), or occasionally, two loops, with the triradius
between them.3

Light is shed upon the question of the hypothenar pattern by
the interpretation of the long-extended human sole as formed from
the hypothenar region, since, if this be true, the pattern, or
elements of it, may thus be looked for on any part of this area.
This would determine at once as hypothenar, not only the narrow
loop found quite commonly upon the fibular edge of the foot, a

1 Cf. Miss Whipple, 1904, p. 292, Fig. 19.

2 1905, pp. 107-109, Figs. 182-184.

3 Schlaginhaufen, loc. cit., Fig. 183, i.

1 68

HARRIS HAWTHORNE WILDER.

little proximal to the ball (Figs. 13, 14, 15) but also the second
loop further back (Figs. 16 and 17), and probably also the enor-
mous figure found in the middle of the arch in the three special
cases mentioned (Figs. 18, 19, 20). Aside from these three

FIGS. 24 and 25. Prints of the two soles of the same individual (Coll. No. 199),
in which an inward rolling of the feet reveals on both sides a rudimentary thenar
pattern.

elements there is also the calcar pattern, which, aside from a single
family, in which it occurs in father, mother and two of the three
children, I have seen but twice. (See below, Part VI.)

Now, at the present state of our knowledge it is impossible to

PALM AND SOLE STUDIES. 169

say whether all of these patterns, or pattern fragments, are parts
of a single hypothenar pattern, which has become spread out,
and its parts disassociated, by the great extension of the part
covered by it, or whether there are added to the genuine hypo-
thenar elements certain "secondary patterns" (Whipple), like
those covering the proximal and medial phalangeal surfaces in
certain apes. The calcar pattern is treated as a secondary one
by Miss Whipple,1 and is figured in two specimens of Cebus,
where it appears definitely distinct from thenar and hypothenar,
although in contact with both.2 On the other hand it involves
no improbability to treat all of these elements, with the possible
exceptions of the calcar, as the more or less disassociated parts
of a long-drawn-out hypothenar, the result of a great extension
of its field in one direction and finds a close analogy in the apical
patterns of the four lesser toes in the human foot, where the
pattern is drawn out laterally to such an extent that there occur
not only long extended S-shaped figures, with the two loops far
apart, but even those with three loops and three associated
triradii. One of the latter is figured by Schlaginhaufen,3 and
the same type is described briefly by Miss Whipple,4 where
"the pattern has become separated into distinct loops and an
accessory degeneration triradius is introduced, that is, a triradius
not originally present in the typical scheme but formed incident-
ally in the process of degeneration of the pattern."

It is thus a priori probable that the three elements represented
by (i) the simple hypothenar loop of the fibular side (Fig. 13),
(2) the second loop occasionally present proximal to the latter
(Fig. 16), and (3) the large loop in the middle of the foot, of which
but three instances have thus far been described (Figs. 18, 19
and 20), are all parts of a degenerated hypothenar, yet the ho-
mologies of these several elements with one another, and the
course of degeneration in the original hypothenar pattern, with
the interpretation of these various existing vestiges, is still a
large problem. For this there is great need of more data, com-
plete prints of the soles of a large number of individuals, each

one including the border areas of the friction-skin, with the

1 1904, p. 361; Taf. VI.

2 P. 334, Fig. 37, b and c.

3 1905, p. 114, Fig. 185.
4 1904, pp. 352-353-

170 HARRIS HAWTHORNE WILDER.

portions so taken that recognizable features of the usual tread
area are included, to allow definite orientation.

From the few sole prints given here there are certain deductions
and surmises concerning these possible hypothenar elements,
that are at once apparent, and may be here mentioned, rather as
suggestions to stimulate comparison than as definite assertions.

1. All the elements here considered are loops, and, with the
exception of one of the loops of Fig. 20, all open toward the medial,
or tibial, margin of the sole. It is thus difficult to explain them
as the two disassociated ends of a long S-shaped figure.

2. In the case of Fig. 16, with two large loops facing the same
way, the more distal is probably the loop commonly found on the
outer margin, near the ball of the foot, while the other is readily
comparable with the large loop in the hollow of the foot, as given
in the cases of Schlaginhaufen and Mme. Loth (Figs. 18 and 19).
If, however, we compare Fig. 16 with Fig. 17, we see this same
proximal loop gradually reduced to a small but evident figure,
on the way towards extinction. This gives us a series, Figs. 19,
18, 16 and 17, in which a loop begins with occupying almost the
entire sole, and ends as a rudiment. One recalls here Schlagin-
haufen's interesting series, in part theoretical, in which he derives
such a case as that of Mme. Loth, from a moderate-sized loop,
found in lower Primates, which crosses the heel region obliquely,
and opens to the medial side (Schlaginhaufen, loc. cit., p. 121).

3. Concerning the calcar pattern, about which so little is
known, a pattern which has been observed here and there, but
has as yet no morphological interpretation, I made recently more
careful studies upon No. 87 of my collection, a subject who has a
good calcar pattern upon each heel. The result of this, in the
case of the right foot, where there is also a good thenar pattern,
is given in Fig. 21 above. In the figure the whole record of
the friction-skin of the foot is spread out as reduced to a plane,
and the tread area is indicated by lines. Unfortunately there
are no indications of the loops here considered hypothenar, so
tha,t a comparison of either thenar or calcar with these is not
possible; on the other hand there is revealed a possible association of
calcar with thenar, suggesting that the two form the two loops of an
S-shaped pattern. This close association of thenar and calcar
patterns is not new, being shown by Miss Whipple in two speci-

PALM AND SOLE STUDIES. I 71

mens of Cebus1 and in Inuus? but here the fields occupied by the
two patterns seem more distinct. This author treats the calcar
pattern as a secondary one, formed probably through the back-
ward extension of the heel, to assist in covering the new area, and
sees a proof of its recent nature from the fact that there is no trace
of such a pattern in the corresponding position in Lagothrix,
where "the calcar region is still covered by epidermic elements
not yet fused into ridges."

An element to be considered in this connection, but one which
adds to the confusion rather than assists in the elucidation, is the
"Fersen-sinus" of Schlaginhaufen, which he figures on the heel
of various Primates, the core of which usually forms a loop
opening to the medial side, as in man. This he shows most
typically in Macacus and Hylobates, but it appears also in Simla,
Gorilla, and Anthropopithecus. As this author does not emphasize
the homologies of the typical patterns as located upon the
original pads, but compares rather the triradii and lines pro-
ceeding from them, one can hardly follow the patterns through
his numerous figures.

IV. THE REDUCTION OF LINE (7.

In 1910 Edward Loth, assisted by Mme. Loth (Jadwiga Nie-
mirycz-Lothowa), published the results of the examination of a
large collection of the palm and sole prints of Russian Poles from
the vicinity of Warsaw. In these they find a number of instances
in which line C, with its triradius, is entirely wanting, and others
in which the line in question is very short, and terminates, after
a perfectly straight course, in a loop (cf. Figs. 33 and 34 below).
These conditions, which are obviously closely related, both desig-
nates in a main line formula by the letter x.

With regard to previous recognition of either of these two con-
ditions he cites Miss Whipple in her paper of 1904, and states
very generously that he makes no claim to priority, yet thinks
that he is the first to indicate it in formulae.

As this condition has been so long known to me, practically
from the beginning of my investigations, I felt sure that I had
described it in detail somewhere, but, to my own chagrin, I can

1 Loc. cit., p. 334.
5 P. 307.

172 HARRIS HAWTHORNE WILDER.

find no direct reference to it in any of my writings. I have
indicated this condition in formulae, however, by the digit 8,
the designation for the radius of origin of the missing line, in-
tending to signify by this that it goes nowhere, yet I am quite
ready to acknowledge that the meaning of this is not clear, and
that Loth's use of the letter x is much clearer.

Since the condition itself is an interesting one, I have recently
taken some pains to estimate its frequency. In the hands of 145
white persons (mostly Smith College students) a complete loss
of line C, with its triradius, was found in both hands in four cases;
in the right alone in five; and in the left alone in eight. That is,
out of 145 individuals, no less than 17 of them showed a complete
loss of the C line in one or both hands; or, put another way, out of
290 separate hands 21 were thus marked. In addition to these,
nine more individuals possessed in one hand a very short and
straight C line, ending in a loop, and as these were none of them
duplicate individuals writh any of the above or with each other,
this gives a total of 26 individuals out of the 145 in which the
term x ( = 8) occurs in one or both of the hand formulae as a
designation of the condition of line C; that is, nearly 18 per cent.

Comparing the palms of races other than white I found eight
cases in 42 palms of the Maya Indians of Yucatan, or 19 per cent,
much as in the whites, while in 118 palms of Liberian soldiers
this condition (including both forms of it) occured but 10 times,
1 1. 8 per cent.

To summarize these results: the condition of line C, in which
it is either wholly absent, together with its triradius, or very
short and ends in a loop, is a fairly common one, apparently in all
races. In the negro palm, however, the marked tendency of all
the lines to run diagonally across the palm, from the bases of the
fingers to a more proximal position on the ulnar margin, which
results so commonly in the formula 7 -5 -5 -5, considerably lessens
the percentage of occurrence of this condition.

In expressing this condition in a formula I would suggest the
general adoption of Loth's abbreviation, x, when the line and
triradius are both wanting, and that of 8, my usage hitherto, to
designate a very short C line, ending in a loop.

(To be continued.)

HEREDITY AND ORGANIC SYMMETRY IN
ARMADILLO QUADRUPLETS.

II. MODE OF INHERITANCE OF DOUBLE SCUTES AND A DIS-
CUSSION OF ORGANIC SYMMETRY.

H. H. NEWMAN.

A. GENERAL STATEMENT.

This paper is in continuation of a study published under the
same general title in July, 1915. In that paper the inheritance
and distribution among quadruplets of more or less extensive
band anomalies were dealt with. All anomalies involving the
presence of two or more consecutive double scutes either in
mother or in one or more offspring of a polyembryonic set were
considered as band anomalies. There remained numerous cases
of inherited anomalies so minute as to involve only isolated double
scutes in parent and in offspring. A detailed study of the in-
heritance and of the symmetrical or asymmetrical distribution
of these double scutes among the fetuses of quadruplet sets forms
the subject matter of the present contribution.

In order to have a compact and truly homogeneous group to
work with, I have decided to confine my study to collections C
and K, omitting several small collections, the data of which are
not so complete. In the C and K collection there are 140 sets of
quadruplets which are sufficiently advanced to show every detail
of the scute pattern. The shells of the mothers of all these sets
are preserved and have been carefully scutinized for anomalies.
Such shells as were badly worn or damaged had to be excluded
from the present study together with the associated offspring.
A study of 140 sets should reveal the conditions typical for the
species, since we have a collection of 700 individuals. Of the
140 sets of quadruplets 73 are female and 67 male. This dis-
crepancy between the sexes is due to two circumstances, first
that three sets of fetuses, in which the sex was obviously male,
were too small to count, the mothers in all cases being normal,

173

174 H- H- NEWMAN.

and second that two sets of offspring, all normal and males,
were discarded because the shell of the mother in each case was
badly damaged. If these five sets of male quadruplets be added
the count would be practically even so far as sex goes, 73 females
and 72 males. Other collections, moreover, have showTn equal
numbers of male and female sets of quadruplets.

I. Three Categories of Offspring.

A survey of the 140 sets of quadruplets reveals that there are
three well-defined classes:

(a) Those in which both mother and offspring have anomalies.
Of these there are 56 sets, 29 female and 27 male.

(6) Those in which the mother is normal but the offspring
have anomalies. Of these there are 41 sets, of which 22 are female
and 19 male.

(c) Those in which both mother and offspring are normal. Of
these there are 43 sets, of which 22 are female and 21 male.

Unless the character in question is strongly inherited as a
dominant we would expect to find a fourth class composed of sets
in which the mother has an anomaly but all offspring are normal.
Only one such case occurs, and this is a doubtful one in which
the anomaly of the mother is a double scute that may be due to
the fusion of two scutes after a local injury. This doubtful
case has been included in class (c). Another case that I at first
thought was in this fourth class was found on examination to
belong to class (a), for one of the fetuses had a double scute in
the last scapular band which had been overlooked in the counts
because we were dealing only with the regular bands. The
facts of this case will be clear from an examination of Set C 65
(bottom of p. 187).

The mode of inheritance of these anomalies does not appear
to be typically Mendelian, for, if the character is a dominant,
with the normal condition the recessive, we would expect a
considerable number of anomalous individuals to be heterozygous
for the character and to produce equal numbers of germ cells with
the anomaly factor and without it; so that on the basis of chance
mating we should often get normal offspring from the mating of
two heterozygous anomalous parents or from one heterozygous

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS.

anomalous parent pairing with a normal parent. That we do not
find this condition means that we have evidently a non-Mendelian
result due probably to factor segregation among the quadruplets,
a process that must evidently take place in order to produce
sets in which some individuals are anomalous and some are
normal. This segregation must also affect the germ cells, so
that an individual that has an anomaly has also germ cells of
only one kind and these homozygous for the anomaly factor.
The same kind of mechanism that quite evidently segregates the
somatic anomaly factor must also be conceived of as responsible
for a segregation of the germinal anomaly factor. The results
do not appear to bear any other interpretation. Further dis-
cussion of this point follows the presentation of data.

Since these anomalies are invariably inherited when present
in the mother we are driven to the conclusion that all these sets
in class (6), which have anomalies, but whose mothers are normal,
must have inherited their anomalies from the father. Further-
more, it is equally obvious that a considerable proportion of
those sets in class (a) must have anomalous fathers as well as
anomalous mothers. Thus class (a) is heterogeneous in that
it is composed of sets inheriting from both parents, on the one
hand, and from the mother alone, on the other.

A calculation may readily be made of the relative sizes of the
various classes of offspring that should result if chance mating
occurs between anomalous and normal individuals. It will be
noted that 56 of the 140 mothers (or 40 per cent.) have anomalies.
Again 115 individual female offspring, or 39+ per cent, and
ill individual male offspring or 41+ per cent., have anomalies.
So we may be safe then in assuming that 40 per cent, of the in-
dividuals of the species have anomalies and 60 per cent, have
none. If 140 chance matings on this basis occur we should expect
the following classes:

(a) 1 6 per cent, or 22.4 with both parents anomalous.
(a'} 24 per cent, or 33.6 with mother only anomalous.

(b) 24 per cent, or 33.6 with father only anomalous.

(c) 36 per cent, or 50.4 with neither parent anomalous.
The sum of classes (a) and (a') is 56, which is exactly the number
of sets in the observed class (a).

H. H. NEWMAN.

The observed class (b), 41 sets, is several sets in excess of ex-
pectation, but not enough to seriously effect the theory of breed-
ing proposed. The observed class (c), 43 sets, is several sets
too low, but it is very probable that the five discarded male sets
referred to above belong to this category. On the whole, then,
the observed and theoretical proportions of the categories of
offspring are as close as could be expected in 140 matings, and
this fact supports the general statement as to the modes of
inheritance of these anomalies stated above.

2. The Genetic Relation Between Scute and Band Anomalies.

There is a very intimate genetic relationship between band
and scute anomalies, as was brought out in the previous paper
(Newman, '15). Sometimes a band anomaly is inherited as such
in some offspring of a set, and as a scute anomaly in others; and
the localization in all cases is so exact that there can be no doubt
about the genetic equivalence of the two anomaly phases.

In Figs. 1-8 is arranged a series of drawings of scute anomalies
leading up to band anomalies. The transition is more readily
made out from the conditions in the bony plates underlying
the scutes. Fig. la represents the appearance from the exterior
where only the scute is visible; Fig. ib shows the underlying plate
condition. Similarly with Figs. 2a and 2b, etc. In Fig. 6, a
and b, we see that the more fundamental anomaly is a transverse
splitting of the bony plate involving only one unit. Fig. 7,
a and b, constitutes an incipient or minimal double band, while
Fig. 8, a and b, represents a band doubling of moderate extent
involving six scutes. Such band doublings may involve more than
half of a band or there may be extensive doubled regions separated
by single or undoubted regions.

Since the bony plates are fully laid down only during post-
embryonic life, it is impossible to study these structures in the
fetuses that are taken from the uteri of the mothers; but the scute
condition, for one who has made a study of the subject, serves as a
very definite index of the condition of the bony plate; and the
scutes are well defined long before birth. For example, whenever
a scute is found with a deep notch above dividing the scute into
two nearly equal cusps (as in Fig. 30), one can be sure that the

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS.

177

bony condition will be like that in 36. Similarly when the notch
is shallow as in 50, we might expect the bony condition to be as
in 5&.

It is an open question as to which doubling is more fundamen-
tal, that of the dermal plate or that of the horny scute. Onto-
getically the scute is the older structure, but I am inclined to
look upon the two structures as phylogenetically coeval, for we
apparently have in the armadillo a case of the persistence in a

16 20 36 45 5

Q/l f\QOf\ f\ QOQ 00 0 00

6b

FIGS. 1-8.

mammal of the ancestral reptilian epidermal scale with its dermal
bony core. The hair group that centers about the scute is
presumably of more recent origin.

I look upon scute and plate doubling as cases of local budding
of normally single primordia, not unlike the division of scutes
and plates in the carapace of modern tortoises, a phenomenon
that has received considerable attention (Newman, '05, and
Coker, '10). The physiological basis of this budding may be a
localized lowering of the rate of growth during scute develop-
ment. If we had a solution of the problem of budding or division

178 H. H. NEWMAN.

in this instance we would have a solution of the general problem
of division that appears everywhere in the field of biology.

Since the inheritance and distribution of double bands was
dealt with in detail in the earlier paper of this series it will not
be necessary to refer further to these results. It may be of inter-
est, however, to compare the facts as to symmetry phenomena in
the two classes of anomaly. Band anomalies are much less
numerous than anomalies of single scutes, but they exhibit the
same mirror-image phenomena and symmetry reversals that are
so frequently seen for double scutes. For details of these phe-
nomena the reader is referred to the published account (Newman,

'15).

The present study deals solely with scute anomalies, double

or split scutes of the various types shown in Figs. I to 6.

3. Frequency and Distribution of Scute Anomalies.

In the present collections (C and K) there are 700 individuals
composed of 140 sets of quadruplets and their mothers. These
should constitute for statistical purposes a reasonably adequate
sample of the species.

In the earlier paper of this series (Newman, '15) 23 sets that
showed band anomalies in one or more members of a set were
dealt with. Many of the individuals (40 in all) had scute
anomalies, some with and some without accompanying band
anomalies. The remaining 117 sets (585 individuals) treated in
this paper contain 199 individuals with double scutes. Whether
we figure on the basis of the entire collection or upon the 117
sets dealt with in the present paper we find that 34+ per cent,
of all individuals have scute anomalies. Add to this 32 (4 +
per cent.) individuals that have only band anomalies and we find
that about 39 per cent, of all individuals have either band or
scute anomalies or both, which is very close to the 40 per cent,
arrived at earlier in this paper. Scute anomalies are, however,
about eight times as numerous as band anomalies and furnish
better material for statistical study.

4. Sex Distribution of Anomalies.

Exclusive of the mothers, exactly 40 per cent, of which have
anomalies, 22 females and 12 males have band anomalies, and

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. 179

87 females and 100 males have scute anomalies. There appears
therefore to be a somewhat pronounced tendency toward band
anomalies as against scute anomalies in females and the reverse
in males. In a collection of this size, however, these differences
may not be significant.

5. Distribution of Scute Anomalies as to Bands.

In the earlier paper of this series it was shown that the band
anomalies are largely confined to the first two bands: 86 per cent,
in band I, 8 per cent, in band 2, and the rest scattered; none oc-
curring in bands 5 or 6, which are in the middle of the banded
region, or in band 9. The reason given for this state of affairs
is that band doubling is normal for the scapular and pelvic regions
and that the bands nearest the scapular and pelvic shields are
naturally more like these regions than are those farther removed
from them.

A somewhat similar condition, though less pronounced, is
brought out by a census of double scutes according to bands:

Band I has double scutes 67 times.
Band 2 " 40 "

Band 3 " 40

Band 4 " 34 "

Band 5 " 9 "

Band 6 " 27

Band 7 " 34

Band 8 " 29 "

Band 9 " 9 "

It is significant that double scutes, like double bands, occur
most frequently in the first three bands and least frequently in
bands 5 and 6 and 9, where band anomalies were entirely absent.
Bands 5 and 6 are farthest from the scapular and pelvic regions
where doubling is the normal condition. Band 9 is really not a
free band at all, but only partially free at the margins; hence it
may be left out of consideration. It is really out of the series of
bands and should be included in the pelvic shield, but, out of
deference to the time honored name "nine-banded armadillo,"
I have treated it as the ninth band.

l8o H. H. NEWMAN.

I interpret the figures for both double bands and double scutes
as I formerly interpreted similar facts brought out by a study of
supernumerary scutes in the carapace of tortoises. These super-
numerary scutes were viewed as atavistic variations or vestiges
of a condition largely outgrown by the species. These anomalies
occur most frequently at the posterior end of the carapace and
progressively less frequently as one goes toward the anterior end,
except that there is a slight increase just at the anterior end.
The hypothesis was ventured that the more frequent the regional
occurrence of an archaic character the more recent has been the
racial suppression of these characters in that region. There
has evidently been an antero-posterior orthogenetic loss of scutes
in the tortoise carapace beginning at or near the anterior end
and ending with the posterior end.

Now in the armadillo I believe that the more generalized and
less regular parts of the carapace, the scapular and pelvic shield,
represent phylogenetically an older condition than does the
banded region. In support of this contention I cite the fact that
the extinct giant armadillo, Glyptodon, had a solid non-banded
carapace. The first banding began in one or two bands at about
the middle of the carapace, where bands 5 and 6 are, and pro-
ceeded posteriorly and anteriorly. Progress anteriorly in the
direction of more bands is probably occurring now, as may be
judged by two facts, first that it is not uncommon to find a part
of the last scute row of the scapular shield separated as a band,
and second that there are not infrequently local fusions between
parts of the first band and the scapular shield. On this hypothesis
band I is the newest band phylogenetically and shows more re-
semblance to the scapular shield than do other bands, both in
band doubling and in scute doubling. Band 2 is much less
affected with these anomalies than is band I, but more so than
any other band. Moreover, bands 5 and 6, which are phylo-
genetically the oldest bands, show the fewest recurrences of
irregular conditions. Scute doublings are viewed as incipient
or vestigial band doublings and we find many more of these
vestigial anomalies in the banded region than fully developed
ones.

In the tortoises the orthogenetic progress in carapace simpli-

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. l8l

fication has proceeded antero-posteriorly and in this single di-
rection. In the armadillo banding has proceeded both ways
from the middle. Various armadillo species today exhibit all
the way from one to twelve bands, a fact that supports the idea
that in our species there has been a gradual increase in the
number of bands.

B. PRESENTATION OF DATA CONCERNING THE INHERITANCE
AND DISTRIBUTION AMONG QUADRUPLETS OF DOUBLE

SCUTES.

The following tabulation gives all of the facts concerning double
scutes, except those brought out in connection with the earlier
paper on double bands (Newman, '15). I know of no method for
presentation of data of this kind that is at once so intelligible
and so compact as the pictorial diagram used in this and the
former paper. When one has read the key to the table there
should be no difficulty in understanding the data thus presented.
The facts are first presented for those sets in which both mothers
and offspring have double scutes, covering pages 182 to 188, and
there follows as an appendix to this paper a table for those sets
in which the mother was without double scutes, but in which one
or more of the offspring have inherited the double scutes from
the unknown father. See pp. 204 to 207.

KEY TO TABULATION OF DOUBLE SCUTES.

Each solid block of bands shows the anomalies in position as
though the bands were straightened and placed with the right-
hand end to the right and left-hand to the left. The number of
the set with the sex of the quadruplets is indicated above and at
the left of each block. Bands of the mother that show anomalies
are marked M; those of the offspring that show anomalies are
marked according to previous custom I., II., III., IV. I. and II.
are a natural pair, II. being a primary bud individual and I. its
twin secondary bud derivative. Similarly III. and IV. are a
natural pair, IV. primary and III. secondary. If the anomaly
occurs in different bands of the same individual a bracket includes
the Roman numerals indicating the individual fetus concerned.
The various types of double scute are drawn in upon the band in as

1 82 H. H. NEWMAN.

nearly as possible the position in which they occupy, but they
take up more space laterally than is actually the case. In addi-
tion to positional location the number of each scute, counting
from either of the margins or from the middle, is given by an
arabic numeral upon the pictured scute, the direction from which
the count has proceeded being indicated by an arrow unless this
is too obvious to need indication. The arithmetical middle of a
band is indicated by a dotted vertical line, except in a few cases
where the right and left halves of two different bands of a single
individual are placed in the same line, in which case a solid line
is used to separate these heterogeneous half bands. This change
of method may be a little confusing at first, but it was introduced
to save space. The band in which the anomaly occurs is num-
bered with an arabic numeral followed by a colon and a figure
indicating the total number of scutes in that band. For instance
6 : 63 means band 6 which has a total of 63 scutes, counting
the double scute as two scutes. L.S. refers to the last scapular
row of scutes, next to the bands. This mode of tabulation admits
of a very condensed record of a large amount of data.

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. 183

C.73

M

X

TC

Et

.Gl

3: 6-8

<r

: €1

•3-.GZ

M

:mr

7:67

C.7/

M

8:<S* /SP^

HE

______

C9^cf

M

5 4--<?3

JT

•^-:^4 ^'°~^

(M

r.C

/

s^-

-^

!M

<-

9

3

:e/

JE

3:(T/

V

z

K6I

M

2 7:64-

8 : 66"

w|->

FIG. 9.

1 84

H. H. NEWMAN.

C.76

M

:r

f*
I*

rnzj

A-'- SO

/ :

Z-60

v

r

U

r.63

: SO

tt

4 :

/j

G

x^

6

/ -

Z • 64

6 : G€

TIL

7 :

8

IF

8 67

M

£

4

t-

;z

3: 6" 3

<r

;5" 2 '• e6~

<-

«

2 :65~

<e

/o

7 * 6^4

3

. (

r4

/o — >•

FIG. 10.

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. 185

n

V

3

8-

<r

8

9:<

IZ

^6

3 --

JL
TL

in

I .

8

I--G4

8'-<

v

IZ

8

il

K. 70 (?

(•x

3T
HE

6

<:

V

6

2:6-8

^

6

3:5-8

3

4:6-8

1-57 "g^

3

£:So

7:6^9 ^->

<-

&

/ -S7

7'.^ 3^

*-

e

2 : 5"7

S'-ffO iQ— >

e. so d"

n

H

7:^3

/ : Go

3E

/:<?/

: 6*0

/•. eo

2-

K.Z

/vj

i ^

X

<- ^7 / - 6"8

ir

•V

l 7:6-0

IE:

&'• G 1 A -?•
\ "

FIG. ii.

1 86

H. H. NEWMAN.

M

n

JL

U

\ 7:64

<-

/3 g-.ee <-

^ g:6€

*riQ

&-G4- <-,2

8--C4

K.I7C?

M
TL

^

g. 3:e4 G--G4

iJ^

3, X* Jt"
• O 4>

/o-^

<

-W 7:^4

^

e--fi3

M

/ -. Gtr

\s
8

->

X

y

(ft3 •$:£/

jn

i

' & I-.66

IE

l:G5-

v
9

-=f

K.IO(f

r

(M

OK

3:

6

-3

v^1
5

->

«-

\x

9

<-

|5

4.5

9

<

V
/2

3:<

r/

C.4

7:6-6

£•>

3-G3

1?

^

7:<T7 /^

-8

card71

V

J

3:6*0

8:6-4

^i

K

M

! ^-'G1/ &^3->

6:G3

10

-^ 1

t

/O

<?:€T3

FIG. 12.

M

i*

IE

IE

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. l8/

K.6"g

V

6 :<o3

<

V
10

8 '-63

8-67

1 : G3

9 -?

• 60

1 --61

-13

M

HL
JZ

2,: 64

/4~*

•v?

/ : C2.

75- -*

M

j: 62- 4

i 1

X

3

7:63 9 '.ei- gX->

'TL

3

/:£4

V

HE

<

- ,5- 3 ; ^-Q

^

^/
4- / : 6/

4:

64 /3

— -?

Z-.G*.

2

2]

8-

ff1

1

/W 3

2:

G

5

4

€/

^v1

M

/

8

- 3 ^-7
/C * ~* /

or

8 : (J61 ^g — ^

<

:<

s-^^

^

v

1 ; (3 ^

IE

2

L.5. Go

FIG. 13.

H. H. NEWMAN.

K4-59

M

<-

/I 3 '65'

X

3 -'66

^S

->

IE

2

9 67

73.

8--70

V
9

7

-*

*

8

/:64'

n

|

3 : 6"8 6 : 5"8

Z

«

4

3.-C/

I

10 -*

c

272

M

1?

Z-.GZ

^

5" £T/ 9
-of X.

nr

^5 3:5-9

3E

z

Z:^

Z-- 6Z.

JBT
IF

/-

Z

I- 6Z

K.S79-

n

2-6-7

/3

K-G89

r

(M

8:62,

V

(
1

i

V

/4

8^

d:6-8

^

-> i

K79

M

I
I

#r 7=« •;

V

6"

/:fi4" *~

\y
33

-> y : 6 S~

i

^r-
1

s»^
/5

-^ 4--tf/

FIG. 14.

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. 189

M

W*

J

f*CV\ ^/4~

I It I ^ " ^^

JE

- __ AQf >/o
fa •(•><. / \/ 1

k)U

C.62 9

M

Is1

/:63

jr

^

7

/ '. e1^

M

/ : <?£

V

5

JI

4 : Gl

V
//

^

C8I

M
I

«-

/£

7 : G3 !

! 8 • 6-4

'v

3

M

3:^

9

-^

I

2,: 60

3

£

OE

V

2. : <?/ .Z

/Tfi4

V

-^

<-

s/1

/O

4-6-3

/ (3-2

\/

4-

->

^

W

4 .6"^

K.3/

K.77?

M

^

V

9

4-6-3 •

j 6:<?6-

s/1

/3

-?•

M

«-

^

8 :6"<

D-

HL

3

:^

V

\/

3

FIG. 15.

H. H. NEWMAN.

I. The Difficulties in the Way of Accurately Analyzing the Data.

It would be interesting to analyze the data given in the tabu-
lation above, but before an intelligent analysis can be made some
of the chief sources of error should be pointed out.

1. Perhaps the most obvious hindrance to accurate analysis of
the data appears in the fact that we cannot be certain, if more than
one anomaly appears, that the inheritance is solely from the
mother, for there must obviously be a fairly large number of sets
of quadruplets in which both parents had anomalies. Some sets
show clearly this biparental influence, but in many cases it is
impossible to decide whether the condition is inherited from the
mother alone or is of biparental origin.

2. Another source of error in presenting a pictorial tabulation
of the localization of these minute elements is the result of a
somewhat arbitrary method of determining the middle of a band,
for although I have chosen to locate the middle by dividing the
scutes into equal numbers on right and left sides, I feel sure that
the arithmetical middle does not always correspond to the mor-
phological middle. There are definite indications in many cases
that there are more scutes on one side of the median dorsal line
than on the other, in which instances the arithmetical middle is
obviously not median. This condition is probably due to the
fact that when two parents with decidedly different scute
counts mate we may get the maternal number in one half band
and the paternal in the other. The middle point therefore is not
a very favorable landmark for scute localization but it seems
necessary to use it in order to indicate unilateral symmetry re-
versals. Localization from the lateral margins of the bands is
much safer than localization from the middle and it is quite obvious
that laterally placed scutes show much closer correspondence in
position than do those near the middle.

3. Another factor that shows the inadequacy of the numerical
localization method inheres in the fact that different bands vary
greatly in number of scutes, ranging from 57 to 70 in the present
collection. Now a scute that is pictured as being 5 scutes from
the margin in a band of 57 scutes is morphologically farther from
the margin than a scute located 5 places from the margin of a
band of 70 scutes. So when we are comparing the locality of the

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. IQI

scutes of one individual with another we should bear in mind the
numbers of scutes in the various bands dealt with.

2. Classification of Types of Inheritance and Resemblance among

Quadruplets.

Analysis of the individual cases reveals several well-marked
categories :

(A) Strikingly close resemblances between mother and offspring,

or among the different fetuses of a set.

(B) Cases of symmetry reversal between parent and offspring.

(C) Cases of mirror-imaging among the fetuses of a set.

(D) Cases of half-band reversals of symmetry between mothers

and offspring.

(£) Cases of half-band reversals between different fetuses of a
set.

(F) Cases of half-band reversals in the same individual.

(G) Longitudinal reduplications down the long axis.
(If) Transverse reduplications in a single band.

Each of the categories will herewith receive separate attention.

A. Strikingly Close Resemblances between Mother and Offspring,
or Between the Different Fetuses of a Set. — Cases are here dealt
with in which the form of the anomaly, its localization and sym-
metrical relations are the same, or very nearly so, in two or more
members of a set, the mother included. The most striking cases
are considered first.

Set K 27 (p. 189) shows the most remarkable case of exact
resemblance. The mother has two double scutes in two different
bands, and one of the offspring (fetus i) has the same double
scutes in the same positions of the same two bands as in the
mother. Here we have a little difficulty due to locating the
morphological middle of the first band. If we take the arith-
metical middle, as has been done, the double scute of band i of
the mother is five scutes from the middle, but that of fetus I. is
four scutes from the middle. If we count from the margin,
however, the scute is number 28 from the left in both cases, for
there are two more scutes to the band in the mother than in the
offspring.

I Q2 H. H. NEWMAN.

Set C 26 (p. 1 88). The mother has a small scute (like Fig. 20)
in band 2, two places from the left hand margin. Fetuses II.
and IV. (one of each pair) have identical anomalies but shifted
to band I.

Set K 57 (p. 1 88). The mother has scute 2 from the middle of
band 2 double, and fetus III. has an identical anomaly in band I.

Set C 30 (p. 185). The mother has a double scute in band I
situated two places from the right-hand margin. Fetuses III.
and IV. have in band 2 an anomalous scute (like Fig. 20} two
places from the right-hand margin. These two elements have
bilaterally symmetrical equivalents on the left hand side of
band I.

Set K 70 (p. 185). The mother has in band 2 a double scute
6 places to the right of the middle. Fetus IV. has an exactly
equivalent anomaly.

Set C 76 (p. 184). Fetus II. has in band 4 a double scute 12
places to the left of the middle. Fetus IV. has an exactly equiva-
lent element.

B. Cases of Symmetry Reversal between Mother and Offspring.—
All cases are to be included here in which an anomaly of the
mother is repeated in the offspring but on the opposite half of the
carapace.

Set C 76 (p. 184). Mother has in band 8 a double scute three
places from the left margin. Fetus III. has in band 8 a double
scute 3 places from the right margin.

Set C 30 (p. 185). Mother has a double scute 2 places from
the right margin of band I. Fetuses II., III. and IV. each have
an anomalous element (like Fig. 20} two places from the left
margin of band i .

Set C 24 (p. 187). Mother has in band I a double scute 4
places from the right margin. Fetus IV. has in band I a similar
element 4 places from the left margin. Fetus II. has in band I
a similar element 3 places from the left, and fetus III. has in
band 3 a similar element 5 places from the left. We have here
several degrees of exactness in the mirror-imaging of the in-
herited anomaly.

C. Cases of Mirror-Imaging among the Fetuses of a Set. — Set
K 70 (p. 185). Fetus I. has a double scute 3 places from the

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. 193

left margin of band 6, and fetus III. has a similar element 3 places
from the right of band 7. Note that these fetuses are not pairs
but face each other from opposite sides of the vesicle. Both are
secondary bud individuals.

Set C 30 (p. 185). Fetuses I. and II., a pair, have striking
mirror-image resemblance. Fetuses III. and IV., the other pair,
have the same anomalous element bilaterally.

Set C 76 (p. 184). Fetus III. has a double scute 12 places to
the right of the middle of band 2. Fetuses II. and IV. each have
identical elements on the left side.

Set K 18 (p. 186). Fetus I. has a double scute 10 places to the
left of the middle of band 6. Fetus III., which faces I. across
the vesicle, has a similar element 10 places to the right of the
middle of band 6.

D. Cases of Half -Band Reversals of Symmetry between Mother
and Offspring. — In this group are included cases in which the
parent and offspring have the anomaly confined to the same,
right or left, half of the carapace, but the symmetry of that half
of one is the mirror-image of the same half of the other.

Set K 70 (p. 185). Mother has a double scute 6 places to the
right of the middle of band 2. Fetus I. has a double scute 6
places from the right hand margin of band I. If this band had
been arranged so that the right hand half were simply shifted
over to the left side it would have been an ordinary case of
mirror-imaging.

Set K 25 (p. 185). Mother has a double scute 6 places to the
right of the middle of band 2. Fetus I. has a similar element 7
places from the left hand margin of band I . This is a case of a
double reversal. Turn either of these half bands end for end and
an ordinary mirror-image case would result.

Set C 4 (p. 1 86). Mother has a double scute 5 places from the
right margin of band 7. Fetus II. has a similar element 5 places
to the left of the middle of band 3, which is evidently the reversed
mirror-image of a similar element in band 7 of the same fetus
situated 8 places from the right-hand margin.

E. Cases of Half -Band Reversals between Different Fetuses of a
Set. — These are cases strikingly like the unilateral reversals of
finger-print patterns described by Wilder for duplicate human
twins.

194 H- H' NEWMAN.

Set K 76 (p. 185). Fetus II. has a double scute 12 places from
the right margin of band i. Fetus II. has a similar element 12
places to the right of the middle of band i. Again fetus III. has
a double scute 8 places to the right of middle of band 2, and fetus
IV. has a similar element 9 places from the right margin of band 4.
Here each pair of fetuses mirrors the other as to the right half of
the carapace.

Set K 56 (p. 184). Fetus I. has a double scute 7 places from
the right margin of band 6. Fetus II. has a similar anomaly 6
places to the right of the middle of band I and a second anomaly
9 places from the left margin of band 2, and a third anomaly 9
places to the left of the middle of band 4. Fetus III. has a double
element 8 places from the right margin of band 7. Fetus IV.
has a similar element 4 places from the left margin of band 2 and
another element 9 places from the right margin of band 8. All
of these anomalies are, I believe, directly inherited from the
mother, who has a double scute u places to the left of the middle
of band 4, but there are numerous reversals within the set in-
volving a shifting up and down the long axis. This is one of the
most complex cases in the collection and would bear careful study
as it illustrates most of the symmetry phases seen in armadillo
quadruplets.

F. Cases of Half-Band Reversals in the Same Individual.—
Set K 56 (p. 184). Bands 2 and 4 of fetus II. have a reversed
symmetry of the left half.

Set K 76 (p. 185). Fetus II. has a reversed symmetry in
bands I and 4; Fetus IV. has a reversed symmetry in the two
halves of band 4.

G. Longitudinal Reduplication Down the Long Axis. — It is
fairly common to find an inherited anomaly appearing not
merely once but repeated in two or more bands. Sometimes an
anomaly of an anterior band is balanced by a similarly located
anomaly in a posterior band. The condition is somewhat like
that which appears in transverse rows of double scutes that are
called double bands, but the tendency for an extended condition
of the anomaly is vertical instead of horizontal and is seldom con-
tinuous.

Set K 32 (p. 184). Fetus II. has a double scute in band 2 and
another exactly in a vertical line with it in band 7.

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. 195

Set K 70 (p. 1 85) . Mother has scute 6 to right of middle double
in bands 2 and 3. Fetus I. has scute 3 from the left margin double
in both bands 4 and 6, also in bands I and 7. Fetus IV. has in
bands 2 and 6 the same double scutes as fetus I. has in bands I
and 7, but there is a symmetry reversal in the half bands. There
seems to be a strong tendency for the offspring in this set to
inherit the vertical reduplication of the mother.

H. Transverse Reduplication in a Single Band. — Here are
included cases in which two or more double scutes occur close
together but not contiguous in the same band. Such cases are
probably in the nature of interrupted band anomalies. Examples
may be seen in sets K 18, K 10, K 43, K 56. There are as many
more cases in sets with normal mothers that are shown in the
appendix.

The interpretations offered in the above analysis may be open
to objection in certain cases but there can be no controversy as
to the nature of the great majority of cases. There is strong
evidence of much mirror-imaging between pairs and between
individuals of a pair and there are numerous cases of half-band
reversals quite like those occasional reversals of finger-print
patterns described for duplicate human twins by Wilder, and
which Bateson believes to be evidences of a former system of
symmetry common to the two individuals before the separation
of their primordia. If these rare and vestigial reversals in du-
plicate human twins are to receive the above interpretation, we
should feel no hesitancy about a similar interpretation of the
quite frequent occurrence of symmetry reversals in armadillo
quadruplets.

C. DISCUSSION AND CONCLUSIONS.
i. The Problems of Inheritance and Segregation.

(a) The Exceptional and Unique Character of the Material.—
No other animal that I know of is so beautifully adapted for
the study of specific and individual variability as is the armadillo.
The strikingly diagrammatic arrangement of the integument into
the five armor shields furnishes an unusual opportunity for
accurate enumeration of units and for a study of inter- and
intra-individual correlation. These characters, moreover, are

196 H. H. NEWMAN.

definite in numbers and arrangement long before birth. How
fortunate a circumstance it is that this species, which is unique
in its polyembryonic method of reproduction, should at the same
time be unique in its possibilities for biometric treatment! I
was especially impressed with this fact lately when I tried to get
some basis for a comparison between cattle twins and could
find nothing satisfactory for the purpose. The same would be
true for sheep, cats or any other mammal. Differences in size,
the only measurable differences in these species, are quite unsafe
criteria for the determination of coefficients of correlation, since
they are notoriously affected by differences in nutrition. Al-
though any part of the integument of the armadillo is superior
for the establishment of correlation constants to anything found
in any other mammal, the banded region, on account of its
regularity, definiteness and specific fixity, is preeminently adapted
for the study of inheritance and of the degrees of resemblance and
difference among individuals of a polyembryonic set.

So regular is the banding that it would be impossible to dis-
cover any symmetry in the bilateral arrangement of integumen-
tary structures were it not for the occurrence of occasional band
irregularities (double bands) and of fairly frequent double scutes.
These peculiarities of the integument furnish landmarks for the
study of symmetry relations that are quite uniquely suitable for
such studies and are more readily comparable than are the finger-
print patterns of human duplicate twins that have been used
for a similar type of study.

(&) Irregularities in the Inheritance of Band and Scute Anom-
alies.— That band and scute anomalies are definitely inherited is
proven by the fact that anomalies are always present in offspring
of mothers that show the anomaly. The real problem is to
explain why all four features of a set of quadruplets do not inherit
these anomalies equally and in the same form. Since they come
from a single fertilized egg they have the same genetic consti-
tution and should be identical unless there exist certain agencies
that result in an irregular distribution of the differentiating factor
responsible for anomalies. That these agencies which produce
irregular distribution of anomalies among the individuals of a
set are not environmental is certain, for there appears to be no

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. 1 97

environmental inequality except that of nourishment, and differ-
ences in nourishment so great as to strikingly affect the size of
the individuals do not at all affect the inheritance of anomalies.
This could be readily shown by a comparison of sets that show
marked size inequality with sets that show size identity.

If then the agency at the basis of irregular distribution of
anomaly factors is not environmental, it must be some sort of
internal agency. I have been unable to think of any mechanism
engaged in either regular or irregular distribution of inheritance
factors except the obvious mechanism of mitotic cell division. If
this mechanism worked in .a perfectly accurate and equitable
fashion there would be no opportunity for any irregularities in
the distribution of differentiating factors; but there are irregul-
arities and hence the mechanism must lack accuracy.

If the factor or factors for anomalies are present in the
oosperm, as we are forced to believe, they must have their seat
either in the nucleus or in the cytoplasm. Since the anomalies are
evidently as strongly inherited from the father as from the
mother, it is hardly likely that the locus of the factors is cyto-
plasmic, for the male cell has an insignificant cytoplasmic organ-
ization. In view of these facts we may safely assume that the
locus of the factor is in one or more chromosomes. When all
four fetuses inherit an anomaly, we must assume that the factor
has been equally distributed to the daughter cells of the first
cleavage and probably for several successive cleavages. If,
however, the anomaly is confined to a pair of embryos derived
from one half of the vesicle, it seems likely that the factor was
so distributed as to be totally absent from one of the first two
blastomeres. When only one embryo inherits the anomaly the
unequal distribution may have occurred during the second or
later cleavages.

After a long search for some more satisfactory explanation of
the facts than that afforded by a resort to the mechanism involved
in cleavage, I have been finally driven back upon this explanation
by the facts immediately to be brought out in connection with
the phenomenon of somatic and germinal segregation.

(c) Somatic and Germinal Segregation. — The particular method
of polyembryonic reproduction in the armadillo is definitely

198 H. H. NEWMAN.

known to be a process involving a precocious dichotomous bud-
ding initiated by the primitive ectoderm. The individual fetuses
are produced agamically as bud products and therefore constitute
a clone; so whatever somatic variations occur among the four
fetuses of a set are obviously types of clonal variation.

Bud or clonal variations have for a long time been propagated
vegetatively by cutting, grafts, etc., and are known to hold true
to their somatic characters. The explanation of bud or clonal
variations that has usually been offered involves the introduction
of two causal factors, one internal and the other external. It is
said that a local peculiarity may be induced by local conditions
outside of the bud tissue itself, such as unfavorable or extra
favorable position on the plant; but such an explanation involves
many theoretical difficulties. A more acceptable explanation
involves the assumption that there is some segregative mechanism
that brings about an uneven distribution of inherited factors, so
that one type of factor may be present in one part of the growing
plant and absent in another. Special cases of somatic segregation
are those cases that we have been accustomed to consider under
the term particulate or mosaic inheritance. Such inheritance
conditions are supposed to be due to a segregation in a single
hybrid offspring of the opposed characters of the two parents.
When, for example, a cross is made between a plant with white
and one with red petals, we may get a hybrid possessing flowers
with some white and some red petals. This is said to be a result
of the segregation of parental color factors so that some cells
get certain factors and others do not.

So far as I am aware it has not been shown whether in bud or
clonal variations, germinal segregation goes hand in hand with
somatic segregation. If a green plant that produces a white bud
variation should be found to produce flowers on this bud that
bred true to the white character, we would have a very clear
case of parallel somatic and germinal variation.

Now in the armadillo we have precisely this state of affairs,
as can be brought out by a study of the inheritance of double
bands and double scutes. For, when some of the fetuses of a
set exhibit the anomaly as shown in the mother and others do
not, we have unequivocal somatic segregation; but we have

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. 1 99

something more than this, since those individuals that have the
inherited anomaly in the soma also must have the factor for the
anomaly in the germ plasm, else how can we account for the fact
that whenever a mother has an anomaly it appears in one or
more of her offspring?

This segregation must have occurred at a period prior to the
separation of primary germ layers, in order to account for the
strict parallelism between the germinal and somatic cells, and I
have reason to look back to the earliest cleavage divisions as the
probable time and occasion of the segregation of determinative
factors. This view is arrived at after considering all of the
available facts and I believe is justified. Strangely enough we
seem once more to be driven back to the cleavage stages for an
explanation of the phenomena of twinning, but there is a vast
difference between merely assuming the blastotomy origin of
twins and accepting the cleavage mechanism as the most prob-
able apparatus responsible for parallel germinal and somatic
segregation of inherited factors.

Moreover segregation of inherited anomalies among individuals
of polyembryonic sets is one problem, while the symmetric or
asymmetric localization of these factors in the soma of the in-
dividual is another — the problem of organic symmetry.

2. The Peculiar Symmetry Relations Among Quadruplets.

(a) Normal and Reversed Symmetry in the Occurrence of An-
omalies and their Significance. — If no complicating factors entered
into the matter of the inheritance of double bands or double
scutes, we would expect to find a double scute of the left margin
of a given band in the mother inherited as a similarly located
double scute by all four offspring. We have attempted to explain
the failure of the anomaly to appear in all of the offspring of a
litter by positing a segregative mechanism operating during cleav-
age. Such a mechanism, however, appears inadequate to explain
the symmetry relations so peculiar to polyembryonic repro-
duction.

The mirror-image type of symmetry is normal for the antimeric
halves of a single individual and exceptions to or breaches of the
mirror-image type of symmetry affords a new problem. Double

2OO H. H. NEWMAN.

monsters are also frequently characterized by mirror-image sym-
metry, due to the fact that the two halves are not physio-
logically isolated and are therefore parts of a single system of
symmetry.

Occasional vestiges of mirror-imaging have been described by
Wilder for the finger-print patterns of duplicate human twins, as
when the pattern of the right index finger of twin A is the mirror-
image of the left index finger of twin B. These symmetry re-
versals in duplicate twins have been emphasized as evidences of
the monovic origin of such individuals and the reversals them-
selves have been interpreted as the last trace of an early sym-
metry formerly common to the two individuals but subsequently
largely outgrown.

Now in the armadillo there are many definite evidences of a
system of symmetry common to all of the quadruplets, upon
which has been superimposed a secondary symmetry system be-
tween twins. This in turn is more or less completely obliterated
later by a tertiary symmetry between the antimeric halves of the
single individuals. In some sets evident traces of the primary
system of symmetry persist as mirror-image relations between
individuals of opposite pairs, but it is more usual to find no trace

•

of the primary system. The secondary mirror-imaging between
pairs is far more commonly in evidence, but is frequently ob-
literated by the tertiary mirror-imaging between antimeric
halves of the same individual, which latter is the prevailing sym-
metry system. An analysis of this intricate interplay of three
grades of symmetry systems is by no means a simple task, but
in the foregoing descriptions of individual sets we have indicated
our interpretation of the various mirror-image and symmetry
reversal phenomena that form the basis of such an analysis. In
general, mirror-imaging between individuals of opposite pairs is
interpreted as an evidence of the early system of symmetry
present in the embryonic vesicle before polyembryonic budding
began. When the primary buds are formed they are the product
of the antimeric halves of the undivided embryo and therefore
should have mirror-image relations, but a partial physiological
isolation of the two buds permits a certain degree of reorganiza-
tion or regulation in the symmetry relations, that tends partially

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. 2OI

to obliterate the original symmetry relations of the undivided
embryo. Similarly, when each primary bud subdivides to form
the secondary buds that are the primordia of the definitive in-
dividuals, a certain residuum of the primary bud symmetry
system is carried over, manifesting itself in mirror-imaging be-
tween the twins derived from the same primary bud. But here
again a certain amount of regulation occurs so that a third system
of symmetry, the bilateral symmetry of each individual, tends to
obliterate former systems of symmetry.

A further evidence of the existence of a primary system of
symmetry relations is seen in the mode of formation of the
secondary buds. Each primary bud gives off to its left a secon-
dary bud as though the whole vesicle had a growth whirl to the
left. It appears to be acting as a single system at the very time
when polyembryonic processes are most active.

(b} The Closer Symmetry of Pairs and its Significance. — One of
the most striking early discoveries resulting from a study of
armadillo quadruplets was that the four fetuses of a set are
paired so that one pair is fixed to the right-hand placental disc and
the other pair to the left-hand placental disc. These discs lie re-
spectively on the right- and left-hand sides of the uterus. The
pairs are more strikingly identical than are the quadruplets as a
whole and they more often exhibit interindividual mirror-imaging
than do the individuals of opposite sides of the vesicle. It at first
seemed probable that this condition was due to the "fact," carried
over from the literature on the origin of human duplicate twins,
that one pair was derived from one and the other from the second
blastomere of the two-cell stage. Now that we know more
about the embryonic history of the armadillo we are less inclined
to relate the symmetry of the vesicle to the symmetry of the
oosperm.

It will be recalled that the early blastodermic vesicle, when it
becomes fixed to the mucosa of the fundus uteri, adheres by means
of a predetermined disc of cells called the "Trager." This disc
is evidently derived from the animal pole of the egg and hence
we have a mechanism for fixing the principal axis of the embryo
and causing it to coincide with that of the mother. The blasto-
dermic vesicle, even if it had a predetermined bilaterality, could

2O2 H. H. NEWMAN.

hardly fix itself so that its dorso-ventral axis would correspond
with that of the uterus, so we are driven to the conclusion that
the embryo at the time of fixation has no bilaterality, but acquires
its bilaterality from that of the mother. The first indication of
bilaterality is seen when the vesicle elongates toward the right
and left oviducal openings of the uterus and when bilateral buds
of ectoderm grow out to the right and left sides. In other words
the bilaterally symmetrical arrangement of the embryonic pri-
mordia is imposed upon the vesicle by the form of the uterus.

These first embryonic buds, the primary buds, stand for two
pairs of twins, each bud redividing to form two secondary buds,
each of the latter going to form a separate embryo. The closer
resemblance between the twins of one side is simply the expected
result of their community of origin, for they are the product of
the longitudinal splitting of a single primary bud of ectoderm.
In other words both have been derived from a rather limited
area of cells which is separated from the primordia of the opposite
pair by a considerable zone of extra-embryonic tissue. On any
theory of unequal distribution during cleavage of differentiating
factors we would expect these twin products of the horizontal
splitting of a common primordium to be closely similar to each
other and to show a relatively large amount of mirror-imaging.

(c) Bateson' s Views Concerning Twinning. — In his book
" Problems of Genetics" Bateson has an interesting and suggestive
chapter on "meristic phenomena" in which he lays great stress
upon the fundamental importance of the problem of cell division
and other division phenomena. He conceives of the phenomena
of bodily symmetry as a direct result of the symmetrical features
of cell division. Twinning in mammals is for him a special case
of the effects of cell division for he adopts the traditional view that
the twins result from a physiological isolation of the blastomeres
of the two-cell stage of cleavage. Bateson accepts Wilder's
position that duplicate twins are in a series with double monsters
of various grades. Double monsters show all degrees of mirror-
imaging, while duplicate twins show only vestiges of this phe-
nomenon. Much emphasis is laid upon the few cases of reversed
finger-print patterns in the duplicate twins described by Wilder.
Any sort of mirror-imaging (symmetry reversal) is an indication

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. 2O3

of a system of symmetry common to both individuals, whether
they are joined or separate.

In closing it may be of interest to state that an examination of a
considerable number of sets of quadruplets reveals no indication
of a reversal of symmetry in the viscera. As in human duplicate
twins symmetry reversals are confined to the integumentary
structures. Why should this be so? Bateson propounds this
inquiry for duplicate twins as follows: "If anyone could show
how it is that neither of a pair has transposition of the viscera
the whole mystery of division would, I expect, be greatly illu-
minated." Now in the armadillo the process of polyembryonic
budding that results in twinning is initiated and carried out in
the ectoderm, while the endoderm becomes involved only passively
and considerably later. Here we probably have the answer to
Bateson's inquiry, for symmetry reversals involve only the tissues
that are primarily concerned in twinning. How much this state-
ment serves to illuminate the mystery of division I am not pre-
pared to say.

I am inclined to believe that duplicate human twins become
physiologically isolated at a considerably earlier period than do
armadillo quadruplets, and my reason for this belief is founded
on the fact that there is so very little mirror-imaging in the former
and so much in the latter. It appears to be a good general rule
that the earlier the separation the more complete is the reorgan-
ization of symmetry relations in the separated individuals and
the less residuum of the original common symmetry. Double
monsters doubtless begin to separate comparatively late in on-
togeny and hence show very pronounced mirror-imaging. Arm-
adillo quadruplets appear to exhibit a condition intermediate
between duplicate twins and double monsters for they have an
interesting combination of the effects of an original common
system of symmetry and of secondary and tertiary systems of
symmetry. Since it seems entirely probable that human dupli-
cate twins are separated at a much earlier period than are arma-
dillo quadruplets, it may not be unreasonable to look for this
separation at some period of cleavage. Or there may be a
division of the inner cell mass into the primordia of two embryos.
The problem of the exact mode of origin of duplicate human twins
is however likely to remain unsolved for a long time to come.

204 H> H' NEWMAN.

APPENDIX.
Sets of Quadruplets Inheriting Double Scutes from Unknown Fathers.

Purely as a matter of record there is included herewith as an
appendix a pictorial tabulation of 31 sets of fetuses from normal
mothers but themselves with anomalies. There is every reason
to believe that these characters have been inherited from the
unknown fathers. There are 15 female sets and 16 male sets.
The same plan of representing the occurrence and location of
these double elements is followed as was used in the earlier tabu-
lation, the key to which is given on page 181.

Since all of these cases are uniparental while many of the cases
discussed earlier are biparental, one would expect a simpler state
of affairs and fewer double scutes per set. This is actually the
case. The interrelations between the different fetuses of a set
are on the average much more clearly denned since there is no
confusing admixture of elements from both parents. These
facts go far to prove that the conclusions reached in the earlier
part of the paper regarding modes of inheritance are well
founded.

Many striking cases are seen of close resemblance between
twins, of mirror-imaging and half-band reversals between twins,
all of which reinforce the conclusions reached on the basis of the
data heretofore discussed. Nothing, however, would be gained
by calling further attention to individual cases. The interested
reader will be able to pick out and classify for himself the various
types of anomaly. The pictorial method of recording these
elements may need a little study on the part of the reader but is
fully justified by the fact that a large amount of valuable data,
the like of which may never again be available, is given in very
compact form.

Correction. — On page 27 of the first installment of this study
(Newman '15) is figured the double band arrangement in set K 2.
This set was classed erroneously among sets the mother of which
had no anomalies. The notebook in which the data was taken
shows the mother of K 2 was misplaced and later found and has
two anomalies. Band 3, which has 64 scutes, has a double scute,
like Fig. 2a, 16 places to the right of the middle. Band 5 (63

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. 2O5

2:63

s

9:65"

6/63

JT
3L

UU

65"

K 39 cf

3-60

T-61

3:56

Kfl/

Y. <35~

HT

r
JL

.65-

K83

1

V

u

-»

<r^

FIG. 16.

206

H. H. NEWMAN.

ll

IDE

JE

IX

JC
3E

r

x

^8

3--JT

.C 96

s

$ 57

4:59

3ZL

F-

'l£

: 60

/ : 6x2,

@^

4: 60

9:65"

IT

HP

/ : 6 /

1:59

s

FIG. 17.

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS.

2O"J

K//9

V

4:60

6 • GZ

.-go

4:6 I

TE

I -61

8-er

3 :

K66

3 '

TE
JJL

3 :

K46

r

br

61

61

2 59

I :6I

C8o

4-: 5*6

3.58

Vx

J

FIG. 18.

208

H. H. NEWMAN.

HE

: 60

I

65-

K59

X

m

Jd

64-

7 :

*•*

3d
ICE

x

7 :

K 66

3 : 6Z

8 •• 64-

C 79 9

IE

3 • 60

4- ' 5-9

C9

4-:6/

6E

FIG. 19.

ORGANIC SYMMETRY IN ARMADILLO QUADRUPLETS. 2OQ

scutes) has a double scute 5 places from the right-hand margin.
These two anomalies mark about the inner and outer limits of
the band doublings seen in the offspring. This set shows almost
perfect mirror-imaging of the whole set of quadruplets. The
two primary bud fetuses II. and IV. mirror each other since both
have the anomaly bilaterally; the two secondary fetuses I. and
III. are also mirror-images, III. having the anomaly on the right
and I. on the left. All of these anomalies occur in band I.

BIBLIOGRAPHY.
Bateson, W.

'13 Problems of Genetics. Yale Univ. Press.
Newman, H. H. and Patterson, J. T.

'09 A Case of Normal Identical Quadruplets in the Nine-banded Armadillo
and its Bearings on the Problems of Identical Twins and of Sex Deter-
mination. BIOL. BULL., Vol. 17, No. 3.

'10 The Development of the Nine-banded Armadillo from the Primitive Streak
Stage to Birth; with Especial Reference to the Question of Specific Polyem-
bryony. Journal Morphology, Vol. 21, No. 3.
'n The Limits of Hereditary Control in Armadillo Quadruplets: A Study of

Blastogenic Variation. Journal Morphology, Vol. 22, No. 4.
Newman, H. H.
'i3a The Natural History of the Nine-banded Armadillo of Texas. American

Nat., Vol. 47, Sept., 1913.

'i3b The Modes of Inheritance of Aggregates of Meristic (Integral) Variates in
the Polyembryonic Offspring of the Nine-banded Armadillo. Journal
Exper. Zool., Vol. 15, No. 2.
'15 Heredity and Organic Symmetry in Armadillo Quadruplets. I. Modes of

Inheritance of Band Anomalies.
Patterson, J. T.

'13 Polyembryonic Development in Tatusia novemcincla. Journal Morphology,

Vol. 24.
Wilder, H. H.

'04 Duplicate Twins and Double Monsters. Amer. Jour. Anat., Vol. 3.

Vol. XXX. March, igi6. No. j

BIOLOGICAL BULLETIN

PALM AND SOLE STUDIES.

HARRIS HAWTHORNE WILDER.

(Continued from page 172.)

V. FRICTION-SKIN CORRESPONDENCE IN PYGOPAGI.

A long series of studies of the palm and sole markings in human
twins, both duplicate and fraternal,1 has established two principles
for twins of the true duplicate type (monochorial, monosexual,
striking facial resemblance, etc.), viz: (i) that the corresponding
palms and soles in the two individuals are strikingly alike in
friction-ridge configuration, even duplicating singular and unusual
features, and (2) that the correspondence in the two sides of the
same individual is far greater than is usually the case, so that the
four hands involved, or the four feet, present so many copies of
practically one picture. To these may be added a third condition,
not absolutely constant, but frequently noted; a reversal of the
pattern of the index figures in the two individuals, affecting either
the two right hands or the two left hands, or occasionally both
sets.

In this connection it is a matter of great importance to ascer-
tain the condition of the friction ridges in "conjoined" twins, that
is, in twins which have never completely separated, but remain
throughout life united by certain common parts (i. e., double
monsters). The numerous specimens of this sort found in
teratological museums have yielded some results, which seem to
accord with the findings in separate duplicate twins, but such
objects are unsatisfactory to study since they are mostly embry-
onic, or newborn, and the ridges in consequence are very soft
and indistinct, while the fluid in which they are preserved renders
a print difficult or impossible.

Recently, however, I have had the opportunity of making

1 Wilder, 1902. 1904, 1911.

211

212 HARRIS HAWTHORNE WILDER.

careful observations of a pair of living conjoined twins, who were
born in May, 1912, and are consequently in their fourth year at
the present writing. These twins, which I studied the first time
a day or two after birth, are typical pygopagi, of the well-known
but rare type represented by Rosa-Josepha Blazek, and Millie-
Christine;1 are united in the sacro-iliac region, but placed some-
what obliquely, so that, instead of looking in directly opposite
directions, they are rotated about 45° towards the same side.

It thus happens that, while lying or sitting they present to the
observer what would be called a front and a back side, considering
the entire double monster as a unit, so that when the observer
is stationed upon the "front" side the two faces are turned 45°

1 Double twins of the pygopagous type are extremely rare. Baudouin, in 1902,
was able to cite but eight authentic cases that have lived, known to the medical
profession, and these span an interval of eight hundred years. These are as follows:

1. The "Biddenden Maids"; Biddenden, Co. Kent, England, b. uoo. d. 1134-

2. Case cited by Lycosthenes, writing in 1665. These were girls, b. in Verona,
Italy, in 1475.

3. The Hungarian Sisters, Helena-Judith; b. near Komorn, Hungary, 1701,
d. 1723-

4. Case cited by Wolff; observed in Russia in 1778, lived but two months.

5. Millie-Christine, negresses; b. in South Carolina, exhibited at Edinburgh in
1856, and later on in the United States. [They died about 1912.]

6. Case of Joly and Peyrat, Jeanne-Marguerite Bombail; b. 1874, at Mazeres,
Dept. Ariege, France. Died 24 hours after birth.

7. Case of Pilat, 1879. Born prematurely and one died during delivery.

8. The Bohemian Sisters, Rosa-Josepha Blazek; b. Jan. 20, 1878, and still
living (1915). One of these is reported to have borne a normal child. These
sisters are well educated, speak several languages with fluency, and have been
extensively exhibited.

To this list must now be added (9) the "Samar Twins" Lucio— Simplicio, two
boys from the island of Samar, in the Philippines, also (10) the subjects of this paper.
The boys are about six years of age at this writing, the girls three and a half. Both
sets were born since the above compilation was made.

That similar cases have occurred among primitive peoples, and in the past, is
evidenced by the occurrence of folk-lore tales, describing such double twins with
some degree of care, and rendering it probable that they were pygopagi. Thus there
are the two sisters in an old Samoan tale already quoted by me in my paper on
"Duplicate Twins, etc."; there is also a Sioux legend of a double woman, cited by
Dorsey in Ann. Rep. Bureau Eth., Vol. n, 1889-1890, p. 480, with a native drawing.

This exhausts the list, so far as known to the writer, excepting fetuses in museum
collections. Any additions to this list would be gladly received. It has been
frequently observed, by the writer and others, that double twins, and probably
separate duplicates, are female in the great majority of cases. This seems well
borne out by the above list.

PALM AND SOLE STUDIES.

213

away from the middle line upon either side, so that the two median
bilateral planes form an angle of 90° with each other. In ac-
cordance with the method employed in former papers, the in-
dividual which is on the left hand of the observer, when he con-

FIG. 26. General pattern for all four hands of the pygopagous twins
This was sketched from the rather unsatisfactory studies made upon the active
hands of infants of two years. All four hands were of this type.

fronts them, is designated as A ; the other, upon his right hand, B.
In this case Mary is A, and Margaret is B.

I studied the palms and soles of these little girls in May, 1914,
within a few days of their second birthday. They were large
and well-developed children, but had not yet learned to walk,

214

HARRIS HAWTHORNE WILDER.

since in attempting it, they would naturally start in nearly
opposite directions. They were sitting in a double high chair
with a platform in front of them, and appeared almost like two
ordinary children, sitting very near together.

I succeeded in taking fairly good prints of their soles, by inking
the feet themselves, and then applying quickly a paper pad held
flat in the hand, and in this manipulation I was aided by the
platform in front of them, which prevented their seeing just what
was going on. Prints of the palms, however, in the case of these
lively and psychologically alert infants, did not seem feasible,
as their hands and ringers were in continual action, and as they
were not old enough to have the matter explained to them. The

FIG. 27. Print of left sole of Margaret (Component A).

hands were therefore studied as we best could, by the aid of a
lens, while the children were being otherwise entertained, and
certain of the more conspicuous details were made out. If the
children live, as is altogether likely, we may hope for actual
prints in three or four years.

Concerning the palms it was fairly definitely made out that

PALM AND SOLE STUDIES.

215

all four are practically alike, and that they conform in both partic-
ulars given above to the condition found normal in twins of the du-
plicate type. In each of the four there is a conspicuous hypothenar
loop opening to the ulnar margin, and in each there is also a large
loop covering the fourth interdigital area, and opening up
between the ring- and little-fingers. Judging from these data
the hand pattern for all four hands cannot be very different from
the sketch here submitted, with lines C and D confluent, and with
line B reaching the margin above (distal to) the hypothenar pattern.
(Fig. 26.) Line A must be left uncertain, as it can perfectly well

FIG. 28. Print of right sole of Margaret (Component A).

follow parallel to B, enter the pattern and become involved in it,
or emerge below it, that is, assume positions 5, 4, or 3. This
will give the formula 8 -6- 5 -5, with a possible variation in the
last term. The correctness of this surmise can be readily tested
as soon as the little girls are old enough to assist in the printing.
The prints of the feet, taken as they were, were necessarily
poor, but enlarged photographs of the results obtained brought
out all the essential details, and it is from these that the outlines
here given were taken. As the photographs were taken by a

2l6

HARRIS HAWTHORNE WILDER.

scientific photographer, and as distortion was avoided, these
outlines may be absolutely trusted. The only improvements
that better prints could show would be in the details of individual
ridges, and other matters which, even in duplicate twins, are
individual, and without significance here.

From a comparison of these four outlines (Figs. 27-30) it will be
seen at once that here is presented the unexpected ; that while three
of the four feet, as well as all four hands (presumably) conform

FIG. 29. Print of left sole of Mary (Component B).

closely to the requirements for typical duplicates, and that in
three of the feet even the curious reduced hallucal pattern, an
extremely rare type, is faithfully reproduced, yet in the fourth,
the right sole of Mary, there is a radical departure from this
type, and a loop of the "A" type is presented instead, i. e., a
loop opening up between the first and second toes. Further-
more, there is also in the same print, but not in the others, a
well-developed fourth interdigital pattern, lying between the
fourth and fifth toes; the presence of two lower triradii, quite
absent in the other three soles, are also to be detected.

Taking up these two differences in order, the hallucal pattern
of the three feet which are alike is of the extremely rare type A C,
in which the two triradii a and c of the typical whorl are absent,

PALM AND SOLE STUDIES.

217

and the ridges consequently escape in both directions through
the points naturally held by a and c, that is, distally between the
great toe and the second, and proximo-laterally towards the
region of the fourth interdigital area and the outer edge. As
shown in the actual prints, or in the tracings from them, here
presented, the b triradius, which is necessary to the full under-

FIG. 30. Print of right sole of Mary (Component B).

standing of the relationship, is not shown, but there is no question
but what it is present, beyond the tread-area, as indicated by the
dotted triradius drawn into each figure.

The difference shown in the aberrant hallucal pattern of Mary's
right foot consists in the presence of the c triradius, thus closing
up all escape for the ridges in that direction, so that consequently
they curve around the core and open with the others between the
great toe and the next. The essential aberrancy of this pattern
consists thus in the presence of a triradius absent in the other
cases.

The other difference in this aberrant foot, that affecting the

21 8 HARRIS HAWTHORNE WILDER.

third interdigital pattern, is also due to the presence of a "lower
triradius" between the third and fourth interdigital areas, thus
rounding the proximal ridges of both into loops. The two proxi-
mal triradii shown in the print may be on the other feet as well,
placed beyond the limit of the actual print, and of this explanation
for the outer one, D, we can be certain, as it is never wanting, but
occasionally beyond the limit of the tread-area. The other,
closing the third area distally, looks like a departure from the
other three soles, thus making the total aberrancy of this foot
due to the presence of three separate triradii, which are wanting
in the other three.

Summarizing the results of this investigation we have, in all
four hands, and in three of the feet, the typical correspondences
found only in true duplicates, also the right and left correspond-
ence usual in such cases. That these two individuals are really
duplicates there is no possibility of doubt since, in addition to the
close facial resemblance, which is typical, and the sex, which is
female, the commoner sex for duplicates, they were monochorial
and had a double, bilobed placenta. As I was fortunate enough
to be on the spot at the time of the birth, I secured the afterbirth,
which had been preserved with great care by the attending phy-
sicians. There was a single chorion, without trace of a separating
partition, and the placenta was bilobed, and nearly as large as
two normal placenta?. The umbilical cord was a single one for
1 1 cm. from the placenta, and proceeded from the margin, at the
point of bilobing, that is, at the notch between the two halves.

This common cord contained the usual two arteries and a single
median vein, and at the forking of the cord to supply the two
umbilici the vein split into two branches, so that one of the arteries
and one branch of the vein continued into each individual cord.
The cord supplying Margaret, now and always the heavier infant,
was ii cm. long from the fork to the ligature; that supplying
Mary was 6. Margaret's cord was also somewhat greater in
caliber. Assuming that the ligatures were made in both cases
at about the same point these latter figures have a meaning,
otherwise they are of little value.

We have thus a conjoined pair of true duplicate twins, mono-
chorial, monosexual, and with striking facial similarity. The

PALM AND SOLE STUDIES. 2IQ

four hands and three of the feet correspond as closely, as has
been found in other cases of separate duplicates; but one of the
feet shows a series of departures from the type, all due to the
presence, or retention, of triradii lost in the other cases. Once
or twice in my other sets of apparently duplicate twins I have
found about the same degree of difference in one of the members,
which has caused me some doubts, and has weakened the argu-
ment in the judgment of others, notably Newman and Patterson,
1911, pp. 856-860.

In this case, however, we have twins who are undoubtedly dupli-
cates, and in which the variation is marked in one member of the
eight involved, the others showing the typical correspondence in
comparing both the two individuals, and the two sides of the
same individual. The case is thus of great service in showing
the limit of hereditary control affecting a case where the in-
dividual differences are expressed as varying degrees of degener-
acy or loss of the features of a complex ancestral configuration.

After the presentation of this new piece of evidence on the
subject of palm and sole correspondences in duplicate twins, it
would seem a fitting time to review briefly the question of the
origin of such cases, especially since a flood of light has been shed
upon the whole subject by the investigations of Newman and
Patterson on the armadillo.

The early stages of normal human individuals during the time
in which we are to expect to find the initial stages of the doubling
process are practically unknown, and it would be an almost
inconceivable chance which would furnish the material for the
study of these stages in human eggs destined to produce twins.
We have long been wont, howrever, to rely in such cases upon the
analogies furnished by related mammals, and to fill in the breaks
in our knowledge by investigation of material taken from other
forms. In this way the normal embryology of single human
embryos has been well made out, and in a way that is generally
accepted, although actual human material, covering important
periods of development, is thus far wanting.

In the armadillo the egg is normally polyembryonic, always
producing monosexual individuals which, by the number and
arrangement of their scales, prove themselves to be duplicates.

22O HARRIS HAWTHORNE WILDER.

In this animal the early stages are now quite fully known, ex-
cepting cleavage, and the first visible signs of multiple indi-
viduals are here seen in a process of budding of the embryonal
anlage. Almost spontaneously comes the impulse to attribute
the origin of human twins and other multiple births to a like
process, as was done by Patterson at the time of his discovery;
writing: "I am therefore bold enough to suggest that the con-
clusions which I have drawn concerning the origin of the embryos
in this mammal may also apply to cases of duplicate twins and
double monsters in the human species." To render this still
more plausible he then refers to certain similarities in the early
embryology of the two animals considered, man and the arma-
dillo, particularly as shown in the Bryce and Teacher embryo,
1908, a similarity which renders this process in man mechanically
possible.

Although this conclusion may seem at first to place the origin
of these multiple individuals at the time the budding process is
first manifest, there is still much evidence to show that this
process rests upon an earlier process wThich conditions it, that is,
that while the visible process of budding first reveals the multi-
plicity of individuals developing from the egg, the true cause is
to be sought further back, or, as Newman and Patterson, as well
as myself, stated at the beginning of our investigation, in some
sort of divison (or we may now say, differentiation) of the early
blastomeres.

It is now clearly evident that all ideas of an actual physical
separation of these blastomeres, a definite blastotomy, does not
take place, yet many things still point to the conclusion that a
similar condition is obtained through some form of differentiation,
and that each of the separate embryonal anlages, be they two
or four or eight or eleven (a possible number in Tatusia hybrida),
is the lineal descendant of a single blastomere, formed during
early cleavage. This was the view of Newman and Patterson
in 1910, who then wrote: "It seems highly probable that the
tissues involved in each of the four quadrants of an embryonic
vesicle do really arise as the lineal descendants of one of the first
four blastomeres." They further believed, as they definitely
expressed farther on, that this differentiation was not a blastot-

PALM AND SOLE STUDIES. 221

omy in the sense of anything involving a separation, and sup-
posed that even if an embryo were found in the 4-cell stage, it
would present to the eye no differences from that of any other
mammalian embryo of four cells. The exact truth of this sup-
position can be known only by a thorough knowledge of the cell-
lineage, from the fertilized egg up to the beginning of the budding
process, and although such a task is well-nigh inconceivable
with our present technique, the results already obtained render
even this possible in the future, perhaps by a culture of eggs
outside of the body of the mother.

In 1913, after his important studies of the early embryology
of the armadillo, Patterson seems to diverge a little from this
view of so early a differentiation, and seems almost to confuse this
with blastotomy, or at least to feel that blastotomy is necessarily
involved in the process, for he says: "The evidence that has been
presented in the first part of this paper [describing the budding
process] makes it certain that polyembryonic development in the
armadillo cannot be explained on the basis of a spontaneous blas-
totomy, in the sense that each embryo is the lineal descendant of a
single blastomere of the four-celled stage, and it causes me to view
with some doubt the conclusions of this same nature that have
been drawn by those who have worked on other polyembryonic
forms." The italics are my own, as I wish to call attention to
what seems to be his interpretation of blastotomy, whereas the
two things, (i) a separation of the blastomeres, and (2) a line of
cell lineage which causes the separate anlages to develop each
as the descendant of one of them, are not at all the same, and have
no necessary connection. The passage is a little obscure, and the
author may not mean to express my interpretation, but it cer-
tainly seems as though he has abandoned any connection between
the numerous embryos and the first four cells, and considers that
the separate individuality commences with the budding process
very much later. Whatever may be this author's opinion on
this subject, which is very likely not as inferred from this sen-
tence, it would seem to me impossible to get such a close corre-
spondence as has been shown in the armadillo, and in human
twins, without starting the separation of identities before there
has been any appreciable differentiation in the material that

222 HARRIS HAWTHORNE WILDER.

controls the development and eventual arrangement of the
parts in question, and I am of the opinion that a carefully estab-
lished cell lineage would substantiate the claims of the two
authors in 1910, quoted above, or their statement of 1909, "In
the case of Dasypus [Tatusia] each embryo probably arises from
one of the blastomeres of the four-celled stage," which does not
open the question of whether these blastomeres were actually
separated from one another or w^ere only the ancestors of the
separate embryonal anlages of a later time. Concerning the
blastotomy theory, that is, an actual separation of the blasto-
meres, I may take this occasion to state that I no longer believe
it, since, among other reasons, comes the very cogent one that
such a separation in man, or in any mammal, would necessarily
produce as many separate and distinct choria, each with its own
attachment to the uterus, a condition which is found in fraternal
births, each from a separate egg, but never in true polyembryony.
In turning to my earlier work, however, I find this theory stated
in unequivocal phrase, as "the total separation of the first two
blastomeres of a single egg" (1904, p. 462) and I take this op-
portunity to recant so far as the "total separation" is concerned,
while still adhering to the first two blastomeres as the probable
ancestors of the two later embryonal buds respectively.

While engaged in noting changes of opinion, I may say further
that the view of a different origin for double monsters with unequal
components, autosite and parasite, which I asserted in my paper of
1904, pp. 462-463,'! rejected utterly later on, 1908, p. 362, and con-
sidered such cases as in origin exactly like other double monsters
(including separate twins), and that the deformed condition had
been brought about by some accident which deprived one of the
components of either its normal amount of nourishment, or pre-
vented a normal growth in some mechanical way. As. my first
view only has been quoted by Newman and Patterson this seems
to be an instance of the very common case in which an error is
published much more widely than its later correction. Such
modifications, or flat rejections, of former views are naturally
incident to all progressive thinking, and in a subject so complex
and so vital as the present investigation are inevitable.

Concerning the causes which produce a double monster, that

PALM AND SOLE STUDIES. 223

is, twins which have some postembryonal parts in common, and
are consequently joined together, the principle of budding, as
seen in the armadillo, furnishes some clues as to the possible
mechanism of the process. Even in normal twins certain of the
embryonal parts are held in common, as always in the armadillo,
but as these common parts are embryonic (trophoblast) and are
abandoned at birth, they are not apparent later. In double mon-
sters the common parts include more than trophoblastic material
and the components are consequently not liberated by the cutting
of the umbilical cord.

In a morphological sense the bands of the armadillo carapace,
with their variations, and the friction ridges of human palms
and soles are partially or wholly homologous structures, so that
their use in determining the degree of similarity of twinned in-
dividuals is equally warrantable in both cases, while the results
may well be compared. Both deal with epidermic structures,
the probable homologs of reptilian scales, placed in rows; in
both are observed the similar phenomena of the forking and con-
sequent doubling of the lines thus formed. Even the double
scale of the armadillo may have its counterpart in a twin sweat-
pore, which indicates the composite nature of the unit to which
they belong.

While, however, the study of the armadillo has the great and
undeniable advantage of allowing the study of the actual em-
bryonic conditions in each case, there are also certain marked
advantages in the study of the human palm and sole configuration
as a similar criterion of similarity of individuals, and in following
the action of heredity. In the first place there is (i) the far
greater complexity of pattern, giving an almost infinite number
of points for comparison. Here the fineness of the lines, and
their occasional participation in the formation of a complex
pattern, allows in number and arrangement, to say nothing of
the minutiae, that is, the composition of the individual ridges, a
complexity far surpassing anything met with in the carapace of
the armadillo, where the details concern only nine rows of scales,
with some 50-60 unit scales in a row. Another very important
advantage is seen in (2) the fact that the records of human twins
may easily be made complete, with prints of both parents, father

224 HARRIS HAWTHORNE WILDER.

as well as mother, and sometimes a second preceding generation,
together with collateral lines, as represented by brothers and
sisters, aunts and uncles. In the work on the armadillo the
mothers only are known, and thus all peculiarities of the offspring
not accounted for by the maternal carapace, are, rightly or
wrongly, attributed to the unknown father. As a third advan-
tage, (3) largely one of convenience, may be mentioned the ease
with which even a very large collection of flat prints may be
filed and handled. This is especially noticed when a series of
prints are to be compared at one time.

On the other hand, there are numerous marked advantages
accruing from the study of the armadillo, so many, in fact, that
the work is indispensible in the study of human friction ridges.
The way in which the results obtained in this field have supple-
mented the work on human twins, rendering vaguely expressed
ideas definite, and raising hypotheses to the category of what
by analogy may be considered facts, has been the subject of the
preceding pages.

In their paper on the limits of hereditary control in the arma-
dillo (1911) the two authors often quoted offer two objections to
my conclusions of a similar nature, as shown in the friction ridge
configuration of human twins. The first, and most valid one, is
that "the origin of the two individuals from a single fertilized
egg is assumed from the facts of resemblance," while the second
is that "the comparison between the twins is made only after
years of post-natal life." Owing to the abundantly proven
unchangeable nature of human friction ridge configuration the
second objection is invalid, so that there is left but the first
one which may become expanded into two questions, that the
early conditions in the embryonic history of human twins is
unknown, and that in individual cases it is not usually possible
to find out whether they were monochorial or bichorial.

As explained above, analogy with the very similar early stages
in the armadillo shows pretty conclusively that human mono-
chorial twins are due to a genuine polyembryony, while in some
cases it is possible to obtain the chorionic relations, as in the
case of the double monster above treated. After collecting the
details of a twin case from any responsible physician the case

PALM AND SOLE STUDIES. 225

need not be watched longer than three or four years to get a set
of satisfactory prints. These objections have, however, some
validity as applied to my definite claims at the time when I first
made them (1902, 1904, 1908), and I am only glad that my
hypotheses of that time have been so ably and satisfactorily
substantiated by the results of the investigation of the armadillo.

That the formation of multiple embryos in man and the
armadillo is due in both cases to a similar cause seems to me
undoubted, and was similarly considered by Patterson, who writes
(1913, p. 560): "There has been considerable speculation as to
how 'indentical twins' and similar types of development have
arisen, and I believe that these studies on the development of
the armadillo will at least indicate how these phenomena may
have come about." He further considers it probable that
"specific polyembryony in the Dasypodidse began by the for-
mation of a set of twins, perhaps at first a sporadic case of gemel-
liparous development such as probably occurs in the production
of duplicate twins in the human species." It is needless to state
that I, too, think the same, and because of this belief can go
farther and state that whatever is learned in the future concerning
the actual causes of polyembryony in the armadillo will solve also
the same question concerning the causation of human duplicate
twins and double monsters.

To close this subject with a pure speculation, suggested by the
work upon the armadillo, Newman and Patterson find (1915)
that the quadruplets may be arranged in pairs, I + 11, and III
+ IV, and that the identity of scale configuration is much greater
in the two members of a single pair than in members of different
pairs. Is it conceivable that human duplicates may bear to
each other these two degrees of similarity found here, and that
the few cases, like my Nos. VII and XIII, where the palm and
sole similarity was not so great as was to have been expected
from the facial resemblance and correspondence in sex, may be
explained by a relationship such as would be found in the arma-
dillo by comparing II with III, or I with IV? This might
involve the early abortion of additional embryos, or might deal
with the question of the two vs. the four-celled stage, as suggested
by Newman and Patterson (1909, p. 186) that "the two in-

226

HARRIS HAWTHORNE WILDER.

individuals derived from one of the blastomeres of the two-cell
stage ought to be more nearly similar to each other than to the
individuals of the other blastomere." Further investigation in

FIG. 31. Print of the right hand of a white male; Coll. No. 96.

the early embryology of the armadillo, particularly the cleavage
stages and the cell lineage, are bound to assist in clearing up these
problems, and will be awaited with interest.

PALM AND SOLE STUDIES.

227

VI. THE HERITABILITY OF FRICTION-SKIN CHARACTERS.
A casual inspection of the palm and sole prints of any family
where two or more generations are represented, and with several
children, indicates that the special configuration of the friction-
skin ridges is to some extent a character, or series of characters,

FIG. 32. Print of the right hand of the six-year-old son of the foregoing, showing
a strong hereditary influence.

transmissible by heredity from parent to offspring. To show this,
two pairs of palm prints are here presented, in each case that of
a father and son, taken from two unrelated white families.

In the first of these (Figs. 31 and 32) the type presented is no
unusual one, yet it is interesting to see how the details of the father
are faithfully copied by the smaller hand of the son. The same
degree of similarity is shown by the handsof theother side, although
the two sides are not alike. Otherwise, so far as may be seen in a
boy of six, the son does not especially resemble the father. It is
also interesting to note that neither the hands of the mother, nor

228

HARRIS HAWTHORNE WILDER.

of the mother's sister, are in any way like those of the father and
son.

The second pair (Figs. 33 and 34) is of still more value as an illus-
tration of direct inheritance, for here the palm of the father presents

FIG. 33. Print of the right hand of a white male; Coll. No. 99.

two unusual features: (i) the reduction of line C, with its triradius,
as described above, and (2) the narrowed, and nearly suppressed,
hypothenar loop, with two triradii, and with its pattern a loop

PALM AND SOLE STUDIES.

229

opening to the ulnar margin ; yet each of these unusual characters
is faithfully reproduced in the little hand of his two-year old son.
The first of these characters occurs in about 18 per cent, of all
cases, although not always to the same degree, as has been stated

FIG. 34. Print of the two-year-old son of the foregoing, showing a strong
hereditary influence, involving in this case two unusual features.

above, but the type of hypothenar here shown is much rarer,
and one might have to look over several hundred sets of palm
prints before meeting with a similar one. This type is given by
Miss Whipple (1904) as B, Fig. 47, p. 349, where its development
from a more typical pattern is explained. As a test of its fre-
quency of occurrence the hand prints of 100 individuals from my

230

HARRIS HAWTHORNE WILDER.

collection were counted, and in eight hands only was the general
type of pattern found, that is, eight out of two hundred, or four
per cent., and each of these differed more in the arrangement of

- 35- Tracings of the hand prints of the two parents, W and N, of the H

family.

triradii and other details, than do these of father and son. It
would be safe to say that the chance of finding two cases with

PALM AND SOLE STUDIES. 23!

hypothenar patterns of the same type and with the details so
similar would be at most one in a hundred, and probably much
less. Multiplying this rough estimate (i in 100) by the chance of
I in 12, the figure already obtained for the occurrence of a com-
plete suppression of line C, and we get I in 1,200 for the probable
occurrence of these two characters in combination, relying upon
chance alone.

The other hands of this same pair, the lefts, correspond almost
as closely, with a like suppression of line C, and with the hypothe-
nar in each reduced to a single triradius, and no other trace of a
pattern. Here the mother's hands also were both of this latter
type with respect to the hypothenar, but the main lines are en-
tirely different.

The results from such isolated instances, however, show merely
that in the friction-skin configuration we are dealing with heritable
characters, and in themselves show little more. A more detailed
line of inquiry is offered by the study of a large family, where the
parents belong to distinctly different types, and where the
children are numerous. Such a family is that of H - - from my
collection, where the material consists of the complete prints
(palms and soles) of the father, W, the mother, N; and six chil-
dren, G, V, M, C, F, and E, together with the palm prints of the
father's three sisters, Fa, Mi, and Lo, which serve as valuable
collateral evidence. Of the six children F is a boy, the other
five are girls.

The hands of the two parents, W and N, are of distinctly dif-
ferent types, that of W, the father, being very complex, with
traces at least of all five palmar patterns, while the palms of the
mother, N, are unusually free from such special features, and
show only an included loop between lines C and D of the left
hand, possibly a fourth interdigital, also a loop acjoss the free
end of the rudimentary line C of the right, the condition desig-
nated in a formula by the digit 8.

These two opposite types, mated, have produced six offspring,
in which the presence or absence of the five palmar patterns are
presented here in tabular form:

The presence of a pattern is represented by a X, the absence
by — , and a rudiment by r. The two designations ia each column

232

HARRIS HAWTHORNE WILDER.

Name.

Thenar.

First Id.

Second Id.

Third Id.

Fourth Id.

Hypoth.

G

X X

'XX

— r.

XX

V

X X

XX

M

x —

— X

X X

X X

C

X X

X X

r. —

F

X X

X X

X X

X X

X X

E..

X-

X-

r. X

X-

r. —

are for the two hands, the designation at the reader's left being
the left.

Aside from the various interdigital patterns, which occur in
about three fourths of the cases (36 out of 48), and in one case
(third) are practically universal, there will be noticed the re-
markable result in the case of the two most prominent patterns,
thenar and hypothenar; the former, a somewhat rare pattern,
which occurs in only about 4 per cent, of the white race, is present
here in 10 out of the 12 hands (83.3 per cent.), while the much
commoner hypothenar, occurring in 20 per cent, of hands of the
white race, and found also in both hands of the father, is actually
present, aside from a rudiment, but once in the 12 (8.3 per cent.).
Now as the father has both thenar and hypothenar patterns upon
both hands, and as the mother has neither, it would suggest that
possibly the thenar region may be here controlled by the male
parent, and the hypothenar by the female.

As the thenar pattern is usually so rare, its almost universal
occurrence in this family is without doubt due to direct inheritance
from the father. For more careful examination the cases are here
figured. The usual, or typical, thenar pattern is a double one,
like that of the father's left, or G's left, with two loops turned
away from each other, but with the loops themselves in contact.
Morphologically the small, upper loop is in reality the first
interdigital, while the lower loop is the thenar proper. This
family, however, although the typical form is not wanting, shows
a strong tendency to suppress the upper, or first interdigital,
loop, and in some cases to modify the lower, the course of .de-
generation following two lines, as follows:

Series i.—W, left, F, right; V, left, C, right.

Series 2.—W, left, M, left; V, right, E, left.

Only upon two palms out of the 12, M, right and E, right, is a
pattern entirely lacking in the thenar region.

PALM AND SOLE STUDIES.

233

It thus seems certain, as already said, that there is here an
unmistakable case of the inheritance of a thenar pattern from
the male parent, and, except for the failure of this in two of the
twelve cases, the character acts like a Mendelian dominant. On
this basis the failure in these two cases may be explained by (i)

FIG. 36. Tracings of the thenar region in each of the six children of W and N.
the individuals illustrated in the preceding figure. The fifth child, F , is a son, the
rest daughters.

that the father himself is heterozygous, or (2) that the character
is not a unit character. As for the first the collateral evidence of
the father's three sisters is to the point, for in these Fa and Mi
have good thenar patterns, on both hands. In the third sister,
Lo, the extreme smoothness of the skin produced an almost il-
legible print, although upon the left hand a part of a double
pattern, the first interdigital loop, is easily made out, rendering
the presence of a thenar figure upon this area absolutely

234 HARRIS HAWTHORNE WILDER.

certain, while it is possible to interpret in a similar way the faint
indication of certain lines on the thenar area of the right. It is
also noteworthy that the type of pattern upon the right hand of
Fa, a quite unusual type, is practically identical with that which
her niece C has upon the same hand. This universality of the
thenar pattern in four individuals of the father's generation (8
hands) wTould suggest dominance.

Concerning the second possibility, that the character here
examined, the thenar pattern, is not in itself a Mendelian unit,
but is conditioned by two or more factors, this is rendered probable
by the many stages of modification, or degeneracy, represented
by the hands here represented, eight in the older generation, and
ten in the younger. It suggests the presence of a factor tending
to reduce the pattern to a series of parallel lines that follow the
strong line of flexion bounding the thenar region, the "Line of
Life" of the palmist; and the presence of the ancestral thenar
pattern, or some stage in its reduction, in so many hands of the
H — family, may be due to the absence of this latter factor fully
as much as to the presence of a positive thenar pattern unit.

Quite aside from all this, the fact that such grades in the re-
duction of a pattern may exist, and that for a given pattern
there may be found several series of them1 is in itself a phenom-
enon of heredity at present inexplicable, for, while in such a
series we may find every step in a definite and continuous process,
the steps themselves do not occur in ontogenetic development, but each
step arises in an individual embryo precisely in the form which
that individual shall retain throughout life. Since, now, this
process of the degeneration of the ridge formation is from a
complex pattern to a series of parallel lines; and since the goal
sought in this process is in countless details, precisely what would
result from the action of external forces upon yielding ridges, as
such authors as Miss Whipple and W. Kidd have abundantly
shown, we have the startling phenomen of a Lamarckian en-
vironmental influence of a mechanical nature, which, while having
no visible effect upon the soma, so modifies the germ-cells that the
effect becomes transmitted to the descendants, and that probably
by Mendelian methods! There has thus developed a long series

1 Cf. Miss Whipple, 1904, pp. 33Q-.352, esp. Fig. 48, p. 350.

PALM AND SOLE STUDIES. 235

of generations, with its beginnings somewhere in the Simian line,
the individuals of which, though with many a retrogression, have
progressed steadily from a series of complicated patterns, de-
veloped in association with raised pad organs, to a simple con-
dition of parallel ridges, more or less conformable to the principal
lines of flexure.

Expressed in another way, the results of the study of friction-
ridges, especially in connection with their possible heritability,
may be collected into a series of definite statements, as follows:

I. Each friction-ridge pattern presents, by taking a large
number of human individuals, every stage in the process of
development, from a whorl of concentric lines with a definite
number of embracing triradii, to a condition in which core
and triradii are completely lost and the surface is covered by
simple parallel ridges, straight or slightly curved.1

II. From comparative study of other Primates it is plain
that the course of evolution has been from the involved condition,
the whorl, which is distinctly Simian, through every stage in the
reduction of the pattern, to its final complete effacement.2

III. A given pattern shows as a rule more than one way in
which it may degenerate, often two or three, and each way may
be shown by complete series of instances taken from individual
cases, and leading to the same final result.3

IV. The ridges forming a given pattern are first seen in an
embryo of approximately four months, since wThich age there is
no indication of change throughout life. This has been absolutely
proven in the case of individuals after birth by separate observa-
tions taken upon the same person, and in a number of instances
these observations have been separated by a period of fifty or
more years; in the embryo the ridges appear in a definite form,
and there is no indication of later modification or of provision for
it.4

1 Miss Whipple, 1904.
" Miss Whipple, 1904.

3 Miss Whipple, 1904, pp. 343-354-

4 The occurrence of ridge rudiments, occasionally found lying between normal
ridges, and reduced often to broken lines without sweat-pores, suggests the pos-
sibility that in an early embryonic stage these may have been as large as the
others, and that they may have become pushed out of existence, or suppressed,
owing to some mechanical pressure during ontogeny. Of this we have no proof,

236 HARRIS HAWTHORNE WILDER.

V. It follows, therefore, that the modifications of the original
typical patterns, as given in II. and III: above, must be affected
through modification of the germ-cells, which determine at each fer-
tilization the form of pattern for the new individual. These pat-
terns are often atavistic, that is, a primitive pattern may oc-
casionally appear, but in the main the progress has been a decided
one, so that the majority of men show a nearly complete efface-
ment of the original complicated condition.

VI. In each given area the characteristic ridge direction of a
typical human hand is precisely that to which the ridges would be
pushed or pressed if considered plastic and capable of being acted
on directly by the objects which naturally come into contact with them
in the ordinary uses of the hands and feet. That is, the ridges are
everywhere arranged to run at right angles to the prevailing
direction along which contact objects would push or slip. In the
hand, which is constantly grasping cylindrical objects, such as
boughs, handles of instruments, ropes, and numberless similar
things, the ridges of the distal half of the palm tend to lie straight
across it, a position which, in the highest degree possible for man,
is expressed by the formula 1 1 • 9 • 7 • 5. It is thus significant that
in the right hands of all human races this formula is found in 22-25
per cent, of the cases, while in left hands this formula forms only
4 per cent, of the cases.1 That the same percentages hold among
such diverse human races as Whites, Negroes, and Maya Indians
marks this as a natural human tendency, and suggests that this
result has been gained since the adoption of right-handedness.

VII. Finally, (i) since the configuration of the friction-ridges
is directly heritable, probably in conformity to the laws of
Mendel; (2) since the countless elements involved in the ridge
arrangement are precisely similar to what would be produced in

and it is not very likely, yet, until otherwise explained, their occurence prevents
the dogmatic assertion that no change can absolutely take place after the ridges
are first laid down. Concerning a modification of this sort after birth, this is
still less likely, and may be practically denied. The author is in possession of a
series of prints of the right forefinger of a child, beginning with the second year,
continued at approximately two-year intervals until the age of twelve, and
still being collected. In this there is an unusual number of these "suppressed"
ridges, yet in the interval covered by the series there has not been the slightest
change in any of them.

1 Wilder, 1904, Amer. Anthrop., Vol. 6, p. 279.

PALM AND SOLE STUDIES. 237

a plastic material by the direct application of the most usual
objects, acting in the most usual way, and (3) since no modifica-
tion of an individual plan of configuration seems possible during
ontogeny, or as a result of continued action upon an individual,
we find ourselves confronted by the phenomenon of a complex series
of arrangements, each corresponding exactly at all points with the
direct result of the action of constantly applied external forces upon
plastic structures; and yet, although these external forces are con-
stantly applied, and are without the slightest visible result upon the
individual, positive results, cumulative in their effects, are actually
transmitted from parent to child. Furthermore, these results must
lie transmitted by the germ-cells, and are probably in accordance with
the Mendelian laws.

One could hardly find a more startling combination of princi-
ples than the above, supporting at the same time the ideas of
Lamarck, Mendel, and Weismann, with Darwin left entirely
out, since with the reduction in the size of the friction-ridges and
the increase in the size of the body, they can be of no real selective
value above the Cercopithecidse; yet from a long and careful
study of a large amount of data (perhaps a thousand hand prints,
including the most varied human races, studied during fifteen
years) every one of the above conclusions seems amply substan-
tiated. The facts stand written upon the palmar and plantar
surfaces of mankind; the detailed record of their physical ex-
ertions and that of a long line of ancestors; these records are
inherited with numerous individual modifications, so that, except
in the case of duplicate twins, the offspring of the same two parents
are never quite alike; it is easily possible to arrange these in-
dividual modifications in developmental series, presumably cor-
responding to their evolution, and, when so arranged, they follow
closely the theoretical series of modifications which would be
induced by direct influence from the usual external mechanical
forces, in case these structures were sufficiently yielding to allow
it.

Such are the conclusions which, from my study of friction-skin
configuration, I have reached at present. They seem inevitable,
and if so, are exceedingly important; and it is because they seem

238

HARRIS HAWTHORNE WILDER.

so that I wish to make an appeal for others to undertake to in-
vestigate the same subject. In the study of friction-skin may
be found taxonomy, morphology, ethnology, genetics, as well as
the practical application to problems of individual identification.
It must be confessed that there is at first a large amount of detail
which must be acquired, as well as a certain amount of nomen-
clature. The bibliography, however, if that which deals with the
special work on apical patterns ("finger-prints") be disregarded,
is not large, and as it is for the most part recent it can be readily

. A/a.

FIG. 37. Tracings of the only calcar patterns in my entire collection, outside
of the two families considered below.

obtained. Unlike most biological material, that concerned here
is readily obtained, and easily kept, a series of prints being
for most purposes more convenient for study than the actual
objects; the prints, too, are more accurate records of the facts
than are any drawings or even photographs.

We who have been at work along this line are in great need of
collaborators. We are firmly impressed with the belief of the

PALM AND SOLE STUDIES. 239

biological importance of the field, and feel that many weighty
conclusions are to be drawn from it. It is with the idea of
facilitating study that I give here, as the final paper in this series,
a bibliography of the subject, which should make the study at
once available to those who care to interest themselves in it.

Still more convincing than the inheritance of the thenar pattern
in the hands of the H-- family is that of the calcar pattern in a
family which I may designate as Wh — . This pattern is the
rarest of all friction-skin patterns known in man, and has been
found once by Schlaginhaufen,1 once by Fere,2 and, outside of
this family, twice by myself (Fig. 37). In the sole prints of
100 Liberian soldiers, collected by Starr, there is not an instance
of it, and Schlaginhaufen states in 1905, including all the obser-
vations to date, that it has never been reported in any non-
European race.

As it appears upon the tread area this pattern is in the form of
a simple loop, with its apex fibulo-proximal and with its opening
directed obliquely towards the tibial and distal directions (Fig.
23). This loop is usually accompanied by a triradius placed
distally to it and on the fibular side, but this does not always
appear upon the tread area, and in general the portion thus
included gives the impression of being a part of a more complete
figure, the remainder of which should lie farther up on the side
of the foot. In fact Schlaginhaufen, who seems to be the only
one who has followed this pattern beyond the tread area, has
traced the loop in his case over the hollow on the tibial side of the
foot and reports that there the lines of the loop converge, turn
about, and form a second loop facing the other way and ac-
companied by a second triradius, so that the entire figure,
thus completed, becomes a long drawn out spiral with two loops,
one of which lies within the tread area.3

Occasionally, though not often, prints are found in which the

1 1905; Fig. 178, p. 100.

2 1900, Les lignes papillaries de la plante du pied, 1900; Fig. 18, p. 615.

3 " Ich verfolgte jedoch die Crura des Sinus in tibiodistaler Richtung und gelangte
zu folgenden Leistenbild. Gegen die Cavitat des Fusses convergieren die Schenkel,
um einer iiber den anderen hinweglaufend umzubiegen; d.h., wir haben eine lang-
gestreckte, linksgewundene Spirale. Ihrem tibiodistalen Ende liegt ein zweiter
Triradius an, dessen Lage in die Fusshohlung fallt." Schlaginhaufen 1905, pp.

100-101.

240

HARRIS HAWTHORNE WILDER.

lines which run across the heel region of the tread area diverge
from one another at a point near the tibial margin of the print, or
perhaps spread out along this border from a shorter portion of
the opposite side, and as this appearance often occurs where there
is some connection with a typical calcar pattern (either on the
other foot of a subject possessing this pattern or upon those
related to such a subject) it gives probability to the surmise
that we have here the vestige of a calcar pattern (Fig. 42). If
this be so it suggests that the tendency to form a calcar loop is

ssue

FIG. 38. Diagram of the relationships of family Wh— — . The squares
represent males, the circles females. A calcar pattern on the right heel is indicated
by a diagonal line drawn through the square or circle, from the right above to the
left below; one on the left heel by the opposite diagonal. Where both diagonals
are used there are two calcar patterns. A divergence, indicating a rudiment of the
pattern, is indicated by a black lower corner on the side of the divergence. A
shaded square or circle indicates that, through decease or otherwise, prints are not
available.

much commoner than the realization, and that a certain config-
uration, although reduced by some unfavorable influence, is very
hard to entirely eradicate.

To proceed now to the conditions obtaining in this family
(Wh — ). The maternal grandfather E and the grandmother 5
show the usual type of heel, save that in the former there is in
each foot a noticeable divergence of the lines towards the inner

PALM AND SOLE STUDIES.

241

margin, which although of no especial importance in itself, may in

connection with the descendants have some significance. (Fig. 42.)

This pair had one son and two daughters, one of which, the

daughter C, possesses a calcar loop upon the right heel, and a

0

X

N,1

FIG. 39. Diagram of the relationships of the family Wo — — , that of the
husband, R, of the double mating. Symbols as in the previous figure. The sister,
Co, is deceased, and her husband, Rs, is not now available, but their daughter, Fl,
is normal. Her two children have not yet been investigated. I have not the
prints of the brother Fr, or of his wife, Et, but again the heels of the two children,
Sm and Je are normal. The source of the double calcar pattern of R remains thus
far unknown.

divergence upon the left. The other daughter, /, shows on the
right heel a noticeable divergence, but no loop, while the left
heel is normal. The son F is deceased, but his wife is normal,

242

HARRIS HAWTHORNE WILDER.

and neither of his two children, Ra and U, show anything unusual
in the heel markings.

The daughter C, who possesses one calcar loop, by a most extra-
ordinary coincidence, married an unrelated man, R, who possesses a
typical calcar loop on both feet, a case unheard of except in this
family, and their three childrem J, M, and L, all possess calcar
loops, J (female) upon the left, M (female] upon both, like the father,

C.

c.a.

FIG. 40. Tracings of the heel patterns of the mating R X C, the first with double
calcar patterns, the second with a single one, the right.

and L (male} upon the right, with a marked divergence on the left.
In the right foot of J there is also some divergence. These results
may be tabulated as follows, using an X for a calcar loop,
and an r for a divergence. The two symbols used for each indi-
vidual signify the two feet, in their natural position. The
enclosed symbols represent males.

PALM AND SOLE STUDIES.

243

X rX

I

Xr

XX

Concerning the source of the calcar loops in the parents, the
father of R was investigated and found to have quite normal heels;
the mother of C is also normal, but the father, E, shows a good
divergence upon both heels. Probably, then, C obtained her
calcar loop from her father rather than from her mother, and in

J. dauber of % **••( C
C.U, M>. &

FIG. 41. Heel patterns of the three children of R and C, shown in the previous
figure. The oldest, J, has a calcar pattern on the left foot; the second, M, on
both, and the third, L, on the right.

•

the same way it may be surmised that the husband R got his
from the side of his deceased mother Is, but for lack of further

244

HARRIS HAWTHORNE WILDER.

data this question cannot be definitely decided. Certain brothers
and sisters of most of the four grandparents, E, S, 0, and Is,
are still living, but to obtain the necessary data from them would
be difficult; there is more to expect from the coming generation,
the offspring of /, M, and L, and of their cousins, Sm and Je,
Ra, and U, but for that several years must elapse.

Leaving the past and the future as open questions, we return

I, sishr .(•
Cia.W*. *

FIG. 42. Heel patterns of the only relatives of C which show even a rudiment of
a calcar pattern; the father, E, and the sister, I.

to the facts here presented, and these are, that a man with a calcar
loop on both feet married a woman with a loop and a divergence, and
that, of their three children, one has a loop on both feet, and two have
a loop on one foot and a divergence on the other, the loop being on
the right foot in one case and on the left in the other. When this
record is considered in connection with the fact that, aside from
this family, but four loops are known to science (that is, in per-
haps 1,000-1,500 individuals of all races) the direct inheritance of
this condition by the mating of XX X rX is beyond question.

PALM AND SOLE STUDIES. 245

Here there is brought forcibly before the reader the question of
unilateral inheritance, i. e., whether a character possessed upon
the right side only in the parent may cross over in the offspring,
and appear on the left side, or whether each side inherits inde-
pendently. The mother C has a loop upon the right foot, and a
divergence only upon the left, and the same condition obtains
in her son L. In the daughter J, however, these conditions are
reversed, and the right loop in both father and mother fail to
govern the right heel of J. The loop which / possesses upon her
left foot may have been inherited from the father, as he has a
loop on each foot, but does the loop upon her mother's right foot
have any influence? Facts thus far, in all the families studied,
tend to show that inheritance does not cross from one side to the
other, but we are yet a long way from stating this definitely, or
even as a plausible hypothesis. The two definite points that
appear as the result of this investigation are (i) that the calcar
loop is heritable, and (2) that its presence upon the right foot of both
parents does not compel its appearance upon the same foot of the
offspring.

VII. BIBLIOGRAPHY OF FRICTION-SKIN CONFIGURATION.
This list is intended to be complete on the subject of the fric-
tion-skin configuration of the palms and soles. On the apical
patterns it makes no attempt to be exhaustive, as the subject is
now in the hands of the police department of all civilized coun-
tries, and has been largely exploited for practical purposes of

*

personal identification, developing a large mass of literature
hardly morphological in the technical sense.

There is also little or nothing upon the structure of the skin as
such, on its development or histology, or on the innervation,
which has been especially a subject of investigation by psychol-
ogists.

At the end there is appended a list of the published investiga-
tions of Newman and Patterson on the subject of polyembryony
in the armadillo, extensively referred to in Section IV. of the
present paper. This subject, as it relates to scute and band
anomalies of the carapace, and their inheritance by twins and
other duplicate individuals, is of special interest here.

246 HARRIS HAWTHORNE WILDER.

D'Abundo.

'91 Contribute allo studio delle impronte digitali. Archivio di Psichiatria.

Pisa.

'94 Le impronte digitali in 140 criminali. Riforma medica.
Alix, M.

'67 Recherches sur la disposition des lignes papillaires de la main et du pied.

Ann. des Sci. Nat., T. VIII., pp. 295-362.
'69 (Same title.) Ann. des Sci. Nat., T. IX., pp. 5-42.
Bruhns, Fanny.
'10 Der Nagel der Halbaffen und Affen. Morph. Jahrb., Bd. XL., pp. 501-609.

(Ridges of finger-tips incidentally mentioned.)
Daae, A.

'94 Ueber Fingerabdrucke und deren Verwendung zur Identitatsfeststellung,
verglichen mit Bertillon's Anthropometrischen System (Uebersetzt von
Teichmann). Zeitschr. fur schweizerisches Strafrecht. Jahrg. VII., p.

3I/+.

Dixey, F. A. On the Epidermis of the Plantar Surface and the Question of Use-
inheritance. Rep. 64th Meeting British Assn. Adv. Sci. (This treats of
the thickness of the epidermis on the fetal sole.)
Evatt, E. J.

'06 The Development and Evolution of the " Papillary " Ridges and Patterns
on the Volar Surface of the Hand. Journ. Anat. and Physiol., Vol. XLI.,
pp. 66-71.
Faulds, H.

'80 On the Skin Furrows of the Hand. Letter in Nature, Vol. XXII. , p. 605.
(This remarkable letter, written from the Tsukiji Hospital, Tokio, antici-
pates every direction of the later investigation; comp. anat; ethnol; indiv.
variations; heredity, etc.)
Faulds, H.

'94 On the Identification of Habitual Criminals by Finger-prints. Nature,
Vol. L. (F. claims priority to Sir Wm. Herschel, because of his letter of
October 28, 1880, the previous title. Herschel's letter to Nature is dated
November 25, 1880.)
Fere, Ch.

(This author has written numerous short papers on the apical patterns
of both fingers and toes. The most important citations are the following:)

'91 (Finger and toe patterns.) C. R. soc. biol., Tome 43, pp. 497-506.

'92 (With Batigny, on the same subject.) C. R. soc. biol., Tome 44, pp. 802-
806.

'93 (Finger and toe patterns.) Journ. de 1'Anat. et de la Physiol. Annee. 29,
pp. 223-237.

'95 (Sensibility of the balls of the fingers.) C. R. soc. biol., Tome 47, pp.
657-660.

'96 (Finger patterns in relation to function.) C. R. soc. biol., Tome 48, pp.
1114-1116.

'98 (Finger patterns in relation to functional aptitude of the hand.) C. R.
soc. biol., Tome 50, p. 837+.

'oo (Hands and finger patterns of certain monkeys.) Journ. de 1'Anat. et de la
la Physiol. normales et pathol. de l'homme et des animaux. Ann. XXXVI.,
pp. 255-267.

PALM AND SOLE STUDIES. 247

'oo (Papillary lines of the palm of the hand.) Journ. de 1'anat. et de la physiol.,

Ann. XXXVI., pp. 376-392.
'oo (Papillary lines of the sole of the foot.) Journ. de 1'anat. et de la physiol.,

Ann. XXXVI., pp. 602-618.

'oo (Imprints of palms and soles.) C. R. soc. biol., Tome 52, pp. 641-643.
'05 (Finger prints of psychopathic subjects.) Journ. de 1'anat. et de la physiol.,

Ann. XLI., pp. 394-410.

'06 (Papillary lines of the heel.) C. R. soc. biol., Tome 61, p. 44.
Forgeot, Rene.

'91 Les empreints latentes. These; Lyons. (Also in same year, in Bull de la

soc. d'anthropol. de Lyon, Tome X., p. 189.)
Frecon, Andre.

'89 Les empreintes en general. These, Lyons.
Fiilleborn, Fr.

'02 Beitrage zur physischen Anthropol. der Nord-Nyassalander. Berlin.
Galton, Sir Francis.

'86 (In Science, Vol. VIII., p. 166 and p. 212.)
'88 Personal Identification. Journ. Roy. Inst.

'88 Personal Identification and Description. Nature, Vol. 38, p. 201.
'90 Patterns in Thumb and Finger Marks. Proc. Roy. Soc., Vol. 48, p. 455.
'91 Patterns in Thumb and Finger Marks. Phil. Trans. Roy. Soc., Vol. 182,

pp. 1-23.
'91 Method of Indexing Finger Marks. Proc. Roy. Soc., Vol. 49, p. 540.

(Also in Nature, Vol. 44; p. 141.)

'91 Identification of Finger Tips. Nineteenth Cent., Vol. 30, pp. 303-311.
'92 Finger Prints. MacMillan, London.
'92 Imprints of the Hand, by Dr. Forgeot of the laboratoire d'anthropologie

criminelle, Lyon. Journ. Anthropol. Inst., Vol. XXL, pp. 282-283.
'93 Identification. Nature, Vol. 48; p. 223.

'93 Finger Prints in the Indian Army. Nature, Vol. 48, p. 595.
'96 Prints of Scars, Nature, Vol. 53, p. 295.

'96 The Bertillon System of Identification. Nature, Vol. 54, p. 569.
'99 Finger Prints of Young Children, Meeting of the British Assn., Dover, pp.

pp. 868-869.

'99 Finger Prints. Journ. Anthropol. Inst., Vol. 29, p, 199.
'93 The Decipherment of Blurred and Indistinct Finger Prints. MacMillan,

London.

'95 Finger Print Directories. MacMillan, London.
Garson, J. G.

'oo A System of Classification of Finger Prints. Meeting of the British Assn.

Bradford, pp. 910-912.

'oo Finger Print Classification. Journ. Anthropol. Inst., Vol. XXX., p. 101.,
Henry, Sir Edward Richard.

'99 Finger Prints as a Method for the Identification of Criminals. Nature,

Vol. 61, p. 42.
'99 Finger Prints and the Detection of Crime in India. Meeting of the British

Assn., Dover, p. 869.
'oo Classification and Uses of Finger Prints, London. (This is the standard

book used in police identification bureaus the world over. There are

numerous later editions both in England and in the United States.)

248 HARRIS HAWTHORNE WILDER.

Hepburn, David.

'93 The Integumentary Grooves on the Palm of the Hand and Sole of the Foot
of Man and the Anthropoid Apes. Journ. Anat. and Physiol., 1893, Vol.
XXVII., pp. 112-130.

'95 The Papillary Grooves on the Hands and Feet of Monkeys and Men. Sci.
Trans, of the Royal Dublin Soc., Vol. V., Series II., pp. 525-537. (Re-
viewed in Nature, Nov. 14, 1895, and in Journ. Anat. and Physiol., Jan.,
1896.)

'97 Note on Dr. Harris H. Wilder's Paper, " On the Disposition of the Epidermic
Folds upon the Palms and Soles of Primates." Anat. Anzeiger, Bd. XIII.,
pp. 435-437-
Herschel, Sir Wm.

'80 Skin Furrows of the Hand, Nature, Vol. XXIII.; p. 76. (This note, often
taken as the first mention of the subject in recent years, and perhaps as the
very first suggesting the use of the ridges in identification, was antedated
by a month by the letter of Dr. Faulds to the same periodical. Faulds's
letter was published in the issue of Oct. 28, 1880 (Vol. XXII, p. 605), and
Herschel's paper appeared in the issue of Nov. 25 of the same year.)

'94 Finger Prints. Nature, Vol. LI., p. 77.
Johnson, R. H.

'99 Pads on the Palm and Sole of the Human Fetus. Amer. Nat., Vol. 33,

pp. 729-734-
Kidd, W.

'03 On Imbrication of the Papillary Ridges in Man. Journ. Anat and Physiol.,
Vol. XXXIX., pp. 413-416.

'05 The Papillary Ridges and the Papillary Layer of the Corium in the Mam-
malian Hand and Foot. Journ. Anat. and Physiol., Vol. XLL, pp. 35-44.

'07 The Sense of Touch in Mammals and Birds, with Special Reference to the
Papillary Ridges. London, A. and C. Black, 176 pp. (The title of this
book is misleading; in reality it treats mainly of the friction ridge configura-
tion in the Primates, and gives good illustrations and descriptions of in-
dividual Primate palms and soles. Review by Inez Whipple Wilder in
Science, Vol. XXVII. , 1908, pp. 582-585.)
Klaatsch, H.

'87 Ueber die Morphologic der Tastballen. Anat. Anz., Bd. II., pp. 400-401.
(Prelimin. Commun.)

'88 Zur Morphologic der Tastballen der Saugertiere. Morph. Jahrb., Bd.
XIV., pp. 407-435. (The first morphological consideration of mammalian
pads.)
Kollmann, A.

'83 Der Tastapparat der Hand der menschlichen Rassen und der Affen in seiner
Entwickelung und Gliederung. Leipzig, pp. 1-75.

'85 Der Tastapparat des Fusses von Affe und Mensch. Archiv fiir Anat. u.

Physiol., Anat. Abteil., pp. 56-101.
Kolossoff, G. and Paukel, E.

'06 Versuch einer mathematischen Theorie der Hautleistenfiguien der Pri-

maten-Palma und Planta. Morph. Jahrb.. Bd. XXXV., pp. 697-708.
Kollman,.

'oo (Discussion concerning impressions of finger patterns on the pottery of the

PALM AXD SOLE STUDIES. 249

Lake-dwellers, in Correspondenzblatt d. deutsch. anthropol. Gesell., No. 10,
Berichte der XXXI. Versamml. in Halle.)
Loth, E.

'09 Dzisiejszy stan wiedzy o filogenii stopy ludzkiej. [The present condition of
our knowledge of the phylogeny of the human foot.] Proc. Soc. Sci.,
Warsaw, II., 8, 1909.

'10 Przyczynek do poznania przebiegu ukladow listewek skornych na stopie
dloni polakow. [Contribution to the knowledge of the arrangement of the
integumental ridges on the palms and soles of the Poles.] Proc. Soc. Sci-
ences, Warsaw, III., 4, 1910.

'10 Anthropologische Untersuchungen iiber das Hautleisten-System der Polen.
Zeitschr. fiir Morphol. u. Anthrop, Bd. XIII., pp. 77-96. (An abridgment
of the above, in the German language.)

'13 Zum Artikel des Herrn Prof. Schlaginhaufen; Beobachtungsblatt und
Anleitung zur Aufnahme von Hand- und Fuss-abdriicken. Correspondenzbl.
d. deutschen Gesell. fur Anthrop., Ethnol., u. Urgeschichte. Jahrg. 44,
No. i, Jan., 1913.
Maack, F.

'01 Das Hautleistensystem an den Fingerspitzen. Wissensch. Zeitschr. fiir

Xenologie, No. 7, pp. 6-15.
Malpighi, M.

1686 De externo tactus organo exercitatio epistolica ad Jacobum Ruffium.
Londini. (The first known reference to friction-ridge patterns; quoted by
Alix, 1867, and by Schlaginhaufen, 1905.)
Martin, R.

"14 Lehrbuch der Anthropologie. Jena, 1914. Cf. pp. 360-367 for methods

of formulating palms and soles.
Meisner,

'oo Scherben mit Fingereindriicken. Correspondenzbl. d. deutsch. anthrop.
Gesellschaft. XXXI Versamml. in Halle. (Short notice of pits made in
pottery of the Lake-dwellers by the finger tips of the potter; farther dis-
cussion by Kollmann. Although in this there is no mention of the friction-
ridges, the paper suggests possible material for investigation. There is
frequent mention in literature of similar impressions.)
Minakata, K.

'94. '95 (Two short letters in Nature, Vol. LI., pp. 199-200, and 274. These
present claims of the use of the " Thumb-stamp " by the Japanese prior to
Sir Wm. Herschel. Cf. in this connection the references to the latter and
to Dr. Faulds, above.) •

Morselli, E.

'74 Sulla dispozisione delle linee papillari nella mano e nel piede del Cercopi-

thecus mona. Ann. di soc. di nat. di Modena, Anno VIII.
Niemirycz-Lothowa, Jadwige (Mme. Loth).

'12 O rzadkim przypadku przebiegu listewek skornych na stopie murzyna.
[A case showing a very rare disposition of the integumental ridges on the
foot of a negro.] Proc. soc. sci., Warsaw, V., 9, 1912.
Paul, Fr.

'03 Sichtbarmachen latenter Finger- und Fuss-abdriicke. Archiv fiir Kriminal-
anthropologie, Bd. XII., pp. 124-129.

25O HARRIS HAWTHORNE WILDER.

'03 Die Kollectivausstellung der Polizeibehorden auf der Stadteausstellung in

Dresden. Arch, fur Kriminalanthropologie, Bd. XIII., pp. 321—323.
Prant, A.

'oo Ueber das Aufsuchen von Fussspuren und Handeabdriicken und ihre

Identifizierung. Arch, fur Kriminalanthrop., Bd. III.
Purkinje, J. E.

'23 Commentatio de examine physiologico organi visus et systematis cutanei,
etc. Vratislav (Breslau). A thesis delivered at Breslau in 1823. This,
except for the brief mention by Malpighi in 1686, is the first paper published
in Europe, at least, treating of the subject of finger-prints. The original is
very rare (a copy in the Surgeon-general's Library in Washington), but the
essential parts of it are given in Engl. transl. in Gallon's Finger Prints, 1892,
pp. 84-88. Cf. also Roscher, 1906, below.
Radl, H.

'95 Fingerabdriicke in Bosnien. Globus, Bd. LXVIL, p. 388.
Retzius, G.

'04 Zur Kenntnis der Entwickelung der Korperformen des Menschen wahrend
der fotalenLebensstufen; in Biol. Untersuchungen, XL, pp. 33-76. (Mainly
the pads.) Preliminary commun. of this in Verh. d. anat. Gesell., i8th
Versamml. Jena, pp. 41-43.
Roscher, G.

'05 Handbuch der Dactyloskopie. Leipzig.

'06 Der Altmeister der Daktyloskopie. Ein Gedenkblatt fur J. E. Purkinje.

Arch, fiir Kriminalanthropol., Bd. XXII. , pp. 326-335.
Sanctis, S. de, e Toscano, P.

'02 Le impronte digitali dei fanciulli normali, frenastenici, e sordo muti. Atti d

soc. Romana di Antropol., Vol. VIII., pp. 62-79.
Schlaginhaufen, O.

'05 Das Hautleistensystem der Primatenplanta mit Beriicksichtigung der
Palma- I. Morph. Jahrb., Bd. XXXIII. , pp. 577-671. II. Morph. Jahrb.,
Bd. XXXIV., pp. 1-125. This paper, with that of Miss Whipple (1904),
appearing almost at the same time, established the principles of comparative
morphology involved in the friction-skin configuration, and are hence in-
valuable to one wishing to go into the subject fundamentally.
'05 Beitrage zur Kenntniss des Reliefs der Planta der Primaten und der Men-
schenrassen. Korrespondenzbl. d. deutschen Anthrop. Gesell., No. 10
pp. 1-4.
'06 Zur Morphologic der Palma und Planta der Vorder-Inder und Ceyloner.

Zeitschr. fiir Ethnol., 1906, pp. 656-706.

'06 Ueber das Leistenrelief der Hohlhand- und Fusssohlen-flache der Halbaffen,
Affen, und Menschenrassen. Ergebn. in der Anat. u. Entwick., Bd. XV.,
J905. PP- 628-662. (An excellent review of the subject to date, with an ex-
haustive bibliography; this has been of much assistance in the present list.).
'12 Beobachtungsblatt und Anleitung zur Aufnahme von Hand- und Fuss-
abdriicken. Korresp. bl. der deutschen anthropol. Gesell., Jahrg. XLIIL,
No. 5, 1912, pp. 33-36.
Schwalbe, G.

'05 Ueber Ballen, Linien, und Leisten der Hand. Strassb. medizin. Zeitung.
Seymour, Lee. (Three books, privately printed, and obtainable from the author,

PALM AND SOLE STUDIES. 251

Los Angeles.) Finger-print -Systems. Finger-print Classification. Finger-
print Identification for Banks.
Varigny, M. H. de.

'91 Les empreintes digitales d'apres M. F. Gallon. Revue Scientifique. T.

XLVIL, pp. 557-562.
Vucetich, Juan.

'01 Conferenzia sobre el sistema dactylosopicodada en la Bibliotheca publica
de la Plata. La Plata, 1901.

'04 Dactyloscopia comparada. La Plata.
Welcker, H.

'97 Die Dauerhaftigkeit des Dessins der Riefchen tmd Faltchen der Hande.
Arch, fiir Anthropol., Bd. XXV., pp. 29-32. (An interesting pioneer paper
of four pages, with prints of the author's palm at 34 and again at 75 years.
This is, perhaps, the first published print of a human palm. The print
taken at the age of 34 must have been in 1858, and thus well antedates any
work on the palm.)
Whipple, Inez L. (Mrs. H. H. Wilder).

'04 The Ventral Surface of the Mammalian Chiridium, with Especial Reference
to the Condition Found in Man. Zeitschr. fiir Morphol. u. Anthropol., Bd.
VII., pp. 261-368. (This is the fundamental paper on the comparative
morphology of the ridge patterns of palms and soles, and includes the study
of the relief of these surfaces in all mammals, and the growth of the ridges
surfaces, as modified by this. This paper, with that of Schlaginhaufen,
1905, are of first importance in the scientific study of human friction-ridges.
The book of Kidd, with the somewhat misleading title of " The Sense of
Touch, etc.," 1907, is also of interest here.)
Wilder, Inez Whipple.

'08 Review of Kidd's " The Sense of Touch in Mammals and Birds, etc."

Science, N. S., Vol. XXVII. , pp. 582-585.
Wilder, H. H.

'97 On the Disposition of the Epidermic Folds upon the Palms and Soles of
Primates. Anat. Anz., Bd. XIII., pp. 250-256.

'02 Scientific Palmistry. Pop. Sci. Monthly, Nov., 1902. (In this the four
main lines were first used, but designated as lines 1-4, and desciibed only as
they limited the areas between them, upon which the attention was mainly
directed.)

'02 Palms and Soles. Amer. Journ. Anat., Vol. L, pp. 423-441. (As in the
previous paper, the description of individual palms and soles was based
upon the areas rather than the main lines. The close similarity in the
configuration of ridges in identical twins was also shown by two sets.)

'03 Palm and Sole Impressions and their Use for Purposes of Personal Identi-
fication. Pop. Sci. Monthly, Sep., 1903, pp. 385-410. (Here is shown for
the first time the formulation of an individual palm by the course of the
four main lines, which are designated as A, B, C, and D.)

'04 Racial Differences in Palm and Sole Configuration. Amer. Anthropol.,
Vol. VI., pp. 244-293. (Gives condition, with comparison, of the palmar
and plantar configuration of Maya Indians, American Negroes, and American
Whites. The Maya prints were collected by Dr. A. M. Tozzer.)

'04 Duplicate Twins and Double Monsters. Amer. Journ. Anat., Vol. III.,

252 HARRIS HAWTHORNE WILDER.

PP- 387-472. (Continues the study of twins, with comparisons of their

palms and soles.)
'08 Zur korperlichen Identitat bei Zwillingen. Anat. Anz., Bd. XXXII.,

PP- 193-200. (Reports a case where an unusually complex palm pattern

is reproduced, with but slight changes, on all four hands. The four feet

are also duplicates.)
'12 Racial Differences in Palm and Sole Configuration: II. Palm and Sole

Prints of Liberian Natives. Amer. Anthrop., Vol. XV., pp. 189-207. (This

work is based upon 100 sets of prints, hands and feet, of Liberian soldiers,

collected by Prof. Frederick Starr.)
Windt, Camillo.'

'03 Ueber Dactyloscopic. Arch, filr Kriminalanthrop., Bd. XII., pp. 101-123.
Windt u. Kodiczec.

'04 Dactyloscopie. Vienna.
Yvert, A.

'05 Identificacion por las impressiones digitopalmares (La Dactiloscopie).

Tesis pres. en la Univ. Lyon. Publ. La Plata.

Bibliography of Polyembryony in the Armadillo; Chronologically Arranged.
Fernandez, M.

'09 Beitrage zur Embryologie der Glirteltiere. Zur Keimblatterinversion und
spezifischen Polyembryonie der Mulita (Tatusia hybrida). Morphol.
Jahrb., Bd. XXXIX., pp. 302-333.
Newman, H. H. and Patterson, H. H.

'09 A Case of Normal Identical Quadruplets in the Nine-Banded Armadillo, and
Its Bearing on the Problems of Identical Twins and Sex Determination.
BIOL. BULL., Vol. 17, pp. 181-187.

'10 Development of the Nine-banded Armadillo from the Primitive Streak Stage
to Birth; with Especial Reference to the Question of Specific polyembryony.
Journ. of Morphol., Vol. 21, pp. 359-423.
'n The Limits of Hereditary Control in Armadillo Quadruplets: a Study of

Blastogenic Variation. Journ. of Morphol., Vol. 22, pp. 855-926.
Patterson, J. T.

'12 A Preliminary Report on the Demonstration of Polyembryonie Development

of the Armadillo. Anat. Anz., Bd. 41, pp. 369-381.
Newman, H. H.
'13 The Natural History of the Nine-banded Armadillo of Texas. Amer.

Nat., Vol. 47, pp. 513-539.

'13 The Modes of Inheritance of Aggregates of Meristic (Integral) Variates in
the Polyembryonie Offspring of the Nine-banded Armadillo. Journ. Exper.
Zool., Vol. 15, pp. 145-192.
Patterson, J. T.

'13 Polyembryonie Development in Tatusia novemcincta. Journ. of Morphol.,

Vol. 24, pp. 559-662.
Newman, H. H.

'15 Heredity and Organic Symmetry in Armadillo Quadruplets. BIOL. BULL.,
Vol. 29, pp. 1-32.

Vol. XXX. April, 1916. No. 4

BIOLOGICAL BULLETIN

THE INFLUENCE OF THE NUCLEUS ON THE
BEHAVIOR OF AMGEBA.1

H. S. WILLIS.
INTRODUCTION.

OUTLINE.

PAGE.

Introduction 253

Material and Methods 255

Movement 256

Normal specimens 256

Fragments 256

Orientation in Light 263

Normal Specimens 263

Fragments 263

Rate of Locomotion 265

Normal Specimens 265

Fragments 265

Possible Causes of Differences in Behavior 267

Size of Fragments 267

Contractile Vacuole 267

Position of Segments in Original Amceba 268

Nucleus 268

The Influence of the Nucleus upon Attachment 268

Summary 269

Literature Cited 270

In the course of experimental work on Amceba in November,
1914, marked differences were observed in the behavior of the
different parts of specimens cut in two. At the suggestion of
Dr. S. O. Mast, an attempt has been made to ascertain whether
these differences depend upon the nucleus and, if so, in what
respects. I wish here to acknowledge my great indebtedness
to Dr. Mast for his helpful and constructive suggestions and for
his kindness in supervising both the experimental work and the
writing of this paper.

1 From the Zoological laboratory of the Johns Hopkins University.

253

254 H- s- WILLIS.

Gruber (1912) found that parts of Am&ba proteus containing
a nucleus behave very much like normal specimens, but that
parts without a nucleus behave very differently; yet, in spite of
this, he held that the nucleus in general has no influence upon
protoplasmic movement. Hofer (1890) in observations on the
same species, obtained results similar to those of Gruber. He
asserts (p. 118) that movements in the parts with the nucleus
are similar to those in normal specimens, as are also those in the
other parts for a period of 15-30 minutes after division, but that
later the movements in the parts without a nucleus differ from
those in normal specimens in rate of locomotion, in regularity
of movement, and in the number and length of pseudopods.

Hofer holds that these differences might be considered to be
due either to the injury received during the operation or to the
influence of the nucleus. But he maintains that, since any
injury sustained from the operation affects both fragments alike,
the operation could not be considered the cause of the observed
difference in behavior. He consequently concludes that the real
cause is to be found in the nucleus. He holds, however, that the
influence of the nucleus may be conceived to be direct or indirect;
that is, that behavior may be due to an impairing of the elemen-
tary functions (such as digestion, respiration, and excretion)
controlled by the nucleus. But Hofer found that the process
of digestion in the absence of the nucleus continues for several
days after division; that respiration takes place in the absence
of the nucleus; and that excretory functions in enucleated
segments continue till death. He therefore comes to the con-
clusion that the nucleus secretes a chemical substance and that
behavior is controlled through this. He thinks a certain amount
of this substance is stored up in the different parts of the proto-
plasm, and that the normal movement of the enucleated seg-
ments for 15-30 minutes after the operation is due to the influence
of the substance thus stored. Hofer maintains that, since
movement occurs in parts without a nucleus, cytoplasm has the
power of movement; but since the movement in these parts is
more irregular and haphazard than it is in nucleated parts, he
holds that the nucleus must have a regulatory function. In
other words, he thinks the nucleus serves as a regulatory "cen-
trum" for behavior.

INFLUENCE OF NUCLEUS ON BEHAVIOR OF AMCEBA. 255

Verworn (1909) agrees with Hofer in holding that there is a
difference in the behavior of parts of Amoeba with and parts
without a nucleus, but does not agree with him in the conclusion
that the nucleus exerts a direct influence on the movement;
that is, he does not think that the nucleus is a regulatory
"centrum" for movement.

Judging from the results of my experiments, it is clear that
there are distinct differences in the behavior of parts of Amoeba
which contain and parts which do not contain a nucleus. Dif-
ferences in such parts have been observed in the character of
movement, in the accuracy of orientation in light and in the
rate of locomotion. It is clear, also, that these differences are
traceable to the nucleus.

FIG. i. Camera sketches showing changes in form and the characteristic
movements in the nucleated and enucleated parts of an amoeba immediately
after being cut in two. The arrows indicate the direction of movement. 1-3,
enucleated part; a—k, nucleated part; mm, projected scale. The sketches of the
enucleated part were made at five-minute intervals; those of the nucleated part
at one minute intervals. Note that the nucleated part progressed regularly in a
given direction, and that the enucleated part changed its form and position only
slightly.

MATERIAL AND METHODS.

The specimens used for the most part in this wTork appeared
in a battery jar containing an old paramecium culture. They
corresponded to descriptions of Amceba proteus and were rather
sensitive to light. Nucleated and enucleated parts were ob-
tained by cutting specimens in two with finely drawn glass rods.

256 H. S. WILLIS.

The cutting was done under a binocular. The parts of each
individual were usually enclosed under a cover-glass by means
of a ridge of vaseline so applied along the edge as to support the
cover-glass and prevent drying. Thus the parts had entire
freedom of movement and both were continuously subjected to
the same environment; and, under these conditions, they were
studied, observations being made both with a compound micro-
scope and a binocular. Under the sealed cover-glasses the
divided amoebee did remarkably well, the nucleated parts living
on an average of approximately ten days, i. e., practically as
long as normal specimens under the same conditions, and the
enucleated parts about half as long. In the following descrip-
tions, the two parts will frequently be designated fragments or

segments.

MOVEMENT.

Normal Specimens. — In the process of locomotion in normal
individuals of the species studied, pseudopods usually appear
alternately on the two sides of the organism near the anterior
end. A pseudopod appears, for instance, on the right, elongates
and enlarges by the flow of protoplasm into it until it constitutes
the main portion of the animal ; then from this there is formed a
new pseudopod on the left side. This, in turn, elongates and
enlarges, after which a new pseudopod again forms on the right
side, etc. Thus the organism takes a zigzag course.

Fragments. — In general it was found that the movement in
fragments containing a nucleus is substantially like the movement
exhibited in normal specimens, and, in some instances, for a
period of 10-20 minutes after division, it was found to be similar
in enucleated parts. Usually, however, the movement is strik-
ingly different in such parts, it being slow and irregular, and
frequently accompanied by contractions. The pseudopods, all

FIG. 2. Series of camera sketches illustrating the difference in the movement
of nucleated and enucleated segments of Amceba. The former is shown in columns
A, the latter in columns B. The numerals between the columns indicate approxi-
mately the intervals of time, in hours, between the cutting of the amoeba and the
production of the adjoining sketches in the two columns. The two sketches, a
and b, in every case were made about one minute apart. They show the changes
in position and in form of the segments during this time. Whenever there is but
one sketch opposite the numerals, it indicates that there was no change in the
organism, a, first position; b, second position, arrows, direction of movement,
mm, projected scale.

INFLUENCE OF NUCLEUS ON BEHAVIOR OF AMCEBA.

A

B

A

257
B

105

I 20

Q

22.S

0. 4-

258 H. S. WILLIS.

of which are usually small, are not formed alternately as they
are in the parts containing the nucleus. Neither do they always
occur near the anterior end. These differences in movement in
the two segments are illustrated in Figs. I and 2.

The nucleated part represented in Fig. i formed pseudopods
within one minute after division. At three minutes it was
attached to the substratum and exhibited the coordinated
movement characteristic of a normal individual (Fig. i, a, b, c,
etc.). This fragment was observed every three to five minutes
for an hour, and the movement was found to be essentially the
same throughout the entire period. The enucleated part of this
individual on the contrary, became globular immediately and
remained so for approximately five minutes during which time a
number of small short pseudopods were directed outwards in all
directions from the body. These became fewer and larger, and
the entire body elongated in the direction of one of the larger
pseudopods. An attachment of the body to the substratum was
then formed and regular movement followed for about one
minute. There was no subsequent locomotion or "streaming
movement"; the organism, however, changed its shape by fre-
quent contractions of the cytoplasm. These changes continued
for an hour when the experiment was brought to a close (Fig. i ,
1-3). Throughout the entire period, the movement in the
enucleated part was very much slower than that in the nucleated
part.

Essentially all of these characteristic differences in the move-
ments of the two fragments in question are further elucidated in
the sketches reproduced in Fig. 2. By referring to this figure it
will be seen that there was considerable locomotion in the nu-
cleated part and that pseudopods were formed more or less
regularly on the opposite sides of the segment; while in the
enucleated segment there was extremely little locomotion and
pseudopods were formed irregularly.

Figs, i and 2 are typical illustrations of the behavior observed
in all of the specimens studied — forty in number. The movement
in all nucleated parts of these specimens was like that in normal
specimens and the movement in all enucleated parts was quite
different from that in normal specimens. The behavior of 17

INFLUENCE OF NUCLEUS ON BEHAVIOR OF AMCEBA. 259

pairs of these parts taken at random is briefly summarized in
Table I. By referring to this table, in which observations on
locomotion are recorded for 13 of the 17 individuals, it will be
seen that locomotion occurred in 13 of the nucleated parts, while
there was no locomotion in the enucleated parts. Other matters
in this table will be considered later.

1.5 9

FIG. 3. A series of camera sketches showing the reactions of a nucleated and
an enucleated part of an amceba in a horizontal beam of light, a-d, nucleated
part; 1.45-2.30 enucleated part; large arrows, beam of light; small arrows, direc-
tion of movement; 1.43-2.00, time at which sketches of nucleated part were made;
1.45-2.30, time at which sketches of enucleated part were made; mm., projected
scale. Note that the nucleated segment oriented fairly definitely; at a (1.43) the
rays of light were at right angles to the moving segment; at 1.44 the segment had
turned and become directed from the source of light. A similar response occurred
at b, c, and d after the direction of the rays of light had been changed in each case.
Note also that the enucleated part did not orient in the light. Both parts were
continuously in the same field and were subjected to the same changes in illumina-
tion.

260

H. S. WILLIS.

4>

•d

£

."2

C

CJ

•3

•a

•3
•d

3

^

c

3

CO

u
rt

rt

'co

flj

•d

O
Q.

IH

CJ

rt

CJ

4)

?

•a

rrt

CJ

^j

tn

-4-J

. fc

a

"rt

rt

CO

_v

CJ

B

CJ

^

3
C

1

C

o

CJ

—

*o

to

*j
"v"
O

CJ

"o

'rt

ft

41

;_

CJ

o

tn

^

•

S3

3

4>

r*»

CJ

^N

'"

*<5

u

rt

rt

UH
O

rt

IH

o

CJ

rr,

•<-*

J3

o

rt

CJ
O

-*->

CJ
C

•r
rt

a

T3

flj

.2

M

4.J

£H

-4-J

CJ

rt

ft

»*H

IH

^CJ

OJ

O

tn

1j

•7*

,f^

;_,

a

•a

rt

ti

s

rt
ft

•d

CJ
CJ

— '

CJ

.

o

CJ

1 •

"rt

"rt

HH

•~~.

TJ

—

W

03

K

^

M-<

O

tn

4->
S

3

4)

_,

*d
rt

^

ragments

M
rt

'— H

CJ

^J

~
4)

"o

pecimens

CO

o

M

'o

to

O

4>
'lo

i

13

S
o

CJ

•4J

S

|H

rt

rrt

4~>

IH

4)

CJ

tn

»^H

IH

to

^

•-H

CJ

CJ

tn

>>

"3

g

CJ

1

C

'o

CJ

-u

"S

CJ

a

o

CJ

CO

i .

CO

CO

CO

"cj

4_>

en

CJ

C

u

<L»

C
4O

•«-»
C

CJ

rt

a

CJ

T)
CJ

•§

O

a

specim

"3

4_)

to

^^

IH

CJ

CJ

.s

o

IH

S

rf]

^j

•a

t-l

£

•a

CJ

S

rt

O

•|

'S.

o
o

g
|

S

3

CJ

•—

_ —

o

"rt

•d

CO

CJ

3

rt
H

fragments

rt

4)

—

IH

to

.0
CO

41

not respon

m

1

_H -a -a ?

rt • rt i~d rt -^

Behavior.

Locomotion.

13

g.-i.l

.2 C J3 C '*J

*e o a o o

0- tn-- g

c o cj o o

*j s s s y

^3 0 rt 0 ^
M U CJ O _
S O CJ O O

en j« j ^

§ l §.§ e ^

§ 1 |s ! :i

*J E -" a •" -d 5
g b 8 ™ 8 g.,2
S I 6 | 6 o >,
oj a *> s w *o •?

§ g 5-2 § g^
S 5 §« § oS

1 rn f

i aj i i

•+J ^ qj

-U r- C, CJ p

< s

<3 *J *" ^ O

*d "°

p T) -d !s "p

c

§.! s

-i 0 P P > CJ

g M rt rt '-3 S

E

'5

'S rt c

'5 .S c c u •?,

3

'5

.-H OC JH O

> OJ S C .in

•^ j> o o cj i*

rt

•— ^< .2 ,j o .2

•~* o co co ij *J

u.

"O

*rt ^i • *^ *"^ 'rrt ^J

tn

•-* o to >

^= C .S .& ."I

J3

3
C/3
O
tn

1

"o

C

^c

4)

•4_>

'rt
C

e

M

T)

4J

Jr! . .- 'S '> -3

fi a "° '> =3 ^
rt -3 h • '-o w Si
. « | .S ' « «
.S «> «» S S . -js rt
Sfed^ .rt^d • -S

s^^lllSil1

= -S g^atJ S^ «£ «

fe^a"0 "° .s s

ii "S rt >-« *- P J3

T3 w (U 4> *H -*-)
w «; rt *J *J IH

60*013 "^ If 3

jj rt rt M «H

^ 1-ls ,.a S£
1 oj^e |s -a

S ^^G- 5^ -3.S

Ji

13

P«i

CJ

rt

4-»
•»->

rt
«j

rt

PH

•d^^-d^-o-d-a^T) .

CJ ,tn4)-t->OJCJCJ-tJCJO
j3-e,^^3 0 & jz ^ Oj3 tn
or^GoCoouCo—.
rty.Hrt_rtrtrt_rt^
-u c cJ_i>djj^j<_)'d-i-> cj
-t_> 2 '^ -*-* •—-*->•*-» -*-> •— -*-> p

<-g^<Q<;«Q<-3

rt-d-Oj2-'2-d .d-d .-dS

-UCJCJ ^CJO-^CJOCJ.

2 ^3 JJ "S J3 <« 2 -13 M -^ £

Poor; o— Po.^or
H rt rt ^H • rt j< _H rt ;P rt *_i
'd4->jjpui-t-<cj-d*jCjHjl5

•- « +J +J JJ 41 4-> C •-« 4-Jp*J™

Q«-6^<'rtQ<rt<^

s o e

Sis EO

<u

^S ^

i_ V 3

V*

N?> 5. -i2 v*

\M

o 'r"! ^

MS.

r-J\ M M ,;\

"\

-5 3^

M W M

Socn c

N

M

CJ

^*

^^

CJ l^

• • IhH bfl

|l

_M

W) 1 3 3

rt 1 1 -^ "^ *— *

1—1 rt O* O* C71

11 1 s

0*0* rt \M

"^ o

CO

_) W W W

W W _) -^

•O ui •

.s t: J

•d

p

•d ^-d^TJ-d-;^

g S gl g g P P

•d -d _: ^ "d ~i "d
p p ^ ^ c ^ p

CJ QJ " " CJ *~* flJ

3 ™ ,_

CJ

<L> CJ CJ CJ

^ ^ ^

° °<

*-< IH U U.

:2 s .2 S .2 .2 S .0

u, ;-. , )-. , U

.2 .2 o o .25 .2

i-* t-i T1 T* *-« "^ *H

crH 2

O QJ W

OJ

O *-"(yl-<cLiCJ!~<*~'

S S CJ 0 S OJ S

U5 -*->cfl'|-Jcnt/3-*-J-*-*

tn tn •*-* ~^ to •*-' to

" X *"" ^H

c

o cofioocn

o o c s o c o

£^.s

<tj

P-i ^p^^finpH-^^

PnPH< < PH< PH

W CO CO 09

tn tn to co

o

3 D -3 ^

33 33

,

CJ * CJ • CJ • QJ

i2 in

c t-1

QJ (^

CO

3

Tt w -7- cn -r-* tf) -T-! co

CJ r-f CJ -^ CJ ^ O ^

~ ^^/ii^fi

•Sg-O §0^-0

3 33 3 cj % £ 3

"o

•— i ^ i— i £ i— i c ~^

o o o u

CJ o O

O

3

O 3O3O^O=J

O 3 O 3 O 3 O

U

£

Z^lz; 2 ^Z g

Hog

O r- co

_W C-St,

oW

INFLUENCE OF NUCLEUS ON BEHAVIOR OF AMCEBA. 26l

C

d

^r

_o

u

Q

'o

.

o

c

S

C

O

C

s

ja

*4J

IH

*J"t

*p

IH

01

O

_S

o

o

ctf

PQ

g

3

g

3

o

M

o

0

M

o

QJ

o

0

O

o

|H
t-H

0

o

)-»

(H

HH

J, _C

C! C

—

_>> 0

1

c

i ja

j, S

A •o

d

o '3
cS M

£
cd

• S ^

u
ea

'3

M •

."j -4->

tJ 3

cj bo

"i g

*e

Relation of Parts to Substratum.

tached 2 min. after cutting. Att
; broken 5 min. later. Attached a

min. and remained so.
tached 2 min. after division.
:hed 5 min. later. Attached at 20 :

division. Unattached at 25 min.

0

T
d

d not attach for an hour,
d not attach for an hour.
;mained attached after first 10 mir
lattached for i hour when it slig
:hed. Remained attached 15 min.

icr attachment,
tached 5 min. after cutting. Att

; broken 3 min. later. Attached a
.in. after division, and remained so

-*-> -|_)

<«
O
. C!

Is

.£H *— *

"ca J5
c c
SB

>o ro

M ^

13 41
4i ^

•52

rt ^

i.
tached 5 min. after division. Att
; broken 3 min. later. Formed a

2 min. and remained so.
.tached 5 min. after division. Att
; broken 3 min. later; then for

i after 20 min. This was lost 2
No further attachment.

•*-• -"

< g

H "^ 4J

IH

1)

rt

s

Q Q PH ID _2

ti<

a s

4> Q

< S

d *^ &

4) <? C
-u ^" 41

Cfl ^3

g

fl 't~'

I4H

rt

0
IH

rt

3

M-i

C M

g

<A 2

•a s

C3 J3

OJ

_e ^

H-o 2

i_ V 3

o ^5 o

r-K

rnS

^C 3 X

M

M M

M

ro

bfj'J*! c

§ •-•

u

N

in 2

> rt

15

— *a

41

"3

'*j ^

3

TO 3

J ,

3

_rtu-.

o-

g o-

cS

o1

*°

W

w W

J

W

"C u> .
.i tl B

•d

•d

G

•§•§•§•§

•3
C

•d

•c
c

~6

QJ

O 3

QJ

4>

4) *" ^ D

UJ

QJ

41

O O ^

o *<:

IH

_o

IH

S .2 .2 S

IH

o

IH

_O

_o

'in

'*-!

C ~ o

41

IH

QJ i i _4_> 41

4)

41

IH

4)

41

III

OJ

O

4-*

*i 03 03 -y

d o o c

tn
O

fl

"c

tn

O

o x"

PH

<

<! PH PH <<

CH

"^

<

CH

to

to to

03

oi

U-i

o

3 3

3

3

.41 . 41

w

tn • — i to • — '

to

01

s "

QJ

3

3 y 3 y

QJ 3 Q) 3

3

O

3

O

c^

*u

C

73 « o e

"u

S

73

C

0

C

3 O 3 O

3

0

3

O

U

^

£

&

^

£

"oJ2 o

f| o2

0\

0 M

N

„

C ^ C 01

M M

M

M

'8 7 3

1 *"

262

H. S. WILLIS.

^3

8

'^»
**-^

S

3

w
J

ffl

$

c

u

S s -1

tt

Behavic

C >-i G C G *-> G

o -g .0 .2 .0 o _o

'Z! o *J *> -u G •"->

O <j O U O O O

I -g 1 1 |o |

O — U G cj u
« 0 0 0 C o

j § j u j^; j

Contraction

t i w J-t i hfl i )H

£J .§ ii •§ 15 § «2 | " .S (^ ^

•g ^ o II | ^ *§ I'l

^ ..S^>-i o;^ "o ._

a

M^C Mr;t2':" ^nx7

G .2 c "° •" >> -2 c .2 ^

rt

CA

^ '> M -| •§ , s 'E ^ 'E u

.0

S '-3 G '> n "^ S '-3 i-1 '-a 3

Cfi

^ •- nj «« _. i-
l- 1-lJcd 13 r< O>^ U, OJ Iij3

S £5^ G ** *• •" v

o

Ul

"C

il ilii> n , *73>-i *tj ^ ts "\

rt rtaJalii'rt 511 -a

CJ

OH

c cdi!'S^-o£2^'o .S .S c^

1+4

o

'§ 'c 'c iS S -2 JJ S "^ G 'S "S ca '§ -a

c
.2

^ H^.s|^|^"5l| 2 s^ •«!

J3

ET-! *^ *O ni

^j

u

. - . _• rt m ni nj )_( -^ fl * M ^ ^ *J

^

£H

•g s -g -g -g * -g g 5 £ o y -g M^i d"S^

s

< -5 < 2 3 cr< S 'S O 6 o < 'S 2 & "> < '3

1

S-^-'-'o i±3 ^x c '"^i fi
rtcSw CT3 rt^ C "O G

+j

03

o

ro « S- S-

j ^

<-.i! s

O T3 O

b£^/3 g

a o

3

0)

N

jj

c« «j

1; jj

U >i

b4 *

S 15 rt

— On

rt 33

rt i^.

. _ E o- cr

"5 0

«

r i r i

"° 2 n

ll|
b oj=

ors<

•a "2 "2 -6 "2 -6 "2 -a

gC G ^G^G
1> CJ O O

g.2 .2 S .2 o.2 S

cr-1 S

'C aj S *S a; '*-• aJ '•-•

O n; rf

j*j I1 ^_j C) 4_> QJ

'" -C C

-WOT m -tJ f)timti

•55 *•£

GO O O C O H

&£*

<U H, fe < PH <A, <

to to to to

"o

3 333
aj . cu o . i>

</> ^

c t:

3^i 3 H 3^3^

2 «

QJ ^J QJ -^ 2j fH ^ ^H

CPn

'o *o *"* *u *"* "u

O

3O 3 O 3O30

U

^ fc ^ ^ ^ !2Z fc

^J2 o

M M M W

c§:=

.22^ „:

«.£ £ £

c-o c rt

bu c-SfL,

'S1"1 -

5|c3

INFLUENCE OF NUCLEUS ON BEHAVIOR OF AMCEBA. 263

It has been thus shown that there is a difference in the general
movements of the segments containing and those not containing
a nucleus. Attention is now directed to the reactions to light
in such segments, especially the reactions resulting in orientation.

ORIENTATION IN LIGHT.

Normal Specimens. — It is well known that certain species of
Amceba when placed in a horizontal beam of light, usually turn
until they are directed from the source of light and then continue
in a fairly direct course; that is, they orient and are negative.
If now the position of the source of stimulation is changed so as
to illuminate the organisms from the side, they again turn until
they are directed from the light, that is, they reorient. In the
form under consideration the reactions to light are marked and
orientation is fairly precise.

Fragments. — Fragments containing a nucleus present precisely
the same kind of reaction as do normal specimens when subjected
to similar light conditions. These parts react readily to light;
the movement is rapid; and the process of orientation is very
much like that found in intact specimens. On the other hand,
enucleated parts usually give no evidence whatever of orientation.
Experiments were made on 25 pairs of segments. In 21 of these
the nucleated parts oriented approximately as precisely as normal
individuals, but in the enucleated parts no indication of orienta-
tion whatever was observed except in three instances, and in
these there was only a mere suggestion of orientation. This
will be described in more detail later.

Typical of the reactions in the 25 experiments mentioned above
are the responses to light of the parts shown in Fig. 3. The
reactions here represented, however, are somewhat more exact
and definite than those observed in some of the other segments.
In this case the enucleated fragment was the larger of the two
and contained the contractile vacuole. Both segments were kept
under the same cover-glass and were subjected to the same
light conditions. A 165-0. P. tungsten lamp was placed 13 cm.
distant from the reacting segment, and a piece of colorless glass
absorbed the heat waves from the lamp. A compound micro-
scope and a camera lucida were used. The temperature at the

264 H. S. WILLIS.

microscope was 29.5° C. The results obtained are in part re-
corded in Fig. 3. It may be observed fronTthis figure that the
nucleated part moved from the light and that with each change
in the direction of the rays of light, there was a corresponding
change in the direction of the movement of the segment, while
on the other hand, there was no indication of orientation in the
enucleated fragment. The two parts were subjected to precisely
the same conditions throughout the entire experiment.

As previously stated, there was usually no indication whatever
of orientation in enucleated segments, and it is questionable
whether any of these parts actually oriented, yet, in a number
of cases, slight movement from the light occurred, and in three
cases there was a change in the direction of movement with a
change in the direction of the light. This movement from the
light may have been a reaction to the light in the form of orienta-
tion or it may have been merely an accidental movement made
without regard to the light. The facts regarding the three cases
are these. One of the segments, while moving at right angles
to the light, formed a pseudopod on the shaded side, and move-
ment occurred in the direction of the pseudopod. With a change
in the direction of the light, a short pseudopod was formed again
on the shaded side, but at this point the fragment assumed a
globular shape. In another segment movement occurred for
several seconds at right angles to a horizontal beam of light, then
the direction was changed, and the fragment moved from the
source of stimulation. Upon a change in the direction of the
light, this fragment changed the direction of its course and again
moved from the light for approximately one minute and then
became globular. In the third enucleated segment substantially
the same reaction was observed. If these fragments actually
oriented in response to the light, the process of orientation in

•

them was essentially different from that in nucleated fragments.
There is, therefore, a difference in nucleated and enucleated
parts of Amoeba in their response to light as well as in the char-
acter of their movements; but there is also a difference in the
rate of locomotion as will be demonstrated presently.

INFLUENCE OF NUCLEUS ON BEHAVIOR OF AMCEBA. 265

RATE OF LOCOMOTION.

Normal Specimens. — The rate of locomotion varies greatly in
different individuals. Some specimens move nearly twice as
fast as others. Quite a number of individuals were observed to
move rather consistently at 0.27-0.3 mm. per minute; others
moved at an average rate of 0.12-0.15 mm. per minute. There
is a fairly definite relation between the size of the moving organ-
ism and the rate of movement. An animal travels a distance
that is approximately two thirds the length of its body in one
minute. The actual distance traveled by a large animal is
greater for a given length of time than the actual distance traveled
by a small animal for the same length of time under the same
conditions. This is clearly demonstrated by the results recorded
in Table II. By referring to this table it will be seen that the
ratio of the distance traveled per minute to the length of the
body of the specimen is approximately the same in all cases.

TABLE II.

Relation between size of Amoeba and rate of locomotion. The average rate of
locomotion was computed from five readings extending one minute each. The
average length was computed from five measurements of the moving organism
at intervals of one minute each. The rate was measured by projecting the moving
organism on a scale with a camera lucida.

Ratio of Distance Traveled

Maximum Length of Specimens Average Distance Traveled per Minute to Length

Observed. per Minute. of Body.

0.36 mm. 0.24 mm. 0.66

0.34 0.20 0.59

0.35 " 0.23 " 0.65

0.37 " 0.24 " 0.65

O.2I " 0.15 " O.?!

0.26 " 0.17 " 0.64

O.2O " 0.14 " O.72

0.26 " 0.17 " 0.64

Fragments. — In the work on the rate of locomotion in frag-
ments, the two parts were on an average approximately the same
size. Observations were made on each pair of segments at about
the same time. Both were kept under the same cover-glass
and in the same environment. It was found that the rate of
movement in nucleated fragments, like that in normal specimens,
bears a definite relation to the size of the body. Large nucleated

266

H. S. WILLIS.

segments usually travel greater distances in a given length of
time than do small ones. This is shown in the results recorded
in Table III. In this table are given the lengths of four intact
specimens together with the rate of locomotion per minute, and
the length of body, and the rate of locomotion in the nucleated
and enucleated parts cut from the four intact specimens.

TABLE III.

Relation between rate of locomotion and length of body in four intact specimens
of Amoeba and in the nucleated and enucleated parts of these four specimens.

Normal Specimen.

Nucleated Part.

Enucleated Part.

Average
Length of
Body.

Rate of Loco-
motion per
Minute.

Average
Length of
Body.

Rate of Loco-
motion per
Minute.

Average
Length of
Body.

Rate of Loco-
motion per
Minute.

.376 mm.
•355 "
•340 "
.360 "

.244 mm.
.230 "
.200
.240 "

.275 mm.

.210 "
.220 "
.200 "

. I^O mm.
• ISO "
.140 "
.128 "

globular

.230 mm.
.230 "

negligible

?

.074 mm.

The table shows that the larger intact specimens and the
larger nucleated parts travel greater distances per minute than
do the smaller ones. It also shows that locomotion in the
enucleated parts is very much slower than it is in the nucleated
parts. This is, however, more clearly shown in Table IV in

TABLE IV.

A comparison of the rate of locomotion in millimeters per minute in the nucleated
and the enucleated parts of ten different individuals.

Nucleated Parts. Enucleated Parts.

O.I5 O.OO2

0.13 0.074

0.13 o.oio

0.18 0.020

O.I3 O.O08

0.15 o.no

0.16 o.ooo

0.19 0.006

0.17 0.006

0.13 0.004

which the rate of locomotion in nucleated and enucleated
segments of the same individual is compared. Each number in
the two columns of this table represents the average rate of

INFLUENCE OF NUCLEUS ON BEHAVIOR OF AMCEBA. 267

locomotion for five trials, each continuing for one minute. The
figures in the left column show the rate of locomotion in the
nucleated parts; those figures immediately opposite, in the other
column, show the rate of locomotion in enucleated parts. For
instance, the nucleated segment of the individual first cut moved
in five trials at an average rate of 0.15 mm. per minute, and the
enucleated part cut from the same individual moved at an
average rate of 0.002 mm. per minute. From the table one
thing is strikingly evident: movement in nucleated parts is
always much more rapid than that in enucleated parts; in no
case does the rate of locomotion in parts without a nucleus equal
that in parts with a nucleus.

POSSIBLE CAUSES OF DIFFERENCES IN BEHAVIOR.

From the experiments cited above, it is evident that there
are differences in the behavior of parts of Amoeba with a nucleus
and parts without a nucleus; differences in the character of
movement, in orientation, and in the rate of locomotion. Now,
to what may these differences be attributed? The two parts
differed in size, in their position in the amceba cut to produce
them, in the possession of a contractile vacuole and a nucleus.
The difference observed in their behavior must be ascribed to
some of these differences in structure.

Size. — In this work nearly 125 amoebae were cut, each into two
parts, one of which contained a nucleus. In approximately one
half of these cases, the part without a nucleus was as large as or
larger than the part with a nucleus. Records of the two parts
of 13 of these amoebae, taken at random, may be found in Table I.
Both parts in all cases were kept under the same cover-glass and
were subjected to the same or similar conditions. All of the
nucleated parts behaved like normal specimens, but none of the
enucleated parts, regardless of the relative size of the two parts.
It appears, therefore, that differences in the size of parts do not
determine the differences in behavior.

Contractile Vacuole. — The contractile vacuole was present in
the nucleated parts of the 125 specimens experimented upon
about as often as it was in the enucleated parts. There was no
observable difference in behavior associated with the presence or

268 H. S. WILLIS.

absence of the vacuole. Enucleated parts with the vacuole
apeared to behave in every way similarly to enucleated parts
without a vacuole. The same is true for nucleated parts. The
vacuole, therefore, cannot be reckoned as a determining factor
in the behavior of the fragments.

Position Occupied by Fragments in the Intact Amceba. — When
an amoeba is cut into two parts, the part containing the nucleus
may be from what was the anterior end or the posterior end of
the original specimen. The behavior was carefully studied in
25 parts, 13 from the anterior and 12 from the posterior end of
the intact individual. The results obtained are recorded in Table
I. No evidence was obtained indicating that the reaction of the
parts depends upon their location in the specimens from which
they were taken. The nucleated parts from the anterior end
responded precisely like those from the posterior end. Ob-
viously, then, the position in the intact specimen is not a
determining factor in the behavior of fragments.

The Nucleus. — We have demonstrated that parts of Amceba
which contain a nucleus behave essentially like normal specimens,
while those which do not contain a nucleus behave -quite differ-
ently, and that this difference in the behavior of the parts is
dependent upon neither their relative size, the presence of the
contractile vacuole, nor the location of the parts in the animals
which were cut to produce them. It must, therefore, in some
way, be related to the nucleus. This will be considered in the
following paragraphs.

THE INFLUENCE OF THE NUCLEUS UPON ATTACHMENT.
Dellenger (1906) made it clear that attachment always accom-
panies and is essential to efficient locomotion in Amceba and
Difflugia. Twenty-five of my experiments on fragments of
Amceba, in which observations were made on attachment,
demonstrate quite clearly that nucleated fragments are usually
continuously attached to the substratum, and that enucleated
fragments are rarely attached. In these experiments some
specimens were observed for only ten minutes; others for a very
much longer period of time, the maximum being 72 hours.
Records of attachment or unattachment in all were made at
definite intervals. The following detailed description of the

INFLUENCE OF NUCLEUS ON BEHAVIOR OF AMCEBA. 269

results obtained in the study of one pair of segments taken from
the same specimen is typical of all. In this case the parts were
approximately equal in size and the nucleated part was cut from
the anterior end of the amoeba.

At five minutes after division both fragments were attached
and moving. Three minutes later the attachment in both was
broken by the sliding of a small glass rod under them. The
nucleated part began immediately to send out pseudopods aim-
lessly. This continued for two minutes; then the segment
attached and remained so for three hours — when the experiment
was brought to a close. This segment was attached continu-
ously to the substratum; it appeared to be attached usually at
three different points and the behavior appeared to be normal in
every respect. The enucleated segment was momentarily at-
tached at several different times, and each time the attachment
was so w^eak that it could not resist even the slightest jar given
the table on which the experiment was made. At no time was
this segment continuously attached longer than two minutes
and it was never attached at more than one point at a time.
There was a slow streaming motion in the protoplasm but no
locomotion at all except for a short time during its first attach-
ment and then it was very slight. The results obtained in all
of the observations on the attachment of segments of Amoeba
are briefly summarized in Table I. By referring to this table it
will be seen that attachment of nucleated parts to the substratum
is continuous; that attachment of enucleated parts is intermittent,
slight and of short duration.

These facts seem to indicate that the nucleus directly in-
fluences the attachment of protoplasm to the substratum and

thus influences behavior.

SUMMARY.

1. Amceba proteus moves regularly and smoothly by alternate
formation of pseudopods on the two sides of the organism.
Locomotion in segments of Amoeba with a nucleus is of the same
general character. Movement in segments without a nucleus,
however, is irregular, jerky, very much slower than that in
nucleated parts, and the pseudopods are not ordinarily formed
regularly or alternately.

2. In a horizontal beam of light, normal specimens direct

270 H. S. WILLIS.

their locomotion from the source of stimulation, i. e., they orient
and are negative. Parts containing a nucleus respond in the
same manner; those without a nucleus, however, do not orient.

3. The rate of locomotion varies greatly in different indi-
viduals. Large specimens move more rapidly than small speci-
mens, the rate of locomotion bearing a fairly definite ratio to the
size of the specimen. The rate of locomotion in nucleated seg-
ments bears essentially the same ratio to their size as it does in
normal individuals. Segments without a nucleus show very little
locomotion and this is always relatively very slow and irregular.

4. The size of the parts, the contractile vacuole, and the
position which the parts occupied in the intact specimens before
division seem to be in no way responsible for differences in the
behavior of nucleated and enucleated parts of Amceba.

5. The only other known respect — aside from those mentioned
—in which the two parts differ concerns the nucleus. Conse-
quently the differences in the behavior of these parts are, in all
probability, in some way related to the nucleus.

6. The regulatory influence of the nucleus on behavior in
Amceba seems to be brought about by some sort of an influence
upon the attachment of the organism to the substratum.

LITERATURE CITED.
Dellenger, O. P.

'06 Locomotion of Amcebae and Allied Forms. Jour. Exp. Zool., Vol. 3, pp.

337-358.
Gruber, K.

'12 Biologische und experimentelle Untersuchungen an Amceba proteus.

Arch. f. Protistenk., Bd. 25, S. 316-376.
Hofer, Bruno.

'90 Einfluss des Kerns auf das Protoplasmas, Jena. Zeit. f. Nat., Bd. 2%, S.

I05-I75-
Jennings, H. S.

'04 Contributions to the Study of the Behavior of Lower Organisms. Carnegie

Inst. of Washington, Pub. No. 16, 256 pp.
'06 Behavior of Lower Organisms. New York. 366 pp.
Mast, S. O.

'n Light and the Behavior of Organisms. New York. 410 pp.
'10 Reactions in Amceba to Light. Jour. Exp. Zool., Vol. 9, pp. 265-277.
Stole, A.

'10 Uber kernlose Individuen und kernlose Teile von Amceba proteus. Arch.

f. Entom., Bd. 29, S. 152-168.
Verworn, Max.

'09 Allgemeine Physiologic, 700 pp.

NOTES ON SUPERFETATION AND DEFERRED
FERTILIZATION AMONG MICE.

F. B. SUMNER,
SCRIPPS INSTITUTION FOR BIOLOGICAL RESEARCH, LA JOLLA, CALIF.

The phenomenon of superfetation, or the occurrence of a
second conception in a mammal already pregnant, has been
reported and discussed by various writers since the time of
Aristotle. The latter1 refers specifically, in this connection,
only to man and the hare. Modern writers2 have described
occasional instances for several animals (man, cat, rat, sheep).
On the other hand, this phenomenon appears to have been
generally regarded as a biological curiosity, while its very
existence has been questioned by some,3 the facts observed being
explained on other grounds. Apparently, it has been very
rarely observed in the rat, for Miss King records only two cases
among the more than 700 normal broods born in her stock.4
For mice, I have thus far found no published records of super-
numerary broods, though two well-known zoologists, who have
reared varieties of the house mouse in large numbers, assure me
that they have encountered cases of this sort. In the absence
of exact records, I am unable, however, to report these instances
in detail.

In the course of several years' experience in breeding \vhite
mice, the present writer has occasionally noted the birth of a
second litter, in cases where an earlier litter had been kept with
the mother for some time after weaning. This was at first
attributed to the precocious sexual development of the earlier
male offspring, one of which was held to be the father of the later

1 See bibliographic references.

2 Marshall, 1910; King, 1913 (also papers cited by the latter).

3 Schultze, 1866; Godlewski, 1914.

4 1 do not here include those cases (known as superfecundation), in which
members of the same litter, derived from the same period of ovulation, are born
one or two days apart. 1, have myself observed a few cases of this phenomenon,
but shall not discuss them here.

271

272 F. B. SUMNER.

brood. In one case of which I have record (Table II., No. i),
this may well have been the true explanation, since the second
conception must have occurred when the three males which
constituted the first brood were about 42 days old. That male,
and even female, white mice may become sexually mature at
this age I am certain from independent evidence. I have
record of at least one case in which a female, mated to a male of
the same brood, became pregnant at an age of about 6 weeks
(between 38 and 44 days).

In one instance, however (Table I., no. 6) such an explanation
was not applicable, owing to the extreme youth of the first litter,
at the time of the second conception. On March u, 1909, a
female gave birth to a brood of two (i cf, i 9). On April 8,
or 28 days later, a second brood of seven was born. Since the
period of gestation of this animal is normally about 20 days,1
the mice of the first litter were scarcely more than a week old
when the second lot began their development. It is needless to
point out that a blind and helpless young male of this age could
not have played the part of a father. Furthermore, the female
had been separated from all other adults of either sex, since the
day of the birth of the first brood of young, 28 days previously.
Under these circumstances, to be sure, the possibility is not
excluded that the second litter resulted from a copulation
occurring immediately after the birth of the first. In this event,
the only anomaly would be the deferred birth of the later brood
of young.

Unfortunately, no further notes on this subject were made
during my experiments on white mice. It is my impression
that other instances similar to the foregoing were observed, but
I find record of only these two cases.

It is the chief object of the present paper to furnish data
relating to superfetation and deferred fertilization which have
been recorded incidentally during the past two years, in the course
of rearing wild mice of the genus Peromyscus. Unfortunately,
my data are in some cases somewhat equivocal, since the experi-
ments were not made with a view to studying these particular
problems. But definite answers can none the less be given to
certain questions, even if not to others.

1 My own observations confirm those of Daniel (1910) on this point.

SUPERFETATION AMONG MICE. 273

My stock comprises four geographical races of Peromyscus
maniculatus , viz., rubidus, sonoriensis, and two quite distinguish-
able races within the assemblage generally known as " gambeli."

In breeding these mice, it is my general practice to place one
male with three to five females. Each male is left with the
respective group of females for a varying period, which ordinarily
does not exceed 20 days. Record is in every case kept of the
day when the male is admitted, and of the day when he is
removed. Each female, when obviously pregnant, is isolated in
a separate cage. The young are ordinarily left with the mother
until they are 6 to 8 weeks old. At this time, they are commonly
separated according to sex, the males and females being thence-
forth kept apart until they are mated, according to the require-
ments of the work. Occasionally, this separating is not done
early enough, however, and matings between brother and sister,
or between mother and son, occur. The youngest mice in which
I have known fertilization to take place were both 42 days old
(possibly a few days younger), the next youngest pair were 55
days old or less. Because of these precocious, and therefore,
unexpected conceptions, the exact date of birth of the resulting
young cannot always be recorded. They were sometimes several
days old when first found. Now it is among very young parents
that the phenomena to be discussed seem to be commonest,
as will be pointed out presently. For this reason, it happens
that in all but two among the eight cases comprised in my first
table the father was still present in the cage, and had access to
the mother up to the time of the isolation of the latter, when her
brood was first discovered. This uncertainty as to the date of
the last coition somewhat complicates the interpretation of
these cases, though, as will be shown, the evidence, even here,
for a deferred fertilization of the ova is not thereby affected.

Before considering the phenomena presented by the tables, it
will be well to make inquiry as to the normal period of gestation
in Peromyscus. Since I have been interested only incidentally
in determining this period, I have in no case restricted the mating
of the parents to a single observed copulation, as has Daniel
(1910), nor have I determined the exact hour of the birth of the
brood, as have Long .and Mark (1911). I have recorded, how-

274 F- B- SUMNER.

ever, the period during which the male had access to the female,
and the date of discovery o.f the brood. During the times
when frequent births were expected, the nests were examined
daily. Hence a given birth must commonly have occurred
between two known observations. It has been my invariable
practice to refer the birth to the day when the brood was dis-
covered,1 except where circumstances were such as to make it
probable that the young were bprn a day or two earlier. In such
cases, this probability is indicated in the records.

Since, in a large series of experiments, it is probable that
successful insemination would occur in some cases pretty soon
after the males were introduced into the cages, it seems likely
that the minimum interval between the introduction of a male
and the birth of a brood represents the actual period of gestation,
at least for the individuals in question. This minimum was
found to be 22 days. In all but five cases, however, out of a
total of 206 broods, for which I have accurate records, the figure
was greater than this. As a control, I have determined, for each
series, the maximum interval between the removal of the male
and the discovery of a brood of young. This maximum figure
is also a fraction over 22 days. In the single instance recorded,
a brood was known to be born as late as some time during the
23d day after the last opportunity for copulation.

From these data, therefore, we know that the period of gesta-
tion of Peromyscus does not, in some cases at least, exceed 22
days. We also know that it may be as much as 22 days. Ac-
cordingly, I have provisionally adopted this figure as representing
the normal term for the species which I am engaged in breeding.
It is likely, however, that this term is subject to some variation.

Tables I. and II. comprise two different classes of cases. In
the former, the interval between the birth of the first and second
litters ranges from 13 to 39 days. In all of these cases the youth
of the first brood of young, at the time of the second conception,
precludes the possibility that a male of the former lot was the
father of the second. Even when the interval was as great as
39 days, conception must have occurred when the first lot was

1 Mice (at least white mice) are said by Long and Mark (1911) to be more
commonly born in the early morning than at any other time of day.

SUPERFETATION AMONG MICE.

275

D
£

O

H
W

W
S3
H

tf
W

S

<
fe

W

g

U

U
M

Q

3 ^

tt o

U

1) •*-• "^

W £ £

-H

s

k, M

o o -H

/) O

•O T3 T3 T3 g T3

Mi

.

O

J3

I

U

a
O

o
o o " "b

^o,

10

to

0\

o\-
o

_ 10 10

M O\ CN

)H OO

» C

— «^ t

M

W

X

o) <u rt
_J CQ_)

-

12 s

too r» oo

2/6 F. B. SUMNER.

not over 17 days old, i. e., when the young animals had just
opened their eyes. At 4 and 6 days, the young are, of course,
blind and helpless. Indeed, in two cases — were further proof
needed — it will be noted that the first brood contained no males,
while in still another, they had not yet been born.

In cases of the second class (Table II.), the first and second
litters were born 63 to 8 1 days apart, i. e., the second conception
occurred when the first brood — which in every case contained a
male — was 41 to 59 days old. In these cases, the possibility is
by no means excluded that a member of the first litter served as
the father of the second. As already stated, there is positive
evidence, both in the white mouse and in Peromyscus, that
fertilization may occur at an age of six weeks. On the other
hand, such precocity is quite exceptional, so that this explanation
of the cases in Table II. seems hardly more probable than that
by which we must explain those in Table I., viz., fertilization by
spermatozoa received from the earlier mate. In the absence of
definite evidence, however, it would be idle to speculate regarding
the relative probability of these two interpretations, and I shall,
accordingly, exclude the cases in Table II. from most of the
ensuing discussion. Nevertheless, it is worth pointing out—
what is not indicated in the table — that the first births recorded
in cases 3 and 4 of Table I. are the same as the second births
recorded in cases 5 and 9 of Table II. If, therefore, these latter
are likewise instances of deferred fertilization, we must assume
for these two mothers the delivery of three consecutive broods,
after separation from their mates. The possibility of such a sur-
prising occurrence surely warrants more carefully controlled
experiments in the future.

In the more familiar discussions of superfetation, it appears
to be invariably assumed that the later fetuses owe their existence
to coition occurring during the first pregnancy. This assumption
is made by Aristotle, in the passages referred to. Marshall
(1910, p. 159) says: "If ovulation takes place during pregnancy,
and if, owing to the occurrence of coition the ova become fer-
tilized, the phenomenon of superfoetation may take place — that
is to say, foetuses of different ages may be present in the same
uterus."

SUPERFETATION AMONG MICE.

277

PQ

Q
fc
O
U

w

B
H

§

w
M

H

a

H

H
(4
M

H
X

H

a

y

s
^

z

HH

tn

M
in

U

v a '"
rj O £

rt Sri.tJ

c ¥ "
3^

0 u

•S £

°§

4> O

o S

rt ,9

n

U

2

o

o JJ "a

-H o

-H

(n

to

1)

LH
O

O O O

\O

OO O OO O

IT) IT) \f> in

^f

ON

\O

b s b*

3 3 3

b
3 v

o

2j8 F. B. SUMNER.

Miss King (1913, p. 385) remarks that "superfoetation . . .
is applied to cases in which ovulation, followed by copulation,
occurred during pregnancy and led to the simultaneous develop-
ment in the uterus of two sets of ova belonging to different
periods of ovulation." In discussing the two instances observed
by her in the rat she says: "The most plausible explanation for
these cases seems to me to assume that the two ovaries acted
independently, ovulation occurring in one ovary some little
time before it took place in the other. If copulation followed
each ovulation two sets of embryos would develop in the uterus
simultaneously, and they would be born at different times,
depending on the interval between the two periods of ovulation"

(p. 390).

I believe, with the writers quoted, and in opposition to certain
others, that these younger fetuses owe their origin to later periods
of ovulation, which may even occur during gestation. I cannot
see the necessity, however, for assuming an independent action
of the two ovaries, as does Miss King. Furthermore, I regard
it as highly improbable that some of the cases which I have
recorded resulted from a copulation occurring at the time of
the later ovulation, whether this last took place during pregnancy
or after the birth of the first brood.

My own observations point, with considerable probability, to two
facts: (i) to a definite periodicity in ovulation, continuing in some
cases throughout pregnancy; and (2), with even greater probability,
to the retention by the spermatozoa of their fertilizing power for days
or even weeks after reception into the uterus or fallopian tubes.

In support of the first proposition I will point to the rather
surprising numerical relations among the figures (Table I.)
representing the intervals between the delivery of the two broods
by the same mother. These eight numbers are all nearly or
quite multiples of 13, viz:

1 X 13 (±)

2 X 13

2 X 13, (+ I?)
2 X 13, (+ I?)

SUPERFETATION AMONG MICE. 2J9

2 X 13, (+2?)

2 X 13, (+2)

3 X 13, (-2?)

3 X 13, (±)

It will be seen that some multiple of a quantity lying between
13 and 14 would fit every case with the exception of one, in
which the figure may have been slightly less than 13. Further-
more, let us add Miss King's two cases for the rat. In one of
these, the second-born brood was delivered "about 14 days"
(but not more than 14 days) after the first. In the other
case, the second delivery was, she says, "about 12 days" after
the first. It must be mentioned, however, that this "newborn"
brood was actually found 13 days after the other.

Here, then, we have altogether 3 instances in which the broods
were born about 13 days apart, 5 instances in which the interval
was about twice as great, and 2 in which the interval was about
3 times as great. Statistically speaking, the case is not, of course,
entirely convincing, and later findings may fail to conform to
the scheme here indicated. But the probability seems to be
sufficiently high to warrant its provisional acceptance.

Referring to the second table, it will be found that the inter-
vals therein recorded could not all be interpreted as multiples of
any one number, though four of them are not far from being
multiples of 13. Were one disposed to have recourse to a not
uncommon type of argument, he could point out that the normal
variability of the single intervals might reasonably be expected
to lead to a cumulative divergence in the course of five or six
periods. But most of us would be reluctant to draw any con-
clusions from this second set of figures.

Regarding the existence of a definite periodicity in the ovula-
tion of mice, there appears to be some difference of opinion.
Most recent observers seem to agree that ovulation occurs
independently of coition in the mouse and rat. Sobotta (1895),
Kirkham (1910, 1913), and Long and Mark (1911) hold that it
takes place not long after parturition. According to the latter
authors the interval between parturition and the ensuing ovula-

28O F. B. SUMNER.

tion ranges from 14! to 28| hours. Their observations have
thrown little light, however, upon the periodicity of ovulation
at other times, though they do not seem to believe that this is
governed by any great regularity. Sobotta (1895) states that
ovulation recurs 21 days after parturition, basing his conclusions
upon the examination of several females. He regards as im-
probable the occurrence of intermediate periods, at intervals
smaller than this.

On the other hand, it is stated by Heape (1900) that "the
usual length of the dicestrous cycle for rodents is ten to twenty
days," that of the rat and mouse (in England) being given as
approximately ten days (p. 26). Since it is also stated (Marshall,
p. 135) that "in the mouse, the rat, and the guinea-pig, ovulation
occurs spontaneously during 'heat,' and generally, if not in-
variably, during oestrus," it would follow that in these animals,
ovulation likewise occurs at intervals of about ten days. This
figure approximates much more nearly than Sobotta's to the
thirteen-day periods indicated by my observations. Presumably,
pregnancy would ordinarily interrupt these periods of ovulation,
and initiate a new cycle at its close, though the phenomena of
superfetation and deferred fertilization are pretty good evidence
that this is not necessarily the case.

If, as seems probable from my data, the second brood of off-
spring resulted from a later period of ovulation, during or after
the first pregnancy, we must suppose that the spermatozoa were
retained in an active condition for many days and in some cases
even for weeks. Fertilization was deferred until the arrival of
viable eggs in the fallopian tubes or the uterus. If we consider
the intervals (Table I.) between the last possible opportunity
for copulation and the birth of the second litter, we have:1 no. i :
13 days ±; no. 2: 26 days; no. 3: 27 days; no. 4: 27 days; no. 5:
47 days (±); no. 6: 28 days; no. 7: 37 days; no. 8: 36 days.

Thus in only a single case (the first) could copulation have
occurred as late as 22 days (i. e., the normal period of gestation)
before the birth of the second brood. In case no. 5, if we suppose

1 In all these cases except no. 5 and perhaps no. i, the first brood was born
unexpectedly, and it is likely that the male was still in the cage at the time of
delivery.

SUPERFETATION AMONG MICE. 28 1

that the fetuses developed at the normal rate, at least one sper-
matozoon must have retained its fertilizing power for 25 days or
more. In cases 7 and 8, even if we admit the probability here
of coition after parturition, the spermatozoa must have remained
alive for 15 and 14 days respectively.

An alternative hypothesis would be to assume that gestation
rather than fertilization had been retarded in these cases. Thus
Schultze (1866), while not denying the theoretical possibility of
superfetation, explains all of the reported cases in man on the
basis of either the death or the retardation of the less advanced
fetus. This would also seem to be the opinion of Godlewski
(1914), who, after pointing out what he believes to be the physio-
logical impossibility of this process, makes the following dogmatic
assertion: "Wir mussen demnach annehmen, dass die Eier,
welche den Mehrgeburten den Ursprung geben, gleichzeitig
befruchtet werden" (781). On this assumption, we should have
to suppose that, at least in cases I and 5, the ova which gave
rise to the second brood had been liberated and fertilized simul-
taneously with those which gave rise to the first. In the remain-
ing cases, on the other hand, they may have been liberated and
fertilized after the birth of the first brood.

Now Daniel (1910) has shown for white mice and Miss King
(1913) for white rats that parturition may be considerably
deferred when the mother is still suckling an earlier brood. This
retardation was found, in some cases, to be as much as 10 days
for the mouse and 12 days for the rat. Both of these writers
assume that fertilization ensues fairly promptly after successful
insemination, and that the delay in parturition is due entirely to
a prolongation of embryonic development.

If we attempt to apply this interpretation to some of the
present examples, we are led into difficulties even greater than
those which confront the hypothesis of deferred fertilization.
For example, in case 5 (Table I.), the last opportunity for
copulation was 19 days before the birth of the first litter, and
about 47 days before the birth of the second. Even on the
assumption that the fertilization of the second lot of eggs actually
took place as late as this last day, the period of gestation was
prolonged by about 25 days, i. e., the normal term must have

282 F. B. SUMNER.

been more than doubled. Now the first brood consisted of only
two individuals, so that any retarding effect due to lactation
would have been almost at a minimum. Daniel found that this
amounted to only about a day for each young mouse suckled.1

If it be contended that the retardation probably resulted from
the gestation of the first litter rather than their subsequent
nursing, it is relevant to point out that this species of Peromyscus
may give birth to 5, 6, or even 7 young, with little, if any,
prolongation of the period of gestation beyond the normal.2
Again, in cases 7 and 8, the birth of the second brood occurred
37 and 39 days respectively after the first. In both of these
cases it is possible that copulation occurred during or shortly
after the first pregnancy, since the male was not removed until
after the delivery of the first brood. Thus, at the least, we must
assume that the second gestation was prolonged more than two
weeks, owing to the suckling by the mother of her three earlier
young.

To me it seems much more credible that the cases here dis-
cussed resulted from the fertilization of recently liberated eggs
by spermatozoa which had been received many days previously.
I do not think it helps us at all to assume that the term of
development, after fertilization, was prolonged to any such
extent as we should here have to suppose. Nor do I believe
that it is necessary to assume a second copulation to account for
the later brood, though this possibility is not excluded in my
experiments. To refer again to case 5, of my first table, the last
opportunity for copulation was 19 days before the birth of the
first brood, i. e., only 3 days after the conception which initiated
the development of this brood. Is it not as easy to admit the
retention of living spermatozoa from this first effective copulation
as it is to admit their retention from another copulation occurring
only three days later?

At first glance, it is not clear why these cases of deferred
fertilization should, in every instance, have followed a previous
pregnancy. In other words, why should not this phenomenon

1 For rats, indeed, Miss King found that retardation was not certain to follow
unless more than five young were suckled.

2 One brood of 6 was born 23 days after the earliest possible fertilization, while
one brood of 6 and one of 7 were born 24 days after the earliest possible fertilization.

SUPERFETATION AMONG MICE. 283

have appeared in females which had earlier merely undergone
insemination, without a normal conception having first ensued?
As a matter of fact, I have never found an instance, among all the
births recorded, in which a first brood1 is known to have been
born more than 22 days after the last possible opportunity for
coition. Why should not the spermatozoa have occasionally
been held in reserve for a period of ovulation occurring some
time after this last opportunity for copulation? In reply, I can
only point out that coition probably occurs only when the
female is in a condition of "heat." If, on the advent of the first
"heat" period after the introduction of the male, fertilization
did not follow insemination, it would, in a large proportion of
cases, point to a sterility (temporary, at least) on the part of one
or both sexes. As a matter of fact, my records show that the
great majority of the females which conceived at all did so
during the first half of the sojourn of the male in the cage.2
As stated above, only one female in the course of my experiments
is known to have become pregnant as late as the last possible
date of copulation.

To what degree the term superf elation is applicable to the cases
comprised in my first table is a matter of definition. Case no. I
(as well as the two cases described by Miss King) would doubt-
less be covered by the term as ordinarily understood, since one
period of gestation, with little doubt, was superimposed upon
another. For reasons stated, it does not seem likely, however,
that two sets of developing fetuses were simultaneously present
in any of the other cases. But in at least one of these instances
(no. 5), the sperm which fertilized the second lot of eggs must
have been received during or prior to the first pregnancy. Per-
haps the term superfetation might conveniently be extended so
as to cover such cases as these last. But the word is plainly
inapplicable to instances in which the second effective copulation
occurred .subsequently to the delivery of the first brood.

Admitting the last named possibility for most of the cases
comprised in my Table I, we none the less find in them instances

1 First, as distinguished from the supernumerary broods here discussed. I do
not refer necessarily to the first offspring of a given mother.

2 This commonly covered 20 days.

284 F. B. SUMNER.

of deferred fertilization, a phenomenon which has been little
recognized among mammals. It is stated by Marshall (1910,
p. 136), on the authority of various investigators, that "in certain
bats copulation is performed during the autumn, whereas ovula-
tion is postponed until the following spring, the animals in the
meantime hibernating, while the spermatozoa are stored up in
the uterus." There is no inherent improbability of the occur-
rence of parallel phenomena in rodents. The presence in the
uterus of active spermatozoa long after copulation ought,
however, to be directly determined for these animals.

A few additional comments seem worth while before closing
these notes. It must be pointed out that, in Peromyscus, at
least, the phenomena discussed cannot be of very rare occurrence.
The seven cases in the first table which relate to deer-mice have
been observed in the course of rearing about 250 broods of young.
Thus nearly 3 per cent, of the litters born were followed by these
deferred or supernumerary broods. Whether or not the small
size of these first litters (i to 3 young) is a matter of signifi-
cance I cannot conjecture.

Furthermore, it should be stated that most of these super-
numerary litters comprised normal, healthy animals, more than
80 per cent, of which survived nearly or quite to maturity. In
the great majority, at least, of the cases recorded for man, one,
if not both, of the fetuses has been dead or imperfectly developed
when born.

Again, it is worthy of remark that in four cases out of eight
the parents of these supernumerary broods were both very young
mice, which had not yet been separated according to sex, while
in two more cases the fathers (though not the mothers) were very
young animals. Since the proportion of all my broods born of
parents of such an early age is very small, it seems likely that
this fact is of some significance.

Finally, the possibility suggests itself that some of the alleged
instances of "telegony" which are recorded from time to time,
may result from the actual retention of spermatozoa received
from an earlier mate. A later copulation with a different partner
might happen to coincide with a conception in which the earlier
insemination was really the effective one.

SUPERFETATION AMONG MICE. 285

LIST OF WORKS CITED.
Aristotle.

a De Generatione Animalium (Translated by Arthur Platt). Oxford,

Clarendon Press, 1910. (IV., 4: 773 b to 774 a.)
b Historia Animalium (Translated by D'Arcy W. Thompson). Oxford,

Clarendon Press, 1910. (VII., 4: 5853.)
Daniel, J. F.

'10 Observations on the period of gestation in white mice. Journal of Experi-
mental Zoology, vol. 9, no. 4, Dec. 1910, pp. 865-870.
Godlewski, E.

'14 Physiologic der Zeugung. (In Bd. III. of Winterstein's Handbuch der

vergleichendenden Physiologie). Jena, 1910-1914, pp. 457-1022.
Heape, W.

'oo The " sexual season " of mammals and the relation of the " pro-oestrum *'
to menstruation. Quarterly Journal of Microscopical Science, vol. 44
(N. S.), no. 173, Nov., 1900, pp. 1-70.
King, Helen D.

'13 Some anomalies in the gestation of the albino rat (Mus norwegicus albinus).

BIOLOGICAL BULLETIN, vol. XXIV., no. 6, May, 1913, pp. 377-391.
Kirkham, W. B.

'07 The maturation of the mouse egg. BIOLOGICAL BULLETIN, vol. XII.,

no. 4, Mar., 1907, pp. 259-265.
Kirkham, W. B., and Burr, H. S.

'13 The breeding habits, maturation of eggs and ovulation of the albino rat,
American Journal of Anatomy, vol. 15, no. 3, Nov., 1913, pp. 291-317.
pis. 1-6.
Long, J. A., and Mark, £. L.

'n The maturation of the egg of the mouse. Carnegie Institution Publication

no. 142, 1911, 72 pp., 6 pis.
Marshall, F. H. A.

'10 The physiology of reproduction. Longmans, Green and Co., 706 pp.
Schultze, B. S.

'66 Ueber Superfoecundation und Superfoetation. Jenaische Zeitschrift fiir

Medicin und Naturwissenschaft, Bd. 2, 1866, pp. 1-22, pi. I.
Sobotta, J.

'95 Die Befruchtung und Furchung des Eies der Maus. Archiv fiir mi-
kroscopische Anatomie, Bd. 45, pp. i5~93. pis. II-VI.

FURTHER DEVELOPMENTS IN OVARIOTOMIZED

FOWL.

H. D. GOODALE,
MASSACHUSETTS AGRICULTURAL EXPERIMENT STATION.

In several castrated Brown Leghorn females certain develop-
ments of particular interest have taken place. These de-
velopments relate not only to the plumage and other external
characters but also to certain structures associated with the
reproductive organs. In brief, these individuals developed male
plumage and other male characters. After a time, however,
certain changes in the plumage of some individuals took place,
best described as a change to or toward the female type, as the
case might be. Still later, the plumage changed again to or
to\vards the male type. Usually, the development of female
plumage in poullards is to be referred to a regeneration of the
ovary but an examination showed that no regeneration had
occurred in these individuals but that instead an organ sui
generis had grown. A portion of the organ has been removed
from each bird and sectioned. Its structure clearly is neither
that of the ovary nor that of the testes. The exact nature of
these organs cannot be determined at present, although their
structure suggests that they have some relation to the epididymis.
On the right side in normal females, both ducks and fowl, a bit
of tissue is sometimes found on the spot corresponding to the
site of the ovary on the other side. From this body a strand
can sometimes be traced posteriorly. While this body has some
resemblances histologically to that of the organs developed in
the castrated females, it is impossible to assert that the latter
results from the hypertrophy of the former, even though there
are many other reasons for drawing this conclusion. To demon-
strate the assumed relationship between the structures will
require a considerable series of stages which are not at present
available and whose collection will require some time.

The history of these cases may now be considered in detail.

286

FURTHER DEVELOPMENTS IN OVARIOTOMIZED FOWL. 287

The first is that of the pullet described in the American Naturalist,
Volume XLVIL, 1913. The chick had been completely cas-
trated when three weeks of age. In due course of time, the bird
developed a good male's plumage with large comb and spurs.
However, there were a number of feathers in the dorsal region
which were shaped and stippled like those of the hen but rather
different in color. (Fig. 3, b, American Naturalist.} With the
coming of the moult it was found that the new feathers were
essentially like the old. That is, the plumage retained its inter-
mediate character. The bird was then killed and dissected.
The conditions found were so remarkable that it was thought
best to await confirmatory data before publishing. On either
side, at the level of the adrenals was an organ which had the
appearance of a small testis, though divided into several lobes.
Histologically, however, it was very different. Leading pos-
teriorly from each of these structures to the cloaca was a yellow-
ish white strand (cord, duct) resembling an immature vas
deferens. The left oviduct in an infantile condition was present.

The presence of the bodies described for 1196 appears to be
more usual in ovariotomized fowl because such organs, with
one possible exception, have not been found among the eight or
ten ovariotomized ducks that have been opened or autopsied at
various times, though found in all fowl thus far examined. Both
species have ranged from a few months to three years of age,
all but one at least a year old. Evidently the possibility of the
development of the bodies in question rests upon some genetic
basis. Nor are they necessary for the assumption of male
characters by the ovariotomized female, since the ducks have
developed as good male plumage as the fowl.

Perhaps the question will be raised regarding the possibility
that all these individuals were really males. It is desirable to
consider this phase of the matter in some detail. There are two
general possibilities of error — first, a possibility that an error was
made in identifying the sex of the individual at the time of the
operation; an error however, that would be equivalent to one
made in identifying the sexes of domestic poultry by their
plumage. The gonads of the male and female are quite un-
like, even before hatching time (cf. Thompson, Arch. Ent.,

288 H. D. GOODALE.

1911). The single ovary is a flat sheet of tissue, roughly tri-
angular in outline, found on the left side only. The two testes
are each cigar-shaped.

The second possibility is that a clerical error was made. It
should be needless to say that great care is taken to avoid this
sort of mistake. However, there are several checks on mistakes
of either sort. First, in ovariotomy an incision is made on the
left side only, never on the right. In the extremely rare event of
there being an ovary on the right side its presence would remain
unknown until much later. The right side is never examined
at the time of operation. Therefore, if a supposed female were
really a male, the presence of the right testis would bring the
mistake to light in due course of time. Second, an infantile
oviduct has always been found in the cases autopsied. These
oviducts are not a vague strand of tissue but on the contrary are
in about the same condition as a four or five month pullet's before
the ovary has begun to enlarge preparatory to laying. Third,
there are a number of peculiarities about the castrated females
that differentiate them from males both normal and castrated.
These are fully considered in another place.1 Finally, it may be
noted that the number of instances on record is fairly large.
During the past year a flock of fifteen ovariotomized ducks,
besides several ovariotomized fowl, have been maintained at
this station.

The second individual for consideration is a rose-combed bird
of a different strain of Brown Leghorns which was castrated as
a four-weeks-old chick in 1914. This bird, number 3840, devel-
oped a beautiful male plumage in due course of time, except that
for some reason the development of the tail was imperfect. The
main tail feathers were present but had an injured appearance.
Number 1196, described above, likewise had the dorsal portion of
the uropygium missing, which gave her a bob-tailed appearance.
Number 3840 also had the same bob-tailed appearance in 1914,
though the feathers were actually present, but lost it with the
new feathers in the autumn of 1915. The comb and wattles
developed more slowly than those of the normal male, but by
late spring they had become large and male-like. The spurs
in press.

FURTHER DEVELOPMENTS IN OVARIOTOMIZED FOWL. 289

also were fully developed. When the bird moulted (1915) the
new feathers were those of the female throughout. As is fre-
quently the case, the bird moulted piece-meal, both kinds of
feathers co-existing at one time. Gradually, however, the male
feathers were replaced by feathers that could not be distinguished
from those of the female. When it is recalled that the Brown
Leghorns are practically identical in color with the Jungle fowl,
the change can readily be appreciated. The bird was not killed
as it was desired to keep her for further observation, but instead
she was opened on each side. Organs roughly similar to those
described for 1196 were noted. On the right side a strand of
tissue was traced posteriorly for a few millimeters but no strand
could be identified on the left. No trace of anything resembling
normal ovarian tissue was found. As the location of the incision
was unfavorable for finding the oviduct, this was not attempted.
A bit of each of the organs were removed and sectioned. The
histological findings were like those described below.

A short time after the operation it was noted that the new
saddle feathers that came in were male-like in that they were
fairly long, and laced with bright yellow, though rather bluntly
pointed at the end and the central stipe though nearly black
was often sprinkled with brown to a greater degree than usual
for a male. The feathers of the tail coverts particularly con-
tained much brown. In the breast feathers, less change was
noticeable, most of the feathers remaining deep reddish salmon
though some of the feathers contained some black. Other
parts of the bird showed feathers much like those of the female
type.

The third instance is that of a bird hatched June 22, 1913, and
castrated August 8, 1913. The juvenile female plumage was
well developed at the time. By early winter, however, this
plumage had given place to one nearly or quite identical with
that of the young male. The spurs developed fully but the
comb never became really male-like, but had rather the general
appearance of a Leghorn female's comb when straightened up.
This comb, however, never- loped but was always erect, an
exception not at all rare among Leghorn females. It was
anticipated that when the bird moulted during the summer

H. D. GOODALE.

the full male plumage would be assumed. Instead, a different
type appeared. The feathers were shaped like those of the hen,
except the tail coverts which in shape resembled those of an
English Campine male. That is, they were rather longer than
those of the hen, curved, and with rounded ends. The dorsal
feathers were dull black with golden shafts, sometimes with a
few minute brown spots on the margins. Ventrally the feathers
were black.

With the moult of 1915, a further change took place in that
the breast feathers were replaced with salmon-colored feathers,
while there was some increase in amount of brown stippling
dorsally, so that this bird, too, was very much like a female.
She was opened on each side in October, 1915. The same
set of organs was found as described for the preceding instances.
This bird, like 3840, also changed the character of its plumage
after the operation. The new breast feathers were black, while
the saddle feathers were good male though not as long and
pointed as is usually the case.

A fourth instance has a history somewhat different from the
one just preceding, although it was a litter sister and castrated
at the same time. The castration, however, proved to be in-
complete in that regeneration of the ovary took place. A minute
bit of the ovary must have been left and when it became suffi-
ciently large the new feathers that developed under its influence
were female. Nearly a year after the first operation, a second
was made and an attempt made to remove the regenerated ovary.
Apparently it was successful for the bird soon after began again
to assume male characters. The spurs became long but the
comb has always remained hen-like, though erect. The present
plumage is a mixture of male and female characters. The breast
feathers are almost black but contain a little salmon in small
patches. The dorsal feathers are not much longer, if any, than
those of the normal female but they are triangular at the apex
and have a margin of golden bristles. The centers, however, are
female colored. An examination of the right side in October,
1915, showed a body similar to those described but rather larger.
The left side seen from the right appeared to be empty but when
an attempt was made to open the bird on the left side, un-

FURTHER DEVELOPMENTS IN OVARIOTOMIZED FOWL. 2QI

expected difficulties were met, so that it was deemed desirable
to proceed no further.

There remain two other birds to be considered. Though
both were of hybrid origin their plumage was that of a typical
Brown Leghorn. At the time they were castrated, the female
juvenile plumage was well developed. After castration both
developed spurs and a perfect male coat of plumage but their
combs remained small.

Number 4290 was killed November 25, 1914. The same sort
of organs on the site of the gonads were noted again. The other,
4471, is still alive and in perfect male plumage. In the summer of
1913, a female-like plumage developed, followed in the fall by
a return to the male plumage. In 1915 no change in plumage
color took place. When opened, October n, 1915, the same sort
of organs were found. Thus, in all but possibly one instance
there has been a development of glandular material on both sides.
The exception relates to the individual that was examined on
only one side. It seems probable that though numbers 3840
and 2087 were operated on just prior to their return to the male
condition the operation as such had nothing to do with the
results secured but rather that a change to this plumage would
have taken place as in the case of 4471. It is of course possible
that the operations accelerated a change about to take place.
It is quite possible, too, that the changes in plumage are cyclic
in nature, like the occurrence of the summer plumage in the
normal drake.

Because it is desired to keep the birds alive for further observa-
tions, the structure of the organs cannot at present be given in
detail. It was thought at first that a small piece would suffice
for a determination of its structure, but the material thus far
examined indicates that the study will have to be made from the
standpoint of the organ as a whole. Owing to lack of material,
this cannot be attempted for some time to come. However,
the material on hand is sufficient to indicate something of its
nature. The following description is provisional: The histo-
logical findings vary from specimen to specimen and also in
different parts of tissue from the same bird. The differences,
however, are probably to be referred to developmental stages,

292 H. D. GOODALE.

though this is by no means certain. In what is supposed to be
the early stages, the cells are small with relatively little proto-
plasm and closely packed. There is a very weak development of
connective tissue which separates the cells into poorly denned
groups. Intermediate stages can be found in which the groups
of cells become well denned. After the first stage, there is a
considerable increase in the connective tissue which in some
instances appears to produce smooth, refractive rods or strands.
Stages have been noted in which the central cells separate some-
what after the manner of thyroid tissue and stain less readily.
These appear to be degenerating. Later the marginal cells also
break down so that small open spaces may be found lined mostly
by connective tissue. They in turn appear to coalesce by the
breaking down of neighboring wralls, so that large open spaces
appear in the tissue, perhaps homologous with the vesicles filled
with a straw-colored fluid observed macroscopically in some
instances. In one part of the organ a group of tubules lined with
a single layer of square cells has been noted. These appear to
be of a different character from the spaces described above.
They may contain a finely granular but otherwise amorphic
substance.

It is rather probable that the present organ is the result of a
development of the Wolfian body. This view, however, is
merely a working hypothesis.

Should further investigation confirm this view, it may throw
light on the structures described by various observers in cock-
feathered females. These bodies are very likely identical with
the bodies found in castrated females. Certainly the presence
of true seminal tubules or spermatozoa together with ova must
be demonstrated before similar individuals found in nature con-
taining these organs can be designated as hermaphrodites.

An explanation of the plumage changes is mainly a matter of
surmise. At first, one is inclined to lay the changes at the door
of the new organ but since they do not appear in all individuals
which have developed the organs, it is evident that if the organs
are concerned with the changes there must be some change in
the activities of the organs either preceding the changes in plum-
age or accompanying them.

FURTHER DEVELOPMENTS IN OVARIOTOMIZED FOWL. 2Q3

Among ducks it has been noted that a similar change in plum-
age is associated with the testes in the male, although the organs
described have not been found in several castrated female ducks
which have been examined. The ovariotomized duck may or may
not undergo a change in plumage, corresponding to that of the
male. Those that undergo such a change have returned in due
course of time to the breeding plumage. Thus their temporary
plumage is like that of the summer plumage of the male.

The change of plumage in the ovariotomized fowl may be due
to the release of a mechanism for changing the plumage but
which has been hidden or rendered latent in some way during
the phylogeny of the race or as a result of domestication. In
this connection it may be noted that laying hens give evidence
of a summer moult. This moult is rarely complete and is
evidenced usually by the shedding of a comparatively small num-
ber of feathers.

THE BEHAVIOR OF THE ACCESSORY CHROMOSOMES

AND OF THE CHROMATOID BODY IN THE

SPERMATOGENESIS OF THE RABBIT.1

L. J. BACHHUBER,
UNIVERSITY OF WISCONSIN, DEPARTMENT OF EXPERIMENTAL BREEDING.

The following studies resulted from a series of experiments
to determine the effect of lead poisoning upon the germ-cells
of the male rabbit as indicated by his offspring. My intention
originally was to attempt to determine the manner in which the
lead-poisoning affected the normal mitosis. The problem of the
normal mitosis in itself proved to be so large that the study of
the effect of lead-poisoning had to be postponed to a future time.

The following work was done under the direction of Dr. M. F.
Guyer, to whom the writer is very much indebted for many
valuable criticisms and kindly help. The writer is also indebted
to the kindness of Professor L. J. Cole for aid given in getting the
necessary material for this study.

All of the rabbit testes used were from animals raised at the
barns of the Department of Experimental Breeding of the
University of Wisconsin. These males were chosen from the
normal stock resulting from the double matings described by
Cole and Bachhuber (1914).

A fairly successful fixing reagent was found in Flemming's
strong. This method brought out the chromosomal and cyto-
plasmic structures better than Gilson's, Zenker's, Hermann's
and possibly Bouin's fixative. Bouin's gave good results in the
study of the accessory chromosomes and the chromatoid body.
A new method reported to be in use in McClung's laboratory
was also tried with considerable success. This fixative employs
urea as a means of more rapid penetration. To one hundred
cubic centimeters of Bouin's, made up of seventy-five parts of

1 Papers from the Department of Experimental Breeding, Wisconsin Agricultural
Experiment Station, No. 6. Published with the approval of the Director of the
Station.

294

SPERMATOGENESIS OF THE RABBIT. 295

aqueous picric acid with fifteen parts formalin and ten parts
glacial acetic acid, were added one and one-half grams chromic
acid and three grams of urea. The solution was then heated to
thirty-seven degrees and the tissue added. This fixative made
the chromosomes stand out somewhat better and clearer than
any of the other fluids.

The sections were cut from four to fifteen micra in thickness
and stained in (i) Delafield's haematoxylin and eosin, (2) Haiden-
hain's iron-haematoxylin and acid fuchsin or orange G, (3)
safranin with gentian violet or lichtgrun. Method (2) proved to
be the most successful and gave some very excellent results.
This method was most valuable in bringing out the detailed
structures, especially the chromatoid body. Cytoplasmic struc-
tures were as a rule satisfactorily stained in Flemming's triple
stain.

Smear preparations were also made. These were fixed in
Flemming's strong or in Bouin's fluid. Haidenhain's iron-
hsematoxylin gave the most satisfactory result when counter-
stained with lichtgrun.

Mammalian spermatogenesis seems to offer greater difficulties
for study than any other form. This is due to the impossibility
of securing by means of existing reagents and methods proper
fixation. In nearly all preparations it has been found that the
chromatic structures have a tendency to mass so that individual
details are lost. The chromosomes in the rabbit have this to a
very marked degree. The various stages in spermatogenesis
were numerous enough, but a great many stages had to be
examined in order to find those in which chromosome counts
were possible.

N. M. Stevens (1911) in the spermatogenesis of the guinea-
pig found that material "very unfavorable and the results are
not so satisfactory as are desired." Montgomery (1912) in his
work with the human spermatogenesis has to say practically
the same thing, "the fixation was not as excellent as might be
desired." Wodsedalek (1913), however, seems to have encoun-
tered less of the massing of chromosomes in the spermatogenesis
of the pig than is present in other mammalian forms.

In both the rat and the guinea-pig, and later in the bull, the

2Q6 L. J. BACHHUBER.

writer has found the same condition, although to a lesser degree
than in the rabbit, namely, the strong tendency of the parts
within the nucleus to mass, hiding many of the individual
structures, and leaving much to be desired in the form of im-
proved methods. In the rabbit, the chromosomes always
agglutinated, permitting but few chromosomal counts. Because
of this, the only stages in which chromosomal counts were
possible with any degree of accuracy were the primary spermato-
cytes. In later stages it was decidedly more difficult to find
stages in which counts were possible.

In the rabbit, the structures to be followed more readily are,
first, the two accesory chromosomes and second, the chromatoid
body. After the spermatogonial stages, the accessory chromo-
somes could nearly always be identified and rather easily traced.
The chromatoid body, similar to that described by Wilson (1913),
whose origin could not be determined, can be easily followed,
beginning with the primary spermatocytes, and can even be
identified when it is cast off from the transforming spermato-
somes.

The other structures displayed in the spermatogenesis of the
rabbit, and the various processes connected with them, do not
differ materially from the corresponding stages described for
other forms. In the following pages, the process of spermato-
genesis, in so far as it could be determined, will be taken up in the
order of the successive stages of spermatogonia, primary and
secondary spermatocytes, spermatosomes, and the fully de-
veloped spermatozoa.

i. ZONES OF PROLIFERATION.

Throughout all the tubules of the testes, there appear small
areas in which active proliferation reveals almost every stage
of spermatogenesis. These areas are not, howrever, confined
to any particular section or part of the tubule. All indica-
tions of the active zones go to show that a certain area may be
active for a period, then halt in its activity while a nearby area
becomes active in spermatogenesis. There are thus, scattered
throughout each of the tubules, areas of active proliferation and
areas of rest. In each of these areas of activity usually there

SPERMATOGENESIS OF THE RABBIT. 297

may be seen the ordinary arrangement of the spermatogonial
cells on the outer margin with the primary spermatocytes
adjacent to them. Further towards the center lie the secondary
spermatocytes while the spermatids, the transforming spermato-
somes, and the fully developed spermatozoa lie in the central
cavity.

2. SPERMATOGONIAL STAGES.

It has been difficult to obtain satisfactory preparations of
the spermatogonial cells showing all of the essential structures,
chromatic and achromatic. Nutritive cells were also compara-
tively scarce although a few may have been identified (Fig. i).
They are very similar to the resting stage of the spermatogonia
with which they may easily be confused and their identity is
never certain. The nuclei of these nutritive cells usually appear
oval or irregular in shape and contain masses of chromatin
scattered throughout This close resemblance to the nucleus
of the resting spermatogonia makes it not at all improbable
that the nutritive cells, if they are such, may be derived from the
spermatogonia. Montgomery (1912) has found this to be true
in man. He was able to trace directly the formation of the
Sertoli cells from the spermatogonia by the presence of a rod-
like body in the cytoplasm, which was, according to him, an
invariable indication of the Sertoli cells. Hegner (1914) takes
the view that the Sertoli cells arise from the primordial germ-
cells. This may also be true in the rabbit although there is
little direct evidence to support either view. The similarity of
the Sertoli cells and the spermatogonia and the relative number
of these cells affords the only evidence.

During the earliest prophases of the spermatogonial stage
(Figs. 2, 3) the nucleus is somewhat elongated and irregular as
a rule, and contains two or more large, spherical karyosomes.
These may be the two accessory elements which can be traced
very accurately after the formation of the primary spermatocytes.
Small linin threads, somewhat granular in appearance, radiate
out towards the periphery of the nuclear wall (Figs. 4, 5).
Slender fibrilke extend throughout the cytoplasm (Fig. 2).
Between these are small areas of a granular appearing substance.
The centrosome and the other cytoplasmic structures cannot be

298 L. J. BACHHUBER.

identified although some of the more granular parts of the
cytoplasm may include the centrosome and the chromatoid body.
This is however doubtful because the staining reactions given
by these structures in the primary and secondary spermato-
cytes ought also be given in the spermatogonia.

Only one spermatogonial stage has been found in which the
chromatoid body appeared to be present (Fig. 6). While it has
the characteristic appearance of this body, it is a rather doubtful
case because it could not be identified in any other of the sperma-
togonial stages.

As development continues, dense masses of chromatin appear
concentrating along the radial linin threads. These masses
vary in number, but never exceed the diploid number of chromo-
somes, making it highly possible that these masses later trans-
form into the univalent chromosomes. The entire nucleus stains
heavily with the basic dyes, indicating a large increase of the
chromatin content of the cell. In this manner, from twelve to
twenty-two masses of chromatin are formed, all united by heavy
linin threads. Gradually the linin threads disappear, and the
chromatin masses assume a more regular appearance (Fig. 7).
The nuclear wall disintegrates as the chromosomes arrange them-
selves in the metaphase stage (Fig. 8). The spindle fibers are
very indistinct although in favorably stained sections they may
be made to stand out more strongly. Because of this excess
of stain, the cells which were stained heavy enough to make the
spindle fibers appear more plainly were useless for the study
of any of the other structures.

The chromosomes at best tend to mass together, and if not
strongly destained, form a huge black mass in which nothing
can be distinguished. In some of the metaphase stages it is
possible to count twenty-two chromosomes (Figs. 8, 9, 10).
The X and the Y elements which are shown to exist from their
subsequent behavior, could not be distinguished from each
other or from the remainder of the chromosomes in the spermato-
gonial division stage. It appears probable that the two large
karyosomes present in the early spermatogonial stages may
represent the accessory elements because they retain their
individuality and later appear to transform directly into two

SPERMATOGENESIS OF THE RABBIT. 2QQ

chromosomes of the spermatogonial division. At times they
may be traced because of a slightly more rounded form than the
other chromosomes. The chromosomes in general are elongated
in shape although the tendency is towards a spherical form.

During the anaphase (Fig. n), the divided chromosomes move
toward opposite poles. As soon as the chromosomes reach the
pole, they go into a resting stage, the new nuclear wall appearing
immediately. The new cell walls may not appear until late in
synezesis. The outlines of the new nuclei are first somewhat
elongated, conforming to the rough outline of the massed chromo-
somes. Gradually these assume a more spherical shape, the
spindle-fibers disappear, and the primary spermatocytes are
ready for the growth period.

3. EARLY MATURATION STAGES.

While the number of chromosomes is not large in the rabbit,
the difficulties encountered in counting chromosomes were great
indeed. In all of the material, the tendency of the chromosomes
to mass was, as already mentioned, very much in evidence, and
the most careful tecnhique in fixing and staining did little to make
the results more satisfactory. There were enough cells in which
chromosome counts were possible to place the probable number
as twenty-two in the spermatogonia and twelve in the primary
spermatocyte.

It was found that twenty-two chromosomes went to each
pole in the spermatogonial division. After synapsis has taken
place, the number is twelve, showing that two of the elements
remain single. These two elements are the accessory chromo-
somes. Thus the twenty ordinary chromosomes of the spermato-
gonia reduce to ten bivalents in the primary spermatocytes,
wrhile the two accessories do not undergo synapsis. As will be
seen, the later behavior of these univalent chromosomes is
entirely distinct from that of the others.

The early prophases of the primary spermatocytes show all
but two of the chromosomes which passed to the poles in the
spermatogonial division to grow irregular in shape, and finally
weave out into fine strands which immediately spread throughout
the nucleus (Fig. 12). All of the irregular masses spin out into

3OO L. J. BACHHUBER.

fine threads constituting the leptotene threads, leaving two, still
rounded chromosomes, which can from this time on be identified
as the accessories. The leptotene threads, in both sections and
smears, are seen to persist as independent units. Their exact
number could not be determined because of their extreme length.

During this stage, the accessory elements remain as small
spherical masses which can easily be identified if the mass of
leptotene threads is sufficiently destained. During the process
of synapsis also, the X and the Y remain unchanged, always close
together, and sometimes apparently connected by thin, dark-
staining threads of chromatin material (Fig. 15). No instances
have been found which at this stage show the accessory elements
in different portions of the cell.

In these prophases, the presence of the chromatoid body
becomes definitely established. It stains similarly to the chro-
matic material in the iron-haematoxylin, the Delafield's haema-
toxylin, and in the Flemming's triple stain. Just where this
body arises is still a question. As before stated, it has been
found in only one spermatogonial cell and that was a rather
doubtful case. In the primary spermatocytes it is absolutely
constant (Figs. 16, 20, 22). In a few rare cases, two and some-
times three have been found, but these extra bodies wTere always
very small and disappeared in later stages. The large body
can be traced through the remainder of the process of spermato-
genesis. Further mention of its behavior will be made later.

In the synezesis stage (Figs. 13-16), the threads drift and mass
at one pole of the cell into a grouping which has apparently no
established order and from which nothing definite could be deter-
mined. Occasionally the threads seem to have a parallel
arrangement suggesting parasynapsis, but when they come out
of the synezesis stage, considerable evidence points to a chiasma-
type synapsis (Figs. 18, 19). The threads are wound around
each other, making possible a division later in which each of the
haploid chromosomes would be made up of alternating segments
of chromatin material, from the respective leptotene threads
of the conjugating pair. That the threads still retain a con-
stancy in number seems highly probable for in favorably stained
sections where their ends can be seen, the number never exceeds

SPERM ATOGENE SIS OF THE RABBIT. 30!

double the number, minus four, of the chromosomes of the
primary spermatocyte.

Previous to the pachytene stage the cells have increased but
very little in size. Now they begin to grow rapidly, complete
fusion occurring between the conjugating threads, and all
gradually drift back to the center of the cell, there to spread
slowly throughout the nucleus. The threads enlarge, become
lighter in staining capacity, due perhaps to the greater volume
occupied by the pachytene thread, and finally form the large
spireme (Figs. 17, 20, 21).

This stage of the primary spermatocytes, with the possible
exception of the synezesis stage, persists longer than any other
stage of the spermatogenesis. This conclusion is reached because
these stages of the primary spermatocytes can be found in prac-
tically any portion of the tubules. Numerous as they are, it is
extremely difficult to gain any knowledge of the processes in
progress in them.

The accessory elements still retain their spherical form, usually
lying closely side by side, but at this time occasionally in different
portions of the cell (Figs. 17, 21). The chromatoid body enlarges
and seems to become more prominent. At the same time it
seems to acquire an activity in the cell although its function
could not be determined.

4. REDUCTION DIVISION.

The spireme of this stage in general appeared to be continuous,
although some of the smears made under considerable pressure
gave indications of segmentation (Fig. 23). The large spireme
now condenses and soon forms the individual chromosomes.
Up to this period of condensation the chromomeres stand out
very distinctly (Figs. 18, 19). These condense to form the large
mass of bivalent chromosomes which, because of their agglutina-
tion, usually stain so dense that the individual structure is
entirely lost.

The X and Y elements, however, can nearly always be iden-
tified in these stages. As soon as the bivalent chromosomes are
formed, they migrate into the equatorial plate. The accessory
elements are nearly always a little to one side of this plate

302 L. J. BACHHUBER.

(Fig. 23), usually close together. It may be that this position
is constant, but the massing of the chromosomes hides the
accessories (Fig. 24). In the division, the X and Y chromosomes
move towards the poles long before the division of the ordinary
bivalents (Figs. 25-29). The X is thus drawn to one pole, the
Y toward the other. In this manner, each of the secondary
spermatocytes will contain only eleven chromosomes, ten ordi-
nary and one accessory (Fig. 30). The ordinary chromosomes,
in their bivalent condition, are approximately twice the size of
the spermatogonial chromosomes.

The split in the bivalent chromosomes is rarely in evidence
except immediately preceding the division, although in a few
isolated cases a V-shape becomes noticeable.

When the chromosomes reach the pole the chromatoid body
seems to acquire more activity. During the formation of the
equatorial plate, this body migrates close to the densely crowded
chromosomes (Figs. 24-27). As soon as these start moving
toward the poles, the X and Y preceding, the chromatoid body
moves in between the two sets of chromosomes, lying approxi-
mately midway between the two centrosomes (Fig. 31). Just
before the new cell wall is formed, this body moves into the
cytoplasm of one of the newly formed spermatocytes. The
possibility presents itself that the chromatoid body always
follows either the X or the Y elements. The two accessories are,
however, so closely related in size that it has been impossible to
determine whether there is any such specific association.

Occasionally, in the division of the primary spermatocytes,
there is found evidence that the accessory elements undergo a
precocious division. The accessories lie in the plane of division
and under certain conditions, they divide, forming two X and
two Y elements. The division then proceeds in the regular
manner with this exception: two X and two Y chromosomes
go toward the poles, again preceding the division of the ten
ordinary chromosomes (Figs. 32, 33).

5. SECONDARY SPERMATOCYTES.

The division of the secondary spermatocytes proceeds without
any appreciable rest stage. In the telophases of the preceding

SPERMATOGENESIS OF THE RABBIT. 303

division the nucleus is somewhat crescent shaped, but soon
assumes a spherical form. This stage develops no spireme or
leptotene threads. The chromosomes, eleven in number, imme-
diately migrate into the equatorial plate to form a very dense
mass which at times appears almost homogeneous, so closely
are the chromosomes drawn together (Fig. 34). In this division
the chromosomes, in migrating toward the poles, often appear
as solid black ring-like masses with apparently no segments
representing the individual elements. Dense fibers seem to be
interwoven in the mass of chromosomes (Figs. 35, 36). It is
seldom that the chromosomes can be distinguished in these
stages. The X and the Y elements also as a rule lose their
identity, but occasionally they can be identified by their presence
in the center of the ring, connected by chromatin strands to the
ordinary chromosomes (Figs. 37, 38).

The chromatoid body appears to undergo no division in these
stages, following either one or the other of the chromatin masses.
We thus find that approximately one-fourth of the total sperma-
tids contain the chromatoid body.

6. TRANSFORMATION OF THE SPERMATIDS.

As soon as the division is complete, a new nuclear wall is
immediately formed. The chromosomes as such disintegrate
but the chromatic material soon condenses at the periphery of
the nucleus (Figs. 39, 40). In this manner nearly all of the
chromatic material is condensed, leaving one part which in
every case is found somewhere near the center of the cell (Figs.
41, 43) and which may represent the accessory element. Some-
times there are one or more smaller bodies which may be remnants
of the disintegrating ordinary chromosomes because they do not
persist for any length of time. The large central body can be
traced through the stages of further condensation and almost
to the completely developed sperm.

After all of the chromatic material has condensed, the centro-
some makes its appearance and moves close to the nucleus (Fig.
41). It here gradually becomes immeshed in strands of chro-
matic material within a slight infolding of the periphery of the
nucleus. By a continuation of this process it soon becomes

304 L- J- BACHHUBER.

entirely surrounded by nuclear material (Figs. 41, 42, 43). The
entire nucleus then migrates to one side of the cell and gradually
leaves behind all, or very nearly all, of the cytoplasm, together
with the chromatoid body (Figs. 43, 44, 45). At the same time
the entire nucleus begins to condense and finally forms a flat
plate, which has at times, however, the appearance of a convex
plate (Fig. 46). In the meantime the sperm tail has been
formed from the cytoplasm with the centrosome in the neck.
The nucleus, now the sperm head, gradually enlarges with a
resultant diminution of staining capacity. In properly stained
sections and smears there can occasionally be distinguished a
number of darker staining bodies one of which is usually larger
than the others. The number of these approaches very nearly
the number of chromosomes which went into the sperm head.
It is thus possible that the chromosomes retain their individuality
even in the fully developed spermatozoa (Figs. 47, 48). As the
sperm head enlarges, these dark-staining bodies gradually diffuse
through the entire head (Figs. 49, 50). Further enlargement
brings about the complete development of the spermatozoon.
The head manifests an even staining capacity, the neck shows
the presence of the centrosome, while the tail takes such a light
stain that it is barely visible (Fig. 51).

7. CONCLUSIONS.

1. The number of chromosomes in the spermatogonium is
probably twenty-two.

2. The number in the primary spermatocytes is placed at
twelve.

3. The number in the secondary spermatocytes is placed at
eleven.

4. Two accessory elements, an X and a Y, are present. One-
half of the spermatozoa contain the X, and the other half, the
Y element.

5. A chromatoid body is present. Its function was undeter-
mined. It underwent no division, and was finally cast off with
the excess cytoplasm in the metamorphosing spermatid.

SPERMATOGENESIS OF THE RABBIT. 305

BIBLIOGRAPHY.

Cole, L. J., and Bachhuber, L. J.

'14 The effect of lead on the germ cells of the male rabbit and fowl as indicated
by their progeny. Proc. Soc. Exp. Biol. and Med., Vol. 12, No. i,
pp. 24-29.
Guyer, M. F.

'10 Accessory chromosomes in man. BIOL. BULL., Vol. 19, No. 4, pp. 219-234,

pi. I.
Hegner, R. W.

'14 Studies on germ cells. Jour. Morph., Vol. 25, No. 3, pp. 375~499. pis- i-io.
Montgomery, T. H.

'12 Human spermatogenesis. Jour. Acad. Nat. Sci. Philadelphia, Vol. 15,

2d ser., pp. 1-22, pis. 1-4.
Stevens, N. M.

'n Heterochromosomes in the guinea pig. BIOL. BULL., Vol. 21, No. 3, pp.

155-167-
Wilson, E. B.

'13 A chromatoid body simulating an accessory chromosome in Pentatoma.

BIOL. BULL., Vol. 24, No. 6, pp. 392-410, pis. 1-3.
Wodsedalek, J. E.

'13 Spermatogenesis of the pig with special reference to the accessory chromo-
somes. BIOL. BULL., Vol. 25, No. i, pp. 8-32, pis. 1-6.

3O6 L. J. BACHHUBER.

EXPLANATION OF PLATES.

All drawings made with camera lucida. Spencer 2 mm. apochromat oil
immersion objective and X 16 compensating eye-piece used. All drawings ar-
shown at a magnification of approximately 1,800 diameters. Drawings are
accurate with reference to the nuclear material and the chromatoid body and their
position in the cell; the cytoplasm, however, is represented conventionally. S,
Sertoli cell; C, chromatoid body.

PLATE I.

FIG. i. Sertoli cell, showing the relation of the metamorphosing spermatosomes
to the nurse cells.

FIGS. 2,3. Early stages of the spermatogonia with the two large karyosomes.
Fig. 2 also shows the slender fibrillae in the cytoplasm.

FIGS. 4, 5. Later stages of the spermatogonial nuclei showing the linin threads
radiating out towards the periphery of the nuclear wall.

FIG. 6. Shows the only spermatogonial cell in which a body was found which
resembled the chromatoid body.

FIG, 7. In this stage a condensation of the nuclear material is taking place,
forming masses which later transform directly into the chromosomes of the sperma-
togonia.

FIGS. 8-10. Metaphase stages of the spermatogonia showing the probable
twenty-two chromosomes. Fig. 10 is taken from a smear preparation.

FIG. n. Shows the divided chromosomes passing towards the poles. The X
and the Y elements cannot be identified at this stage.

FIG. 12. Prophase of the primary spermatocyte showing the chromosomes
from the previous division weaving oat into fine strands which immediately spread
through the nucleus.

FIGS. 13-15. Synezesis stages of the primary spermatocytes showing the
massing of the fibers, accompanied by a slow condensation, as shown in Fig. 15.
This also shows the two accessory elements, which retain their spherical shape,
while the ordinary chromosomes weave out into the leptotene threads.

BIOLOGICAL BULLETIN, VOL XXX

PLATE I.

J$ii$&

; ••• -,ri ••<-;; -./'Y.--''i- •_.<,•

\:^^:S'M.

^IlilfP1

:&jKf?A-

..- ivV"'-
• "-.;' '.

•/:-.?^4&^

L. J. BACHHUBER.

308 L. J. BACHHUBER.

PLATE II.

FIG. 16. Shows a further condensation of the leptotene threads, which finally
results in the large spireme. This stage also shows the first appearance of the
chromatoid body.

FIG. 17. The two accessory elements lie a little to one side of the condensing
leptotene threads. The accessories can nearly always be identified in these stages.

FIGS. 1 8, 19. These stages show the leptotene threads coming out of the
synezesis stage and giving evidence of a chiasmatype synapsis.

FIGS. 20, 21. The accessory elements are very conspicuous in the large spireme
stages.

FIG. 22. Cell from a smear, showing evidence of a segmentation of the large
spireme.

FIG. 23. The accessory elements lie a little to one side of the chromosomes
massed in the equatorial plate stage, just previous to division. The chromatoid
body has migrated into the same plane as the chromosomes.

FIG. 24. Shows the massing of the chromosomes, which usually hides the
accessory elements. The chromatoid body is very prominent.

FIGS. 25-28. The two accessory chromosomes travel towards the poles long
before the division of the ordinary chromosomes. The chromatoid body is still
in the plane of the equatorial plate.

BIOLOGICAL BULLETIN, VOL. XXX

'••.-•?••'•

,.., ......,,

' '"'''-^ >?lv ^'X •:"•" "•-''•/"•i:.':;-; •"

22

C--*:-

-:.v'.- •.., .

'-:.-•• ' - -

23

* .•'•••*•"-":.•"'. •\V>---V

':../''-r;:--^: • -^ ;

24

&;C#£f|H^r*

.IP'"

' ; • : , '••••-•' '

"* '•-.'

'. . :•..•*. ." . •. ~'V.-

s •'..'':••.:"••. '.'.'•.'.-. '.'.'..-• '
\v . !-. : • ;•*.:, -;..•••

26

L. J. BACHHUBER.

27

3IO L. J. BACHHUBER.

PLATE III.

FIG. 29. Again showing the accessory elements migrating towards the poles
in advance of the ordinary chromosomes.

FIG. 30. A stage following division of the chromosomes of the primary sperma-
tocyte showing the eleven chromosomes at the pole, the accessory in the center.

FIG. 31. Shows the chromatoid body migrating in between the two sets of
chromosomes immediately after division.

FIGS. 32, 33. These cells give evidence of a precocious division of the X and
the Y elements. In this case, two X elements and two Y elements travel towards
the poles.

FIG. 34. The chromosomes of the secondary spermatocytes have immediately
lined up in the equatorial plate stage, ready for the next division.

FIG. 35. The chromosomes of the secondary spermatocytes divide and pass
to the poles in ring-like masses, practically losing their identity as individual
elements. This stage also shows the chromatoid body.

FIG. 36. Shows the chromosomes after reaching the poles and before formation
of the spermatids. The chromatoid body is still present.

FIGS. 37, 38. Show the occasional identification of the accessory elements in
the center of the closely interwoven mass of chromosomes, giving a ring-like
appearance.

FIGS. 39, 40. Condensation of nuclear material at the periphery of the nuclear
wall. Also shows the presence of the chromatoid body.

FIGS. 41-43. Further condensation of chromatin around the periphery of the
nuclear wall. The centrosome is meshed in one side of the nucleus. The chroma-
toid body may lie in any portion of the cytoplasm and may be very irregular in
shape.

FIGS. 44-46. Still further condensation, with the gradual escape of the nucleus
from the excess cytoplasm. The chromatoid body is cast off with the excess cyto-
plasm.

FIGS. 47, 48. Show the presence of darker staining masses of chromatin
material which may represent the individual chromosomes in the sperm head.

FIGS. 49, 50. The darker staining masses now diffuse through the sperm head
and finally form an even-staining mass of chromatin.

FIG. 51. Fully developed spermatozoon, showing the head, the middle-piece
with the centrosome, and the long, thin, lightly staining sperm tail.

BIOLOGICAL BULLETIN, VOL. XXX.

PLATE III.

x'*:'ro:--'.-^;'-;'::'--

30

. s
^HK^

c

• i ' \ '.r\^A--.

#*%$

'•'•/••I-'' 'i •:lltfci;'

5S»f • ' ,

, ^--^... , . .

;,

>
•:.

.

,;;'v:j^
J5

'^'' ;

>

fit

i

; ,

»A

"•;• • "•- '—•'./

L. J. BACHHUBER.

Vol. XXX. May, 1916. No. 5

BIOLOGICAL BULLETIN

THE THEORY OF ANESTHESIA.

RALPH S. LILLIE,

i
BIOLOGICAL DEPARTMENT, CLARK UNIVERSITY, WORCESTER, MASS.

Anaesthesia, also termed narcosis, is a physiological condition
in which the normal responsiveness or automatic activity of the
living system — organism, tissue, or cell — is temporarily decreased
or abolished. The subjective accompaniment of this change in
higher animals is a more or less complete suppression of conscious-
ness, with consequent insensibility to pain; the term "anaesthe-
sia" refers more directly to this condition. By "narcosis" is
usually meant a temporary paralysis or anaesthesia produced by
chemical substances; this term has a more objective connotation;
and is the one usually employed in purely physiological discus-
sion. It is especially noteworthy that the condition may show
all gradations of degree, ranging from a comparatively slight
inhibition or insensibility to a state of profound depression in
which the organism is completely inert and shows no response to
even the strongest stimuli. Yet on the removal of the anaesthe-
tizing agent the normal properties and activities return. Reversi-
bility is thus an essential characteristic of the condition; this
peculiarity distinguishes it from the irreversible change of death.
There are, however, significant resemblances between these two
states, and in fact transitions from the one to the other are fre-
quent. Too prolonged or too profound anaesthesia may pass
into death; and most anaesthetic substances, if present in too
high concentration, soon cause irreversible and cytolytic changes
in cells. There is in fact evidence that in many instances
anaesthetic and toxic effects have the same essential physico-
chemical basis. The same cell-structures — especially surface-
structures, e. g., plasma-membranes — are primarily affected in

312 RALPH S. LILLIE.

both cases, but in the one case the change produced is reversible,
in the other irreversible. The degree of reversibility is however
itself subject to variation. In many colloidal systems changes
which are reversible in their earlier stages may become irrever-
sible later; and the fact that anaesthesia, especially if profound,
cannot be prolonged indefinitely without danger to^ life, may
find its explanation here.

In any theoretical discussion of anaesthesia it is important to
recognize from the first that normal or physiological conditions
of reversible inhibition or suspended activity are in no sense
unusual among organisms. In both animals and plants irrita-
bility and automatic activity are fluctuating properties, with a
wide range of strictly physiological variation. Thus in higher
animals we have conditions ranging from the profound narcosis
of sleep — a state due apparently to the accumulation of fatigue-
products — to one of complete mental and physical alertness or
wide-awakeness. Generally speaking, responsiveness is largely
a matter of metabolic condition; and most vital activities are
subject to inhibition or enhancement according to the physio-
logical requirements. Variability of this kind is in fact a neces-
sary condition of adaptation to the changing conditions of life.
Thus the activities of animals as a class are influenced to a
marked degree by variations in the food-requirements. In
general they become sluggish and irresponsive when well fed,
and show heightened activity when deprived of food. In other
words, both the automatic motor activity and the responsiveness
to the stimuli of food-substances — the physiological condition
expressed in consciousness as hunger — are increased when the
supply of energy-yielding material is depleted and vice versa.
For example, the fresh water Hydra shows restless swaying move-
ments when hungry; these movements increase the area swept by
the tentacles, which respond promptly to the contact of small
organisms or food-particles by capturing and conveying to the
mouth.1 When well fed the creature is quiescent, and the ten-
tacles are indifferent to such contact; they are, as it were, in an
anaesthetized condition ; this state passes off as the organic demand
for food reasserts itself. Such an instance illustrates the regula-

1 Cf. S. J. Holmes, "The Beginnings of Intelligence," Science, N. S., 1911, Vol.
33- P- 473-

THE THEORY OF ANESTHESIA. 313

tory role which fluctuations in the general responsiveness of an
animal play in its normal life. Similar variations of neuro-mus-
cular responsiveness occur throughout the animal kingdom. This
is well illustrated by sleep, which is an instance of a normal or
"physiological" narcosis, characterized by a definite periodicity
and by affecting especially certain parts of the central nervous
system; the use of opiates illustrates how readily a chemically
induced narcosis may pass into the physiological form. From
such facts we must conclude that the essential basis of anaesthesia
consists not in a purely artificial modification of nervous or other
irritability, but in some normal or physiological modification
which is capable of being intensified and prolonged by the use of
certain physical and chemical agencies; these are the various
anaesthetizing agencies, such as the electric current, cold, or nar-
cotizing substances. From this point of view, anaesthesia is to be
regarded not as an essentially abnormal or artificial phenomenon,
but simply as an intensification of a normal physiological condi-
tion; and in investigating its essential conditions we are led first
to consider the normal inhibitions and depressions shown by all
living cells.

Instances of such normal inhibitions are innumerable. The
motor neurones innervating any group of muscles become in-
excitable during the activity of the antagonist groups, as Sher-
rington has shown; the respiratory nerve cells cease automatic
activity with over-oxygenation of the blood ; vasomotor, cardiac,
glandular, and muscular activities are subject to various forms-
of inhibition, partly nervous and partly chemical in origin.
Such inhibitory mechanisms play in normal life a part whose
importance is daily more widely recognized by physiologists.
Mechanisms of the inverse kind, which exercise sensitizing and
reinforcing influence on various functions, are also frequent in
organisms. A large part of these normal inhibitions and excita-
tions are now known to be due to chemical substances (hormones)
present in the blood and derived from ductless glands or other
sources of internal secretion. The regulation and integration of
bodily activities are thus largely under direct chemical as well as
nervous control. Such normal chemical inhibitions are probably
of the same nature as artificial inhibitions due to anaesthesia.

314 RALPH S. LILLIE.

In both cases the same kind of physico-chemical modification
in the irritable element appears to form the essential determining
condition.

The phenomena of anaesthesia have thus the widest biological
interest; they belong chiefly in the class of chemical inhibitions
or desensitizations. The inverse phenomenon of sensitization—
enhancement of irritability or responsiveness — is equally wide-
spread and plays an equally important physiological role.
Although its study has received less attention than that of
anaesthesia, its physiological interest is no less great. Irrita-
bility may in fact be altered reversibly either in the direction of
increase or decrease.

It is important to note that the same substance may cause
either increase or decrease of irritability or spontaneous activity,
•according to the conditions of concentration, temperature, physio-
logical state of the organism, etc. In the group of lipoid-solvent
substances, which include most of the anaesthetics in common
use, weak solutions very generally increase excitability; stronger
solutions, within a certain range of concentrations, produce
typical reversible narcosis; while still stronger solutions cause
cytolysis. The basis common to all of these effects requires to
be determined. The problem of the general nature of anaesthesia
is in fact inseparable from the wider problem of the nature and
conditions of irritability in general. The essential question may
be expressed thus: what is the physico-chemical basis of this
property of irritability, and what conditions determine its rever-
sible increase or decrease by chemical or other agents? This
problem is one of the most fundamental in biology; and the
phenomena of artificial anaesthesia are of general physiological
interest largely because of the light which they throw on this
larger problem.

Instances of increase in irritability or spontaneous activity
under the influence of low concentrations of anaesthetic sub-
stances are frequent in both animals and plants. One of the
most familiar is the general nervous excitement caused by small
•doses of ether, alcohol and other narcotics. Automatic rhythm-
ical activity, as of cilia, spermatozoa, or the heart beat, is very
generally heightened in weak solutions of alcohol and other

THE THEORY OF ANESTHESIA. 315

narcotics. The nerve-cells controlling the heart beat of Limulus
show a faster rhythm in weak solutions of alcohol, chloral
hydrate, choretone, and chloroform.1 Hamburger has shown
that many lipoid-soluble substances — iodoform, chloroform, tur-
pentine, benzol, chloral hydrate, camphor, fatty acids, soaps-
increase the amoeboid and phagocytic activity of leucocytes,
while stronger solutions decrease this activity.2 A similar rule
appears to hold for the respiratory center of vertebrates.3 Ac-
cording to Vernon,4 weak solutions of narcotics increase the con-
sumption of oxygen in isolated tissues like the kidney. Tashiro
and Adams find that low concentrations of urethane and chloral
hydrate increase the excitability of nerve as well as its output of
carbon-dioxide; in higher concentrations both are decreased.5
The staircase phenomenon in irritable tissues is probably due
to the stimulating action of small quantities of substances
("fatigue-substances") which in higher concentrations decrease
irritability. Small quantities of alcohol increase the responsive-
ness of voluntary muscle and the energy of its contractions.6
The musculature of medusae shows increased response to mechan-
ical stimuli in sea water containing a little alcohol.7 Similar
facts are met with in plants. Many depressant substances, when
present in low concentration, increase the rate of growth.8
Traces of ether have an accelerating or forcing influence on plant

1 A. J. Carlson, Amer. Journ. Physiol., 1906, Vol. 17, p. 182.
- Hamburger, "Archives Neerlandaises des Sciences Exactes et Naturelles,"
Serie III, B, 1911, p. i; A rchiv fiir Anatomie und Physiologic, Physiol. Abth., 1913,

P- 77-

3 Cf. Hamburger, "Koninklijke Akademie van Wetenschappen te Amsterdam,"

1915, Vol. 17, P- 1325-

4 H . M. Vernon, "The Function of Lipoids in Tissue Respiration," Journ. of
Physiol., 1912, Vol. 45, p. 197.

5 Tashiro and Adams, Internal. Zeilschr. f. physik-chem. Biol., 1914, Vol. i,
p. 450.

6 Cf. Lee and Salant, " The Action of Alcohol on Muscle," Amer. Journ. Physiol.,
1902, Vol. 8, p. 61.

7 Cf. Bethe, "Allgemeine Anat. u. Physiol. d. Nervensystems," Leipzig, 1903,
p. 359. One half per cent, alcohol decidedly increases the mechanical irritability
of the isolated central portion of the medusa Cotalorrhiza. F. S. Lee observed that
in the Woods Hole medusa Gonionemus the spontaneous contractions of the swim-
ming bell are markedly increased by small quantities of alcohol (1/16 to 1/4 per
cent.); cf. Amer. Journ. Physiol., 1903, Vol. 8, p. xix.

8 Numerous instances of this effect are cited by Czapek, Biochemie der PJlanzen,
Jena, 1913. P- 148.

316 RALPH S. LILLIE.

growth, — a fact of which practical use is made by horticulturists.
Increase in oxygen-consumption under the influence of chloro-
form and ether has been observed by Elfving and others; higher
concentrations decrease oxygen-consumption.1 Demoor and
others have observed an acceleration of protoplasmic rotation in
plant cells during the early stages of chloroform and ether
narcosis; alcohol also causes this effect.2 Traces of ether increase
the irritability of sensitive plants (Mimosa} ;3 higher concentra-
tions cause typical anaesthesia.4

A probably related phenomenon is seen in certain artificial
modifications of response induced in various organisms by weak
solutions of anaesthetics. A striking instance is the reaction of
many lower animals to light. Loeb has found that Daphnice,
wrhich normally show little or no directive light-response, become
positively heliotropic in weak solutions of alcohol and other
narcotics, in concentrations of a third to a half of those required
for anaesthesia.5 Similarly I have found that the larvae of the
marine annelid Arenicola, which normally exhibit strong positive
heliotropism, become negative in weak solutions of various
anaesthetic substances. Similar observations have been made
by Torrey, A. R. Moore, and other observers.

The phenomenon of reversible decrease of activity or respon-
siveness is anaesthesia. The vital processes' subject to such
reversible arrest are of the most varied kind. They include

1 Cf. Czapek, loc. oil., p. 159, for instances of this effect. Tashiro and Adams
(loc. cit.) cite observations of Kosinski showing that respiration in yeast cells is
increased in presence of 0.5 per cent, ether; 5 per cent, reduces respiration one half,
while 7 per cent, almost stops it. Baer and Meyerstein find increased oxidation
of oxy-butyric acid to acetone in the perfused liver under the influence of various
compounds which in higher concentrations check oxidations, e. g., tricholor-q^cohol,
p- and ra-oxy-benzoic acid, ^-oxy-benzaldehyde (cf. p. 458 of their paper in Arch,
exper. Path. u. Pharm., 1910, Vol. 63).

• Cf. the instances cited by Czapek, p. 161. H. Nothmann Zuckerkandl has
also observed this effect with low concentrations of alcohol and ether (cf. footnote
2, p. 317)-

3 Personal communication from Professor J. M. Macfarlane, of the University
of Pennsylvania.

4 Cf. Claude Bernard, "Lecons sur les phenomenes de la vie communs aux
animaux et vegetaux," Paris, 1878. Anaesthesia of plant-growth was also studied
by Bernard.

'" J. Loeb, Biochem. Zcitschr., 1909, Vol. 23, p. 93.

THE THEORY OF ANESTHESIA. 317

amoeboid movement;1 protoplasmic rotation in plant cells;2 all
processes depending on response to stimulation, like muscular
contraction and stimulation and conduction in nerve; automatic
rhythmical activities like the heart beat or the motion of cilia
or spermatozoa; cell-division;3 the artificial initiation of develop-
ment in unfertilized eggs;4 the stimulating, cytolytic or other
physiological action of salt solutions;5 various fermentative and
oxidative processes;" light-production, e. g., by luminous bac-
teria;7 typical metabolic processes like the assimilation of carbon
dioxide by plants;8 growth processes in plants and animals, and
developmental processes dependent on growth and cell-division.
It is especially worthy of note that not only motor activity and
responsiveness are subject to control of this kind, but also
processes like growth and development. The growth of seedlings
may be temporarily arrested by ether in sufficient concentration,
as Claude Bernard showed.9 Cell-division in the eggs of sea-
urchins is checked by anaesthetics in concentrations of the same
order as those required for neuro-muscular anaesthesia in Areni-
cola larvae.10 It is thus not surprising that developmental

1 Cf. Hamburger: loc. cit.

2 Cf. H. Nothmann-Zuckerkandl: Biochem. Zeitschr., 1912, Vol. 45, p. 412.

3 Cf. (e. g.) my observations on anaesthesia of cleavage in sea-urchin eggs,
Journ. Biol. Chem., 1914, Vol. 17, p. 121. The development of astral radiations in
dividing egg-cells is prevented by etherization, and existing radiations are sup-
pressed: cf. E. B. Wilson: Arch. f. Entwicklungsmechanik, 1901, Vol. 13, p. 353.

4 R. S. Lillie, Journ. Ex per. Zool, 1914, Vol. 16, p. 591.

6 Cf. my papers on antagonisms between salts and anaesthetics; Amer. Journ.
Physiol., 1912, Vol. 29, p. 372; Vol. 30, p. i; 1913. Vol. 31, p. 255.

6 Cf. the papers of Warburg: Zeitschr if t /. physiol. Chemie, 1910, Vol. 69, p. 452,
and Vol. 70, 1911, p. 413; Pfliiger's Arch., 1914, Vol. 155, p. 547; Warburg and
Wiesel, 1912, Vol. 144, p. 472; Usui, ibid., 1912, Vol. 147, p. TOO; Meyerhof,
Pfliige/s Archiv, 1914, Vol. 157, p. 251. Claude Bernard describes the reversible
inhibition of yeast-fermentation by anaesthetics (cf. footnote 9, below).

7 E. N. Harvey, BIOLOGICAL BULLETIN, 1915, Vol. 29, p. 308.

8 Cf. Claude Bernard, loc. cit.; Overton, "Studien iiber die Narkose," p. 182.

9 Loc. cit. In Bernard's address, "La Sensibilite," given in 1876 before the
French Association for the Advancement of Science, and published in his book,
"La Science Experimentale," Paris, 1890, he cites instances of anaesthesia of the
most various vital processes, including photosynthesis, germination and growth
in plants, fermentation by yeast, development of the hen's egg, — and concludes:
"We may say that everything living is sensitive and can be anaesthetized; whatever
is not sensitive is not living and cannot be anaesthetized "(p. 224).

10 R. S. Lillie, Journ. Biol. Chem., 1914, Vol. 17, p. 121.

31 8 RALPH S. LILLIE.

processes, depending as they do on cell-division and growth,
are similarly subject to inhibition by anaesthetics. Stockard
and McClendon1 have shown that such substances induce ab-
normalities like cyclopia in developing fish eggs, an effect which
is to be referred to the arrested development of certain portions
of the central nervous system, especially the anterior region of
the fore-brain between the optic vesicles. Abnormalities of
growth and development as well as of irritability may thus be
produced under the influence of anaesthetics. Since an automatic
powrer of growth — i. e., increase in specifically organized and
metabolically active material — is perhaps the most fundamental
manifestation of vital activity, the fact that it is subject to
reversible arrest by anaesthetic substances is of the greatest
biological significance, and illustrates in a striking manner the
unity of the conditions which control the most various cell-
processes. We may infer that in the general course of con-
structive as well as of destructive metabolism, processes are
concerned which are identical writh those underlying the ordinary
manifestations of stimulation. These latter, however, are almost
certainly dependent on surface-changes, of which the most
essential are probably variations in the electrical polarization
of the plasma-membranes (see below, p. 365). The controlling
influence of membrane-processes in such fundamental physio-
logical activities as growth and assimilation is thus indicated by
this susceptibility to arrest by anaesthetics.

In any complete theoretical discussion of anaesthesia it is
necessary first to consider the chief conditions under which
living cells in general undergo reversible decrease or loss of
irritability. This change occurs under a variety of external
conditions, mechanical, thermal, electrical and chemical. Me-

11 Cf. C. R. Stockard, Archiv f. Entwicklungsmechanik, 1907, Vol. 23, p. 249;
Anatomical Record, 1909, Vol. 3, p. 167 ("The Artificial Production of One-eyed
Monsters .... by the Use of Chemicals"); Amer. Journ. Anal., 1910, Vol. 10,
p. 369 ("The Influence of Alcohol and other Anaesthetics on Embryonic Develop-
ment"). Also McClendon ("Physical Chemistry of the Production of One-eyed
Monstrosities") in Amer. Journ. Physiol., 1912, Vol. 29, p. 289. Stockard observed
the production of these abnoimalities first with magnesium salts, later with lipoid-
solvent anaesthetics (alcohol, ether, chloroform, chloretone). Developmental
defects are also produced in mammals by alcohol (cf. Stockard, Arch. f. Entwick-
lungsmech., 1912, Vol. 35, p. 569; Amer. Naturalist, 1913, Vol. 47, p. 641).

THE THEORY OF ANAESTHESIA. 319

chanical shock may cause temporary loss of irritability. This is
probably an effect of over-stimulation and due to prolongation
of the refractory period ; it resembles in some respects the effect
produced in voluntary muscle by poisons like veratrin, which
greatly prolongs the relaxation phase and the recovery of irrita-
bility following contraction. The paralysis due to mechanical
shock differs however from that of anaesthesia in important
respects; it represents an injury from which the cell can recover,
while true unmixed anaesthesia is quite without injurious action.
Certain effects of altered temperature have a closer resemblance
to anaesthesia. Most cells and tissues, within the range of
temperature in which they show normal activity, show decreased
automatic activity with decrease of temperature. Thus accord-
ing to Snyder1 the heart of the tortoise shows eighteen beats per
minute at 20° and thirty-five at 30°. Observations on the hearts
of other animals have given similar results.2 Within the physio-
logical range of temperature the rate is doubled or trebled by a
rise of 10°. This rate of change of velocity with temperature,
or temperature-coefficient, is characteristic of chemical reactions
in general and is not a distinctively physiological phenomenon.
Metabolic and hence vital activity is slowed by cooling just as
any other chemical process is slowed. The same temperature-
coefficient is shown by a large number of physiological processes
including cell-division, rate of conduction in nerve, enzyme action
and many others.3 Thus the above effect of cold is dependent
simply on a slowing of chemical processes in cells and has in it
nothing distinctively vital. It is important, however, to con-
sider this effect in relation to the problem of anaesthesia, for a
simple decrease in reaction-velocity, due to the presence of anti-
catalytic substances, is held by various investigators to be the
essential condition of anaesthesia. Decrease in the rate of a
physiological process, like the heart beat, or muscular contrac-
tion, or the spread of the excitation-wave in nerve, is not how-

1 University of California Publications, Physiology, 1905, Vol. 2, p. 125.

• Cf. C. D. Snyder, Amer. Journ. Physiol., 1906, Vol. 17, p. 350; Zeitschr. f. allg.
Physiol., 1912, Vol. 14, p. 263; Robertson, BIOL. BULL., 1906, Vol. 10, p. 242;
C. G. Rogers, Amer. Journ. Physiol., 1911, Vol. 28, p. 81; BIOL. BULL., 1914,
Vol. 27, p. 269; Loeb and Ewald, Biochem. Zeitschr., 1910, Vol. 28, p. 340.

3 For instances cf. Snyder, "Temperature-coefficients of Various Physiological
Actions," Amer. Journ. Physiol., 1908, Vol. 22, p. 309.

32O RALPH S. LILLIE.

ever necessarily associated with a change in the irritability and
other vital properties of the tissue; in fact moderate cooling
may increase the irritability of nerve. Irritability and rate of
metabolic processes represent in fact two independent variables.
We infer that anaesthesia is not simply an expression of a decrease
in the velocity of certain chemical reactions, such as oxidations,
but that some other factor enters, probably physical in nature.
Certain other effects of temperature bear a closer resemblance
to true anaesthesia. Various irritable tissues become reversibly
insensitive at temperatures slightly below or above the normal
physiological range. Thus the frog's heart shows an accelerated
rate with rise of temperature up to 36° or 37°; it then becomes
temporarily inactive and insensitive ("heat-standstill"), but
resumes beating if the temperature is lowered. Similarly the
musculature of tropical medusae becomes irresponsive at 40°
and recovers on lowering the temperature.1 This condition of
reversible heat-paralysis has certain suggestive resemblances to
anaesthesia. Cooling may produce a similar loss of sensitivity
in cells whose normal temperature is high, as those of tropical
marine animals2 or warm-blooded vertebrates. Sensory nerve
endings, musculature, etc., lose sensitivity if cooled sufficiently,
and recover on warming. In these effects structural alterations
due to modification of the colloids of the cells (as gelation) are
probably concerned; and, as will be shown later, there are indica-
tions that similar changes form part of the essential basis of true
anaesthesia. The fact that changes of temperature may thus
alter the irritability of the tissue independently of their influence
on reaction-velocity as such, is highly important to the general
theory of narcosis; and it appears unfavorable to those theories
which refer anaesthesia to a simple change in the rate of chemical
processes like oxidation. Recent experiments by Loeb and
Wasteneys3 on sea-urchin eggs illustrate this. They found that
during a condition of narcosis sufficient to arrest cell-division
completely, the rate of oxidation is lowered by only 10 per cent.;

1 Cf. E. N. Harvey, Carnegie Institution Publications, No. 132, 1910, p. 32.

2 Cf. A. G. Mayer, "Effects of Temperature upon Tropical Marine Animals,"
Carnegie Institution Publications, No. 183, 1914, p. i.

zjourn. Biol. Chem., 1913, Vol. 14, p. 517; Biochem. Zeitschr., 1913, Vol. 56,
p. 295-

THE THEORY OF ANESTHESIA. 321

the same effect on the rate of oxidation results from a simple
lowering of temperature by 2° to 3°, a change which only slightly
retards cell-division. Decrease in the rate of oxidation as such
is thus quite insufficient to account for the inhibitory effect.
The fact that, e. g., in frogs' muscle a lowering of temperature
of 20° (e. g., from 35° to 15°) — which reduces the rate of oxidation
to one fifth of its former value — leaves irritability unimpaired,
indicates that any explanation of anaesthesia based on simple
decrease in reaction-velocity is inadmissible. A similar decrease
in the rate of oxidation can be produced by lipoid-solvent
anaesthetics only in concentrations which are much higher than
those requisite for anaesthesia.

The constant electric current produces in many irritable tissues
effects closely resembling true anaesthesia. Many physiological
inhibitions may be caused by passing a constant current through
the tissue. There is indeed reason to believe that many of the
normal inhibitions, e. g., in the neurones of reflex arcs, are elec-
trical in their nature.1 The anti-stimulating or desensitizing
action of the constant current thus deserves careful consideration
in any general theory of anaesthesia. As is well known, the
action of the current on irritable tissues like nerve and muscle
is polar; where the current enters the tissue there is decreased
irritability, depression, or inhibition (anelectrotonus) ; where it
leaves there is excitation or heightened irritability (catelectro-
tonus). Thus a nerve becomes inexcitable near the anode when
the constant current is passed ; under similar conditions the heart
is inhibited and voluntary muscle relaxed. The condition is
reversible, and in fact constitutes a typical local anaesthesia.
The essential basis of the effect appears to be an altered electrical
polarization of the cell-surface. Near the anode, where the
current enters the cell or irritable element, the normal outer
positivity of the semi-permeable plasma-membrane is increased.
Apparently this change renders the membrane irresponsive to
stimulation. Variations in the electrical polarization of the
plasma-membrane are in all probability constantly associated
with variations in irritability. The facts of electrotonus show
that such changes of polarization may profoundly alter the

1 Cf. my discussion of this possibility in Amer. Journ. Physiol., 1913, Vol. 31,
pp. 284 seq.

322 RALPH S. LILLIE.

irritability and automatic activities of the cell. This general
conception is of the greatest importance in the theory of anaesthe-
sia, and will be reconsidered later.

The most important instances of anaesthesia are those produced
by chemical substances. First it should be noted that substances
belonging to the most various classes may have anaesthetic effects.
This fact is overlooked in theories like those of Overton and
Meyer, Traube, and others, which refer anaesthesia to the special
properties of lipoid-solvent substances, which are regarded as
acting either by dissolving in the lipoid constituents of the cell
or by adsorption at the surfaces of membranes or other structures.
The anaesthetic influence of certain neutral salts shows, however,
that lipoid-solubility or surface activity is not essential to narcotic
action; magnesium sulphate has long been used by naturalists
to narcotize marine animals; more recently it has been applied
by Meltzer to produce spinal anaesthesia in mammals. Similar
reversible depressant effects are produced by potassium salts.
Salts of calcium and strontium also cause reversible desensitiza-
tion of isolated nerve and muscle. In most animals the calcium-
content of the medium has marked influence on irritability and
automatic activity; this is well shown in the case of vertebrate
muscle; lowering the ratio of calcium to sodium in indifferent
media like Ringer's solution has a sensitizing effect, and if the
calcium falls too low the muscle twitches spontaneously; increas-
ing the calcium-sodium ratio has a desensitizing action; these
effects are reversible.1 Calcium also antagonizes the stimulating
and sensitizing action of pure solutions of sodium and other salts
on muscle and nerve. Similarly the heart beats best in media of
a certain calcium-content. In marine medusae (Rhizostoma ac-
cording to Bethe) the rhythmical beat ceases when the animal is
transferred to calcium-free sea water, and is restored if calcium
is added; still further addition of calcium again arrests the move-
ment.2 These facts make it clear that alteration of the salt-
content of the media may have effects essentially identical with
anaesthesia. This is a fact of much theoretical interest, since it
indicates that the general condition of the colloids of the cell,

1 For instances of these various effects cf. J. Loeb's article on "Physiological
Actions of Ions," in Oppenheimer's " Handbuch der Biochemie," 1909, Vol. 2, p. 104.

2 Cf. Bethe, Pfliiger's Archiv, 1909, Vol. 124, p. 561.

THE THEORY OF ANESTHESIA. 323

especially of the surface-layer or plasma-membrane, is a chief
factor in determining the irritability and automatic activity of
the living cell. Further evidence of this will be given later.
Modification of the properties of this layer may result from an
alteration in the state of either its lipoid or its protein con-
stituents, and if this alteration is reversible a temporary inhibi-
tion, or anaesthesia, may result. A related condition is seen in
the irritable tissues of higher animals, such as muscle and nerve.
In these tissues irritability depends on the presence of certain
salts in the media; simple withdrawal of salts and replacement
by indifferent non-electrolytes like sugar is followed by a tem-
porary loss of irritability; the latter is restored by return to
media containing salts, especially sodium salts.1 The muscula-
ture of marine animals (e. g., Arenicola larvae) is similarly in-
activated in isotonic solutions of non-electrolytes, and regains
irritability in isotonic solutions of various neutral salts. Solu-
tions of sodium salts, together with a small proportion of calcium,
are especially favorable. Sodium may be partly replaced by
lithium, but not by other metals.2 Thus the presence of certain
salts in the medium is necessary for normal irritability, — hence
the effects of isotonic sugar solution, which are due to the absence
of salts, not to any special action of the non-electrolyte. The
salt-content of the medium may be reduced to a small fraction-
one tenth or less — of the normal by diluting the physiological
salt solution with isotonic sugar solution, without causing loss
of irritability. But with the complete withdrawal of salts
irritability soon disappears. In cases like this, where normal
irritability is dependent on the salt-content of the media, modifica-
tion of the latter may induce a reversible desensitization closely
resembling anaesthesia. Probably several factors enter in the
production of this effect, of which the two chief are, a direct
change in the properties of the plasma-membrane (colloidal con-
sistency, electrical polarization), and a lowering of the electrical
conductivity of the medium.

The reaction of the medium (H-ion concentration) also has
profound influence on the irritability and automatic activity
of many cells; and a reversible suspension of function akin to

1 Cf. Overton, P finger's Archiv, 1902, Vol. 92, p. 346, and 1904, Vol. 105, p. 176.

2 R. S. Lillie, Amer. Journ. Physiol., 1909, Vol. 24, p. 459.

324 RALPH S. LILLIE.

anaesthesia may result from a slight change in this reaction. In
higher vertebrates the normal reaction of the blood plasma is
not far from neutral, and varies only slightly from a constant
normal value (CH ;= 0.35 X io~7 to 0.5 X io~7); but certain
cells of the central nervous system are especially sensitive to
such variations. The activity of the respiratory center is ap-
parently regulated by the variations in the H-ion concentration
of the blood, cessation of activity resulting from a slight decrease
(i. 6., increased alkalinity), and increased activity from a slight
increase.1 Reversible cessation of activity may thus result from
a slight change in reaction, due, e. g., to loss of CO2. Similar
conditions are known to exist in certain marine animals; thus
according to Bethe,2 slight increase in the alkalinity of the sea
water arrests, while slight acidulation accelerates, the rhythmical
contraction of medusae. On the other hand, the activity of
many cells and tissues is favored by slight increase in external
alkalinity, and depressed by slight acidulation. The irritability
and automaticity of living cells are thus largely a function of the
reaction of the medium, and this fact has an intimate bearing on
the question of the mechanism of anaesthetic and other inhibi-
tions. The precise physico-chemical basis of this action is un-
certain, but it probably depends chiefly on alterations in the
electrical polarization of the cell-surface. Slight variations in
alkalinity or acidity are known to produce marked effects on
the electrical polarization of surfaces bathed by media of approxi-
mately neutral reaction.3

The chief chemical substances exerting a reversible depressant
influence on a wide range of vital activities are those numerous
and chemically diverse organic compounds of which the most
evident common property is a solvent action on, or solubility
in, fats and fat-solvents. Substances of this class form the
majority of anaesthetics in common use; they include alcohols,
ethers, esters, aldehydes, ketones, nitrites, amides, various normal
and substituted hydrocarbons (chloroform, benzol, etc.) and
other related compounds. Most of these bodies are members of

1 For a recent review and discussion of the evidence cf. Winterstein, Biochem.
Zeitschr., 1915, Vol. 70, p. 45.

2 Pfliiger's Archiv, 1909, Vol. 127, p. 219.

3 Cf. Haher and Klemensiewicz, Zeilschr.f. physik. Chemie, 1909, Vol. 67, p. 385.

THE THEORY OF AN.ESTHESIA. 325

homologous series; and it is highly characteristic of such series
that the ratio of oil-solubility to water-solubility (oil-water
partition-coefficient) increases regularly with increase in molecu-
lar weight. At the same time the narcotizing power increases;
i. e., in any single series (e. g., alcohols) the higher the molecular
weight the lower the concentration required for narcosis. It was
this general parallelism that led Overton and Meyer to the view
that anaesthetic powrer, in the case of any substance, is a direct
function of its solubility in the fat-like or lipoid constituents of
the cell. That a connection exists between the fat-solvent and
the anaesthetic properties of a compound had previously been
suggested by Bibra and Harless in 1847, and the same view was
later expressed by Hermann, C. Bernard, Richet, Ehrlich and
others.1 The first systematic studies of this relationship were
however those of Overton and Meyer, the results of whose experi-
ments, carried on independently, were published about the same
time (1899).

In a study of the permeability of animal and plant cells to
various types of compounds, Overton2 had reached the conclusion
that solubility in lipoids was the chief factor determining the
ready entrance of a compound into cells; compounds with well-
marked power of penetration belonged chiefly to the narcotic
group; and in a later extensive investigation on narcosis in
tadpoles3 a far-reaching parallelism wras found between the oil-
water partition-coefficients of a wide range of organic compounds
and their narcotizing action. The nature of Overton's results
may be best seen from the following series, which gives the
concentrations required to narcotize tadpoles in the case of }he
ethyl esters of the first five fatty acids. (See Table I.)

The narcotic action is seen to increase steadily with decrease
in the water-solubility, — i. e., increase in the ratio of partition
between oil and water. Each member of the series is from two
to three times as effective as its immediate predecessor. This
rule appears to hold very generally for members of homologous

1 For an account of these earlier views cf. Overton, "Studien liber die Narkose,"
June, 1901.

2 Overton, "On the General Osmotic Properties of Cells," etc., Vierleljahrsschr.
d. naturf. Gesellsch. in Zurich, 1899, Vol. 44, p. 88.

3 "Studien iiber die Narkose," 1901.

326 RALPH S. LILLIE.

series, and a large number of similar instances have been collected
by Traube and other recent investigators.1 Numerous other
experiments with alcohols, hydrocarbons, aldehydes, ketones,
etc., showed a similar increase in narcotic action with increase
in the oil-water partition-coefficients. Overton accordingly drew
the conclusion that narcotics act by dissolving in certain sub-
stances, contained especially in nerve-cells, which resemble fats
in their solvent properties; these substances are the lipoids,
especially lecithin and cholesterin, which appear to be essential
constituents of protoplasm; it is the physical modification of
these substances, due to their being charged or impregnated
with the lipoid-soluble narcotic, that forms the essential condition
of anaesthesia. Meyer's conclusion wras similar;2 the narcotiza-
bility of cells is thus related to the nature and the proportion of
the lipoids present in the protoplasm; the high susceptibility of
nerve-cells is probably dependent on their high lipoid-content.
The unequal action of different narcotics depends on their unequal
partition-coefficients, which determine their distribution in a
mixture of water and lipoid substances. The greater the relative
lipoid-solubility the larger the proportion of the anaesthetic
present in solution in the lipoid cell-constituents when the
partition-equilibrium is reached. Hence, if the lipoid-solubility
of a substance is very high, extremely dilute solutions may exert
anaesthetic action. Overton, for example, found that phenan-
threne could narcotize tadpoles in dilutions so low as one part
in 1,500,000 of water.

TABLE I.

Narcotizing Concentration.
Ester. (Mols per Liter). Solubility in Oil and Water.

Ethyl formate o.ojm-.ogm Oil: water =4:1

acetate o^m In 15.2 parts \vater; in all parts oil

propionate oim-.oi2m 50

butyrate 0043^ " 190

( isobutyrate) 0057^ " 140

valerianate 00197/2 ' 500

Overton's study of permeability had led him to the conclusion
that the outer layer or plasma-membrane of cells consists largely
of lipoid material; in this way he explained the ready entrance

1 Cf. Traube, "Theorie der Narkose," Pfluger's Archiv, 1913, Vol. 153, p. 276.

2 Hans Meyer, Arch. f. exper. Path. u. Pharm., 1899, Vol. 42, p. 109.

THE THEORY OF ANESTHESIA. 327

of lipoid-soluble substances into cells. Now it is an evident
corollary of Overton's hypothesis that if the anaesthetic acts
by changing the physical state of the lipoid cell-constituents it
must affect the properties of a lipoid-rich cell-structure like the
plasma-membrane. Overton, however, does not refer narcotic
action specifically to a modification of the plasma-membrane
alone, but to a general modification in the physical state of all
cell-lipoids, wherever situated. Recently, however, much evi-
dence has accumulated indicating that the essential influence is
that exerted on the plasma-membrane, and that it is the modifica-
tion in the properties of this structure which determines the
characteristic anaesthetic effect. This evidence and its implica-
tions will be considered later.

The hypothesis of Overton and Meyer has received wide
acceptance. It is not clear, however, why simple solution of
chemically indifferent substances in the lipoids of the tissue
should so modify its irritability; and Overton and Meyer do not
attempt to explain this connection. The parallelism between
lipoid-solubility and narcotic action is not an exact one, and
many exceptions to the rule are known. The powerful narcotic
action of chloral hydrate, which is several times more soluble in
water than in oil, is not thus explained; and lipoid-insoluble
neutral salts of magnesium and other metals may exert typical
narcotic action. Evidently other factors than solubility may
enter. Yet the evidence adduced by Overton and Meyer, as
well as by more recent investigators, leaves no doubt that in the
case of organic anaesthetics high lipoid-solubility is typically
associated with marked narcotic action. The reversibility of
anaesthesia corresponds to the reversibility of the process of
solution. The chemical indifference of many anaesthetics is thus
not surprising, since the substance acts not by chemical combina-
tion but by simple solution in the cell-lipoids.

According to Overton and Meyer's hypothesis it is this solution
of the narcotic in the lipoid which determines anaesthetic action.
This view has recently been attacked from various sides. Ac-
cording to Traube,1 the anaesthetic acts not by dissolving in the

1 I. Traube, "Theorieder Narkose," Pfluger's Archiv, 1913, Vol. 153, p. 276, and
Vol. 160, 1915, p. 501; also "Theorie des Haftdrucks und Lipoidtheorie," Biochem-
Zeitschr., 1913, Vol. 54, p. 305, and other papers cited there.

328 RALPH S. LILLIE.

cell lipoids, but rather by undergoing surface-condensation or
adsorption at the physiologically active surfaces within the
living system ; these may be the surfaces of special cell-structures
or of colloidal particles, whether lipoid or protein. The catalytic
activity of these surfaces is thus decreased, and the reaction-
velocities of essential chemical processes, especially oxidations,
is lowered. A corresponding depression of cell-functions results.
Whether this effect is to be attributed to a displacement of
metabolically active water-soluble substances like sugar, whose
surface-activity is relatively small, or to a direct alteration in
the catalytic properties of the physiologically active surfaces
themselves, is uncertain. The essential feature of Traube's
view is that it regards the surface-activity of a narcotic com-
pound, i. e., its influence in lowering surface-tension — rather than
its lipoid solubility — as the determining factor in its depressant
action. This surface-activity determines the degree of adsorp-
tion, and hence, indirectly, of anaesthetic action. It is well known
that the surface-tension of such a solvent as water, in contact
with air or with another liquid or a solid, is greatly influenced
by the presence of dissolved substances. This influence is
usually in the direction of a decrease. A few substances like
inorganic salts and sugars increase the surface-tension of water,
although the effect is slight; but the majority, especially of
organic substances, cause well-marked and often great decrease.
This is especially true of substances whose water-solubility is
limited; and in general the more soluble a substance is in oils
or other wrater-insoluble organic solvents, and the less soluble
in water, the greater is its influence (for a given molecular concen-
tration) on the surface-tension of water. In a capillary tube the
level of pure water or of an aqueous solution is raised, by the
contractile force or tension of the surface of the water-film lining
the walls of the tube, to a certain height above the level of the
water outside. This height (h) is proportional to the surface
tension of the water (cr), and inversely proportional to the
radius of the tube (r} and the specific gravity (g) of the liquid
(h •-- 2<T/rg). The relative surface-tensions of aqueous solutions
may thus be determined by measuring the heights to which the
column of solution is raised by capillarity in a given tube. This

THE THEORY OF AN/ESTHESIA.

329

height is decreased by surface-active substances, and the degree
of capillary activity or tension-lowering action of different sub-
stances can thus be determined. Now in any homologous series
this action (for equimolecular solutions) is found to decrease as
the molecular weight increases. The surface-tension in milli-
grams per linear centimeter (i. e., the pull exerted by a strip of
surface one centimeter wide) of ml\ solutions for the first five
aliphatic alcohols at 15° is given by Traube as follows: the
surface-tension of the m/q. solution of dextrose, a physiologically
important surface-inactive compound, is given for comparison.

TABLE II.

Liquid (Solutions = m/4).

Surface-tension
(Milligrams per
Centimeter).

Molecular Concen-
trations of Isoca-
pillary Solutions.

Concentration for
Narcosis of Tad-
poles (Overton).

Water

7?

7K/4 dextrose

72.7

Methyl alcohol

7O $

Id O

0 527W— 0 62m

Ethyl

67 T

c o

o T]m—o 31 in

w-propyl

eg Q

i 6

o i im

/-butyl "

44.0

0.46

o oa^m

i-amvl

™.<;

0.14

o.Q2T,m

The second column gives the concentrations required to effect
a definite lowering of surface-tension. It will be observed that
the surface-activity of each member (as measured by the recip-
rocals of the isocapillary concentrations) is approximately a
third of that of its immediate successor. The third column
shows that the narcotic activity increases from each member to
the next following in a closely similar proportion. Results of
this kind are on the whole typical for homologous series. The
question arises as to their general physiological significance.

According to Traube the essential physico-chemical factor in
these physiological effects is the characteristic influence which
surface-tension has upon the distribution of dissolved substances
in any polyphasic system. The general principle of Willard
Gibbs and J. J. Thomson states that substances which lower the
surface-tension of any solvent attain, when equilibrium is reached,
a higher concentration in the surface-layer than in the interior
of the solvent; a surface-condensation or adsorption thus results,
which is the greater the greater the surface-activity of the dis-

33O RALPH S. LILLIE.

solved compound. Hence substances having a high degree of
surface-activity are as a class readily adsorbed. The effect is
the same as if a relatively slight coherence existed between the
solvent and the dissolved substance. Hence Traube conceives
of a surface-active substance as one in which the union or adhesion
between solute and solvent is slight; i. e., relatively little work
is required to separate the substance from solution; and he has
introduced the expression "Haftdruck" (adhesion-tension or
solution-affinity) to designate this condition. The capillary
activity of any substance in a given solvent varies inversely
with its solution-affinity (Haftdruck) relatively to that solvent.
The lower the solution-affinity relatively to water the greater is
the tendency of any substance to pass out of its solution in water;
this tendency favors the entrance of capillary-active substances
into other adjoining solvents or media, e. g., into and through
the membranes bounding cells.1 The ready penetration of such
substances into living cells is in fact referred by Traube not to
lipoid-solubility, but to low solution-affinity in relation to the
medium bathing the cell. A tendency to surface-condensation
or adsorption is a characteristic accompaniment of low solution-
affinity to water; the marked physiological activity shown by
surface-active substances as a class is a direct consequence of
this tendency.

According to data already cited, the narcotic activity of organic
substances shows a parallelism with both capillary activity

1 Traube's attempts to apply this conception to a general explanation of osmotic
phenomena are not convincing. The fact that lipoid-solubility varies in the same
direction as capillary activity makes the question difficult to decide by experiments
on living cells. But in the case of dead cells, as well as of more permeable partitions
like parchment paper or collodion, surface-active and surface-inactive substances
appear to penetrate with equal ease. The difference between the rate of penetra-
tion of dissolved lipoid-insoluble substances into living and into dead cells shows
that the permeability of the partition (/. e., relative resistance to diffusion) is the
deciding factor in their entrance. Furthermore, instances are well known where
the rate of penetration of a substance through an artificial partition varies directly
with its solubility in the material composing the partition. Flusin's experiments
with rubber membranes are a good instance of this. The diffusion of substances
through the surface-films bounding the cells implies a passage either through or
between the membrane-constituents; and in the case of the living cell, solution of
lipoid-soluble substances ill the lipoids of the membrane is probably the main
factor in their entrance, although this entrance may be favored by adsorption due
to surface-activity.

THE THEORY OF ANESTHESIA.

331

and lipoid-solubility. Other physiological effects (e. g., mem-
brane-formation in sea-urchin eggs,1 reversal of the sense of
heliotropism, sensitizing action, cytolytic action) show a similar
parallelism. The question of whether the particular physio-
logical effect under consideration is determined by one or the
other factor, or by the interaction of both, has to be decided by
further evidence. In favor of the view that lipoid-solubility
rather than surface-activity is the essential determining factor
in the action of lipoid-soluble narcotics, is the fact that the action
varies with temperature in the same direction as the oil-water
partition-coefficient. This is shown in a remarkable manner in
the following table from Hans Meyer.2 The concentrations
required to narcotize tadpoles were determined at the two tem-
peratures 3° and 30° using (a) narcotics whose relative solubility
in oil decreases with rise in temperature, and (&) where it increases.
The critical anaesthetizing concentrations for the following six
anaesthetics are given in the table.

TABLE III.

Anaesthetic.

Critical Concentration for
Anaesthesia.

Oil/Water Partition Coefficients.

At 3°.

At 30°.

At 3°.

At 30°.

A. Salicylamide .

W/I30O
m/soo
mlgo
m/3
TO/SO
ml?.

m/6oo
m/2oo
mho
mil
m/2$o
mh

22.23
0.67
O.OQ9
O.O26
0.053

O.IA6

14
0-43
0.066
0.047
0.236
0.2^

Benzamide

Monoacetin

B. Ethyl alcohol .

Chloral hydrate

Acetone. .

In the first three compounds the relative lipoid-solubility
decreases and the narcotizing concentration increases with rise
of temperature; while with the others the conditions are reversed.
Thus simple cooling suffices to restore activity to tadpoles anaes-
thetized in chloral hydrate at 30°. If adsorption under the
influence of surface-tension were the main factor in these effects,
such a change of narcotic power with temperature would be
inexplicable, since surface-tension is influenced in a totally
different manner by change of temperature. The fact that

1 Cf. J. Loeb, "Artificial Parthenogenesis and Fertilization," University of
Chicago Press, 1913, Chapter 14.

2 H. Meyer, Arch. f. exper. Path. u. Pharm., 1901, Vol. 46, p. 338.

332 RALPH S. LILLIE.

anaesthetics collect in cells in higher concentration than in the
medium also favors the partition rather than the adsorption
theory of narcosis. Chloroform, ether, and esters undergo con-
centration in nervous and other tissues, as Pohl1 and Hedin2
have shown. Warburg and Wiesel3 have also found in the case
of yeast that the tendency of compounds to concentrate in cells
increases with increase in their narcotic power. In solutions
that diminished fermentative activity by one half, phenyl
urethane was found to be three times and thymol nine times
more concentrated in the cells than in the medium. These facts
strongly suggest a distribution according to relative solubilities.
What Traube especially insists upon is that effects similar to
narcosis are shown in cases where lipoid-solubility can play no
part. Thus according to Warburg and Wiesel4 the fermentative
and oxidative activities shown by lipoid-free preparations of
dried microorganisms are influenced by the lipoid-solvent anaes-
thetics in the same manner as in the intact organisms. It is to
be noted, however, that the effective concentrations are much
higher in the case of such preparations than in that of living cells.
Traube cites a large number of observations made with solutions
of various surface-active substances, showing that with both
animal and plant cells, as well as with enzymes, the degree of
narcotic and cytolytic action — of inhibition and destruction in
the case of enzymes — is nearly proportional to the surface-
activity of the solution.5 Solutions of widely different sub-
stances, provided they have the same surface-tension ("iso-
capillary" solutions), have equal physiological action. In the
following table I have collected a number of observations illus-
trating the various physiological effects produced by members of
the aliphatic alcohol series. In each instance the molecular
concentrations required to produce a definite physiological effect
are given ; the molecular concentrations which cause equal lower-
ing of surface-tension (isocapillary concentrations) are given at
the end of the table.

1 Pohl, Arch. f. exper. Path. u. Pharm., 1891, Vol. 28, p. 239.
z Hedin, P finger's Archiv, Vol. 68, p. 229.

3 Warburg and Wiesel, Pflugers Archiv, 1912, Vol. 144, p. 472.

4 Loc. cit., p. 471.

6 Cf. Traube, "Theorie der Narkose," loc. cit.

THE THEORY OF AN/ESTHESIA.

333

w

_1

m

U

O

•c-}- ir

O M M C
O O O C
000 C

>

• o
• o

Si

N .2

ft '5

5> I

I

H.

V

'.'.'. r-

M

. . . 0
. . . O

. M

. o
; o

3

"o <~

> °

1? 1

O\ »J

M 03

bo

'5

3

i*

j{ '5

ro rt

u

• o

aj

u

ffi

. 0

ft 42
_ -*->

<2

Ui

V

3

sf

ro r-~
IN ro ro

O O O

T|
C

• M I/

• o c

)

r'
cs

> r-

1— 1

c

S 1-
C5

r
i-

h

"3 '^

K* a;

u
o.

""

O 0 0

O

. 0 C

O

c

o

c

C

C

<N aj

Os ft

M

"3

"o

ta

OO -O u*
ro OS 00 -)
O O O O

• <N

co

^ M I/

O ro H-

5 rj
i/

I/

) C

O

o

1 tN

r^

\
\f

i "d

M

- t-

" p)

•sC

<s

'

to

•<F *o

•2 -^

o" "^

_o

000 0

o

o c

C

c

0

0

c

c

C

"8
en

a
o

ca

0.

o

IH

PH

1-1 ^ t— r»
M ro r-i M

)
t>

M

6

IO 00
00 OC

r<

r*
2 P

IY
t-

5 1
\J

^

•) 1-

sC

h-

I/

IsC

r. Journ
rations

a

0 O 0 C

C

0 M C

>-

C

u

c
o
U

5

Os r- M

W M CO ^

•sc

tN (N s£

i/

(M

1 "

oc

5 I
IOC

OC
(*

r-
) "•

> f

5 C

^ 0

- c

0) o
— o

w

O w O O

c

O ro ^

r^

: c

^

O)

(V

) 1-

u

}

C/2 H

]p

t^> • H
10 tN r~

sO ro C

C

h-

^

M
I
\S

) r-

^

\f

) C

C

V- (
(M

l-l

s

OP). O

t-

O r-- IY

1 »-

0

M

\f

i r-

• f

; -

"

ft

. . . M

. . -c

• O

to
c3

g

•o

Cr

. . . <u

o

; 3

<u
•c

0
10

03
"a3

^3

y^

M ^-1
O O
M K*1

a;

>>

^-i

^

o

'rt

Sb

o
o

Narcosis of tadpoles (Overton)1
Narcosis ot Arenicola larvae2
Prevention of cleavage in Arbacia eggs3.
Checking of development in Strongylocenln

ftTiihnpr'H

ta

X—

-
t

—

v»,

_5

J

-C
C

C

h

H

4 C

x ;
C
£

Production of heliotropism Daphnia (Loeb)6
Haemolysis (Fiihner and Neubauer;6
Decreasing oxidations by 50 per cent, i

^^»-T-»,,c?^loc /'AAfarVmror V

Inhibition of fermentation by ether-extraci

CWo.-Hiirfr'tS

Depression of isolated tortoise ventricle b

>«f ^\7o^r,r^Tl^9

C

c
c

a.

0.

t/
a

•c
'£

:
a
^
C
c
- ~

;;=

H :

- :

:.;

j 1

5 I

J c
0

C

Precipitation of nucleo-proteid of liver (Ba

QfornMl

Destruction of oxydon of ox-muscle (Bat

Cfor-nMl

a.
>
a.

~

•*~.

a.
w

a

-t~
\-
a

_ =

•^
t/

C!

a
k

•_
C

c

4.

c

c

I

al

j|

i

a

' -
C

I

1

! C
2

1

5j

H
il

H •*

1

^

i

•»

i

i

!

)

]

^
3

0

5

j
a

j

;
j

-«

;

1 Overton, "Studien iiber die Narkose," p
3 R. S. Lillie, Journ. Biol. Chem., 1914,

a

jy

03
+J

0)

to

M

Os

ro
O\

a

a ro

O C\;

O

- N S

•^ ^

H -^

'< :q

s o

IO

l^a

• Os

S >

<o -J

^^ J

CO - °^

« <0 Co

aj ?> -^

m -s; £•

a; o Si,

.—i MH ^o

> :=s

?* rJ f*-

•c i ^

s ^ .^

M

— _ ^

ro n i?

. fe ^ S

ro !> •>• ^

ro oo => N

ro

a

o

<N
M

a

ri

tl d 2 > "

"? (X, H M X

I >

tn os

'S r

9. "o

^ - *>
S "S N

a

42
3

to

M
c/)

N

c

42 42

CH Co

vj

O
•

o c

*•> 1-1

' D

» +J

§ M

£ S

aj 03

— "^

*5 *J

<H -t-*

, 03

ta

334 RALPH S. LILLIE.

These data show that the increase of surface-activity observed
on passing from one member of the series to the next is very
generally associated with a proportionately similar increase of
physiological activity. In general each member has from two to
three times the capillary activity of its immediate predecessor;
and the same holds true in a general way for its physiological
activity. It is also to be noted, however, that in general the
same holds true for lipoid-solubility.

If physiological activity is in fact a function of capillary
activity, solutions of equal surface-tension ought to exhibit equal
narcotic or other physiological effects. Traube cites various
observations indicating that this is frequently the case. Thus
Czapek1 has determined for a large number of organic substances
the surface-tensions of the solutions which have equal effects in
liberating tannin from plant cells (the leaves of Echeveria} ; this
effect is analogous to haemolysis and depends on increase in the
permeability of the plasma-membrane. The surface-tensions of
equally effective solutions (against air) were found to approach
a fairly constant value, about two-thirds of that of pure water.
Kisch2 also found that isocapillary solutions had equal effects in
preventing germination of yeast; and H. Zuckerkandl3 obtained
similar results for the inhibition of protoplasmic streaming in
plant cells. The results of observations by Fiihner and Neubauer
and also by Traube himself on haemolysis are similar. Thus,
taking again the series of alcohols: the surface-tensions of the
least concentrated solutions which free tannin from Echeveria
leaves and which inhibit the germination of yeast cells are as

follows: (water = i).

TABLE V.

Critical Surface-tensions of Solutions Causing

A. Exomosis of Tannin B. Inhibition of

Alcohol. from Echeveria Cells. Growth of Yeast.

Methyl 0.7 0.51

Ethyl 0.67 0.48

w-propyl 0.675 ca. 0.49

t-propyl 0.69

w-butyl 0.69

i-butyl 0.665 ca. 0.495

j-amyl 0.665 °-49

1 Cf. Czapek, "Uber eine Methode zur direkten Bestimmung der Oberflachen-
spannung der Plasmahaut von Pflanzenzellen," Jena, G. Fischer, 1911.

2 Kisch, Biochem. Zeilschr., 1912, Vol. 40, p. 152.

3 H. Nothmann-Zuckerkandl, Biochem. Zeitschr., 1912, Vol. 45, p. 412.

THE THEORY OF ANESTHESIA. 335

In each instance equal physiological effects are produced by
solutions of approximately equal surface-tension. Czapek finds
that the same rule of equal action for isocapillary solutions
holds good for ketones, esters, urethanes, and other compounds.
Lately, however, Vernon,1 Hober,2 and others have pointed out
various exceptions to this rule; thus chloroform, chloral hydrate,
nitromethane, and ethylene glycol begin to set free tannin in
solutions of much higher surface-tensions than the above. And
with the higher alcohols the surface-tensions of the effective
solutions are lower than the theory requires. These deviations
from the rule of isocapillarity are referred by Traube partly to
chemical influences (e. g., acid in chloral hydrate solutions),
partly to incorrect determination of surface-tension in solutions
of volatile substances like chloroform and ether, partly to the
influence of viscosity. In general it appears that the more
viscous compounds, e. g., the higher alcohols, show equal physio-
logical action, e. g., haemolysis, in solutions of lower surface-
tension than Traube's theory demands; Traube, however, be-
lieves that this difference is due simply to slowness of penetration,
incident to the high viscosity of the adsorbed layer of the narcotic
at the surfaces (e. g., of red corpuscles) where the essential action
takes place. For solutions of approximately equal viscosity the
rule of equal action with equal capillarity appears to hold true.
In general isocapillary solutions of surface-active substances have
less hsemolytic or other physiological action the greater their
viscosity; hence equal action for isocapillary solutions is to be
expected only when the viscosities are similar.3 Even this,
however, is not always the case. For instance, Loeb4 has found
that weak solutions of fatty acids, as well as weak solutions of
narcotics, produce positive heliotropism in daphnids; the least
effective concentrations for the first six members of the acid
series were: .oo6« formic, .oo6w acetic, .00572 propionic, .004^
butyric, .004^ valerianic, .OO2« caproic. The increase in effec-
tiveness with increasing molecular weight is much more gradual

1 Vernon, Biochem. Zeitschr., 1913. Vol. 51, p. i.

2 Rudolf Hober. " Physikalische Chemie der Zelle und der Gewebe," 4th edition.
1914, pp. 415 seq.

3 Cf. Traube, " Influence of Viscosity and Surface-tension in Biological Proc-
esses," Internal. Zeitschr. f. physik-chem. Biol., 1914, Vol. i, p. 275.

4 J. Loeb, Biochem. Zeitschr., 1909, Vol. 23, p. 95.

336 RALPH S. LILLIE.

than the increase in capillary activity. The influence of the
H-ion concentration enters here, and probably constitutes the
preponderant factor. The part played by purely chemical action
seems to be underestimated in Traube's theory. Where this
factor enters, surface-activity may be of subordinate importance.
Thus in the case of neutral salts solutions of equal surface-ten-
sions may have entirely different action on colloids and hence
on living cells. Traube's rule is at best an approximation; but
its significance is not to be underestimated on this account.
Adsorption and surface-condensation undoubtedly run parallel
with capillary activity; this is a matter not only of deduction
from the Gibbs-Thomson principle, but also of direct observation;
and surface-forces play so large a part in biological processes
that it is not surprising to find a frequent parallelism between
the physiological effects of solutions and their surface-activity.

Traube in fact recognizes that the lipoid-content of cells may
have an influence on the rapidity of intake of the narcotic
(since lipoid-solubility will naturally favor penetration), and
hence may be a factor in the narcotic action; but the narcosis
itself does not depend on this solution in lipoids; "the lipoid-
content influences narcotic action but does not determine it;
even lipoid-free cells may be narcotized."1 This conception,
however, does not seem adequate in view of the observations of
Meyer cited above, on variation of narcotic action with tempera-
ture.

It is important to note that surface-active substances affect
not only biological processes but also catalyses of various kinds,
especially those due to enzymes and other colloidal catalyzers
(platinum, etc.), where the action is probably dependent on the
character and extent of the surface between catalytic agent and
medium (heterogeneous catalyses). Enzymes are colloidal cata-
lyzers formed in cell-metabolism; as colloids it is to be expected
that surface-conditions will largely influence their action. Re-
cently Bayliss2 has emphasized the importance of such conditions.

1 "Theorie der Narkose," loc. cit., p. 305.

2 Cf. Bayliss, "The Physiological Importance of Phase-boundaries," Science,
N. S., 1915, Vol. 42, p. 509. For the influence of anaesthetics on invertase and on
colloidal platinum, of. Meyerhof, Pfliigcr's Archiv, 1914, Vol. 157, pp. 251, 307;
cf. also Warburg, ibid.. Vol. 155, p. 547, for their influence on the catalytic activity
of finely divided carbon (animal charcoal).

THE THEORY OF ANAESTHESIA. 337

He has shown that various enzymes (lipase, emulsin, urease,
trypsin) retain their activity when suspended in media in which
they are insoluble. This fact is best explained on the view that
increased concentration of the substrate at the surface of the
enzyme particles- is mainly responsible for the increased reaction-
velocity in its presence. This view does not explain specificity,
which is probably a matter in which selective adsorption and
stereochemical conditions enter as factors; but it leads to the
general expectation that readily adsorbable, i. e., surface-active,
substances will as a class have marked influence on enzyme action.
In many organisms oxidations are the chemical processes
which are most evidently influenced by narcotics; and the view
that narcotic action consists essentially in a suppression of intra-
cellular oxidations has gained wide favor, and has been sup-
ported chiefly by Verworn in Germany, and by Mathews,
Loevenhart and others in America. The view that this anti-
oxidative action may be exerted directly upon the oxygen-
catalyzers of the cell is supported by Traube. Considerable
evidence consists favorable to this view. Thus Warburg finds
that inorganic iron salts accelerate oxidations in disintegrated
sea-urchin eggs, and he further finds that this accelerative
action is checked by urethane.1 _ This interesting observation
suggests the possibility that catalysis by iron plays a part in
intracellular oxidations; this catalysis is checked by ansesthetics,
and it is to be inferred that oxidative processes under the in-
fluence of organic catalyzers or oxidases would be similarly
affected. In support of this conception Traube cites various
instances where oxidative processes are checked by surface-
active or narcotic substances.2 Such instances include the de-
composition of hydrogen peroxide by platinum (Bredig), the
oxidation of sodium sulphite by free oxygen (Bigelow, Titoff,
Young; Young finds this process checked by traces of morphine,
brucine, nicotine, and especially quinine; and Traube even refers
the antipyretic action of quinine to its inhibiting action on oxida-
tion) ; the oxidation of phosphorus and phosphorus trioxide by
free oxygen (Centnerszwer, Scharff) ; oxidation of stannous
chloride (Young) ; oxidation by tissue-oxidases (Vernon, Baer

1 Warburg, Zeitschr.f. physiol. Chem., 1914, Vol. 92, p. 231.

2 Traube, "Uber Katalyse," Pfluger's Archiv, 1913, Vol. 153, p. 309.

338 RALPH S. LILLIE.

and Meyerstein) ; catalytic oxidation of oxalic acid by animal
charcoal (Warburg).1 Thus not only oxidations under the in-
fluence of heterogeneous or colloidal catalyzers may be checked
by surface-active substances, but also oxidations in homogeneous
solution. This would indicate that surface-activity is not the
only factor involved. On the basis of this and other facts
Traube puts forward the hypothesis that narcotics are essentially
negative catalyzers, especially in relation to oxidative processes.2
The question of how this anti-oxidative effect is produced within
the living cell is the essential one. Traube and others have
suggested that the direct action of surface-active and narcotic
substances on colloids may be a chief factor. Moore and Roaf3
have investigated the precipitation of serum by such substances,
an effect which shows a general increase with surface-activity.
These authors, however, refer the effect to the formation of loose
chemical combinations between the narcotic and the proteins;
the quantity of chloroform and other anaesthetics dissolved by
serum is several times greater than by water; they regard the
excess as held by chemical union, and they attribute narcotic
action to such loose protein-anaesthetic compounds which limit
the chemical activities of the protoplasm, including presumably
the oxidations. Warburg and Wiesel also find that narcotic
substances have a precipitating action on the press-juice of yeast,
and that the anti-fermentative action runs parallel with the
precipitating action; and they recall the older view of Claude
Bernard, according to which a semi-coagulation of the cell-
colloids forms the basis of narcosis.4 Battelli and Stern5 find
that the nucleo-proteins of cell-extracts are also precipitated by
lipoid-solvent anaesthetics, and that the precipitating action runs
parallel with the influence in checking the activity of cell-

1 Warburg, Pfliiger's Archiv, 1914, Vol. 155, p. 547.

2 Winterstein ("Heat-paralysis and Narcosis," Zeitschr. f. allg. Physiol., 1905,
Vol. 5, p. 323) had earlier compared narcotics to the anticatalyzers or " Paralysa-
toren" of Bredig.

3 Moore and Roaf, Proc. Roy. Soc., 1904, Vol. 73, p. 382; 1906, Vol. 77-B, p. 86.

4 Cf. Warburg and Wiesel, loc. cit.; also Claude Bernard, "Lecons sur les anes-
thesiques et sur 1'asphyxie," Paris, 1875, p. 154. Anaesthetics, however, do not
affect the osmotic pressure of protein solutions, according to Meyerhof (see footnote
I, p. 346).

6 Loc. oil.

THE THEORY OF ANAESTHESIA. 339

oxidases or oxydones. Traube also cites observations by Schry-
ver1 in support of this general point of view; surface-active
substances retard the gelation of certain colloidal solutions, e. g.,
of sodium cholate under the influence of calcium salts, and the
degree of retardation runs parallel to capillary activity and
narcotic action. Physical alterations of the cell-celloids may
thus lie at .the basis of the anti-oxidative action which, according
to this view, conditions the narcotic action. Other instances
of this effect will be considered later. Traube expresses his
essential view as follows: "the physical alterations of the cell-
colloids — and by no means of the lipoids alone — form one of the
most essential conditions for the slowing of chemical processes
in cells, and hence also for narcotic and other toxicological
processes. These physical alterations are a consequence of the
depressant influence which narcotics exert upon surface-tension,
and upon the internal pressure of the cell contents."1 According
to this conception the physical alteration of the colloids would
be the primary effect of the narcotic, and decrease of oxidation
secondary; this view is more consistent with the membrane-
theory of narcosis, about to be described, than with the previously
quoted view which refers narcosis to a direct anti-catalytic action.
According to the membrane-theory, it is the plasma-membrane
which is primarily affected; and the decrease of oxidations (when
this occurs) is a secondary consequence of the change in the
membrane. This view would make the direct action of the
anaesthetic on oxidation-processes relatively unimportant. To
regard anaesthesia as dependent on a direct anti-catalytic action
seems insufficient, especially in view of the fact that the effective
anti-catalytic concentrations are much higher than those required
for narcosis. It should also be remembered that magnesium
sulphate and other salts can act as anaesthetics — such salts can
have no such direct anti-catalytic action — also that the electric
current may show the same influence. The action of anaesthetics
on oxidases will shortly be considered in more detail.

The general fact that surface-active substances alter the
physical condition or state of aggregation of many colloids is,
however, highly important in any general theory of anaesthesia.

1 Schryver, Proc. Roy. Soc., B, 1914, Vol. 87, p. 366.

2 "Theorie der Narkose," p. 302.

34O RALPH S. LILLIE.

Probably this effect is to be related to their influence on the
electrical potential difference normally existing between the
colloidal particles and the medium. Gouy1 found that the
potential-difference between mercury and sulphuric acid in the
capillary electrometer is lowered by the presence of many surface-
active or narcotic substances; similar observations were made
by Abl2 for cadmium amalgam cells, and by Grumbach3 for
various contact-potentials; and according to Traube the order
of relative action in all of these cases is essentially that of capillary
activity. Now precipitation or increased aggregation of colloids
is typically associated with decrease in the electrical polarization
of the colloidal particles; and capillary-active substances which
produce this latter effect ought therefore to further such precipi-
tation. A similar influence of anaesthetics on the potentials
shown by organic membranes like apple-skin against salt solu-
tions was observed by Loeb and Beutner;4 the concentrations
required for appreciable lowering of potentials were, however,
much higher than those ordinarily required for anaesthesia.
Notwithstanding this difficulty Traube suggests that a decrease
in contact-potentials, as well as of surface-tension at the active
surfaces in tissues like nerve, may be an important factor in the
action of narcotics. To quote from Traube's recent paper on
narcosis: "The narcotic substances, in collecting at the boundary-
surfaces of cell-wall and cell-fluid, lower there the electrical
contact-potentials, and in so doing directly prevent the trans-
mission of motor and sensory impulses by means of nerve-centers.
. . . This retarding or inhibiting action, exerted by substances of
low solution-affinity (Haftdruck) to water, upon the oxidations
and other intracellular processes conditioned by cell-colloids, and
also upon the electrical phenomena at boundary-surfaces, is the
cause of that condition which we designate as narcosis."5 A
somewThat similar view had previously been expressed by A. B.
Macallum: "Chloroform, ether, alcohol, and chloral lower sur-

1 Gouy, Annales de chimie el de physique, 1906, Ser. 8, Vol. 8, p. 291, and Vol. 9,

P- 75-

- Abl, Dissertation; Bonn, 1907 (cited from Traube, Pfliiger's Archiv, 1910,
Vol. 132, p. 521).

3 Grumbach, Annales de chimie et de Physique (8), 1911, Vol. 24, p. 463.

4 Loeb and Beutner, Biochem. Zeitschr., 1913, Vol. 51, p. 303.
6 " Theoric dcr Xarkose," p. 306.

THE THEORY OF ANESTHESIA. 34!

face-tension in aqueous solutions, in blood plasma and lymph,
and in all probability also the surface-tension of all cells, but
especially of the nerve-cells. This would make them incapable
of receiving or transmitting a nerve-impulse."1

The general view that narcosis is essentially a phenomenon of
asphyxia or retarded oxidation is an old one, suggested by
Claude Bernard and others, and has been revived in somewhat
different form of recent years, chiefly through the influence of
Verworn2 and his pupils. Decrease in oxidations is in fact
frequently observed during narcosis. Thus Alexander and
Cserna3 showed by direct analysis of blood a marked decrease
in the oxygen-consumption of the brain during ether and mor-
phine narcosis; oxidation-processes in the liver are also decreased
under the influence of various narcotics.4 But whether this
decrease is simply a consequence of a paralysis of metabolic as
well as of other functions, or whether it is the primary and
determinative condition, is a question which has been answered
differently by different investigators. Verworn has identified
narcosis with asphyxia chiefly because of certain similarities
between the physiological behavior of asphyxiated and narcotized
tissues and cells. A summation of the effects of narcosis and
of asphyxia is seen under certain conditions; thus Winterstein
found that frogs asphyxiated by perfusion with oxygen-free salt
solution until reflex activity was lost showed no recovery from
the asphyxia if supplied with oxygen while still in a state of
anaesthesia.5 The condition of narcosis appears to render oxygen
unavailable to the cells. Frohlich6 found the same rule to hold
for nerve-trunks; normally oxygen revives irritability which has
been lost in an oxygen-free medium (nitrogen atmosphere) ;

1 A. B. Macallum, "Surface-tension and Vital Phenomena," Univ. of Toronto
Studies, 1912, No. 8, p. 70.

2 Cf. Verworn, "Die Narkose," Jena, 1913; cf. also Harvey Lectures, 1911-12,
p. 52. Cl. Bernard ("Legons sur les anesthesiques et sur 1'asphyxie," pp. 92 seq.)
discusses the question whether anaesthesia is a form of asphyxia, but rejects this
view on the ground that anaesthetized animals show no signs of general asphyxia

(p- 97)-

3 Alexander and Cserna, Biochem. Zeitschr., 1913, Vol. 53, p. .100.

4 Cf. Joannovics and Pick, PJliiger's Archiv, 1911, Vol. 140, p. 327; Baer and
Meyerstein, Arch. f. exper. Path. u. Pharm., 1910, Vol. 63, p. 441.

6 Winterstein, Zeitschr. f. allg. PhysioL, 1902, Vol. i, p. 19.
6 Frohlich, Zeitschr. f. allg. PhysioL, 1904, Vol. 3, p. 75.

342 RALPH S. LILLIE.

but oxygen has no such effects on narcotized nerves; similar
observations were made by Nagai1 on ciliated epithelium. There
is thus no recovery from asphyxia during narcosis, even with a
good supply of oxygen. Nerves subjected to prolonged narcosis
show the same physiological changes as after exposure to lack of
oxygen (Frohlich,2 Boruttau3) ; the rate of conduction is slowed,
the refractory period is prolonged, and repeated stimulation
causes definite fatigue-effects; oxygen then restores the normal
properties, but only in the absence of the anaesthetic. All of
the phenomena of asphyxia appear during narcosis even in the
presence of a good supply of oxygen, just as they do in an oxygen-
free atmosphere in the absence of the narcotic (Frohlich, Bondy,4
Heaton).5 Experiments by Ishikawa6 on amoebae gave analogous
results; recovery from the inhibited or non-irritable condition,
whether due to simple lack of oxygen or to the presence of a
narcotic, requires the same condition, namely the presence of
free oxygen. Other forms of inhibition, such as the heat-paralysis
of the frog's central nervous system, are promoted both by lack
of oxygen and presence of narcotics (Winterstein) ;7 i. e., anaes-
thesia acts in the same direction as lack of oxygen, — an indication
that both conditions produce essentially the same physiological
effect. Mansfeld8 found that in the absence of oxygen tadpoles
succumb more readily to anaesthesia than in its presence; the
same is true of simple protoplasmic streaming in plant cells at
temperatures of 30° and over (Zuckerkandl).9 All of these facts
seem to indicate that lack of oxygen produces essentially the
same effects as anaesthesia; that the two actions are largely
interchangeable, and hence capable of summation.

In general, however, it may be said, in criticism of such con-
clusions, that inhibition or prevention, under the influence of
narcotics, of physiological processes which require oxygen, does

-1 Nagai, Zeitschr. f. allg. Physiol., 1905, Vol. 5, p. 34.

2 Frohlich, ibid., pp. 455, 468.

3 Boruttau and Frohlich, P finger's Archiv, 1904, Vol. 105, p. 444.

4 Bondy, Zeitschr. f. allg. Physiol., 1904, Vol. 3, p. 180.
6 Heaton, ibid., 1910, Vol. 10, p. 53.

6 Ishikawa, ibid., 1912, Vol. 13, p. 339-

7 Winterstein, ibid., 1905, Vol. 5, p. 342.

8 Mansfeld, Pjliiger's Archiv, 1909, Vol. 129, p. 69.

9 H. Nothmann-Zuckerkandl, Biochem. Zeitschr., 1912, Vol. 45, p. 412.

THE THEORY OF ANESTHESIA. 343

not demonstrate that narcotics act directly and primarily upon
oxidation-processes. The proper inference is rather that vital
processes, including those which require free oxygen, are inhibited
during anaesthesia. But anaesthesia may also inhibit physio-
logical processes which are independent of free oxygen, as Winter-
stein has shown in his experiments on the narcosis of anaerobic
animals *iike Ascaris.1 Similarly the growth of yeast under
anaerobic conditions is checked by anaesthetics in the same
manner as in the presence of oxygen;2 and the nerve cord of
Limulus, which continues to send out impulses in the absence
of free oxygen, is anaesthetized by ether in a typical manner.3
Some more general condition which determines the rate of
oxidations, as well as of other metabolic processes and cell-activi-
ties, is more probably the one directly affected in anaesthesia.
This latter view would regard the suppression of oxidations as
secondary rather than primary, — an effect rather than a cause of
narcosis. During anaesthesia those processes which are directly
dependent on oxidations are arrested, together with those not
so dependent.

Various attempts have been made to refer narcosis to a decrease
in the external supply of oxygen, or to an inability of oxygen to
enter the cells. Thus anaesthesia as well as sleep were at one
time popularly attributed to a condition of cerebral anaemia—
an obviously untenable view, since neither condition is confined
to animals with brain and circulation. That the anaesthetic
hinders the entrance of oxygen into cells has recently been sug-
gested by Mansfeld;4 the solubility of oxygen in the lipoids of
the plasma membrane — and hence its rate of entrance into cells-
was held to be diminished by the solution of lipoid-solvents in
the lipoids; and E. Hamburger5 attempted to show that narcotic
substances actually decreased the solubility of oxygen in lipoids.
These views however must be regarded as unfounded (see
Winterstein's criticism).6

'Winterstein, Biochem. Zeilschr., 1913, Vol. 51, p. 165.

2 Cf. Warburg and Wiesel, loc. cit., p. 480.

3 A. P. Mathews, cf. the article of Tashiro and Adams, loc. cit., p. 451.

4 Loc. cit.

6 E. Hamburger, Pfliiger's Archiv, 1912, Vol. 143, p. 186.
• Winterstein, Biochem. Zeilschr., 1913, Vol. 51, pp. 158 seq.

344

RALPH S. LILLIE.

The facts which offer best support to the oxidation theory of
narcosis appear to be those which demonstrate an actual decrease
in oxygen-consumption by isolated cells and tissues under the
influence of narcotics. Thus Warburg and his associates have
shown that oxygen-consumption by living cells of various kinds
(red blood-corpuscles, sea-urchin eggs, bacteria, liver-cells, yeast-
cells, the central nervous system) is decreased by various anaes-
thetics,1 and the different narcotic compounds show the same
order of relative action as in anaesthesia. For example, the
several alcohols lower the oxygen-consumption of birds' erythro-
cytes by 50 per cent, in solutions of the following concentrations.2

TABLE VI.

Alcohol.

Concentration of Solution Depressing
Oxidations 50 Per Cent.

Concentration for
Anaesthesia of
Tadpoles (Overton).

Per Cent,
(by Weight).

Molecular.

Methyl

16
7-3
5
I.I

i.i

O.4

5m
i.6m

o.&m
o.isw
o.iSm
o.QASm

0.52-0.62
0.27-0.31
O.II
0.038
0.045
O.O2^5

Ethyl

Propyl .

w-butyl

i-butyl

Amvl.

It is to be noted that these concentrations are much higher
than those usually required for anaesthesia, as comparison with
Overton's results (third column) will show. Vernon also found
that anaesthetics decreased the oxidation of the indophenol
reagent by fresh tissues. An especially interesting fact is that
anaesthesia causes a similar though less marked decrease in
oxidation by dead cells and by tissue-extracts, as has been
demonstrated by both Warburg and Vernon.3 This suggests
that the anaesthetizing influence is exerted directly upon the
oxygen-catalyzers of the cell; and recently a number of investi-
gators have devoted special study to the inhibiting action of
anaesthetics on oxidases.

1 For a summary of these researches of. Warburg, Miinchen. med. Wochenschr.,

1911, Vol. 58, p. 289; for the case' of isolated tissues cf. Usui, Pfliiger's Archiv,

1912, Vol. 147, p. 100; for microorganisms, Warburg and Wiesel, loc. cit.

2 Warburg, Miinchen. med. Wochenschr., loc. cit.; Zeitschr. f. physiol. Chemie,
1910, Vol. 69, p. 452.

3 Warburg and Wiesel, loc. cit.; Vernon, Journ. Physiol., 1912, Vol. 45, p. 197.

THE THEORY OF ANAESTHESIA. 345

Vernon1 has investigated the influence of various anaesthetics
upon the activity of the indophenol oxidase of the vertebrate
kidney. Enzymes which accelerate the oxidative formation of
the blue dye, indophenol, from a mixture of a-naphthol and
para-diamino-benzene are widely distributed in organisms; and
Vernon's investigations on the distribution of this oxidase in
the tissues of vertebrates indicate that a relation exists between
the oxidase-content of a tissue and its general oxidative activity.2
Various anaesthetics were found to decrease the oxidation and
eventually to destroy the oxidase. The concentrations required
to decrease activity by one half under otherwise constant condi-
tions showed a close parallelism with those required to haemolyze
red corpuscles. In both cases the order of relative action was
the same as for anaesthesia. The parallelism of lipoid-solubility
with narcotic action appeared closer than that of surface-
activity; and Vernon inclines to the belief that the narcotics
exert their action chiefly by dissolving in the lipoids of the plasma-
membrane and so altering the properties of this structure (pos-
sibly by interfering with the interaction of oxidase and per-
oxidase).3 The oxidase-inhibiting concentrations are, however,
far higher than the narcotizing, and correspond rather with the
cytolytic concentrations, so that a direct connection seems
doubtful. The work of Battelli and Stern4 on the influence of
anaesthetics on other tissue-oxidases (e. g., a liver-oxidase which
oxidizes succinic to malic acid) shows in general similar relations;
in this case the inhibiting action was found to run closely parallel
with surface-activity, — more so, according to Battelli and Stern,
than with lipoid-solubility. It showed also a striking parallelism
with a precipitating action on the nucleo-proteins of the tissue.
As already stated, a similar precipitating action of anaesthetics
has been investigated by Moore and Roaf, who agree with
Battelli and Stern in regarding this action as an important factor
in anaesthesia; the authors also attribute the action of anaesthetics
not to their influence on lipoids alone, but rather to an alteration

1 Loc. cit.

2 Vernon. Journ. Physiol., 1911, Vol. 42, p. 402.

3 Loc. cit., 1912.

4 Battelli and Stern, loc. cit.

346 RALPH S. LILLIE.

of the proteins of the tissue.1 It seems clear that anaesthetics
may directly inhibit oxidation-catalysis in tissues. But the
objection again rises that the concentrations required for these
effects greatly exceed those required for anaesthesia in living cells.
The relation of oxidases to cell-respiration is still obscure.
Present opinion inclines to the belief that these bodies are
essentially peroxide-forming compounds (oxygenases) which are
activated by other accessory substances present in cells. These
bodies (activators or co-enzymes) may be other organic com-
pounds (peroxidases) ; it appears also that in some cases inorganic
salts, especially iron salts, may play this role (cf. Warburg).2
There is some evidence that the combination of haemoglobin
with oxygen is of the nature of a peroxide; both haemoglobin and
haematin may cause bluing of guaiac and exhibit other oxidase-
like properties.3 Compounds which form unstable peroxide-like
unions with oxygen are regarded by certain investigators as
forming the essentially irritable part of the cell; the temporary
stabilization of such compounds by any physical or chemical
influence would thus be equivalent to anaesthetization. This
view has recently been supported in this country by Mathews;4
he regards anaesthesia as resulting from the formation of chemical
unions between the anaesthetic and protoplasm; these unions are
due to the residual valences of the anaesthetic (i. e., the reserve
powers of union left over in many compounds after the ordinary
valences are satisfied, as seen in the formation of double salts,
hydrates, etc.) ; the number of residual valences is variable, but
tends to be higher in compounds with well-marked anaesthetic
property. Mathews finds a general though not complete parallel-
ism between the number of such valences in a compound and its
narcotic power. He proposes the following explanation of anaes-

1 Cf. Moore and Roaf, loc. oil. But Meyerhof (Pfluger's Archiv, 1914, Vol. 157,
p. 273) finds that anaesthetics, in concentrations of the anaesthetizing or inhibiting
order, have no influence on the osmotic pressure of protein solutions. This ob-
servation appears to be incompatible with the view that anaesthetics act by altering
the aggregation-state of proteins.

" Zeitschr.f. physiol. Chem., 1914, Vol. 92, p. 231.

3 Cf. Moitessier, Compt. Rend. Soc. Biol., 1904, Vol. n, p. 373; Czylharz and
Fiirth, Beitr. zur chem. Physiol. u. Path., 1907, Vol. 10, p. 358; McClendon, Journ.
Biol. Chem., 1915, Vol. 21, p. 275.

4 A. P. Mathews, Internal. Zeitschr.f. physik-chem. Biol., 1914, Vol. i, p. 433.

THE THEORY OF ANESTHESIA. 347

thesia: "The irritable substance in protoplasm is a molecular
oxygen-protoplasmic union or a peroxide union, unstable and
similar to oxy-haemoglobin. By stimulation this unstable mol-
ecular union passes by molecular rearrangement into a stable
form, oxidation taking place and carbon dioxide being directly
or indirectly produced. The anaesthetic produces anaesthesia by
occupying the oxygen-receptors of the cell, thus forming a non-
irritable, dissociable, anaesthetic-protoplasm compound. The va-
rious facts of anaesthesia are explicable on this theory."

Now it is a striking fact that the concentrations of the lipoid-
solvent anaesthetic required to inhibit oxidations under the
influence of enzymes, inorganic catalysts, or dead cells are far
higher than those required for true anaesthesia, and are closely
similar in their order of magnitude to the cytolytic concentra-
tions. As already mentioned, Vernon found that the concentra-
tions at which the activity of the kidney-oxidase began to be
decidedly lowered corresponded closely with those found by
Fuhner and Neubauer for haemolysis; i. e., in Vernon's words,
"those which dissolve the lipoid membrane of the corpuscles."
The concentrations required for anaesthesia in tadpoles are from
eight to ten times lower. It thus seems doubtful that the
anaesthetic effects can be referred to a direct inhibitory action on
oxidases. The prompt and complete reversibility of anaesthesia
is also a fact unfavorable to this view. According to Vernon's
results, tissue-oxidases are rapidly destroyed by those concentra-
tions of lipoid-solvent anaesthetics which reduce their activity
to half its original value. We find, in fact, that in those cases
where anaesthesia is associated with decrease of intracellular
oxidations, the inhibitory effect is obtained in relatively low
concentrations; while in order to induce an equal decrease of
oxidations in tissue-extracts, or in disintegrated tissues or cells,
the required concentrations must be much higher. In other
words, destroying the structure of the tissue destroys its sensitive-
ness to the inhibiting action of low concentrations of the anaes-
thetic. This fact seems to indicate that the anaesthetic acts on
living cells primarily by altering certain organized or structural
elements, upon the condition of which the normal rate of oxidation
and of other metabolic processes depends. Experiments on

348 RALPH S. LILLIE.

tissue-extracts indicate that the normal oxidations in living cells
like muscle-cells are far more rapid and complete than can be
effected through the simple agency of the oxidases present
(Fletcher and Hopkins, Warburg, Battelli and Stern)1; the role
of oxidases in cell-respiration may therefore well be a subsidiary
one; and if so, the fact that anaesthetics arrest cell-activities in
concentrations which are without direct influence on the oxidases
may be understood. The essential change in anaesthesia would
then be a reversible alteration of certain structural elements that
control oxidations as well as other cell-processes.

Such a view would regard oxidases as accessory rather than
primary factors in cell-oxidations. If this is true, there should
be no direct parallelism between decrease of oxidations and
anaesthesia; and it should be possible in certain cases to secure
anaesthesia without influencing oxidations. There are in fact
numerous instances of complete and typical anaesthesia unaccom-
panied by any essential decrease in the rate of oxidations.
The anaesthesia of anaerobic animals and yeast-cells has already
been cited ; these instances, however, may be considered equivo-
cal,— since oxidations are equally essential to metabolism in
these organisms even though molecular oxygen may not take part
in the reactions. But many cases are known where the activity
of aerobic organisms or tissues may be profoundly inhibited by
anaesthetics, while the rate of oxidation is unaltered or only
slightly decreased. Such lack of parallelism was observed by
Rhode and Ogawa for the influence of chloral hydrate on the
isolated heart.2 The case of cell-division in developing eggs is
an especially clear one; here the rate of oxidation is relatively
slight compared with that of active muscle-cells. Warburg found
that phenyl urethane in concentrations of m/2,ooo arrests cell-
division completely in sea-urchin eggs (Strongylocentrotus) , while
leaving oxygen-consumption essentially unchanged ; in order mate-
rially to decrease oxidations (by 40 per cent.) several times the
minimal anaesthetic concentration was needed (m/'^oo).3 Similar

1 Fletcher and Hopkins, Journ. Physiol., 1907, Vol. 35, p. 287; Warburg,
Zeitschr. f. physiol. Chem., 1911, Vol. 70, p. 413; Pfluger's Archiv, 1912, Vol. 145,
p. 277; Battelli and Stern, Biochem. Zeitschr., 1914, Vol. 67, p. 443.

" Rhode and Ogawa, Archiv /. exper. Path. u. Pharm., 1912, Vol. 69, p. 200.

'" Warburg, Zeitschr. f. physiol. Chem., 1910, Vol. 66, p. 305; 1911, Vol. 70, p. 413.

THE THEORY OF ANESTHESIA. 349

observations were made by Loeb and Wasteneys1 on the eggs of
another sea-urchin (Arbacia). In order to arrest cleavage by
cyanide (which directly inhibits oxidations) a concentration suffi-
cient to lower the rate of oxidation to one third the normal was
needed. The oxidations could be reduced to one half the normal
without arresting cleavage. But in solutions of various anaes-
thetics (chloral, urethane, chloroform, methyl, ethyl, and propyl
alcohols) of concentration sufficient to prevent cleavage entirely,
the rate of oxidation was found to be only slightly decreased,—
on the average by less than 10 per cent. In solutions of urethane,
during the complete arrest of cleavage, the rate of oxidation was 98
per cent, of the control. When it is considered that oxidations
may be decreased by much more than 10 per cent, (by means of
cyanide, or by lowering the temperature a few degrees) without
arresting cleavage, it seems clear that the slight decrease of
oxidation observed in these experiments can stand in no causal
relation to narcosis. It is an accessory and apparently an unessen-
tial effect. Very similar results were found in experiments with
young fish embryos (Fundnlus) at a stage when the musculature
was well-developed, so that active contractions could be evoked
by external stimulation (e. g., by acidulated sea water). If
unstimulated, the embryos lie quiet within the egg-envelope;
the disturbing effects of variations in muscular activity are thus
absent. It was found that complete chloroform-narcosis had
little or no influence on the rate of oxidations; ether and butyl
alcohol caused some decrease in oxidations (25 per cent, to 30
per cent, at the narcotizing concentrations); but in order to
render the animals insensitive by direct inhibition of oxidation
through cyanide, it was found necessary to reduce the oxidations
to one ninth of their normal rate. In marine medusae (Gonio-
nemus) paralysis by cyanide required a decrease in oxidations
from three to six times greater than that accompanying urethane
narcosis.

The insensitivity of many cells and tissues to simple abstrac-
tion of oxygen or presence of cyanide is in striking contrast to
their sensitivity to anaesthetics. Thus nerve-trunks only gradu-

!Loeb and Wasteneys, Journ. Biol. Chem., 1913, Vol. 14, p. 517; Biochem.
Zeitschr., 1913, Vol. 56, p. 295.

35O RALPH S. LILLIE.

ally lose irritability and conductivity in an oxygen-free atmo-
sphere, or in cyanide solutions of considerable concentration
(Dontas) j1 while the desensitizing effects of anaesthesia are rapid
and complete. That the two effects are essentially different is
further shown by the difference in the rate of recovery, which is
much prompter in the case of anaesthesia. Apparently narcosis
may decrease the oxidations in resting nerve trunks, as indicated
by the output of carbon dioxide; this is seen in the experiments
of Tashiro and Adams cited above, but the effect is compara-
tively slight and probably unconnected with the loss of irrita-
bility, since, as just seen, the nerve retains irritability in cyanide
solutions for a long time. Compare also Winterstein's results,
about to be described. Other similar instances are ciliary move-
ment and protoplasmic rotation, both of which are only gradually
checked by lack of oxygen or by cyanide, but instantly by
narcotics. Recently Winterstein2 has shown that the reflex
irritability of the frog's isolated spinal cord may be entirely
lost under anaesthesia without affecting the general oxidation-
rate of the tissue; in fact during alcohol narcosis there was a
slight but regular increase in oxygen-consumption. The nar-
cotized cord differs from the non-narcotized in one chief respect;
in the normal and incompletely narcotized cord stimulation
causes increased oxygen-consumption; but during complete nar-
cosis no such effect is seen; the oxygen-consumption during
stimulation is the same as in the resting non-narcotized cord.
Oxygen is, however, essential for the normal irritability of the
cord ; complete recovery from narcosis requires not only removal
of the anaesthetic but also the presence of free oxygen. There
may, however, be partial recovery even in the absence of oxygen.
These facts illustrate how important oxygen is for the normal
activity of the nerve-cell; but they also show that, given a
supply of free oxygen, the consumption in the normal resting
cord may be the same as in the narcotized cord. If narcosis
were simply asphyxia, such a result would be inexplicable. A
further fact inconsistent with the "asphyxia hypothesis" of
narcosis is that after a narcosis lasting for days (9 days in one of

1 Dontas, Arch. f. exp. Path. u. Pharm., 1908, Vol. 59, p. 430.

2 Winterstein, Biochem. Zeitschr., 1914, Vol. 61, p. 81.

THE THEORY OF ANESTHESIA. 35!

Winterstein's experiments with urethane) reflex irritability re-
turns promptly on the removal of the anaesthetic. There is no
evidence that nerve-cells can resist lack of oxygen for any such
time. There is, however, ample evidence from other sides that
during narcosis the normal resting oxidations of tissues continue
uninterruptedly. Winterstein also finds that the oxidative re-
moval of acid products of asphyxia takes place equally readily in
normal and in narcotized nerve-cells. The central nervous sys-
tem of the frog, which normally exhibits an alkaline reaction to
litmus, becomes acid during asphyxia; if oxygen is then restored
the alkaline reaction returns; but narcosis was found to have no
influence on this effect; clearly therefore, narcosis does not inter-
fere with these oxidations.1

Such observations should be correlated with those of Vernon,
Warburg, Battelli and Stern cited above, indicating that tissue-
oxidations are directly influenced only by relatively high con-
centrations of anaesthetics. Taken in conjunction with the other
instances just cited, of anaesthesia with essentially unaltered
rate of oxidation, they indicate that a direct suppression of
intracellular oxidations (or asphyxia) is probably not the essential
basis of anaesthesia. Apparently the latter condition does not
depend on any alteration of purely chemical conditions, but on
some influence exercised by the anaesthetizing agency on the
structural or organized (or "living") substratum in which the
chemical processes take place and by which their character and
rate are controlled. The evidence of this will now be considered.

There appears to be a general relation between degree of
organization and susceptibility to narcosis. Plants and lower
organisms require higher concentrations of anaesthetic than
higher animals (Overton) ; in vertebrates the cells most suscep-
tible to narcosis are those of the higher brain-centers. Such cells
are distinguished by high irritability and rapid variations in their
activity, peculiarities which are undoubtedly a function of their
special structure. It is true that if organization is destroyed
many of the chemical processes of protoplasm (oxidations and
fermentations) may still continue (autolysis), and may then be
slowed by anaesthetics; but as already shown, much higher con-

1 Winterstein, ibid., 1915, Vol. 70, p. 130.

352 RALPH S. LILLIE.

centrations are required to produce such effects than to anaes-
thetize the intact living cells. Such facts indicate that when
anaesthetics influence oxidative and other metabolic processes
within the cell, they do so not directly, but through their influence
upon some specially sensitive intermediary, which is a part of the
organized structure of the cell and itself controls the rate of the
intracellular chemical processes. It is this intermediary which
is directly influenced by the anaesthetic. Its part may be com-
pared to that of a sensitive starter or relay in a complex mechan-
ical or electrical system.

Various general considerations support this view. When an
irritable element, e. g., a muscle-cell, is stimulated by a mechanical
impact, it is difficult to suppose that the primary effect of this
impact is to hasten oxidative processes; it is true that an increase
in oxidations does follow, but this effect represents a later stage
in the complex sequence of interdependent processes constituting
the response to stimulation (and of which the contraction is the
most evident). If the muscle is previously treated for a short
time with an anaesthetic, or if the magnesium or calcium-content
of the medium is sufficiently increased, contraction no longer
results. The entire sequence of processes normally following
stimulation is prevented. It seems more probable that the
primary event in the physiological sequence is the one directly
interfered with; if this is prevented so also are the others. It
further seems clear that in an irritable cell this primary or deter-
minative change must be a surface-change; obviously that part
of the irritable element which is directly in contact with the
medium is the one first affected by the stimulus; and there is
ample evidence that the direct action of many stimuli is confined
to this surface-layer. Indirectly the activity of the whole cell
is of course affected ; but this must be by means of some influence
transmitted from the surface throughout the cell-interior. The
nature of this influence forms in fact the chief problem of the
physiology of stimulation.

The fact that a surface-effect is sufficient to set in motion the
whole complex apparatus of response in the cell-interior is a
cardinal one in any theory of anaesthesia. Unmistakable evi-
dence that this is the case is seen in the delicacy of the response

THE THEORY OF AN/ESTHESIA. 353

which sensitive organisms like protozoa show to the contact
stimuli of food particles or prey; also in the prompt response of
many living cells to the presence of substances which are known
from experiments on permeability not to penetrate the plasma-
membranes. In general these membranes in their normal state
are impermeable to the neutral salts of the alkali and alkali
earth metals, as Overton showed; yet variations in the pro-
portions of such salts in the tissue-media profoundly influence
the activity and irritability of living cells. Increase in the
magnesium or calcium salts of the medium may cause typical
anaesthesia in muscle and . nerve cells. Warburg and Harvey1
have shown that cell-division in sea-urchin eggs may be arrested
by weak solutions of sodium hydrate without the alkali pene-
trating the cell. A. J. Clark has made similar observations for
heart muscle cells, and Harvey for the contractile cells of medusae
and for protoplasmic rotation in plant-cells.2 Irritability and
automatic activity may thus be abolished by substances to
which the" cell-surface is impermeable. The general facts of
electrical stimulation also indicate that alteration in the electrical
condition of the cell-surface is the primary event in stimulation.
The investigations of Nernst and his successors in the theory of
electrical stimulation show that the electrical current stimulates
by changing the relative concentrations of ions on the opposite
faces of the semi-permeable membranes enclosing the irritable
elements, — i. e., by altering the electrical polarization of the
surface-film or the plasma-membrane. A characteristic electrical
variation, the action-current — which is best explained as the
result of alterations in the electromotor properties of the cell
surface — also accompanies all forms of stimulation, and this
electro-motor variation is prevented by anaesthetics. It may
be held, therefore, with a high degree of probability that the
primary event in stimulation is a surface-process, consisting in
some physico-chemical alteration of the modified protoplasmic
surface-film (plasma-membrane) which delimits irritable cells.3

1 Warburg, loc. cit.; E. N. Harvey, Journ. Exper. Zoo/., 1911, Vol. 10, p. 507.

2 A. J. Clark, Journ. Physiol., 1913, Vol. 46, p. xx; Vol. 47, p. 66; E. N. Harvey,
loc. cit., and Yearbook of Carnegie Institution, No. 10, 1911, p. 128.

3 For a general discussion of the evidence bearing on this problem cf. my article
"The Relation of Stimulation and Conduction in Irritable Tissues to Changes in

354 RALPH S. LILLIE.

Lately many investigators have concurred in this view that
alterations in the physico-chemical properties of the plasma-
membrane form the essential basis of anaesthesia. Whatever
condition alters this structure so as to make it less capable of
undergoing the changes of permeability and of electrical polariza-
tion which normally accompany stimulation — and apparently
other forms of cell-activity — has an inhibiting or paralyzing
effect on the cell. This general view has been reached as the
outcome of a large number and variety of investigations. Over-
ton in 1904 pointed out that the paralyzing action of potassium
salts on voluntary muscle must be referred to their action on the
membrane, since plasmolytic experiments indicate that such salts
do not penetrate into the cell interior.1 The view that the
physiological action of neutral salts is due primarily to their
influence on the plasma-membrane was later strongly supported
by Hober2 on the basis of experiments on the influence of neutral
salts on the demarcation-current potential of muscle. This
may be influenced in the direction either of increase or decrease
by treatment with isotonic solutions of sodium and other salts.
Salt solutions, like those of potassium salts, which decrease this
potential — i. e., produce local negative variation — apparently
do so by altering the colloids of the plasma-membrane and so
increasing its permeability; in general such increased permea-
bility to ions involves decrease in the electrical polarization of
the membrane (i. e., negative variation); increased polarization
means an alteration of the membrane in the reverse direction,
i. e., of decreased permeability. These changes of polarization
and permeability result from the altered condition of the colloids
forming the membrane. The colloidal system of the membrane
acquires in the presence of certain salts a less dense consistency
("Auflockerung") associated with increased permeability; and a
denser consistency ("Verdichtung") in the presence of others,
e. g., sodium iodide, etc. The microscopic appearances observed

the Permeability of the Limiting Membranes," in Amer. Journ. Physiol., 1911,
Vol. 28, p. 197. Also, for a more elementary treatment, the articles "The Role of
Membranes in Cell-Processes," and "The General Physico-chemical Conditions
of Stimulation," in the Popular Science Monthly, 1913 and 1914.

1 E. Overton, Pfliiger's Archil), 1904, Vol. 105, p. 176; cf. p. 207.

2 R. Hober, Pfliiger's Archiv, 1905, Vol. 106, p. 599.

THE THEORY OF ANESTHESIA. 355

in nerve-fibers treated with pure solutions of sodium and potas-
sium salts support this conception.1 Changes in the electrical
condition of the cell are thus indices of changes in the colloids
forming the plasma-membrane, and especially in the lipoids.
Now, since the stimulation-process is always accompanied by
electrical variations, it seems probable that this process is itself
dependent on changes in the colloids of the plasma-membrane,
involving changes of permeability; correspondingly, artificial
alterations in the condition of these colloids should modify the
irritability of the cell.2 In an important paper published in 1907,
entitled "Contributions to the Physical Chemistry of Stimulation
and Narcosis,"3 Hober applies this general conception to the
problem of narcosis essentially as follows: Stimulation is asso-
ciated with an alteration in the condition of the colloids of the
plasma-membrane, involving a general increase of permeability;
narcotics are those agents which prevent this alteration in the
condition of the protoplasmic colloids and hence prevent stimula-
tion. The colloids chiefly concerned are the lipoids; narcosis
depends on the collection of lipoid-soluble substances in the
lecithin of the plasma-membrane to a certain critical concentra-
tion, and on a prevention, by means of these substances, of
the colloid-process normally concerned in stimulation. Hober
showed that various anaesthetics (ethyl and phenyl urethane,
chloral hydrate, chloroform, hypnon) do in fact check the action
of rubidium and potassium salts in causing negative variation in
frogs' muscle. They also prevent the effect of potassium sulphate
in causing structural changes ("Auflockerung") in nerve. The
anaesthetic thus antagonizes the salt-action, — just as it is known
to antagonize the stimulating action. Similar effects may be
produced by alkali earth salts, especially of calcium. According
therefore to Hober's hypothesis, the physico-chemical basis of
these antagonisms, and hence of anaesthetic action in general,
is an alteration of the colloids, especially the lipoids, of the

1 Hober, Zentralblalt f. PhysioL, 1905, Vol. 19, p. 390.

2 For a fuller account of Hober's views see his general article on "the physico-
chemical processes in stimulation" in Zeitschr.f. allg. PhysioL, 1910, Vol. 10, p. 173.
Also his textbook, " Physikalische Chemie der Zelle und der Gewebe," 4th edition,
1914.

3 PJliiger's Archiv, Vol. 120, p. 492.

356 RALPH S. LILLIE.

plasma-membrane, and a consequent change in the properties
of this structure. The temporary increase of permeability,
which according to Bernstein's theory of the bioelectric varia-
tions, forms an essential part of the stimulation-process, is thus
rendered difficult or impossible.

Very clear and concrete indication that the stimulation of
muscle is in fact associated with a temporary increase in the
permeability of the plasma-membrane, and that anaesthetics act
by preventing this increase, wTas afforded by my own experiments
on the larvae of the marine annelid Arenicola, carried out at
Woods Hole in IQOS.1 This larva is a free-swimming trochophore
one third of a millimeter long, possessing a well-developed
musculature and swimming by cilia, and is peculiar in having its
body-cells permeated by a brown water-soluble pigment. This
pigment normally remains writhin the cells, but under conditions
of increased permeability, as on death or under the influence of
cytolytic substances, it diffuses readily into the sea-water and
imparts a yellow tinge to the latter. It serves therefore as a
convenient indicator of increase of permeability. If larvae are
brought suddenly from sea water into a pure isotonic solution of
a sodium salt (e. g., o.6m NaCl), a strong muscular contraction at
once results, accompanied by a well-marked loss of pigment;
at the same time the cilia cease movement and soon afterwards
undergo breakdown, and other toxic effects follow. If, instead
of a pure solution of NaCl, a solution containing a little CaCl2
or MgCl2 is used, both changes are simultaneously prevented;
neither contraction nor loss of pigment is shown ; the cilia remain
active, and normal swimming movements continue for a time;
the general toxic action of the pure salt solution is also prevented.
Thus the calcium prevents at the same time both the stimulating
and the permeability-increasing action of the sodium salt. It also
greatly diminishes the injurious action of the latter, i. e., exerts
anti-toxic action. Pure solutions of KC1 also cause strong mus-
cular contractions accompanied by loss of pigment; and the
effect is similarly checked by the addition of MgCl2. In mixtures
of KC1 and MgCl2 both effects vary with the Mg-content of the

1 Cf. Amer. Journ. Physiol., 1909, Vol. 24, p. 14. Cf. also ibid., 1911, Vol. 28,
pp. 210 seq.

THE THEORY OF ANESTHESIA. 357

solution in a closely parallel manner; solutions with relatively
high Mg-content cause slight stimulation and slight loss of
pigment, while in those of relatively high K-content both effects
are well marked.

Entirely different effects from those of pure solutions of
potassium and sodium salts are produced by pure solutions of
magnesium salts. These exert typical anaesthetic action on
Arenicola larvae, as on other marine animals. When brought
suddenly into pure isotonic MgCl2 solution the larvae show no
contraction or loss of pigment; all muscular contraction imme-
diately ceases, the body remains rigid and extended; the cilia
are more resistant and remain active, and slow undirected
swimming movements continue. On return to sea water muscu-
lar movement and other normal activities are at once restored.
If larvae are brought into isotonic MgCl2 solution for a few
minutes, and are then transferred into pure NaCl solution, the
characteristic effects following transfer to the solution from sea
water — stimulation and loss of pigment — are no longer seen ; the
organisms remain motionless and without apparent change.
If they are then returned to sea water they show prompt revival.
Apparently the treatment with MgCl2 renders the plasma-
membrane more resistant than normally to the permeability-
increasing action of the pure NaCl solution, and at the same time
the muscle-cells become refractory to stimulation. The following
was the general conclusion drawn at the time from these facts:
"MgCl2 and similarly acting solutions appear to decrease the
permeability of the tissues and so prevent the ionic transfer on
which stimulation depends. The general action of anaesthetics
consists in decreasing the normal permeability ; stimulating agencies
on the other hand have the reverse effect."1 Later experiments
with a variety of lipoid-solvent anaesthetics — alcohols, esters,
ether, hydrocarbons — gave essentially similar results;2 solutions
of anaesthetics in pure NaCl solution, in every case where they
prevented the stimulating action of the solution, also prevented
the permeability-increasing action as indicated by loss of pig-
ment; when they did not prevent this effect, they did not
prevent stimulation. A general parallelism between prevention

1 Loc. cit., p. 44.

2 Amer. Journ. Physiol., 1912, Vol. 29, p. 372; 1913, Vol. 31, p. 255.

358 RALPH S. LILLIE.

of permeability-increasing action and prevention of stimulation
was thus shown. Anaesthetics were also found to prevent other
effects depending on increase of permeability, such as the toxic
effects of pure solutions of Na and K salts on sea urchin and
starfish eggs, as well as on Arenicola larvae,1 and also the activa-
tion of unfertilized sea-urchin eggs by pure salt solutions (KCNS,
Nal).2 In general the anaesthetics appear to exert a stabilizing
influence on the plasma-membrane, rendering it more resistant
than normally to influences that tend to increase its permeability.
To this stabilizing action both the inhibitory or anti-stimulating
(anaesthetic) and the protective (anti-toxic) actions of both salts
and lipoid-solvent anaesthetics are due.

Recently a large body of evidence has accumulated from
various sides indicating that anaesthetics either decrease the
permeability normal to the resting cell, or render the plasma-
membrane more resistant than normally to increase of permea-
bility. Thus, according to Lepeschkin,3 the entrance of dyes
into plant cells (Spirogyra) is checked in the presence of low
concentrations of ether and chloroform, and according to Szucs,4
also by neutral salts. I have made similar observations on
Arenicola larvae.5 These effects indicate a decrease in the
general permeability of the plasma-membrane to diffusing sub-
stances; this change is apparently associated with a character-
istic alteration in the density or physical consistency of the
membrane, rendering it more than normally resistant to dis-
integrative or toxic agencies in general. Hence anaesthetics as
well as neutral salts protect the cilia, pigment-cells and muscula-
ture of Arenicola larvae against the injurious action of pure Na-
salt solutions; the same is true of sea-urchin and starfish eggs.
Similar observations are described by other authors. Arrhenius
and Bubanovic6 find that blood-corpuscles may be protected by
anaesthetics against cytolysis in hypo tonic solutions; and Traube

1 Ibid., 1912, Vol. 30, p. i.

-Journal of Experimental Zoology, 1914, Vol. 16, p. 591. Cf. also Journ. Biol.
Chem., 1914, Vol. 17, p. 121.

3 Lepeschkin, Ber. d. deutsch. Bot. Gesellsch., 1911, Vol. 29, p. 349.
Szucs, Jahrbuchf. wissenschaftliche Botanik, 1912, Vol. 52, p. 85.
Ainer. Journ. Physiol., 1909, Vol. 24, p. 26.

6 Publications of Nobel Institute, 1913, No. 32 (quoted from Hober's "Physik.
Chem. d. Zelle," p. 466).

THE THEORY OF ANESTHESIA. 359

has made similar observations for solutions of cytolytic sub-
stances (haemoglobin).1 The increase in permeability caused
by pure solutions of Na-salts in fish eggs is also hindered by
alcohol and ether, according to McClendon;2 and Loeb has de-
scribed similar effects of alcohol on Fundulus eggs.3 The most
conclusive evidence of a decrease in permeability during narcosis
is however derived from experiments on the electrical conduc-
tivity of narcotized cells; this appears to be decreased during
narcosis (McClendon, Osterhout, Hober and Joel). McClendon
in 1910 found the conductivity of sea-urchin eggs to be decreased
by chloroform, but did not investigate the phenomenon in detail.
More extensive experiments have been carried out on plant cells
by Osterhout,4 who has succeeded in showing clearly that in the
presence of moderate concentrations of anaesthetic the cells of
the marine alga Laminaria undergo increase in electrical resis-
tance, indicating decreased permeability to ions; on removing
the anaesthetic the original conductivity returns. If too high
concentrations of anaesthetic were used the result was quite
different; conductivity underwent marked increase and the
effects were irreversible; the tissue had in fact undergone
injurious or cytolytic alteration. Evidently the change cor-
responding to anaesthesia is the reversible change of decreased
conductivity, indicating decreased permeability. Similar obser-
vations on blood-corpuscles have recently been made by Joel
under Hober's direction.5

Taken in its entirety the foregoing evidence indicates that
under the influence of a narcotizing agent the plasma-membrane
undergoes an increase in its general stability or resistance to
alteration; stimulation is prevented because this effect requires
ready and rapid variations of permeability, and of these the
stabilized membrane is no longer capable. Associated with this
general stabilization is a decrease in the permeability to diffusing
substances; the cell is more completely shut off from the disturb-

1 Traube, Biochem. Zeitschr., 1908, Vol. 10, p. 371. See also the experiments of
G. H. A. Clowes, Proc. Soc. Exper. Biol. and Med., 1913, Vol. n, p. 8.

2 McClendon, Amer. Journ. Physiol., 1915, Vol. 38, p. 173.

3 J. Loeb, Biochem. Zeitschr., 1912, Vol. 47, p. 127; cf. p. 155.

4 Cf. Osterhout, Science, N. S., 1913, Vol. 37, p. in; also "Quantitative Re-
searches on Permeability" in "The Plant World," 1913, Vol. 16, p. 129.

6 Joel, Pfluger's Archiv, 1915, Vol. 161, p. 5.

360 RALPH S. LILLIE.

ing effects of environmental changes. There is a possibility
that the permeability to gases like carbon dioxide and oxygen is
also decreased, but this remains uncertain at present.

It is important to note that changes in the resistance of plasma-
membranes, probably essentially similar to those underlying
anaesthesia, may occur under completely normal conditions, as
I have recently found in experiments on dividing sea-urchin eggs.
It had been shown earlier by Lyon,1 Spaulding,2 and Mathews3
that during the normal cycle of cell-division these cells are much
more susceptible to poisons (cyanide, acids, ether, and K-salts)
at the time when the cleavage-furrow is forming, than in the
period preceding or following cleavage; during cleavage there is
also an increased output of CO?.4 A rhythm of susceptibility to
poisons and of CO2-production is thus associated with the rhythm
of cleavage. This condition suggested the possibility that the
essential underlying condition of this rhythm might be a rhyth-
mical change in the properties of the plasma-membrane; and in
a series of experiments with dilute sea water it was found that
the eggs do in fact undergo cytolysis in hypotonic media far more
readily at the time when the furrow is forming than during the
intervals between cleavage.5 In other words, the membrane is
relatively sensitive to osmotic disruption during cleavage, and
relatively resistant during the intervals between cleavage, i. e.,
at the resting times when the cell is relatively highly resistant to
the action of poisons. This normal state of relatively high
stability may be compared to the condition which is imparted to
the membranes of irritable cells by anaesthetics. Increased
resistance to hypotonic solutions in the presence of anaesthetics
has been observed by Arrhenius and Bubanovic in blood cor-
puscles, as already cited.

How does the anaesthetic produce this alteration in the prop-
erties of the membrane? Attempts to find parallels between the
effects of anaesthetics on living cells and on colloidal suspensions
of lipoids have not given entirely consistent results. Hober

1 E. P. Lyon, Amer. Journ. Physiol., 1902, Vol. 7, p. 56.

2 E. G. Spaulding, BIOL. BULL., 1904, Vol. 6, p. 224.

3 A. P. Mathews, BIOL. BULL., 1906, Vol. n, p. 137.

4 Lyon, Amer. Journ. Physiol., 1904, Vol. n, p. 52.

'R. S. Lillie: Amer. Journ. Physiol., 1916, Vol. 40, p. 130.

THE THEORY OF ANESTHESIA. 361

and Gordon1 found that lecithin suspensions containing ether,
chloroform, chloral, or amyl alcohol were less readily precipitated
by calcium and barium salts than the control suspensions, i. e.,
were protected against precipitation or stabilized by the anaesthe-
tic; and they refer to this phenomenon as "narcotization of the
plasma-membrane colloid lecithin." Koch and McLean,2 how-
ever, find that no such effect is general with anaesthetics; they
find chloral and ether indifferent, while chloroform protects
lecithin against precipitation by CaCl2; on the other hand,
alcohol and paraldehyde further precipitation. These discrep-
ancies are probably due to differences of concentration. Recent
experiments of my own with a large number of anaesthetics have
shown that with the great majority of such compounds lecithin
emulsions may be protected to a greater or less degree against
the precipitating action of CaCl2 and HC1; i. e., a concentration
of electrolyte just sufficient to cause precipitation in the absence
of the anaesthetic, fails to do so or does so incompletely in its
presence. The concentrations required to produce this stabiliza-
tion are however much higher than the anaesthetizing concentra-
tions, and the different compounds vary greatly in effectiveness.
The usual increase of action with increasing molecular weight
in members of homologous series (alcohol, esters) was seen.
But with certain compounds little or no protection was found,
while a few (ethyl and methyl alcohols, and in part isopropyl
alcohol, acetonitrile, and in part paraldehyde) definitely furthered
precipitation. The compounds which distinctly prevented pre-
cipitation included higher alcohols (w-propyl, w-butyl, i-amyl,
capryl), esters (ethyl nitrate, propionate, butyrate; methyl,
ethyl, and phenyl urethanes), hydrocarbons (chloroform, carbon
tetrachloride, nitromethane, benzol, toluol, xylol) ; ethyl ether,
chloral hydrate, chloretone, chloralose, acetanilide, phenyl urea.
In general, therefore, lipoid-solvent anaesthetics exhibit a stabiliz-
ing influence against precipitation by electrolytes, but this
influence is not always present. It may depend upon altered
electrical polarization of the colloidal particles, or possibly upon
increase of viscosity; but it is doubtful if in itself it forms a

1 Hober and Gordon, Beitr. zur chem. Physiol. u. Pathol., 1904, Vol. 5, p. 432.

2 Koch and McLean, Journ. of Pharm. and Exper. Therapeutics, 1910, Vol. 2,
p. 249.

362 RALPH S. LILLIE.

factor of any importance in true anaesthesia. The effective con-
centrations are too high and the effect is too variable.

A probably more significant physical change caused by lipoid-
solvent anaesthetics is an increase in the viscosity of lecithin
suspensions. Handowsky and Wagner1 observed such an in-
crease of viscosity in the presence of alcohol; and A. Thomas,
working in my laboratory, has confirmed and extended these
observations.2 Thomas observed that in the case of ether the
increase of viscosity in concentrated emulsions of lecithin might
go so far as to cause true gelation. I have found that this
effect is very general; it is well shown in lecithin emulsions of
10 to 12 per cent, concentration; these are highly viscous, but
still fluid in consistency; on the addition of many anaesthetics
this consistency changes to that of a soft more or less coherent
and elastic gel, in some cases of sufficient firmness to permit the
inversion of the test-tube without spilling. Well-marked gelation
was observed with the following compounds: alcohols (w-propyl,
•i-propyl, w-butyl, i-amy\, capryl), esters (ethyl formate, acetate,
propionate, butyrate, nitrate), ethyl ether, ethyl chloride, ethyl
bromide, chloroform, carbon tetrachloride, benzol, toluol, xylol,
chloretone, chloral hydrate, chloralose, paraldehyde. On the
other hand, certain very efficient narcotics had no such effect,
e. g., methyl, ethyl and phenyl urethanes, ethyl alcohol, nitrome-
thane and acetonitrile. Gelation, however, is to be regarded as
simply an end-effect of increase in viscosity; the latter change is
the essential, and this is perhaps the most general and significant
of the purely physical changes produced by lipoid-solvent
anaesthetics in colloidal suspensions of lipoids. Such a change
will have in general a retarding influence on physical and chemical
processes taking place in such a system; it will decrease diffusion-
rates and hence reaction-velocities depending on such rates;
more energy, mechanical or other, is required to produce any
kind of change in a highly viscous system. A general hindrance
to diffusion would express itself in a decrease of permeability
and of electrical conductivity. The influence of changing viscos-
ity on the electrical conductivity of lipoid suspensions remains

1 Handowsky and Wagner, Biochem. Zeitschr., 1911, Vol. 31, p. 32.

2 A. Thomas, Journ. Biol. Chem., 1915, Vol. 23, p. 359.

THE THEORY OF ANAESTHESIA. 363

to be studied. Recently Loewe1 has investigated the influence
of a number of anaesthetics on the electrical resistance of solid
artificial membranes impregnated with lipoids, and he has found
that in some cases they cause decided decrease of conductivity,

—a change to which he refers as "narcosis of the membrane."
These results recall those of Osterhout on living plasma-mem-
branes, but whether the anaesthetic effect in living cells is so
direct as Loewe's experiments would indicate seems doubtful.
Clowes's recent interesting conception of an interchangeability in
the relative positions of the lipoid and aqueous phases of the
protoplasmic emulsion — so that either the one or the other, ac-
cording to conditions, may form the external or continuous phase

—may throw a much-needed light on these phenomena. He
suggests "that anaesthetics function by promoting the continuity
of an external fatty or lipoid phase. The solubility of this
lipoid film in adjacent aqueous phases being lowered, its per-
meability to water-soluble substances would be diminished.
Since certain vital processes presumably depend upon inter-
mittent intercommunication between adjacent aqueous phases,
it may well be imagined that a temporary interruption in this
communication would result in anaesthesia" (loc. cit., p. 10).
Such a change would involve both decreased permeability to
water-soluble substances and greater stability of the membrane.
On the whole it appears highly probable that lipoid-solvent
anaesthetics cause their effects through some purely physical
change in the cell-system — more particularly in the plasma-
membrane. Their chemical inactivity as a class indicates this.
Processes like solution and adsorption, rather than chemical
combination, probably determine their action in most cases. It is,
however, best not to be too dogmatic on this point, since the
distinctions between solution, chemical combination, and adsorp-
tion are. probably not absolute. There is always the possibility
that in certain cases the anaesthetic may form some chemical
combination which interferes with chemical or other changes
necessary to stimulation; the inactivation of haemoglobin by
carbon monoxide may serve as an illustration of how this is
possible. Hober2 has recently suggested that in true reversible

1 S. Loewe, Biochem. Zeitschr., 1913, Vol. 57, p. 161.

2 Hober, Deutsche medizinische Wochenschrift, 1915, No. 10.

364 RALPH S. LILLIE.

narcosis the direct effect of the anaesthetic is limited to a surface-
action or adsorption affecting all of the colloids of the membrane;
in higher than the narcotizing concentrations the solvent action
of the anaesthetic upon the lipoids of the membrane assumes im-
portance and leads to disruptive and hence irreversible effects.
This view attributes the destructive action of anaesthetics in high
concentrations to a process different from that underlying true
anaesthesia. Meyer's experiments on the effects of temperature,
already cited, indicate, however, that solution of the anaesthetic
in the lipoids is an essential factor in the total physiological
effect. It is conceivable that a slight solution of this kind, com-
bined with the general increase in the viscosity of the proto-
plasmic surface-layer, may increase the stability of the mem-
brane; while a more pronounced solution may lead to direct
solvent or other structure-altering effects which destroy its semi-
permeability and so cause cytolysis.

It must not be forgotten, however, that lipoid-solvents form
only one class of anaesthetics. Apparently all substances or
conditions which stabilize the membrane in a reversible manner
may exert anaesthetic action. Anaesthesia due to neutral salts,
cold, or the electric current, may be understood from this point
of view; salts or low temperature may stabilize the membrane
by causing gelation or other direct changes of aggregation in the
colloids; the electric current by altering its state of electrical
polarization. Some years ago I expressed this general view as
follows: "Anaesthetic action is due primarily to a modification
of the plasma-membrane of the cells or irritable elements, of such
a kind as to render these membranes more resistant towards
agencies which under the usual conditions rapidly increase their
permeability; cytolysis and stimulation, both of which depend
on such increase of permeability, are hence checked or prevented.
Decrease in the readiness with which the permeability is increased
thus involves for an irritable tissue decreased irritability; this
effect is produced by various salts, e. g., of magnesium, and by
ether and other lipoid-solvent anaesthetics in certain, not too
high, concentrations. ... It seems clear that for irritable tissues
the state of the lipoids in the plasma-membrane largely deter-
mines the readiness with which changes of permeability — and

THE THEORY OF AN/ESTHESIA. 365

of the dependent electrical polarization — are induced by external
agencies. Slight permeation of the lipoids with a lipoid-solvent
apparently often facilitates such changes, and hence increases
irritability; the present of more lipoid-solvent renders a change
of permeability difficult, hence the protective or anaesthetic
action; while concentrated solutions of lipoid-solvents disrupt
the membrane and produce cytolytic or irreversible alterations
in the cells; hence such substances in higher concentrations are
markedly toxic."1

The question of just how this stabilizing influence is exerted is
the critical one. An irritable element like a nerve-fiber or muscle-
cell responds to a slight local electrical stimulation or mechanical
impact; this response is apparently associated with a rapid and
reversible increase of membrane-permeability; to this latter
change the electrical variation is apparently due. It is this
membrane-change, with the associated variation of electrical
polarization, which appears to be the primary physiological
event in stimulation; it spreads rapidly over the whole mem-
brane, and the other consequences of stimulation (contraction,
increased oxidation, etc.) follow upon this surface-change. The
question thus involves the whole problem of the physiology of
stimulation. This is not the place for a detailed discussion of
the various questions implicated in this central problem. It is
evident, however, that the whole process of stimulation depends
on the local initiation of the excitation-state, and on the rapid
conduction of this state from the point of stimulation so as to
affect the entire element.2 All of these processes depend on the
physical and chemical condition of the membrane; hence altering
this condition alters the whole stimulation-process.

According to this conception the sensitivity of the membrane
to changes of electrical polarization is its most characteristic
peculiarity.3 The basis of this sensitivity remains to be deter-

1 BIOLOGICAL BULLETIN, 1911, Vol. 22, p. 328; cf. p. 331.

2 For a more special discussion of the conditions of conduction, as distinguished
from the other features of the stimulation process, cf. my two recent papers in
the Amer. Journ. Physiol., 1914, Vol. 34, p. 414, and 1915, Vol. 37, p. 348.

3 Cf. the concluding section of my paper on antagonisms between salts and
anaesthetics, Amer. Journ. Physiol., 1912, Vol. 29, p. 391 seq. "In anaesthesia
it is to be assumed that the membrane is so altered that it fails to respond to a
change in its electrical polarization by an increase in its permeability" (p. 393 )„

366 RALPH S. LILLIE.

mined. It would appear that the peculiar properties of the
membrane depend upon its being a living structure, the seat of a
specific metabolism. That the characteristic semi-permeability
depends on this latter peculiarity is seen in the fact that the
death-process, however induced, is always associated with a
marked increase in the general permeability and electrical con-
ductivity of cells. In other words, the normal semi-permeable
condition — involving, as it must, a certain constancy in the
composition and physical state of the surface-film — is maintained
only so long as the cell remains alive. This fact shows that semi-
permeability, and the electrical polarization which is associated
with this, are not simply static properties of the plasma-mem-
brane, but are functions of a specific metabolic activity — includ-
ing probably oxidations in most cases — which maintains constant
the physico-chemical characteristics of the surface-layer of proto-
plasm. In the irritable element this metabolism appears to be
altered in a definite manner by changes in the electrical polariza-
tion of the membrane; and along with these chemical alterations
go alterations of permeability and, secondarily, of electrical
polarization. These latter involve the production of local elec-
trical circuits which traverse and hence stimulate the adjoining
inactive portions of the irritable element; in this manner the
state of excitation spreads, and the whole element is stimulated.1
But this is the case provided only that the membrane retains
its normal sensitivity to changes of electrical polarization; if it
has previously been rendered resistant by an anaesthetic, no such
effect follows; the element as a whole then shows itself irre-
sponsive to stimulation.

1 For a fuller discussion cf. my papers on the conditions of conduction in irritable
tissues, already cited.

THE NATURE OF THE POLYHEDRAL BODIES FOUND

IN INSECTS.1

R. W. GLASER AND J. W. CHAPMAN.

CONTENTS.

1 . Introduction 367

2. The polyhedral bodies 369

3. Origin of the polyhedral bodies '. 373

4. Bio-chemical observations 376

5. Summary 382

6. Bibliography 383

7. Explanation of Plates 386, 388

INTRODUCTION.

A very large group of lepidopterous larvae is subject to a class
of infectious diseases known as the "polyhedral diseases."
Whether the organism concerned in their production is identical
or not for all the species of insects affected must remain a matter
of conjecture till further work allows us to venture an interpreta-
tion. One thing, however, is certain, namely that curious
crystal-like structures called "polyhedral bodies" or "polyhedra"
are always associated with the type of diseases we are here dis-
cussing. Although these polyhedra may vary considerably in
size and somewhat in shape in the different species of insects,
nevertheless, they are always specific for a certain type of malady.

Wahl, followed by Prowazek and Escherich, consider the
polyhedral diseases as distinct and absolutely divorce them from
the fungous, protozoan and bacterial affections of insects. We
believe that the erection of a separate group to embrace all of
the polyhedral diseases is an excellent plan and receives our
sympathetic endorsement, for the reason that the confusion of
all of the insect diseases is still common amongst entomologists.
We venture to say that there are scarcely two entomologists in
America who know the difference or similarity between any of
the diseases expressed by such terms as Muscardine, Pebrine,

1 Contribution from the entomological laboratory of the Bussey Institution in
cooperation with the U. S. Bureau of Entomology (Bussey Institution No. 115).

367

368 R. W. GLASER AND J. W. CHAPMAN.

flacherie, lethargia, maladie de morts-blancs, schlaffsucht, faul-
sucht, fettsucht, polyhedral disease, wilt, wipfelkrankheit, jaun-
dice and gelbsucht.

We do not wish to dwell upon the differences or similarities
between all of these maladies, but will confine ourselves solely
to the group of polyhedral diseases under which are included the
four commonest manifestations, viz., wilt, wipfelkrankheit, jaun-
dice and gelbsucht. Wilt is the vernacular term used in America
for the polyhedral diseases. It is a name suggestive of post-
mortem aspects, but unfortunately a number of caterpillar
diseases distinct from wilt on superficial examination also have
a similar post-mortem appearance. Moreover, a number of
diseases common to plant pathology are labelled with the same
term. We do not think it wise to eliminate the name, however,
for the reason that it has been used so long and "wilt" certainly
means more to field entomologists and foresters than "polyhedral
disease" which is only significant to laboratory workers. The
term "Wipfelkrankheit," which is used for a similar affection of
nun moth caterpillars in Germany, is a very suggestive name
for the reason that when the animals are in the last stages of the
disease, they congregate in masses at the tops of the trees or
"wipfeln" and die hanging by their prolegs. The name "Wip-
felkrankheit" has the further advantage in that it is not applied
to any other plant or animal disease. "Gelbsucht" or "jaun-
dice" are terms used to designate a polyhedral disease in silk-
worms. The two terms are descriptive of the clinical picture
of the diseased worms and are not used for any other affection
known to pathology.

The larval stages of the following species of lepidoptera have
been examined by us and found to be susceptible to the poly-
hedral diseases. Animals which can only be infected experi-
mentally have been omitted, i. e., the list comprises only those
which besides being capable of experimental infection also have
the disease or diseases in a state of nature.

I. Saturnidae.

1. Hemileuca maia Drur.
II. Arctiidse.

2. Apantesis virgo Linn.

NATURE OF POLYHEDRAL BODIES FOUND IN INSECTS. 369

III. Noctuidae.

3. Leucania nnipuncta Halw.

4. Noctua clandestine, Harris.

5. Autographa brassicce Riley.

IV. Lymantridae.

6. Porthetria dispar L.

7. Lymantria monacha L.

8. Orgyia leucostigma S. and A.
V. Lasiocampidae.

9. Malacosoma americamim Fabr.
10. Malacosoma disstria Hubn.

VI. Bombycidae.

n. Bombyx mori L.
VII. Dioptidae.

12. Phryganidia calif ornica Packard.
VIII. Pieridae.

13. Colias philodice Godart.

From a perusal of the above list it will be seen that the poly-
hedral diseases are very widely distributed and affect some of
our most important economic insects. We have estimated that
epidemics of polyhedral diseases at certain times kill off from 30
to 70 per cent, of some of our most noxious pests. This is
especially true, as we have found in connection with our studies
on the gipsy moth, tent caterpillars and army worms. The
polyhedral diseases contribute much more to the control of
certain of our noxious caterpillars than the combined efforts of
all their hymenopterous and dipterous parasites. Therefore,
we believe that these diseases merit a serious consideration from
all points of view.

THE POLYHEDRAL BODIES.

Caterpillars dead from wilt are usually found on some elevated
place hanging by their prolegs. Dead nun moth caterpillars
are found hanging from the very highest branches of a conifer,
dead gipsy moth caterpillars are found hanging anywhere on the
trunk of a tree or on a branch ; the favorite dying places of the
American tent caterpillars being usually in close proximity to
the nest on the branches of an apple or cherry tree. The army

370 R. W. GLASER AND J. W. CHAPMAN.

worm seeks the tip of a grass blade and succumbs thereon. An
innate desire to reach an elevated place on their favorite food
plant always seizes the diseased insects prior to death. We
have never observed animals in the last stages of wilt descend
their food plants. So far we are unable to offer any explanation
which would satisfactorily assist in analyzing this ascending
instinct. A short time after death, the animals become deliques-
cent. At the slightest touch the skin ruptures and a dark brown
liquid oozes out. In some species such as the American and
forest tent caterpillars this liquid is pink shortly after death and
becomes dark brown later. The corpses will be practically
odorless if they have hung but a short time and before septic
bacteria have gained a foothold.

If some of the brown liquid from a dead caterpillar is examined
microscopically with a high-power dry or oil-immersion lens, it
will be found to contain, besides the elements of disorganized
tissues, myriads of polyhedral bodies of various sizes. (Fig. I,
Plate i .) Certain polyhedra have been found to measure ^ M
and less in diameter while still others reach the size of 15 M-
The average polyhedron of the gipsy moth caterpillar measures
3.4 fj. in diameter. The bodies in this species are larger than
those in any other form we have hitherto examined. The average
size of nun moth caterpillar polyhedra measure 2.65 /JL in diameter;
those of the forest tent caterpillar 2.6 p and so on until we come
to Phyrganidia californica and the tussock moth caterpillars in
which the average diameter has been found to be 1 .6 /* and 1 .5 /*
respectively. Thus it is seen that the average polyhedron varies
greatly in size in the different species. As stated previously the
sizes of the polyhedra within one species or even within one
animal (gipsy moth: ^M~I5M) varies also. There exists a
striking similarity between the shapes of these bodies in the
different species but some variation within a particular species
or even within the same animal can be observed. In general
the form is that of a polyhedron with more or less rounded angles.
They never assume the shape of a perfect sphere, and an actual
geometric outline has never been observed except in the silk
worms where almost perfect octahedra are found. The poly-
hedra are highly refractive, and on focusing are seen to have a

NATURE OF POLYHEDRAL BODIES FOUND IN INSECTS. 371

denser center differentiated from a somewhat lighter periphery.
Sometimes within the bodies concentric layers like those of an
onion are observable. Often two polyhedra are seen adhering
to one another as if in the act of dividing, but an actual division
in a hanging-drop has never been observed. When pressure is
applied to the cover glass, the polyhedra crack very readily into
a number of pieces, and often without the application of pressure
the same fragmentation may be observed to occur somewhat
more slowly. In the latter case a notch appears at one side of
the polyhedron which gradually lengthens into a line progressing
slowly toward the other side, much like the cracking of ice.
Usually before the line has completely separated the two halves
other lines appear, and soon the entire polyhedron is divided
into a number of pieces, which may separate or may stick together
in a rosette-like fashion. At no time was anything observed to
come out of the polyhedra when they cracked in this manner.
If the cover glass is moved while applying a little pressure, one
half of the polyhedron may sometimes be folded upon the other
half without the cracks appearing, showing that it is composed
of a tough substance and is not at all brittle like inorganic crystals.

The only objects in a fresh preparation with which one could
possibly confuse the polyhedra are the fat globules and urate
crystals, but with a little practice these may be readily dis-
tinguished. Fat globules are perfectly spherical and are there-
fore unlike the polyhedral shape of the bodies in question; but
when in doubt, Sudan III was used, for in this stain the fat
globules become red, while the polyhedra remain colorless. The
urate crystals are often more acutely angular or are of an entirely
different shape from the polyhedra and are frequently traversed
by radiating lines.

Besides polyhedral bodies, fat globules, and urates, a smear
from a newly "wilted" caterpillar contains cellular debris, hairs
and pigment granules. The pigment granules must not be con-
fused with bacteria, for many of them superficially resemble
these organisms very closely. When a preparation is dried,
mounted, and examined under oil, the pigment granules of the
gipsy moth may easily be confused with small micrococci,
owing to the fact that they are usually arranged in pairs. As a

372 R. W. GLASER AND J. W. CHAPMAN.

matter of fact, a smear made from a recently wilted caterpillar
is almost devoid of bacteria, and in many cases none at all can
be found. If bacteria are present, they have escaped into the
body cavity through rupture of the intestine and bear no direct
etiological relation to the disease.

In fixed and stained smears a number of things can be demon-
strated to advantage within the polyhedra. Fixation was accom-
plished either by passing the preparation through a flame or by
placing it in absolute alcohol for a few minutes. The smears
were then stained in Giemsa's solution for 12 hours or were stained
for a shorter period with one of the following dyes: Methylene
blue, trypan blue, gentian violet, carbol fuchsin, Bismarck brown,
or iron haematoxylin. When iron hsematoxylin was used, the
preparation was first mordanted in a 4 per cent, ferric-alum
solution for two or three hours. After staining, the preparations
were sometimes quickly passed through the alcohols to xylol
before mounting. This not only clears everything, but dissolves
away all the fat on the slide and thus increases the transparency
of the preparation. Gipsy moth polyhedra are rather resistant
to stains in general and usually color along the periphery only,
unless the stain is applied for a long time. When this is done
one may succeed in staining the entire polyhedron, /especially
after the use of some mordant like ferric alum before hematoxylin
or anilin water before gentian violet. Steaming the preparation
with a stain like carbol fuchsin has also given good results.
When properly stained, one of three conditions is observed:
First, the polyhedral bodies are uniformly stained so that nothing
can be detected within them; or second, a uniformly darker
staining central mass can easily be differentiated from an almost
unstained outer substance; or, third, many little refractive,
reddish granules are seen within the polyhedra. An actual
differentiation between what might be interpreted as nuclear
and cytoplasmic material within the polyhedra never occurs.
Therefore, in accounting for the staining reactions we believe
that at times the polyhedra have a central granular or homo-
geneous substance easily distinguishable from an outer tougher
substance which is more resistant to the dyes. This varies a
great deal, however, and sometimes the periphery takes the

NATURE OF POLYHEDRAL BODIES FOUND IN INSECTS. 373

stain more readily than the underlying strata. From these
staining reactions it becomes apparent that the polyhedra are
complicated in structure, and do not therefore differ essentially
from what Bolle and Prowazek found to be true of the silkworm
polyhedra. Our observations on the staining reactions also show
that morphological studies do not enable us to regard the poly-
hedral bodies as organisms. We believe that the polyhedra, are
protein degeneration-products of the disease. The staining re-
actions have demonstrated that they are not simple crystals,
but complicated in structure, and have a tough outer layer.
Consequently, and for a number of other reasons, we do not
believe them to be true crystals and therefore choose to call
them pseudo-crystals. The variations in their staining reactions
which one obtains at times can well be accounted for by assuming
that one is dealing with different stages in the synthetic process
of pseudo-crystals. Another matter militating against the idea
that the polyhedra are organisms is the fact that Glaser ('15)
and Chapman and Glaser ('16) have experimentally demon-
strated the possibility of infecting healthy gipsy moth cater-
pillars with wilt material from which the polyhedral bodies were
removed by passing the virus through Berkefeld Grade "N"
candles. We have shown that wilt is caused by a filterable virus
and believe that the polyhedra arise as a reaction against the
invasion of this virus.

ORIGIN OF THE POLYHEDRAL BODIES.

Studies on sectioned gipsy moth, army worm and tent cater-
pillar material have shown that the polyhedra originate within
the nuclei of the hypodermal, fat, tracheal matrix, and blood
cells. (Fig. 2.) In the true army worm, however, polyhedra
are at times formed within the nuclei of intestinal epithelial
cells. We have been utterly unable to find the bodies within
the nuclei of muscle tissue, Malpighian tubes, ganglia, nerves,
cenocytes, salivary glands, gonads and within the intestinal
epithelial cells of all forms except the true army worm.

The formation of the polyhedral bodies within the nuclei of
the four tissues above mentioned and the visible changes taking
place within these nuclei may be described as follows: The first

374 R- w- GLASER AND J. W. CHAPMAN.

indication of a diseased nucleus seems to consist in the flowing
together of the chromatin into a lump in the middle. Then out
of the achromatic substance the polyhedra arise as very minute
individuals. (Fig. 2, Plate I.) They gradually increase in size
and probably most of the chromatin is also used up during the
synthetic process. As the polyhedra grow, they become more
and more refractive, stain with difficulty, and the nucleus be-
comes hypertrophied. The late stages of these hypertrophied
nuclei are more than twice as large as the largest normal nucleus.
(Fig. 2, Plate I, and 3, 4 and 5, Plate II.) This swelling of the
nucleus is due to the increase in size of the polyhedral bodies which
stretch the nuclear membrane. During the earlier stages of the
disease the polyhedra are somewhat rounder than the larger ones
found later on prior to death. This can be accounted for by the
fact that, as the polyhedra grow, they become so closely packed
within a nucleus that they press upon one another and thus the
more or less polygonal shape is produced. As the polyhedra
grow and become more refractive the remains of the chromatin
lump disappears and there remains simply the nuclear membrane
enclosing the polyhedra. (Fig. 2, Plate I, and 3, Plate II.)
Finally the nucleus disintegrates, and the polyhedra are found
free in great numbers in smears of dead caterpillars. (Fig. 6,
Plate II.) We believe (and this belief is based on morphological,
chemical and experimental evidence) that the polyhedral bodies
are degeneration-products of the disease — products of nuclear
disintegration. This view may not seem so improbable if one
reviews some of the literature dealing with a few of the diseases
in higher animals.

HYDROPHOBIA (RABIES).

In 1903, Negri described certain bodies occurring in the nervous
system of animals dying of rabies. The bodies seem to be specific
to the disease, and are of great assistance for diagnostic purposes.
The Negri bodies vary in size, measuring from .5 to 25 n- They
are round, oval or angular in outline and are found in the proto-
plasm of the nerve cells and their processes. The bodies occur
in all parts of the nervous system, but are most common in the
Purkinje cells of the cerebellum and especially in the cells of
the cornu Ammonis. The virus of hydrophobia passes through

NATURE OF POLYHEDRAL BODIES FOUND IN INSECTS. 375

the coarser Berkefeld and Chamberland filters, and these filters
exclude the Negri bodies. The most generally accepted view
at present is that the Negri body is a cellular reaction against the
invasion of the filterable virus. The virus of hydrophobia has
not been cultivated.

VARIOLA AND VACCINIA (SMALLPOX AND COWPOX).

In 1892 Guarnieri described certain cellular inclusions in
variola and vaccinia which are specific for these two diseases.
The bodies are from 1-8 /x in diameter, and round, oval, or sickle-
shaped. They lie in the cellular spaces, often in close proximity
to the nucleus, and can be demonstrated in vaccine pustules as
well as in the experimental lesions produced in the rabbit's
cornea. The virus of variola and vaccinia passes through the
coarser porcelain (Chamberland) filters. Most authorities regard
the Guarnieri bodies as the effect of a specific reaction of epi-
thelial cells against the virus. The virus has not been cultivated.

TRACHOMA.

In trachoma certain granules are found in the cytoplasm of
the inflamed epithelial cells which cover the conjunctiva. Bodies
similar to the trachoma bodies have also been found in other
inflammations of the conjunctiva, and are therefore thought
not to be specific of trachoma. The trachoma bodies are re-
garded as the product of mucous secretion under pathological
conditions. The virus passes through Berkefeld filters. It has
not been cultivated.

Cellular inclusions not regarded as parasitic by some workers
have been found in a number of other diseases. In scarlet
fever cellular inclusions occur in the skin lesons; in foot and
mouth disease inclusions are found in the vesicles; in fowl pest
inclusions are found in the brain and in epithelioma contagiosum
or fowl diphtheria in the epithelium. Cytoplasmic inclusions
accompanying many of the diseases of higher animals, and re-
garded as non-parasitic in most cases are not at all uncommon,
but we have been unable to find any account in the literature of
vertebrate diseases which are accompanied by the formation of
nuclear inclusions as is the case with the polyhedral diseases of
insects.

376 R. W. GLASER AND J. W. CHAPMAN.

BIO-CHEMICAL OBSERVATIONS ON THE POLYHEDRAL BODIES.

According to Tubeuf, Krassilschtschik, Prowazek, Wahl, Wolff,
Glaser and Chapman the polyhedral bodies are regarded as
being reaction products; towards bacteria (Tubeuf, Krassilsch-
tschik) or towards Chlamydozoa (Prowazek, Wolff) or towards
an unknown virus (Wahl) or towards a filterable virus (Glaser
and Chapman). According to Bolle, Fischer, Marzocchi, and
Knoche the polyhedral bodies are stages of a protozoan ; accord-
ing to Escherich and Miyajima they are bearers of an unknown
virus.

It seems that the views are very diverse as regards the true
nature of the polyhedra. For this reason it was thought advis-
able to submit some of our bio-chemical work on this subject in
further support of our contention, obtained from morphological
and experimental studies, that the polyhedra are merely organic
degeneration-products of the disease.

During the gipsy moth season great quantities of diseased
material can be obtained. For this reason all of the bio-chemical
investigations were performed with gipsy moth polyhedra. These
bodies are heavier than water and consequently can be obtained
in bulk by centrifuging aqueous emulsions of diseased material.
By repeated washing, filtering and centrifuging most of the fat,
cellular debris, etc., can be eliminated. After this treatment
the polyhedra were always washed with ether in order to free
them from any possible remnants of adhering fat. Thorough
attention to this cleansing operation will yield polyhedra in a
fairly pure state for chemical tests. The material was allowed
to dry naturally in the centrifuge tube, after which the lump
that formed at the bottom could be loosened and transferred to
a mortar where it was pulverized. This pulverization if done
gently does not crack or injure the polyhedra in any way. After
this procedure the mass of polyhedra look very much like pul-
verized chalk. It is comparatively easy to obtain 2 or 3 grams
of polyhedra from about one or two hundred caterpillar cadavers.

As the polyhedra do not blacken with osmic acid, and do not
stain with Sudan III., it seems unlikely that they contain fat.
They stain with picric acid, however, and this gave us the clue
to their possible protein nature. The color tests for dry proteins

NATURE OF POLYHEDRAL BODIES FOUND IN INSECTS. 377

were then applied and we obtained positive reactions with xantho-
proteic, Millon's, biuret, Adamkiewicz's and Lieberman's tests.
It was next found necessary to obtain the polyhedra in solution
so that the various coagulation or precipitation tests could be
performed.

Before testing the solubility of the polyhedra in various re-
agents they were first rubbed energetically in an agate mortar
with the addition of a little sea sand. This grinding was found
necessary for the reason that the outer surface of the bodies is
composed of more resistant material than the underlying strata.
By grinding with sea sand the polyhedra are fragmented and
this offers more delicate surfaces to the action of the reagents.

The polyhedra were found to be insoluble in hot or cold water,
alcohol, chloroform, ether, or xylol. They dissolve readily in
strong acids and alkalies, but these reagents were thought to
produce too great a hydrolytic cleavage of the protein molecule,
and since we did not wish to alter our material to any appreciable
extent a number of milder reagents were tried. Moreover, from
the standpoint of the classification of proteins it is important
to determine just what will and what will not dissolve the
material.

The following solubility tests were performed. Two grams of
ground polyhedra were divided into four parts. To Y^ gram
water was added; to ^ gram .5 per cent, and to another ^ gram
2 per cent. NaCl solution, and to the fourth ^ gram 10 per cent.
Na2COa were added. The tests were kept over a water bath
for 133/2 hours at a temperature varying between 55° and 58° C.
At the end of this time the solutions were filtered and the various
tests for soluble proteins applied. All of the tests (acetic acid,
nitric acid, cupric sulphate, mercuric chloride, acetic acid with
potassium ferrocyanide and ammonium sulphate) were negative
showing that nothing went into solution.

Two grams of polyhedral material were again divided into four
parts and treated respectively with H2O, .5 per cent, and 2 per
cent. NaCl and 10 per cent. Na2COs. The tests were placed
over a small direct flame for two hours. At the end of this time
the solutions were filtered and the tests for soluble proteins
applied. Negative tests were obtained with the water and salt

378 R. W. GLASER AND J. W. CHAPMAN.

solutions. The material treated with the 10 per cent. Na2CO3
solution gave slight coagulation tests with acetic and nitric
acids showing that a small amount of protein went into solution.
The polyhedra treated with the carbonate were examined micro-
scopically and it was found that some fragments had partially
dissolved while most of the polyhedra, which had apparently
resisted fragmentation in the mortar had swollen to double their
normal size.

Concentrated HC1 (37 per cent.) used both hot and cold seems
to dissolve the polyhedra with difficulty. The solubility of the
bodies in boiling HNO3 seems to lie between 15 and 20 per cent.
We began with 4 per cent. HNOs in which the polyhedra are
not affected and worked up to 31 per cent. HNO3 in which they
dissolve instantaneously on boiling. The fact that the liquid
clears when some of the lower percentages of HNOs are used is
not sufficient evidence that all of the polyhedra have been
dissolved. For this reason the solubility of the bodies towards
the acid was checked by microscopical examinations.

(NH),iOH does not seem to affect the polyhedra, but the other
alkalies such as KOH and NaOH dissolve them readily. It was
found that as low a percentage as 1/16 per cent. NaOH will
dissolve polyhedra if they are boiled in the solution. For con-
venience 2 per cent. NaOH was used for the following tests.
Two grams of ground bodies were dissolved in the alkali by means
of heat. The solution was then dialyzed in order to get rid of
the alkali. The dialysis was usually complete after 24 to 48
hours. At the end of this procedure the proteins remain in
solution (i. e., on dissolving in alkali, after which, although the
alkali be removed, the polyhedra proteins remain soluble) and
the tests for soluble proteins can be applied. We obtained
positive reactions with acetic acid, nitric acid, cupric sulphate,
mercuric chloride, acetic acid with potassium ferrocyanide and
ammonium sulphate.

It might be well to mention that the percentages of NaOH used were accurate
and the material pure. In making up the solution we did not rely on the so-called
purity of the hydroxide sticks, but always eliminated every trace of Na2COs by
precipitation with Ba(OHl2.

After determining that the polyhedra are protein in nature it

NATURE OF POLYHEDRAL BODIES FOUND IN INSECTS. 379

was next thought advisable to ascertain whether or not they are
nucleoproteins. It seemed likely that this would be the case
for the reason that they are formed in the nuclei of certain tissue
cells. Two grams of whole (unrubbed) polyhedra were digested
for one week in artificial gastric juice. After this time a micro-
scopic examination failed to reveal any polyhedra. Theoret-
ically, everything should have been decomposed excepting the
nucleins which have a high phosphorus content. At the end of
one week's digestion the material was filtered through and dried
on ash free filter paper. The paper containing the residue was
then cut into fine pieces and put into a platinum crucible with
an oxidizing mixture (Na2CO3 (2 parts) + KNO3 (i part)).
The material was slowly ignited and the residue dissolved in
weak HNO3. This solution was then warmed with the addition
of some NH4NO3 in order to make the expected precipitate less
soluble. Lastly 5 per cent, molybdic acid was added. The
material was placed in an incubator and on standing a pronounced
yellow precipitate (ammonium-phospho molybdate) was formed.

On the basis of this and the other tests described the polyhedra
meet all of the requirements of the nucleoproteins. Nucleo-
proteins give all the color reactions, are soluble in water contain-
ing a small amount of alkali (1/16 per cent, in case of polyhedra)
and are precipitated from this solution by acetic acid. The
nucleins which have a high phosphorus content are not decom-
posed by gastric juice, and are obtained as an insoluble residue
after the artificial digestion of nucleoproteins with pepsin.

Since iron is an element known to be contained in chromatin it
was further thought advisable to determine whether the poly-
hedra during their synthesis from the chromatin and other
substances in the nuclei embodied any iron. It will be needless
to go through all the details of the analysis. Suffice it to say
that every precaution to prevent iron contamination from water,
air, etc., was used. Reagents known to be free from iron con-
tamination were employed. Furthermore, the polyhedra were
not rubbed with sand for fear of introducing iron in this manner.
.0988 of a gram of polyhedra were used and .00049 of a gram of
iron was found.

On dissolving polyhedra in alkali and after dialyzing away

380 R. W. GLASER AND J. W. CHAPMAN.

the alkali, three fractional precipitations with magnesium sul-
phate or sodium chloride can be obtained. Whether this means
that the polyhedra are composed of three separate proteins or
three groups of proteins, we are not prepared to say. The three
precipitates, however, do demonstrate that the polyhedra are
complex and not at all simple, a fact which does not seem strange
when one reflects on the complexity of cellular or nuclear material
in general.

As stated previously we regard the polyhedra as degeneration
products formed during the course of the disease in the nuclei
of certain tissue cells. The bodies are nucleoprotein crystal-like
(pseudo-crystalline) aggregates. The idea suggested itself to
us that it might be possible to dissolve the polyhedra and re-
crystallize them again after dissolution. If this should prove
to be possible, it would militate seriously against the views held
by Bolle, Fischer, Marzocchi, Knoche, Escherich and Miyajima
that the bodies are organisms or the stages of an organism. Our
experimental attempts at recrystallizing the polyhedra are a bit
varied and so we do not wish to overemphasize our results as yet,
but submit them with a full appreciation of their preliminary
value.

Polyhedra were dissolved in 2 per cent. NaOH by heating.
The process of dissolution was followed by an examination of
samples microscopically, and when traces of the bodies could no
longer be observed, the material was evaporated over a water
bath and examined before it became entirely dry. Besides long
NaoCOs crystals we found many small and large bodies which
resembled polyhedra. As a check we evaporated ordinary 2
per cent. NaOH, but we could not find the polyhedra-like
crystals obtained with the protein solution. This seemed to be
a result which offered possibilities, so we proceeded more carefully
with another experiment.

Two grams of polyhedra were dissolved by heating in 2 per
cent. NaOH. This solution was filtered and washed with ether
in order to rid the material of any traces of fat. The ether was
then eliminated by means of a separating funnel. The solution
was next dialyzed to get rid of the alkali and a few protein tests
were performed with some of the solution just to convince our-

NATURE OF POLYHEDRAL BODIES FOUND IN INSECTS. 381

selves of the presence of proteins. This protein solution freed
from the alkali was now slowly evaporated. On partial evapora-
tion, we found beautiful single and double crystals which simu-
lated the polyhedra very closely. If the material is evaporated
completely it is difficult to find the crystals owing to the presence
of coagulated and other protein material which hides them. If
one shoots water under the cover-slip, however, the crystals
again become visible just as soon as the coagulated sediment
softens and becomes transparent. This fact that a coagulated
residue remains shows that the entire protein material contained
in the original polyhedra is not used during the formation of these
new crystals. The majority of the crystals produced in this
way simulate polyhedra very closely, but some are rounder and
larger. Double forms are very common and are absolutely
indistinguishable microscopically from ordinary polyhedra. The
staining reactions of the crystals are similar to those of the poly-
hedra and Millon's reaction is identical. We have as yet not
obtained a sufficient amount of these new crystals to submit them
to all of the protein tests applied to the polyhedra. The crystals
are not quite as stable as the polyhedra. They seem to lack the
more resistant outer layer and therefore are more easily soluble in
alkali and other reagents. For this and other reasons we do not
claim to have reproduced typical polyhedra after their disintegra-
tion, but we firmly believe that the results are suggestive. It
seems unreasonable, after submitting proteins to the violent
hydrolytic action of both heat and alkali, to expect to reproduce
the identical proteins. However, in the material under con-
sideration, there seems to be a tendency for this particular
protein or group of proteins to crystallize out in the shape char-
acteristic of the polyhedra. These observations seem to support
our view that the polyhedra are merely degeneration-products
and not some inexplicable, unclassifiable organisms as supposed
by many workers. An organism certainly could not be dissolved
and its original form again reproduced or very nearly reproduced
on evaporation.

So far we are unable to obtain the crystals after the dissolution
of the polyhedra at every trial. Out of possibly ten trials, one
usually succeeds four or five times. Undoubtedly some condition

382 R. W. GLASER AND J. W. CHAPMAN.

of which we are at present ignorant is responsible for the frequent
failures. We have performed a sufficient number (15) of these
recrystallization experiments, however, to warrant a report of
the results.

Crystallizable proteins are, of course, not uncommon. Haemo-
globin is perhaps the best known example in animals and the
aleurin grains a well known example in plants. By fractional
precipitation with magnesium sulphate or sodium chloride two
or three separate crystalline proteins can be obtained from
the albumen of the hen's egg. In insects, by the evaporation
of blood with a trace of acetic acid, beautiful protein crystals
can be obtained, different in every species.

Our view regarding the nature of the polyhedral bodies may
therefore be summarized as follows: During the course of the
disease the virus disintegrates the nuclear material in such a
way that crystal-like bodies called polyhedral bodies or polyhedra
are synthesized out of the disintegrating proteins. Just how the
process from nuclear material to polyhedra takes place .is at
present unknown. At any rate, from our morphological observa-
tions, experimental infection data (published elsewhere) and
from our chemical studies here presented, it seems clear that the
polyhedra are nucleoprotein degeneration-products and not
organisms responsible for a series of insect diseases.

f SUMMARY.

1. Polyhedral bodies are found in many different species of
lepidopterous larvae.

2. The bodies are specific for a certain type of disease.

3. The polyhedra vary in size in the different species.

4. There exists a striking similarity in shape between the
polyhedra found in different species.

5. The polyhedra are structurally complicated.

6. They arise in the nuclei of certain tissue cells.

7. Cytoplasmic inclusions are found in certain diseases of
higher animals.

8. Nuclear inclusions have not been known previously.

9. The polyhedra are nucleoprotein crystal-like degeneration-
products and not organisms.

NATURE OF POLYHEDRAL BODIES FOUND IN INSECTS. 383

10. The polyhedra contain iron and phosphorus.

1 1 . On dissolving polyhedra in alkali and after dialyzing away
the alkali and evaporating the protein solution crystals are
obtained which simulate the original polyhedra.

BIBLIOGRAPHY.

Chapman, J. W. and Glaser, R. W.

'15 A Preliminary List of Insects which Have Wilt, with a Comparative Study

of their Polyhedra. Journal Econ. Ent., Vol. 8, No. i, pp. 140-149, i pi.

'16 Further Studies on Wilt of Gipsy Moth Caterpillars. Journal Econ. Ent.,

Vol. 9, No. i, pp. 149-167. Tables 21.
Escherich, K. and Miyajima, M.

'u Studien iiber die Wipfelkrankheit der Nonne. In Naturw. Ztschr. Forst u-.

Landw. Jahrg. 9, Heft 9, s. 381-402, 6 fig.
Glaser, R. W. and Chapman, J. W.

'13 The Wilt Disease of Gipsy Moth Caterpillars. Journal Econ. Ent., Vol. 6,

No. 6, pp. 478-488.
Glaser, R. W.

'14 The Bacterial Diseases of Insects. Psyche, Vol. 21, No. 6, pp. 184-190.
'15 Wilt of Gipsy Moth Caterpillars. Journal of Agri. Research, Vol. 4, No. 2,

pp. 101-128, 17 fig., 4 pi.
Huntemiiller

'13 Filtrierbare Virusarten. Ztschr. f. Chemotherapie, Bd. 2, Heft i, pp. 56-

69, Tafeln 9.
Osborne, Thomas B.

'99 Crystallized Vegetable Proteids. Am. Chem. Jour., Vol. 14, No. 8, pp.

1-28, fig. 3.

'99 Egg Albumin. Jour, of Am. Chem. Society, Vol. 21, No. 6, pp. 477-485
'99 On Some Definite Compounds of Protein-bodies. Jour, of Am. Chem.

Society, Vol. 21, No. 6, pp. 486-493.
Osborne, Thomas B., and Campbell, George F.

'oo The Proteids of the Egg Yolk. Twenty-third Report of the Conn. Agri.

Ex. Sta., pp. 339-348.
'oo The Protein Constituents of the Egg White. Twenty-third Report of

Conn. Agri. Ex. Sta., pp. 348-375-
Osborne, Thomas B., and Harris, Isaac F.

'03 The Precipitation Limits with Ammonium Sulphate of Some Vegetable

Proteins. Jour, of Am. Chem. Society, Vol. 25, No. 8, pp. 837-842.
Prowazek, S. von.

'07 Chlamydozoa. II. Gelbsucht der Seidenraupen. In Arch. Protistenk. ,

Bd. 10, Heft 2/3, s. 359~364. 2 fig.
'12 Untersuchungen iiber die Gelbsucht der Seidenraupen. In Centbl. Bakt.

(etc.), Abt. i, Orig., Bd. 67, Heft 4, s. 268-284, i fig., 2 pi.

'12 Gelbsucht (Polyederkrankheit) der Raupen. In Prowazek's Handbuch
der Pathogenen Protozoen, Bd. I., s. 156-161, 2 fig., Verlag von Johann
Barth. Leipzig, 1912.
Tubeuf, Carl von.

'92 Die Krankheiten der Nonne. (Liparis monacha.) In Forstl. Naturw.
Ztschr., Jahrg. i, Heft i, s. 34-3?. fig. 9-16; Heft 2, s. 62-79.

384 R. W. GLASER AND J. W. CHAPMAN.

'92 Weitere Beobachtungen iiber die Krankheiten der Nonne. In Forstl.

Naturw. Ztschr., Jahrg. i, Heft 7, s. 277-279.
'n Zur Geschichte der Nonnenkrankheit. In Naturw. Ztschr. [Forst u-.

Landw., Jahrg., 9, Heft 8, s. 357~377-
Wolbach, S. B.

'12 The Filterable Viruses, a Summary. Jour, of Med. Research, Vol. 27,

No. i (New Series, Vol. 22, No. i), pp. 1-25.
'n The Nature of Trachoma Bodies. Jour. Med. Research, N. S., Vol. 19,

No. 2, pp. 259-264. 3 pi.
Wolff, M.

'10 Ueber eine neue Krankheit der Raupe von Bupalus piniarius L. In

mitteilung des Kaiser Wilhelm Institutes fur Landwirtschaft in Bromberg,

Bd. Ill, Heft 2, s. 69-92.
Wahl, Bruno.

'09-' 1 2 Ueber die Polyederkrankheit der Nonne (Lymantria monacha L.).

In Centbl. Gesam. Forstw., Jahrg. 35, Heft 4, s. 164-172; Heft 5, s. 212-215;

Jahrg. 36, Heft 5, s. 193-212; Heft 8/9, s. 377~397; Jahrg. 37, Heft 6, s.

247-268; Jahrg. 38, Heft 8/9, s. 355-378.

386 R. W. GLASER AND J. W. CHAPMAN.

PLATE I.

FIG. i. Photomicrograph of a smear showing free polyhedra.
FIG. 2. Photomicrograph showing various stages during the formation of
polyhedra in tissue nuclei of a gipsy moth caterpillar. X 720.

BIOLOGICAL BULLETIN VOL XXX

PLATE I.

FlG. I.

FIG. 2.

fl. W. GLASER AND J. W. CHAPMAN.

388 R. W. GLASER AND J. VV. CHAPMAN.

PLATE II.

FIG. 3. Photomicrograph of army worm tissue showing fully developed poly-
hedra in nuclei of fat cells. X 300.

FIG. 4. Photomicrograph of army worm tissue showing normal hypodermal,
fat, muscle and blood cells. X 300.

BIOLOGICAL BULLETIN, VOL. XXX.

PLATE II-.

FIG. 3.

FIG. 4.

R. W. GLASER AND J. W. CHAPMAN.

3QO R. W. GLASER AND J. W. CHAPMAN.

PLATE III.

FIG. 5. Photomicrograph of army worm tissue showing normal, hypodermal,
fat, muscle and blood cells. X 300.

FIG. 6. Photomicrograph of army worm tissue showing disintegration of
nuclei and cells with liberation of polyhedra. X 300.

BIOLOGICAL BULLETIN, VOL. XXX.

PLATE III.

FIG. 5.

FIG. 6.

R. W. GLASER AND J. W. CHAPMAN.

AXIAL SUSCEPTIBILITY GRADIENTS IN THE EARLY
DEVELOPMENT OF THE SEA URCHIN.

C. M. CHILD.

(WITH 20 FIGURES.)

Axial gradients in susceptibility to cyanides and various other
agents have already been demonstrated in Planaria (Child, 136)
various infusoria (Child, '14), the early developmental stages of
the starfish (Child, '150), in a number of species of Oligo-
chetes (Hyman, '16), in a number of algae and in various
other animals the data for which are as yet unpublished. Thus
far such gradients have been found, at least in 'the earlier stages
of development in all forms examined, comprising more than
sixty species and including algae, ccelenterates, flatworms, echino-
derms, annelids, and vertebrates. The relation of these gradients
to developmental gradients of other kinds and the problem of
their significance for the physiological individual has also been
considered (Child, '156, Pt. Ill, 'i$c).

During the summers of 1913 and 1915 at the Marine Biological
Laboratory, Woods Hole, Mass., axial susceptibility gradients
were demonstrated in the early developmental stages of the sea
urchin, Arbacia punctulata, and the control and modification of
development by means of the differential susceptibility along
these gradients was found to be possible. In the present paper
only the direct evidence for the existence of such gradients,
based on the progress of death and disintegration along the axis,
is considered and this is incomplete in certain respects. The
indirect evidence from the control and modification of develop-
ment, with which I was chiefly concerned and which is of greater
interest, will be presented in another paper.

METHODS OF DEMONSTRATING SUSCEPTIBILITY GRADIENTS.

The method employed in demonstrating the gradients has
already been described (Child, '130, '156, Chap. III.) and con-
sists in directly determining the differences in susceptibility along

392 C. M. CHILD.

the axis or axes to cyanides and various other agents used in
concentrations sufficient to kill in the course of a few hours,
but not high enough to kill immediately and not low enough to
permit the organisms to become acclimated or acquire a tolerance
to them. The agents used in these studies on the sea urchin
were potassium cyanide, ammonium hydrate, ethyl alcohol and
hydrochloric acid in sea water. The various developmental
stages were placed in concentrations of these substances deter-
mined by preliminary experiment and the progress of death
along the axis was observed.

In many of the lower animals the progress of swelling, cy tolysis,
separation and disintegration serves directly as an indication of
the progress of death. In the developmental stages of the sea
urchin changes of this sort occur as the cells die, but they differ
somewhat with different reagents and different stages of develop-
ment. In KCN the cells swell, become spherical, and separate
from each other as they die and the region concerned breaks down
into a shapeless mass of these spherical cells which soon dis-
integrate. If motor activity is still present in other parts the
dead cells may be progressively left behind as the living portion
moves about. This disintegration of the body is more marked
in the earlier stages of the blastula and gastrula than in later
stages where supporting tissues have differentiated, but even in
the later stages extensive disintegration can be brought about
by return to sea water after a sufficient length of time in KCN.
These death changes are somewhat accelerated and intensified
by the return to water and death is marked by very complete
disintegration. Thus when KCN is used, the progress of death
can either be followed directly under the microscope in the KCN
solution, at least in the blastula and gastrula stages, or lots may
be returned to water at stated intervals and the progress of
death determined by the comparison of dead and living portions
of the body in successive lots. Both methods have been used,
but the latter is more satisfactory in many cases, because after
return to water the dead portions disintegrate rapidly and com-
pletely while the parts which are still alive may recover and
resume motor activity where the stages in which movement
occurs are concerned. The progress of death can also be made

AXIAL SUSCEPTIBILITY GRADIENTS. 393

visible by staining with neutral red before placing in KCN.
When death occurs the neutral red color changes to yellow, as
the alkali of the KCN solution penetrates the cells, and then
disappears.

The death changes in ammonium hydrate are very similar to
those in KCN. The cells swell and in the blastula and gastrula
stages separate, but in later stages the cells cohere more or less
after death and the visible death changes are merely swelling
and rounding of the cells and a consequent increase in size and
greater translucency of the whole. Here likewise, return to
water increases the disintegration and so makes it easier to follow
the progress of death, and neutral red may also be used as an
indicator of death in the same way as with KCN.

Ethyl alcohol is used in the same way as KCN and NH4OH,
but of course in much higher concentration. Disintegration of
blastula and gastrula stages is very complete and in the plutei
only the supporting tissues retain the body form.

In hydrochloric acid, however, the behavior of the cells is
different, as might be expected. The cells shrink and do not
separate and death may occur with very little visible change
except in size. But return to sea water after a sufficient length
of time in HC1 brings about disintegration and the dead and
dying cells swell and separate so that the progress of death can
be followed without difficulty by removing lots from HC1 to
sea water at regular intervals.

As regards concentration of the reagents used, a wide range is
possible according to the stage of development and the length
of survival time desired. Since the susceptibility increases very
greatly from fertilization up to the blastula stage as physio-
logical rejuvenescence occurs (Child, '15^, pp. 412-418), much
higher concentrations can be used for the former than for the
latter stages. At a temperature of 22-24° C. unfertilized eggs
begin to die after 4-5 hours in KCN m/ioo, while blastulse and
later stages begin to die after the same length of time in KCN
w/i,ooo. The other reagents were used only on the blastula
and later stages. In NH4OH m/^oo death of these stages begins
in half an hour to an hour, in m/i,ooo in 1-2 hours. The same
stages in alcohol 4 per cent, (roughly m 2/3) begin to die in 3-4

394 c- M- CHILD.

hours and after 5-10 minutes in HC1 777/400 death begins on
return to water, but it is difficult to determine just when it
occurs, if the stages are left in HC1. Much lower concentrations
of KCN and NH4OH can be used without the occurrence of any
appreciable degree of acclimation, but acclimation to alcohol
and HC1 takes place much more rapidly and in relatively high
concentrations, alcohol 1.5 per cent., HC1 777/2,000, so that for
direct demonstration of the susceptibility gradient with these
reagents the range of concentration is not so great.

The significance of differences in susceptibility to cyanides,
various narcotics and certain other agents has been considered
elsewhere (Child, '13^, '156, Chap. III.). It has been found
that in concentrations high enough to kill without acclimation
the susceptibility varies in general directly with the rate of
metabolism or of certain fundamental reactions, while in con-
centrations low enough to permit acclimation the higher the
metabolic rate the greater the degree and rapidity of acclimation
consequently in the long run the susceptibility to these concen-
trations varies inversely as the metabolic rate.

Although the susceptibility method has proved in certain
cases to be a very satisfactory means of distinguishing differences
in general metabolic activity, it is actually of course a rather
crude method and merely makes it possible to compare in a
rough way metabolic differences in different individuals or body
regions and does not of course give us any exact quantitative
data. Moreover, the nature of the method makes it evident
that we cannot expect to distinguish with certainty the minuter
metabolic differences, because the reagents used decrease to some
extent the differences which they are expected to show. Definite
and constant differences in susceptibility must mean considerable
differences in rate of reaction, but the absence of such differences
in susceptibility does not necessarily mean the complete absence
of differences in rate of reaction. Notwithstanding these limita-
tions, however, the method is useful and the positive results
obtained by means of it afford ample proof of its value.

One further point requires some consideration. The method
has been criticized, more particularly in personal conversation,
as involving certain assumptions concerning the action on living

AXIAL SUSCEPTIBILITY GRADIENTS. 395

chemically active protoplasm of cyanides and other agents used.
This, however, is not the case. There is no reason for believing
that different agents which retard or inhibit metabolic activity
in living protoplasm all act in the same way. Protoplasm is a
complex system in which both the physical substratum and the
chemical reactions play a part, but the important point is that
it is a system, i. e., that the different processes, changes and
conditions in it are not independent of each other but mutually
correlated and dependent to a greater or less degree. It is not
in the least improbable that different agents may give the same
general results, as regards susceptibility, even though one acts
primarily on the aggregate condition of the colloids or let us say
the permeability of membranes, another on the production or
constitution of enzymes, and still another on the chemical re-
actions of oxidation. My observations on susceptibility to
cyanides, various narcotics, acids, alkalies, metabolic products,
and even temperature have convinced me that the relation
between susceptibility to retarding or inhibiting agents and
conditions and general metabolic rate or the rate of certain
fundamental metabolic reactions is a very general relation and
there seems to be absolutely no reason for believing that it is
dependent upon any one particular method of action on the
protoplasmic system of the agent or condition employed. It still
remains of course for future investigation to determine the
exact method of action of each agent and condition, to formulate
the general rule and to discover and account for exceptions if
they exist. Since our present knowledge indicates both that
the oxidations are fundamental metabolic reactions and that
they are more or less dependent on various conditions in the
protoplasmic system; we may expect to find that their rate is
altered by a great variety of external agents and conditions
even though these do not enter directly into the chemical reac-
tions of oxidation.

In certain cases, as in the green plants, the energy for certain
of the synthetic or anabolic reactions is derived directly from
sources outside the organism and in such cases these reactions
may be to a considerable degree independent of the energy-
producing reactions in the organism. Such reactions, however,

396

C. M. CHILD.

are in a sense only preliminary, although of course essential to
the fundamental processes of life and the relation between
susceptibility and metabolic rate is primarily concerned, not
with them, but rather with the reactions which play a funda-
mental role in setting free the energy characteristic of living
organisms.

In the sea urchin the differences in susceptibility along the
axis so far as observed, are essentially the same whether cyanides,
alcohol, alkalies or acids are used as reagents. All these sub-
stances interfere with the action of the protoplasmic system in
some way, but by no means necessarily in the same way, yet
the general results as regards susceptibility are the same. This
fact is highly significant and indicates to some extent the general
character of the relation between susceptibility and the funda-
mental conditions and processes of life.

THE EGGS AND CLEAVAGE STAGES.

Since experiments on the control and modification of develop-
ment through differential susceptibility along the axes had shown
very conclusively the existence of axial gradients in the early de-
velopmental stages and had indicated their probable presence even
in the unfertilized egg, and since it was soon found that the
character of the death changes made it difficult to reach definite
conclusions concerning death gradients from direct observation,
but little time was spent on these earlier stages and the evidence
obtained, so far as it has any value, is merely contributory.
I regret to state, however, that I neglected to employ as a means
of orientation the method of making the micropyle visible by
colored suspensions in the water. This might have made the
data on these earlier stages somewhat more conclusive.

AXIAL SUSCEPTIBILITY GRADIENTS.

397

The unfertilized eggs in KCN m/ioo show death changes of
the character indicated in Figs. 1-3. First, one or more clear
droplets or masses which contain some colloids and much water
appear on the surface of the egg and grow larger, while the
granular portion decreases in size. This process may go on, the
clear masses often uniting, until the clear and granular portions
are almost equal in size, but finally the granules spread and the

4

5

distinction between clear and granular areas disappears. The
only indication of a susceptibility gradient is the appearance
of the clear droplets only or earlier on one hemisphere or at one
pole. Within certain limits of concentration this has been
observed in some 70—80 per cent, of the eggs, but in some eggs
the clear droplets appear at opposite poles or over various parts
of the surface at the same time. The fertilized eggs before
cleavage (Figs. 4-6) behave in KCN essentially like the un-
fertilized.

The blastomeres of the earlier cleavages show a similar separa-
tion into clear and granular areas and here again the clear areas
or droplets usually appear first or only on one hemisphere or at
one pole of the egg. In eggs placed in KCN after the appearance
of the micromeres I believe that the apical hemisphere is in
general more susceptible than the basal.

An interesting gradient in the first cleavage appears in certain
concentrations of cyanide. In such cases the first cleavage plane
at its first appearance extends as a furrow only half way or less
around the egg and cuts through the egg in one direction without
the appearance of any furrow on the opposite side (Figs. 7-10).
In some cases a larger or smaller portion of the cytoplasm is
separated from the two blastomeres at the end of this cleavage

398

C. M. CHILD.

by the division of the cleavage furrow into two (Fig. n). Such
cleavages occurred in about 40 per cent, of eggs placed in KCN
w/i,OOO for io}/2 hours before fertilization then well washed and
kept in sea water and fertilized one hour after removal from KCN.
Figs. 7-10 are drawn from such eggs. After this treatment there

8

10

ii

is no elevation of a fertilization membrane from the egg surface,
and the blastomeres frequently become entirely separated after
cleavage. Eggs placed in KCN w/ioo fifteen minutes after
fertilization and washed and returned to sea water after three
hours also showed cleavage of this kind in 30-40 per cent. In
the other eggs of these lots cleavage is normal or incomplete and
often irregular or else delayed with simultaneous formation of a
number of blastomeres. Apparently then these one-sided cleav-
ages represent the first step in departure from normal cleavage.
The indirect evidence shows that a susceptibility gradient is
present in these stages of development and that the region of
highest susceptibility and therefore of highest metabolic rate is
the apical pole. It is evident that cyanide inhibits cleavage,
and if the apical pole is most susceptible, we should expect to
find cleavage most completely inhibited there, and least inhibited
at the basal pole. It is probable, therefore, that these cleavage
furrows start from the basal and proceed toward the apical pole,

AXIAL SUSCEPTIBILITY GRADIENTS. 399

and that in cases like that in Fig. n, where a portion of the
cytoplasm is cut off, it is the apical region, which is so much
injured that it cannot give rise to a cleavage furrow.

With the same methods of treatment a general gradation in
the size of the blastomeres was observed in 30—40 per cent, of the
eggs in stages from 32 cells onward, though it became less marked
in later stages as recovery proceeded. In these cases also it is
probable that the region of most rapid cleavage and therefore of
the smallest blastomeres is the basal region and the region where
cleavage is most inhibited and therefore the blastomeres are
largest is the apical region.

The small size and protoplasmic character of the micromeres
at the basal pole suggests the possibility that their metabolic rate
may be higher than that of the cells about them, but I have riot
been able to discover that their susceptibility is greater than
that of the adjoining cells, and the indirect evidence shows that
the mesenchyme cells, which are believed to be at least in part
descendants of the micromeres are among the least susceptible
if not the least susceptible of all the cells.

These observations on the earlier developmental stages are not
conclusive. Taken by themselves they are of little value, but
considered in the light of the indirect evidence and of the more
definite results obtained in later stages they possess a certain
significance as contributory evidence. My observations on these
stages were only incidental to other work and only KCN was
used as reagent.

THE BLASTULA AND GASTRULA.

As soon as the blastula begins to elongate the direction of
movement, and in more advanced stages the greater thickness
of the cellular wall in the basal region make it possible to dis-
tinguish apical and basal ends without difficulty. In these
stages death begins at the apical end and proceeds basally (Figs.
12-15), advancing in many cases somewhat more rapidly down
one side (Figs. 13, 14), probably that side which later becomes
the anterior end of the pluteus. The progress of death in the
basal direction is very regular, though a few cells here and there
may swell and push out of the body-wall earlier than others

400

C. M. CHILD.

about them. The susceptibility gradient is the same with all
the reagents and methods of use noted in the preceding section
so that there can be. no doubt of the existence of an apico-basal
gradient which is fundamentally related to the activity ofjthe

12

living protoplasm. The more rapid progress of death down one
side of the blastula which is frequently observed is probably an
indication of differences in other axes, but on this point certainty
is impossible.

,O

14

In the gastrula stage the same gradient appears (Figs. 16-18),
the apex of the gastrula, which represents the apical region of the
egg and blastula, being most susceptible, the basal least suscep-
tible. The susceptibility of the entoderm and the blastopore
region is very much lower than that of the other ectodermal
regions. The entoderm is still intact after practically the whole
ectoderm has disintegrated (Fig. 18) and in concentration's where

AXIAL SUSCEPTIBILITY GRADIENTS. 40!

ectodermal disintegration occurs in three or four hours dis-
integration of the entoderm is usually not complete until one or
two hours later or in some cases even a longer time.

By returning to sea water after the proper length of time in the

reagent it is possible to stop death at any level of the body and
to bring about recovery and further development of the parts still
alive. In the blastula it is more or less of the basal portion which
remains alive, and this may close up and gastrulate, producing
dwarf gastruke with a disproportionately large enteron. In the
gastrula stage also partial dwarf gastrulse with large enteron
result from partial disintegration and recovery, because the more
apical ectoderm may be in large part destroyed while the basal
region of the gastrula and the entoderm remain intact. The
forms of larvae which develop from these partial basal blastulae
and gastrulae will be described in another paper.

LATER STAGES.

In the early stages of transformation into the pluteus, the
gastrula loses its apparent radial symmetry and becomes tri-
angular in outline in basal or anal view, the base of the triangle
representing the anterior region of the future pluteus. In side
view the apex of the gastrula is seen to be shifted toward the
anterior end as compared with earlier stages (cf. Figs. 19 and 16).
In the further transformation this apical region becomes the
oral lobe of the pluteus and the long anal arms develop from the,
basal region of the anterior end (Fig. 20).

402 C. M. CHILD.

The susceptibility gradient in the earlier stages of this trans-
formation is the same as in the gastrula, apico-basal (Fig. 19)
and the same difference in susceptibility between ectoderm and
entoderm persists. Later, when the larva begins to elongate in

20

the antero-posterior direction, and the anal arms begin to develop,
these arms and the posterior end both appear as secondary
regions of high susceptibility (Fig. 20) though the susceptibility
of the anal arms is in general somewhat less than that of the oral
lobe and that of the posterior end somewhat less than that of the
anal arms. On the oral lobe and over the body death progresses
in the basal and posterior direction and in the anal arms from
tip to base of the arms.

In the fully developed pluteus the susceptibility gradients are
less marked. The ectoderm of the oral lobe and anal arms is still
somewhat more susceptible than that of other regions but the
differences are less conspicuous. In all these later stages, how-
ever, the entoderm remains much less susceptible than the ecto-
derm and apparently the mesenchyme is least susceptible of all
parts.

It is probable that the gradual fading out of the metabolic
gradients in the pluteus is a physiological change which precedes
and makes possible the development of the axial gradients of the
mature sea urchin which have been previously inhibited by the
existing axial relations. If this suggestion is correct these changes
are the factors which determine, or rather permit metamorphosis.

The development of the arms and the posterior elongation
has been shown to be dependent on the development of the
skeleton. That being the case the appearance of high suscepti-
bility in the anal arms and the posterior end is not self-determined
in these parts but results from skeletal growth.

AXIAL SUSCEPTIBILITY GRADIENTS. 403

DISCUSSION AND SUMMARY.

The existence in the apico-basal axis of the blastula, gastrula
and later stages of a definite and conspicuous gradient in suscepti-
bility which is the same for cyanide, alcohol, acid and alkali is
a significant fact, particularly in the light of the relations between
susceptibility to cyanides and various narcotics and general
metabolic rate. The results with HC1 and NH4OH show that
this relation is the same for these substances. The further
evidence for the existence of this gradient in the earlier develop -
mental stages, together with the difficulty of conceiving how an
apico-basal gradient could arise de novo during the earlier develop-
ment, leave little doubt that this axial gradient persists from the
unfertilized egg to the mature pluteus. This conclusion will be
confirmed by the indirect evidence presented elsewhere.

As regards gradients in other axes the evidence is less con-
clusive. It is probable that the asymmetry in the progress of
death frequently observed in the blastula (Figs. 13, 14) is an
indication of a difference between anterior and posterior regions
and this probability is strengthened by the appearance of a
similar asymmetry in the gastrula and prepluteus stages (Figs.
17 and 19).

It must be remembered, however, that this method of deter-
mining susceptibility is far from being a perfect method for the
demonstration of differences in metabolic rate. It can be ex-
pected to show only the grosser differences, for the reagents used
tend to decrease the differences which they are used to demon-
strate, and if the differences are originally slight they may disap-
pear so rapidly in the reagent that no appreciable or constant
differences in the time of death appear.

It is evident from the course of development that the apico-
basal gradient is the most strongly marked and in the early
stages even this is not very clearly defined by differences in
susceptibility as determined by time of death, though it becomes
more distinct later. It is not to be expected that the minor
differences along the axes of symmetry should appear as clearly
by this method as the differences along the major axis. The
important facts are that the major axis appears so distinctly as

404 C. M. CHILD.

a susceptibility gradient and that indications of susceptibility
differences in the minor axes have been observed.

The apico-basal susceptibility gradient is the same as that in
the starfish (Child, '150) and in all other forms examined, at
least in early developmental stages and in many cases throughout
life. That this gradient is of fundamental significance cannot
be doubted and I have attempted elsewhere (Child, 150) to
present some of the evidence which seems to me to indicate that
a physiological axis is fundamentally such a gradient in metabolic
rate.

The chief points are summarized as follows:

1. A distinct gradient in susceptibility to potassium cyanide,
ethyl alcohol, ammonium hydrate and hydrochloric acid is present
along the apico-basal axis of blastula, gastrula, and later stages
of larval development of Arbacia and indications of a gradient
are found in earlier stages.

2. In the apico-basal gradient the susceptibility is highest at
the apical end of the axis and lowest at the basal end. Since
susceptibility to these reagents varies in general directly with
metabolic rate, the susceptibility gradient indicates the existence
of a gradient in rate of metabolic activity in which the rate is
highest in the apical region and decreases basally.

3. Some indications of gradients in other axes appear in the
differences of susceptibility, but these are much less distinct
than the differences along the apico-basal axis.

4. The anal arms and the posterior end of the larval body ap-
pear as secondary regions of high susceptibility after they begin
to develop.

5. In the fully developed pluteus these susceptibility gradients
become less marked and doubtless disappear as metamorphosis
begins.

HULL ZOOLOGICAL LABORATORY,
UNIVERSITY OF CHICAGO,
February, 1916.

AXIAL SUSCEPTIBILITY GRADIENTS. 405

REFERENCES.
Child, C. M.

133 Studies on the Dynamics of Morphogenesis and Inheritance in Experi-
mental Reproduction. V. The Relation Between Resistance to Depressing
Agents and Rate of Reaction in Planaria dorotocephala and its Value as a
Method of Investigation. Jour. Exp. Zool., XIV.

'i3b Studies, etc. VI. The Nature of the Axial Gradients in Planaria and
their [Relation to Antero-posterior Dominance, Polarity and Symmetry.
Arch. f. Entwickelungsmech., XXXVII.

'14 The Axial Gradient in Ciliate Infusoria. BIOL. BULL., XXVI.

'153 Axial Gradients in the Early Development of the Starfish. Amer. Jour.
Physiol., XXXVII.

'i5b Senescence and Rejuvenescence. Chicago.

'150 Individuality in Organisms. Chicago.
Hyman, L. H.

'16 An Analysis of the Process of Regeneration in Certain Microdrilous Oligo-
chetes. Jour. Exp. Zool., XX, 2.

Vol. XXX. June, 1916. No. 6

BIOLOGICAL BULLETIN

THE MARINE BIOLOGICAL LABORATORY

EIGHTEENTH REPORT; FOR THE YEAR 1915
TWENTY-EIGHTH YEAR.

I. TRUSTEES (AS OF AUGUST, 1915) 407

II. ACT OF INCORPORATION 408

III. BY-LAWS OF THE CORPORATION 409

IV. THE TREASURER'S REPORT 411

V. THE LIBRARIAN'S REPORT 415

VI. THE DIRECTOR'S REPORT 418

Statement 418

1. The Staff 422

2. Investigators and Students 425

3. Tabular View of Attendance 431

4. Subscribing Institutions 431

5. Evening Lectures 432

6. Members of the Corporation 433

I. TRUSTEES

EX OFFICIO

FRANK R. LILLIE, Director, The University of Chicago.
GILMAN A. DREW, Assistant Director and Vice- President, Marine

Biological Laboratory.

D. BLAKELY HOAR, Treasurer, 161 Devonshire Street, Boston, Mass.
GARY N. CALKINS, Clerk of the Corporation, Columbia University.

TO SERVE UNTIL 1918

CORNELIA M. CLAPP Mount Holyoke College.

E. G. CONKLIN Princeton University.

Ross G. HARRISON Yale University.

CAMILLUS G. KIDDER 27 William Street, New York City.

407

408 MARINE BIOLOGICAL LABORATORY.

M. M. METCALF ......... Oberlin, Ohio.

WILLIAM PATTEN ........ Dartmouth College.

JACOB REIGHARD ........ University of Michigan.

W. B. SCOTT ............ Princeton University.

TO SERVE UNTIL 1917

S. F. CLARKE ............ Williams College.

CHARLES A. COOLIDGE. . . .Ames Building, Boston, Mass.

C. R. CRANE ............ \Voods Hole, Mass., President of the Board.

ALFRED G. MAYER ....... Carnegie Institution.

C. E. McCLUNG ......... University of Pennsylvania.

T. H. MORGAN .......... Columbia University.

ERWIN F. SMITH ......... United States Department of Agriculture.

E. B. WILSON ........... Columbia University.

TO SERVE UNTIL

H. H. DONALDSON ........ W7istar Institute of Anatomy and Biology.

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

C. W. HARGITT .......... Syracuse University.

H. S. JENNINGS .......... Johns Hopkins University.

GEORGE LEFEVRE ........ University of Missouri, Secretary of the

Board.
A. P. MATHEWS ......... The University of Chicago.

G. H. PARKER ........... Harvard University.

HENRY B. WARD ......... University of Illinois.

TO SERVE UNTIL 1915

H. C. BUMPUS ........... Tufts College.

R. A. HARPER ........... Columbia University.

W. A. LOCY ............. Northwestern University.

JACQUES LOEB ......... . .The Rockefeller Institute for Medical Re-

search.
F. P. MALL ............. Johns Hopkins University.

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

L. L. NUNN ............. Telluride, Colo.

JOHN C. PHILLIPS ........ 299 Berkeley Street, Boston, Mass.

II. ACT OF INCORPORATION

No. 3170.

COMMONWEALTH OF MASSACHUSETTS

Be It Known, That whereas Alpheus Hyatt, William Sanford
Stevens, William T. Sedgwick, Edward G. Gardiner, Susan Minns,

BY-LAWS OF THE CORPORATION. 409

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 estab ishing and maintaining
a laboratory or station for scientific study and investigation, and a
school for instruction in biology and natural history, and have com-
plied 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 privileges, and
subject to the limitations, duties, and restrictions, which by law apper-
tain 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 HUN-
DRED and EIGHTY-EIGHT. HENRY B. PIERCE,

Secretary of the Commonwealth.

[SEAL.]

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 Tustees 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 Direc-
tor and Assistant Director, who shall be chosen by the Trustees, shall
also be Trustees, ex officio.

4IO MARINE BIOLOGICAL LABORATORY.

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
publication 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
affairs of the Corporation; they shall present a report of its condition
at every annual meeting; they shall elect one of their number Presi-
dent and 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 busi-
ness. The Board of Trustees shall have power to choose an Execu-
tive Committee from their own number, and to delegate to such
Committee such of their own powers as they may deem expedient.

VII. The President shall annually appoint two Trustees, who shall
constitute a committee on finance, to examine from time to time the
books and accounts of the Treasurer, and to audit his accounts at the
close of the year. No investments of the funds of the Corporation
shall be made by the Treasurer except approved by the finance com-
mittee in writing.

VIII. The consent of every Trustee shall be necessary to dissolu-
tion of the Marine Biological Laboratory. In case of dissolution, the
property shall be given to the Boston Society of Natural History, or
some similar public institution, on such terms as may then be agreed
upon.

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.

TREASURERS REPORT. 4! I

IV. TREASURER'S REPORT

Inasmuch as the records of the Laboratory have been reor-
ganized on an income and expense basis, taking into account
the bills receivable and payable, as well as the cash receipts and
disbursements, the treasurer makes a departure from the form
of report previously employed in order to be in accord with the
new methods.

The cash statement of operations is, therefore, now presented
in consolidated form and the income and expense schedule follows
accompanied by a balance-sheet, showing the condition of the
Laboratory on December 31, 1915, as far as the figures are avail-
able. Figures for the plant and inventory valuations are in
process of compilation.

CONSOLIDATED CASH STATEMENT OF RECEIPTS AND DISBURSE-
MENTS FOR THE YEAR ENDED DECEMBER 31, 1915

OPERATING ACCOUNTS

Cash on hand January I, 1915 $1,686.68

Receipts from departments $46,676.99

Receipts from donations 35,500.00

Receipts from work done for others 1,402.16

Receipts from loans repaid 125.00

Total receipts for year $83,704.15

Payments for departmental ex-
penses $61,856.38

Payments for improvements . . . 16,863.51
Payments for work done for

others i,537-<>5

Payments for new town landing 761 .50

Payments for Loan 25.00

Total payments 81,043.44

Excess of receipts for year 2,660.71

$4.347-39
Less Oklahoma warrant included in cash receipts but

not yet paid 135.23

Cash balance December 31, 1915 $4,212.16

412 MARINE BIOLOGICAL LABORATORY.

CASH RECEIPTS AND PAYMENTS ON ACCOUNT OF INVESTMENTS
FOR THE YEAR ENDED DECEMBER 31, 1915

RESERVE FUND

Cash on hand January i, 1915 $210.30

Receipts:

Interest —$3,000 American Telephone &
Telegraph Company, 4 per

cent bonds . . . ." 60.00

% of $500 Western Telephone

& Telegraph Company 5's . . . 18.74
Dividend — 6 shares American Smelting &

Refining Company, preferred 42.00
8 shares General Electric Com-
pany 64.00

14 shares United Shoe Machin-
ery Corporation, preferred.. 21.00
2 shares Massachusetts Gas,

preferred 8.00

Interest on bank deposit 1.04

$425-08
Payments:

Purchase of 2 shares Massachusetts Gas

Companies, preferred 180.25 $244.83

LUCRETIA CROCKER FUND

Cash on hand January i, 1915 $ 99.04

Receipts:

Interest --1/5 of $1,000 American Tele-
phone & Telegraph Company

4 per cent, bond 8.00

Dividend — 18 shares Vermont & Massa-
chusetts Railroad Company. 108.00
i share West End Street Rail-
way Company 3.50

I share American Telephone &

Telegraph Company 8.00

2^/2 shares General Electric

Company 20.00

$246.54
Payments:

2 Scholarships 100.00 146.54

TREASURER'S REPORT. 413

LIBRARY FUND

Cash on hand January i, 1915 $174.16

Receipts:

Interest -—4/5 of $1,000 American Tele-
phone & Telegraph Company

4's 32.00

J4 of $500 Western Telephone &

& Telegraph Company, 5's. . 6.26
Dividends — 3 shares American Telephone

& Telegraph Company 24.00

1 share American Smelting &
Refining Company, preferred 7.00

2^/2 shares General Electric

Company 20.00

5 shares United Shoe Machin-
ery Company, preferred. . . . 7.51

2 shares Massachusetts Gas
Companies, preferred 8.00

$278.93
Payments:

Purchase 2 shares Massachusetts Gas

Companies, preferred 181.26 97.67

Cash on hand December 31, 1915 $489.04

INCOME AND EXPENSE FOR YEAR ENDED DECEMBER 31, 1915

Expense Income

Administration expense $ 6,805.36

BIOLOGICAL BULLETIN 2,594.69 $ 1,523.36

Boat department expense 6,629.63

Carpenter department expense 983.09

Chemical department expense 1,430.82

Dormitories 1,489.49 1,691.04

Fish trap 1,159.21 667.21

Instruction 3.435-98 5,150.00

Philosophical lectures 100.00

Library department expense 2,628.64

Mess 15.753-01 16,165.48

Membership dues i ,000.00

414 MARINE BIOLOGICAL LABORATORY.

Maintenance, buildings and grounds. . . . 5,220.23

New laboratory expense 1,927.76

Pumping station expense 464.70

Research income 3,175.00

Sundry expense and income 2,165.48 3,255.25

Supply department 12,822.40 19,150.77

Interest on notes payable 150.00

Total expense $65,760.49

Total income 51,778.11 $51,778.11

Excess of expense $13,982.38

Contribution by Mr. C. R. Crane 20,000.00

Excess carried to balancing account $ 6,017.62

BALANCE-SHEET, DECEMBER 31, 1915

Assets Liabilities and Capital

Cash, bank $ 4,212.16 Accounts payable. $ 1,394.02

Petty cash 200.00 Note payable. . . . 3,000.00

Accounts receivable 7,642.37 Trust funds 8,222.55

Investments 10,733.51 Balancing account 33,997-98

Investment cash. .. 489.04

Inventories

Plant account

Improvements, 1915 23,337.47

$46,614.55 $46,614.55

D. BLAKELY HOAR, ESQ., Treasurer,
Marine Biological Laboratory

Woods Hole, Mass.
161 Devonshire Street, Boston.

Dear Sir: We have audited the accounts of the Marine Bio-
logical Laboratory as kept at Woods Hole and of the trust funds
and accounts as kept at your office, 161 Devonshire Street, for
the year ended December 31, 1915.

We have checked the report of the Treasurer submitted above
and find it correct and in accord with the books. The extent
of the audit together with the supporting schedule is set forth
in a detailed report under date of February i, 1916.

Very respectfully,

HARVEY S. CHASE & Co.,
Certified Public Accountants

LIBRARIAN'S REPORT. 415

V. LIBRARIAN'S REPORT

AUGUST, 1915

Since the last annual report the reorganization of the library
has been carried on by Miss Scott, who has brought each depart-
ment up to a state of system and efficiency.

We can now make a more definite statement of the contents
of the library, including a number of accessions not previously
listed: Total number of accessions 10,046. Old and new ac-
cessions recorded since last report 6,746. Of these there have
been added since 1914:

934 volumes loaned by the American Museum of Natural
History,

466 volumes given by the Wisconsin Academy of Sciences,

200 given by Mr. Crane,

720 bound during year.

These volumes may be conveniently divided as follows:

Journals, for which most of our appropriation is expended and which form the
bulk of the library, in bound sets 95

Serials and publications of academies, museums, laboratories, and societies
(many of these are from exchanges which are steadily increasing) 188

Books:

Zoology 321

Botany 134

Physiology 100

Hygiene and medicine 74

Chemistry 18

Evolution 82

Psychology 36

Philosophy 104

Miscellaneous 198

Total 1,067

We can buy very few books and are dependent upon gifts from
authors and publishers, who recognize that this is a particularly
useful place to have their works examined.

The separates number 4,000. Our reprint collection might
easily be notably increased through the aid of authors and
friends. Such duplicates are in great demand.

The following advances have been made. First, and very
important, is the considerable number of missing parts which

41 6 MARINE BIOLOGICAL LABORATORY.

have been secured by Miss Scott's enterprise toward completing
back volumes of sets.

These have been completed: Nature, Wilson Bulletin, Uni-
versity of California Publications in Zoology, Colorado University
Studies, Bergens Museum Publications.

These have been added to but are still incomplete:

Annales d. Sciences Naturelles — Zoology,

Bulletin Museum of Comparative Zoology, Harvard,

Proceedings Boston Society of Natural History,

Proceedings American Philosophical Society,

American Naturalist,

American Journal of Science and Arts,

Journal Medical Research,

Field Columbian Museum Publications,

Bureau of Fisheries, Bulletin and Report,

Contributions U. S. National Herbarium,

Proceedings American Association for the Advancement of
Science,

Proceedings Iowa Academy of Sciences,

Proceedings Indiana Academy of Sciences.
The following new sets have been added to the library:

Transactions American Microscopical Society,

Transactions Wisconsin Academy of Sciences,

Bulletin Torrey Botanical Club,

American Journal of Botany,

Proceedings Academy of Natural Sciences, Philadelphia,

Journal of Parasitology,

Illinois Biological Monographs,

Unpopular Review,

Annals Missouri Botanical Garden,

Indiana University Studies,

U. S. Bureau of Education, Report and Bulletin,

U. S. War Department, Bulletin,

Maine Agricultural Experiment Station, Report and Bulletin,

Museum of Brooklyn Institute of Arts and Sciences, Science

Bulletin.

Seventy-six volumes of Hoppe-Seyler's Zeitschrift fur Physio-
logische Chemie were purchased from Miss Koch for $200, and

LIBRARIAN'S REPORT. 417

the remainder of the set, which is now in the 94th volume, has
been secured.

Exchanges now number 29; six have been arranged this year:
Indian Museum; American Journal of Botany; New York Academy
of Sciences; Academy of Natural Sciences, Philadelphia; Biologische
Untersuchungen, Stockholm; and Zoologiska Bidrag, Upsala. New-
subscriptions have been placed for the Popular Science Monthly
and the English Journal of Physiology.

Here we must report a partial disposition of the Journal Fund
which was begun in 1912. Since then, a number have subscribed
to this fund which is to enable us to secure new files of journals.
Drs. Mayer, Knower, Meigs, Rice and Just have subscribed
from $5 to $10 a year for a period of five years. This has already
furnished $55 and will still accumulate from new gifts.

In addition, last year Mr. Tashiro and a number of other in-
vestigators, interested in securing the English Journal of Physi-
ology, subscribed amounts from 50 cents to $i each, thus securing
$7.50. This sum has now been increased to $12. For this, we
must thank Drs. Tashiro, Heilbrunn, Gould, Wherry, Packard,
Sturtevant, Bridges, Harvey, Metz, and Morgan, and Misses
Hoge, Medes, Browne, and Dunn. Such cooperation is most
encouraging. It makes possible a purchase long needed. We
have subscribed to the Journal with this gift and have used the
amount already accumulated for the Journal Fund to purchase
a number of back volumes, hoping gradually complete to the set.

Further growth in this spirit of cooperation in building up the
library is shown in the list of important gifts.

Various publishers have given a total of 16 books this year.
Miss Fay has given several quite valuable books on mushrooms.
Mr. Crane gave us two hundred volumes from his personal
library, as well as two very interesting Russian curios. Dr.
George T. Moore gave duplicates from his library, 28 bound
volumes and 21 separates. Dr. Philipp Fischelis donated 53
valuable papers by Kowalewsky and other Russian authors.
Dr. Conklin has given his book on "Heredity and Environment,"
Dr. Parker, his book on "Biology and Social Problems," and Dr.
Abbott has ordered a copy of his book on "General Biology"
sent to us. Dr. Calkins also presented his texbook on "Biology."

41 8 MARINE BIOLOGICAL LABORATORY.

The Wistar Institute gave a set of the "Biological Lectures,"
Mrs. Gardiner donating Volume I. to complete our set. Dr.
Ward gave a subscription to the Journal of Parasitology, and Dr.
Bumpus to the current volume of the American Naturalist and
the Unpopular Review. Dr. Gustav Retzius has promised to
send a complete set of his works as soon as transportation is
safer. Three hundred and twelve reprints have been presented
to the library this year by various authors.

Through the cooperation of Dr. George Wagner, a large number
of duplicates from the library of the Wisconsin Academy of Sci-
ences are being transferred to this library; 466 volumes have
been received and the number will probably reach 1,000.

The Prince of Monaco has sent us about 75 photographs of the
buildings and equipment of the Institut Oceanographique,
together with pamphlets telling about the foundation and
progress of the institute. A number of other stations have sent
various literature and photographs, and it is hoped to secure
material of this kind from all biological stations in the world.

We wish to acknowledge the cooperation of Drs. Moore and
Duggar in securing the Bulletin of the Torrey Botanical Club,
and of Dr. Ivey Lewis in arranging and preparing for the binder
a most confusing set of plates.

We are really in need of a number of journals that we do not
have, and we should complete as soon as possible certain sets
that have long been defective. The library needs an increased
appropriation for handbooks, monographs, and general works
like the Encyclopedia Britannica, the Century Dictionary, etc.

Above all, we must point out the great value of reprints and
books sent here by individuals, and the importance of personal
efforts in behalf of the library on the part of the biologists who
use it. Such interest has already done much to build it up.

VI. THE DIRECTOR'S REPORT

To THE TRUSTEES OF THE MARINE BIOLOGICAL LABORATORY:

Gentlemen: The season of 1915, the twenty-eighth session of
the Marine Biological Laboratory, was marked by a farther

DIRECTOR'S REPORT. 419

advance in many of the statistics of the Laboratory as will appear
from the appended lists (i) of the Staff, (2) of the Investigators
and Students, (3) the Tabular View of Attendance, (4) the Sub-
scribing Institutions, (5) of the Evening Lectures, (6) of the
Members of the Corporation. The Treasurer's report and the
Librarian's report also show a healthy condition of growth.
We have much cause for congratulation, and reason for deep
thankfulness to Mr. Crane without whose whole-hearted financial
support and intelligent sympathy with our objects, our best
efforts must have been comparatively ineffective.

The number of investigators in attendance was 137 as compared
with 129 in 1914, 122 in 1913 and 93 in 1912; the number of
students was 105 as compared with 89 in 1914, 69 in 1913 and
67 in 1912. The total attendance was 242 as compared with 218
in 1914, 191 in 1913 and 160 in 1912. These figures bear testi-
mony to the growing appreciation of the facilities furnished by
the Laboratory both in research and in instruction. The number
of institutions represented by these workers was 79, which is
almost the same as in the two preceding years ; the increase was
therefore due to larger attendance from certain institutions.
Indeed no considerable increase in the number of institutions
represented is to be expected, because practically all of the larger
institutions of higher education in the East and Middle West
are represented each year, and the variations from year to year
are accounted for by fluctuations of a more or less fortu* _ -ecu's
character among the smaller institutions represented. »

The receipts from subscribing institutions and fees Were
$8,325 as compared with $7,300 in 1914, $6,160 in 1913 and $5,y 75
in 1912. The Supply Department has also shown an encouraging
increase, having filled 282 more orders than in 1914 with total
paid receipts of $16,932.00 as compared with $14,003.35 in 1914.

Three important additions to the property and equipment of
the Laboratory were made during the year: (i) The so-called
Bake House property adjoining the Laboratory property on the
east was purchased; this piece has about 103 feet frontage on the
Eel pond and on the main street by 100 feet in depth, and its
title carries with it the private roadway previously controlled
jointly by the Laboratory and the owners of this property. The

42O MARINE BIOLOGICAL LABORATORY.

old house on the property has been fitted up as a workshop for
the carpenters' and plumbers' department, which was very badly
needed. The frontage on the eel pond is especially valuable
for future development. (2) The Laboratory also purchased the
grounds and house known as the Ritter property, 78^/2 by 100
feet, adjacent to the Whitman house, and opposite the lecture
hall. The house furnished much needed enlargement of our
dormitory facilities. (3) The old "Homestead" hitherto used
as the matron's and helpers' house, adjoining the mess, was torn
down and replaced by a much larger, modern and attractive
dwelling house with accommodations for 43 persons. As all of
this space was not needed for the mess, part of it was used for a
woman's dormitory. We owe these three splendid additions
to the facilities of the Laboratory, as I need hardly say, to the
continued generosity, and thoughtfulness in the matter of all
Laboratory needs, of the President of the Board of Trustees,
Mr. Crane. Other improvements during the year were as follows:
The steamer Cayadetta was thoroughly overhauled and her deck
raised 18 inches at a cost of over $2,000, so that she is a much
stauncher and more seaworthy boat than ever before. The filling
in of the Laboratory harbor frontage, including the Yacht Club,
has been completed; this will be graded and planted in the
spring and the building moved to the east end of the frontage
away from the center of the new laboratory building. Very
considerable additions were made to the kitchen and laundry of
the ri^ess, power machinery electrically operated being installed

for all the major operations, such as dish-washing, washing and
drying of laundry, freezing ice-cream, etc. The work has thus

greatly lightened and facilitated. The Laboratory also
transferred a 4O-foot strip at the extreme north and west end of
our Eel pond frontage to the town of Falmouth for the purposes
of a boat landing, which has now been built at a cost of about
$1,500, equally divided between the town and Laboratory.
This was in pursuance of an agreement entered into with the
town at the time of the construction of the drawbridge over the
Eel pond entrance. It is therefore now possible to use the Eel
pond as a harbor, and to secure delivery of supplies by this back
entrance to the Laboratory property.

DIRECTORS REPORT. 42!

With all of the enlargements and additions of the past few
years the Laboratory is still badly crowded during the height of
the season. This applies to a certain extent to the actual accom-
modations for workers in the laboratory buildings, but more
particularly to the housing accommodations in the village. We
are planning to meet the former condition by restricting numbers
in the larger classes; applications for admission will be received
up to May, and appointments then made in accordance with the
number of working places; if places are still available after such
assignments later applicants can be admitted. It is probable
that by means of various adjustments we can continue to care
for all qualified investigators.

The housing accommodations in the village, however, raise a
more serious question. They are entirely inadequate, and there
is a consequent tendency for the prices of rooms to advance,
which results in discouraging attendance. Investigators who
wish to carry on regular work at the Laboratory are unable to
rent houses at reasonable rates for the accommodation of their
families. Conditions have become increasingly discouraging in
these respects for the past several years. Moreover, land is
almost unavailable for those families who wish to solve the living
problem by building. It is undoubted th?* these conditions will
exercise an inhibiting influence on the work of the Laboratory
in the future, and one of the most important problems before
us is to devise some means of counteracting them.

As an organization we have laid the principle of cooperation
at our foundation, and we have attempted to build it into every
one of our activities. Our Board of Trustees is representative of
the institutions of learning of the country; our corporation re-
presents the workers in biology on the broadest lines we can secure ;
our staff of instructors is widely drawn from different institutions ;
our students and investigators come from all parts of the ac-
cessible territory, and some from abroad ; we ask only that they
have a fit preparation and exhibit the spirit of scholarship. We
are not tied down to any one institution or small section of the
country. In past years we have considered these principles
worth struggling for, and we have more than once sacrificed
security to freedom of action. Our recent years have been free

422 MARINE BIOLOGICAL LABORATORY.

and comfortable. We must not allow ourselves to forget that the
principles for which we stand are never entirely won; and I
appeal therefore to every member of the Board of Trustees
and to all members of the corporation not to allow a sense of
security to dull the edge of devotion, to keep the interests of the
Laboratory at heart and to work for their support in all possible
ways.

i. THE STAFF

1915

FRANK R. LILLIE, DIRECTOR,

Professor of Embryology, and Chairman of the Department of
Zoology, The University of Chicago.

OILMAN A. DREW, ASSISTANT DIRECTOR,
Marine Biological Laboratory.

ZOOLOGY

I. INVESTIGATION

GARY N. CALKINS Professor of Protozoology, Columbia Uni-
versity.

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

GILMAN A. DREW Assistant Director, Marine Biological Lab-
oratory.

GEORGE LEFEVRE Professor of Zoology, The University of

Missouri.

FRANK R. LILLIE Professor of Embryology, The University

of Chicago.

C. E. McCLUNG Professor of Zoology, University of Penn-
sylvania.

T. H. MORGAN Professor of Experimental Zoology, Co-
lumbia University.

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

II. INSTRUCTION

CASWELL GRAVE Associate Professor of Zoology, Johns

Hopkins University.
W. C. ALLEE Assistant Professor of Zoology, University

of Oklahoma.

DIRECTOR'S REPORT. 423

GEORGE A. BAITSELL Instructor in Biology, Yale University.

RAYMOND BINFORD Professor of Biology, Earlham College.

E. J. LUND Instructor in Protozoology, University of

Pennsylvania.
T. S. PAINTER Instructor in Biology, Yale University.

EMBRYOLOGY

I. INVESTIGATION (see Zoology)
II. INSTRUCTION

WILLIAM E. KELLICOTT. ..Professor of Biology, Goucher College.

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

(Absent in 1915.)

J. F. ABBOTT Professor of Zoology, Washington Uni-
versity.

CHARLES PACKARD Instructor in Zoology, Columbia Univer-
sity.

CHARLES C. ROGERS Professor of Zoology, Oberlin College.

PHYSIOLOGY

I. INVESTIGATION

ALBERT P. MATHEWS Professor of Physiological Chemistry, The

University of Chicago.

RALPH S. LILLIE Professor of Biology, Clark University.

HAROLD C. BRADLEY. . . Assistant Professor of Physiological

Chemistry, University of Wisconsin.
SHIRO TASHIRO Instructor in Physiological Chemistry, The

University of Chicago.

II. INSTRUCTION

RALPH S. LILLIE Professor of Biology, Clark University.

WALTER E. GARREY Associate Professor of Physiology, Wash-
ington University Medical School.

FRANK P. KNOWLTON Professor of Physiology, Syracuse Univer-
sity.

EDWARD B. MEIGS Associate in Physiology, Wistar Institute

of Anatomy and Biology.

PHILOSOPHICAL ASPECTS OF BIOLOGY AND ALLIED SCIENCES

LECTURES

EDWARD G. SPAULDING. . . . Professor of Philosophy, Princeton Uni-
versity.

424 MARINE BIOLOGICAL LABORATORY.

BOTANY

GEORGE T. MOORE Director, Missouri Botanical Garden and

Professor of Botany, Washington Uni-
versity.

B. M. DUGGAR Physiologist, Missouri Botanical Garden

and Professor of Plant Physiology,
Washington University. (Absent in

I9I5-)

IVEY F. LEWIS Professor of Botany, University of Mis-
souri.

R. H. COLLEY Instructor in Botany, Dartmouth College.

A. R. DAVIS Lackland Research Fellow, Missouri Bo-
tanical Garden.

W. H. WESTON Graduate Student, Harvard University.

LIBRARY

H. McE. KNOWER Professor of Anatomy, University of Cin-
cinnati, Librarian.
MARY E. SCOTT Assistant Librarian.

Chemical Supplies

OLIVER S. STRONG Instructor in Anatomy, College of Physi-
cians and Surgeons, New York City,
Chemist.

Supply Department

G. M. GRAY Curator.

JOHN J. VEEDER Captain.

E. M. LEWIS Engineer.

A. W. LEATHERS Collector.

A. M. HILTON Collector.

F. G. GUSTAFSON Collector in Botany.

EDNA E. WELLS . . Clerk.

F. M. MACNAUGHT Business Assistant.

HERBERT A. HILTON Superintendent of Buildings and Grounds.

DIRECTOR'S REPORT. 425

2. INVESTIGATORS AND STUDENTS

1915

A. ZOOLOGY
Independent Investigators

ABBOTT, JAMES F., Professor of Zoology, Washington University.

ADDISON, WILLIAM H. F., Assistant Professor of Normal Histology and Embry-
ology, University of Pennsylvania.

ALLEE, WARDER C., Professor of Biology, Lake Forest College.

ALLEN, EZRA, Professor of Biology, Philadelphia School of Pedagogy.

BAITSELL, GEORGE A., Instructor in Biology, Yale University.

BECKWITH, CORA J., Associate Professor of Zoology, Vassar College.

BINFORD, RAYMOND, Professor of Zoology, Earlham College, Richmond.

BORING, ALICE M., Associate Professor of Zoology, University of Maine.

BROWNE, ETHEL N., 510 Park Ave., Baltimore, Md.

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

CASTEEL, DANA B., Associate Professor of Zoology, University of Texas.

CHIDESTER, FLOYD E., Associate Professor of Zoology, Rutgers College.

CHILD, C. M., Associate Professor of Zoology, University of Chicago.

CLAPP, CORNELIA M., Professor of Zoology, Mt. Holyoke College.

CLARK, ELEANOR L., Columbia, Mo.

CLARK, ELIOT R., Professor of Anatomy, University of Missouri.

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

COPELAND, MANTON, Professor oi Biology, Bowdoin College.

COWDRY, E. V., Associate in Anatomy, Johns Hopkins Medical School.

COWDRY, N. H., Johns Hopkins Medical School, Baltimore, Md.

CRAMPTON, H. E., Professor of Zoology, Barnard College, Columbia Univ.

DANCHAKOFF, WERA, Woods Hole, Mass.

DEXTER, JOHN S., Professor of Biology, University of Saskatchewan.

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

DREW, GILMAN A-i Assistant Director, Marine Biological Laboratory..

DUNN, ELIZABETH H., Woods Hole, Mass.

GEE, WILSON, Professor of Biology, Emory University, Oxford, Ga.

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

GOLDSCHMIDT, RICHARD, Member of Kaiser Wilhelm Institut Fur Biologie.

GRAVE, CASWELL, Associate Professor of Zoology, Johns Hopkins University.

GREGORY, EMILY R., Professor of Biology, University of Akron.

GREGORY, LOUISE H., Instructor in Zoology, Barnard College.

GROSS, ALFRED O., Bowdoin College, Brunswick, Maine.

HARMAN, MARY T., Assistant Professor of Zoology, Kansas State Agricultural Col-
lege.

HARVEY, BASIL C. H., Associate Professor of Anatomy, University of Chicago.

HEILBRUNN, LEWIS V., Associate in Embryology, University of Chicago.

HEGNER, ROBERT W., Assistant Professor of Zoology, University of Michigan.

HOGUE, MARY J., Instructor in Zoology, Wellesley College.

JACOBS, MERKEL H., Assistant Professor of Zoology, University of Pennsylvania.

426 MARINE BIOLOGICAL LABORATORY.

KELLICOTT, WILLIAM E., Professor of Biology, Goucher College.

KNOWER, HENRY McE., Professor of Anatomy, University of Cincinnati.

LANE, HENRY H., Professor of Zoology, State University of Oklahoma.

LEFEVRE, GEORGE, Professor of Zoology, University of Missouri.

LEWIS, MARGARET R., Carnegie Institution.

LEWIS, WARREN H., Professor of Physiological Anatomy, Johns Hopkins Medical
School.

LILLIE, FRANK R., Professor of Embryology, University of Chicago.

LUND, ELMER J., Assistant Professor of Zoology, University ot Minnesota.

MAJLONE, EDWARD F., Associate Professor of Anatomy, University ot Cincinnati.

MACKLIN, CHARLES C., Johns Hopkins Medical School, Baltimore, Md.

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

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

MORGAN, ANNA H., Associate Professor of Zoology, Mt. Holyoke College.

MORRIS, MARGARET, Fellow in Zoology, Yale University.

PACKARD, CHARLES, Instructor in Zoology, Columbia University.

PAINTER, THEOPHILUS S., Instructor in Biology, Yale University.

PAPPENHEIMER, ALWIN M., Assistant Professor of Pathology, Columbia Univ.

PATTERSON, JOHN T., Professor of Zoology, University of Texas.

RICHARDS, A., Instructor in Zoology, University of Texas.

ROBERTSON, W. REES B., Assistant Professor of Zoology, University of Kansas.

SPAETH, REYNOLD A., Instructor in Biology, Yale University.

SPAULDING, E. G., Professor of Philosophy, Princeton University.

STOCKARD, CHARLES R., Professor of Anatomy, Cornell Medical College.

STREETER, GEORGE L., Research Associate, Carnegie Institution of Washington.

STRONG, OLIVER S., Instructor in Anatomy, Columbia University.

STURTEVANT, ALFRED H., Columbia University, New York City.

TENNENT, DAVID H., Professor of Biology, Bryn Mawr College.

WELCH, PAUL S., Assistant Professor of Entomology, Kansas State Agricultural
College.

WOODWARD, ALVALYN E., Fellow in Zoology, University of Michigan.

ZELENY, CHARLES, Associate Professor of Zoology, University of Illinois.

Beginning Investigators

ADKINS, W. S., Graduate Student, Columbia University.

ALTENBURG, EDGAR, Assistant in Botany, Columbia University.

BINKLEY, LELIA T., University of Texas.

BRIDGES, CALVIN B., Columbia University.

COHN, EDWIN J., University of Chicago.

COLLETT, MARY E., Instructor in Biology, Carnegie Institution of Technology.

FISH, J. Burton, Teacher of Biology, Boys' High School, Brooklyn.

FREIBERG, HENRY B., University of Cincinnati.

GLOBUS, JOSEPH H., Assistant in Anatomy, Cornell Medical College.

GOLDMAN, AGNES, Cornell University.

GOODRICH, H. B., Union College, Schnectady.

GOULD, HARLEY N., Fellow in Biology, Princeton University.

HANCE, ROBERT T., Fellow in Zoology, University of Pennsylvania.

HASTINGS, GEORGE T., Teacher of Biology, DeWitt Clinton High School, Yonkers,

N. Y.

\

DIRECTOR S REPORT. 427

HAYDEN, MARGARET A., Instructor in Biology, Carnegie Institute of Technology.

HOLT, CAROLINE M., Graduate Student, University of Pennsylvania.

HICKERNELL, Louis M., Instiuctor in Zoology, Syracuse University.

HOY, WILLIAM E., JR., Research Student, Princeton University.

JOHNSON, SYDNEY E., Fellow in Zoology, Northwestern University.

LYNCH, CLARA J., Instructor, Smith College.

MARKS, BERNICE, 4 East 94th St., New York City.

MEDES, GRACE, Graduate Student, Bryn Mawr College.

MINOURA, TADACHIKA, Zoology Department, University of Chicago.

MOORE, CARL R., University of Chicago.

OSTERUD, HJALMAR L., Instructor in Zoology, University of Washington

PHILLIPS, RUTH L., Associate Professor of Biology, Western College.

REAGAN, FRANKLIN T., Fellow in Biology, Princeton University.

ROOT, FRANCIS M., Johns Hopkins University.

SAFIR, SHELLEY R., Teacher in New York High School, New York City.

VANNEMAN, AIMEE S., Technician, University of Texas.

WARE, CLARA C., Graduate Student, Columbia Univeisity.

WEINSTEIN, ALEXANDER, Columbia University.

WHEAT, FRANK M., Columbia University.

WHITING, PHINEAS W., University of Pennsylvania.

WHITE, EDITH G., Columbia University.

WILKINS, LAWSON, Johns Hopkins University.

B. PHYSIOLOGY

Independent Investigators

BOGACKI, KAMIL J., Physician, Western Reserve University.

CHAMBERS, ROBERT, JR., Assistant Professor of Histology and Comparative
Anatomy, University of Cincinnati.

EDWARDS, DAYTON J., Instructor in Physiology, College of the City of New York.

GARREY, WALTER E., Associate Professor of Physiology, Washington University.

HYDE, IDA H., Professor of Physiology, University of Kansas.

JUST, ERNEST E., Professor of Physiology, Howard University.

KINGSBURY, FRANCIS B., Instructor in Physiological Chemistry, University of
Minnesota.

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

LILLIE, RALPH S., Professor of Biology, Clark University.

LOEB, JACQUES, Head of Department of Experimental Biology, Rockefeller In-
stitute for Medical Research.

MATHEWS, ALBERT P., Professor of Physiological Chemistry, University of Chicago.

MEIGS, EDWARD B., 1722 H. Street, N. W., Washington, D. C.

MOORE, ARTHUR R., Associate Professor of Physiology, Bryn Mawr College.

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

SCOTT, ERNEST L., Associate in Physiology, Columbia University.

TASHIRO, SHIRO, Instructor in Physiological Chemistry, University of Chicago.

WALLER, JOHN C., Graduate Student, University ot Chicago.

WARREN, HOWARD C., Professor of Psychology, Princeton University.

WASTENEYS, HARDOLPH, Associate in Experimental Biology, Rockefeller Institute
for Medical Research.

WERBER, ERNEST L. Sessel Research Fellow, Yale University.

428 MARINE BIOLOGICAL LABORATORY.

Beginning Investigators

ATWOOD, W. G., Dartmouth College.
CHAMBERLAIN, MARY M., Bryn Mawr College.
LEVY, AUGUSTUS, University of Chicago.
CATTELL, McKEEN, Harvard Medical School.

C. BOTANY

Independent Investigators

BLAKESLEE, A. F., Professor of Botany and Genetics, Connecticut Agricultural

College.

COLLEY, R. H., Instructor in Botany, Dartmouth College.
HIBBARD, R. PERCIVAL, Plant Physiologist, Michigan Agricultural College.
LEWIS, IVEY F., Professor of Botany, University of Missouri.
MOORE, GEORGE T., Director, Missouri Botanical Garden, St. Louis.
WESTON, WILLIAM H., JR., Instructor in Botany, Harvard University.

Beginning Investigators

COBB, RUTH, Falls Church, Virginia.

SMITH, PEARL M., Assistant in Botany, University of Wisconsin.

STUDENTS

1915

i. ZOOLOGY

ADAMS, A. ELIZABETH, Instructor in Zoology, Mount Holyoke College.
ANDRUS, EDWIN C., Student, Oberlin College, Oberlin, Ohio.
ANDRUS, WILLIAM D. W., Student, Oberlin College, Oberlin, Ohio.
BAIN, THERESE S., Vassar College, Poughkeepsie, N. Y.
BALDWIN, IMOGEN, Student, Mt. Holyoke College, So. Hadley, Mass.
BELL, CHARLES E., Student, Ursinus College.
BLANCHARD, NELLIE P., Head of Biological Dept., Hood College.
CATTELL, OWEN, Garrison-on-Hudson.
CATTELL, PSYCHE, Student, Sargent, Cambridge, Mass.
CHARLTON, HARRY H., Assistant in Zoology, Yale University.
COCKS, EDMUND, Graduate Student, Columbia University, New York City.
COBB, MARGARET C., Barnard College, New York City, N. Y.
COWDERY, LAWRENCE T., Student, Oberlin College, Oberlin, Ohio.
CROSS, HOWARD B., Instructor, Oklahoma University, Norman, Okla.
DIBELL, MABEL E., Instructor of Biology, Western College.

EHRENFELD, FREDERICK, Assistant Professor of Geology, University of Penn-
sylvania.

GILBERT, MARION, Student, Oberlin College, Oberlin, Ohio.
GREISHEIMER, ESTHER M., Clark University, Worcester, Mass.
HARVEY, HELEN F., Oberlin College, Oberlin, Ohio.
HARRIS, COLEMAN J., Lewisburg, Pa.

DIRECTOR S REPORT. 429

HAYDEN, RUTH, Goucher College, Baltimore, Md.

HEA, EMILY M., Teacher in High School, Morristown, N. J.

HEEMAN, HARRIET M., Student, Oberlin College, Oberlin, Ohio.

HOWE, MARION G., Mt. Holyoke College, So. Hadley, Mass.

HUGHES, DOROTHEA M., Simmons College, Boston, Mass.

HULST, MYRA M. Vassar College, Poughkeepsie, N. Y.

IRVINE, HELEN, Student, Mt. Holyoke College, So. Hadley, Mass.

JOSEPHS, H. W., Assistant in Chemistry, Harvard University, Cambridge.

KOSTIR, WENCEL J., Instructor, Ohio State University, Columbus, Ohio.

LUDWIG, ALBERT P., Student, Oberlin College, Oberlin, Ohio.

MARINUS, CARLETON J., Syracuse University, Syracuse, N. Y.

MAHR, ERNST F., JR., Syracuse University, Syracuse, N. Y.

McGRATH, JULE G., Laboratory Assistant, Hunter College, New York City.

MOSES, GERTRUDE, 2002 Bolton St., Baltimore, Md.

OUDESLUYS, HORTENSE, Teacher, Western High School, Baltimore, Md.

PENNYP ACKER, FRANCES W., Sweet Briar College.

RIPPLE, CLARENCE V., Student, University of Pennsylvania, Philadelphia.

ROGERS, MILDRED, Student, Goucher College, Baltimore, Md.

RUNYON, PAUL M., Student, Princeton University, Princeton, N. J.

SANFORD, ELDON W., Graduate Student, Yale University, New Haven, Conn.

SHELDON, PAUL B., Student, Oberlin College, Oberlin, Ohio.

SMITH, SUE F., Carnegie Institute of Technology.

SMITH, CHRISTIANNA, Assistant in Zoology Laboratory, Mt. Holyoke College, So.

Hadley, Mass.

SMITH, INEZ C., Student, Mt. Holyoke College, So. Hadley, Mass.
SPENCE, MARGARET, Instructor La Crosse State Normal School, La Crosse.
WILDER, KATHERINE, Science Assistant, The Newton High School.
WRIGHT, HELEN G., Student, Mt. Holyoke College, So. Hadley, Mass.

2. EMBRYOLOGY

ASHMAN, RICHARD, Rutgers College, New Brunswick, N. J.

ALLARD, ANN D., Teacher, So. Boston, Mass.

BLANCHARD, FRANK H., Instructor, Tufts College, Mass.

BLAU, JULIA E., Princeton, N. J.

CAHN, ALVIN R., Assistant in Zoology, University of Wisconsin, Madison, Wis.

CARROLL, MITCHEL, 617 So. i6th St., Philadelphia, Pa.

CLOSSON, JAMES H., Student, Princeton University, Princeton, N. J.

COBB, DOROTHY, Student, Radcliffe College, Cambridge, Mass.

CUTTER, ELIZABETH, Assistant in Zoology, Vassar College.

FARNUM, ALICE R., Student, Smith College, Northampton, Mass.

FISH, GORDON T., Assistant, Yale University, New Haven, Conn.

HAMLIN, HOWARD E., Student, Harvard University, Cambridge, Mass.

HENRY, EDNA M., Student, Barnard College, New York City, N. Y.

HERRICK, JOSEPH C., Professor of Biology, St. Joseph's Seminary, Yonkers, N. Y.

HUFFORD, CLARENCE E., Student, Oberlin College, Oberlin, Ohio.

HURLIN, RALPH G., Instructor, Clark College.

JAMIESON, JANET P., Student, University of Pennsylvania.

KAKIUCHI, SAMURO, Yale University, New Haven, Conn.

LEWIS, ELSIE M., Student, Oberlin College, Oberlin, Ohio.

430 MARINE BIOLOGICAL LABORATORY.

MACY, CORA F., Student, Syracuse University, Syracuse, N. Y.

MALLARD, AGNES K., Teacher, 1658 Columbia Road, So. Boston, Mass.

MYERS, MAE L., Associate Professor of Anatomy, Woman's Medical College of
Pennsylvania.

MILLIKAN, FRANCES, Student,. Smith College, Northampton, Mass.

MONTGOMERY, PRISCILLA B., 105 So. 41 St., Philadelphia, Pa.

PARMENTER, CHARLES L., Assistant Professor, University of Wisconsin.

PATTEN, MARY W., Goucher College, Baltimore, Md.

RICHARDS, LYMAN G., 259 Prospect St., Fall River, Mass.

RICHARDS, GEORGE L., 124 Franklin St., Fall River, Mass.

SHERWOOD, HELEN L., Student, Vassar College, Poughkeepsie, N. Y.

STOCKING, RUTH J., Professor of Biology, Agnes Scott College, Decatur.

STOCKING, BESSE E., Student, Goucher College, Baltimore, Md.

STRAUS, AUBREY H., Associate Professor of Bacteriology, Medical College of Vir-
ginia, Richmond, Va.

THOMAS, ANNA M., Student, Carnegie Institute of Technology.

WEBSTER, LESLIE T., Amherst College, Amherst, Mass.

WILLIAMS, G. HUNTINGTON, Graduate Student, Johns Hopkins Medical School.

WILLIAMS, JAMES W., Lincoln St., New Haven, Conn.

WINSLOW, MINA L., Graduate Student, L'niversity of Michigan.

3. PHYSIOLOGY

ADLER, FRANCIS H., University of Pennsylvania, Philadelphia, Pa.

BENSING, LERUE P., Research Assistant, Research Commission of Natural

Dental Association, Cleveland, Ohio.

BODINE, JOSEPH H., University of Pennsylvania, Philadelphia, Pa.
FENN, WALLACE O., Harvard University, Cambridge, Mass.
EGGSTEIN, A. A., Instructor in Pathology, Vanderbilt College.
HOLT, EMMETT, 14 West 55th St., New York City, N. Y.
HUGHES, WALTER S., Massachusetts Institute of Technology.
LOEB, ROBERT F., Student, University of Chicago, Chicago, 111.
LUNDGREN, C. ALBERT, Assistant in Biology, Clark University.
METCALF, JOHN T., Instructor in Psychology, Princeton University.
MINNICH, DVVIGHT E., Harvard University, Cambridge, Mass.
PRICE, WESTON A., Chairman, The Research Commission of the National Dental

Association, Cleveland, Ohio.

REEVES, PRENTICE, Assistant in Experimental Psychology, Princeton Univ.
SCHNEIDER, PETER A., Instructor in Zoology, University of Vermont.
THOMPSON, MARTHA, Teacher of Biology, Bay Ridge High School, New York

City. N. Y.

4. BOTANY

MCCONNELL, LOUISE J., 341 Rector St., Perth Amboy, N. J.
MONG, GRACE E., Student, Oberlin College, Oberlin, Ohio.
OAK, DOROTHY, Student, Barnard College, New York City, N. Y.
SEVERY, J. WARREN, Student, Oberlin College, Oberlin, Ohio.
YASUI, KONO, Student, Radcliffe College, Cambridge, Mass.
YOUNG, Anna R., Student, Smith College, Northampton', Mass.

DIRECTOR'S REPORT. 431

3. TABULAR VIEW OF ATTENDANCE

1911 1912 1913 1914 1915

INVESTIGATORS — Total 82 93 122 129 137

Independent:

Zoology 42 44 58 62 69

Physiology 18 14 17 22 20

Botany 8 10 n 10 6

Under Instruction:

Zoology 12 21 21 31 36

Physiology 2 2 7 I 4

Botany 2 7 3 2

STUDENTS — Total 65 67 69 89 105

Zoology. 26 24 33 43 47

Embryology 20 15 22 21 37

Physiology 6 n 8 10 15

Botany 13 17 7 15 6

TOTAL ATTENDANCE 147 160 191 218 242

INSTITUTIONS REPRESENTED—

Total 57 80 77 79

By investigators.. 37 43 50 51 59

By students 31 36 41 47 42

SCHOOLS AND ACADEMIES REPRESENTED.

By investigators 3 2 3 I 3

By students 9 I 6 5 9

4. SUBSCRIBING INSTITUTIONS— 1915

AMHERST COLLEGE

BARNARD COLLEGE

BOWDOIN COLLEGE

BRYN MAWR COLLEGE

CARNEGIE INSTITUTE OF TECHNOLOGY

CARNEGIE INSTITUTION, WASHINGTON

CLARK UNIVERSITY

COLUMBIA UNIVERSITY

DARTMOUTH COLLEGE

ELSE SERINGHAUS SCHOLARSHIP, HUNTER COLLEGE, NEW YORK

CITY
GOUCHER COLLEGE

432 MARINE BIOLOGICAL LABORATORY.

HARVARD UNIVERSITY

JOHNS HOPKINS UNIVERSITY

KANSAS STATE AGRICULTURAL COLLEGE

LUCRETIA CROCKER SCHOLARSHIPS

MOUNT HOLYOKE COLLEGE

NORTHWESTERN UNIVERSITY

OBERLIN COLLEGE

PRINCETON UNIVERSITY, DEPT. OF BIOLOGY

PRINCETON UNIVERSITY, DEPT. OF PSYCHOLOGY

RADCLIFFE COLLEGE

ROCKEFELLER INSTITUTE FOR MEDICAL RESEARCH

RUTGERS COLLEGE

SMITH COLLEGE

SYRACUSE UNIVERSITY

UNIVERSITY OF CHICAGO.

UNIVERSITY OF ILLINOIS

UNIVERSITY OF KANSAS

UNIVERSITY OF MICHIGAN

UNIVERSITY OF PENNSYLVANIA

UNIVERSITY OF TEXAS

UNIVERSITY OF WISCONSIN

VASSAR COLLEGE

WELLESLEY COLLEGE

WESTERN COLLEGE

WISTAR INSTITUTE

YALE UNIVERSITY

5. EVENING LECTURES, 1915

Friday, Ju'y 2,

PROF. C. R. STOCKARD.." Experimental Production of Racial De-
generacy by Alcohol Poisoning."
Tuesday, July 6,

PROF. C. H. PARKER.. . ." The Seals of the Pribiloff Islands."
Friday, July 9,

PROF. G. L. STREETER . ." Some Experimental Studies on the De-
velopment of the Membranous Laby-
rinth in the Tadpole."
Tuesday, July 13,

PROF. E. G. CONKLIN. .." Effects of Centrifugal Force on the

Structure and Development of the Egg."

DIRECTOR S REPORT. 433

Friday, July 16,

PROF. R. S. LILLIE " The Nature of Intelligent and Purposive

Action from a Physiological Point of
View."
Monday, July 19,

PROF. SIMON FLEXNER. ." The Control of Infection as Affected by

Variation Among Parasitic Microor-
ganisms."
Saturday, July 24,

PROF. G. N. CALKINS. . ." Protozoa and the Cancer Problem."
Friday, July 30,

PROF. GEORGE SHULL. . ." Inheritance of Sex in Lychnis.'"
Tuesday, Aug. 3,

PROF. C. B. DAVENPORT " Heredity of Criminality."
Friday, Aug. 6.

DR. MARTIN EDWARDS.." The Story of Bubonic Plague."
Tuesday, Aug. 10,

DR. ALFRED G. MAYER." The Role of Adsorption in Nerve Con-
duction."

6. MEMBERS OF THE CORPORATION

i. LIFE MEMBERS

ALLIS, MR. E. P., JR., Palais Carnoles, Menton, France.
ANDREWS, MRS. GWENDOLEN FOULKE, Baltimore, Md.
BILLINGS, MR. R. C., 66 Franklin St., Boston, Mass.
CAREY, MR. ARTHUR ASTOR, Fayerweather St., Boston, Mass.
CLARKE, PROF. S. F., Williams College, Williamstown, Mass.
CONKLIN, PROF. EDWIN G., Princeton University, Princeton, N. J.
CRANE, MR. C. R., Woods Hole, Mass.
DAVIS, MAJOR HENRY M., Syracuse, N. Y.
EVANS, MRS. GLENDOWER, 12 Otis Place, Boston, Mass.
FARLOW, PROF. W. G., Harvard University, Cambridge, Mass.
FAY, Miss S. B., 88 Mt. Vernon St., Boston, Mass.
FOLSOM, Miss AMY, 88 Marlboro St., Boston, Mass.
FOOT, Miss KATHERINE, 80 Madison Ave., New York City, N. Y.
GARDINER, MRS. E. G., Woods Hole, Mass.
GARDINER, Miss EUGENIA, 15 W. Cedar St., Boston, Mass.
HANNAMAN, MR. CHARLES E., 103 ist St., Troy, N. Y.

434 MARINE BIOLOGICAL LABORATORY.

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.
KENNEDY, MR. GEO. G., 284 Warren St., Roxbury, Mass.
KIDDER, MR. C. G., 27 William St., New York City, N. Y.
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, 171 Marlboro St., Boston, Mass.
MARRS, MRS. LAURA NORCROSS, 9 Commonwealth Ave., Boston,

Mass.

MASON, MR. E. F., I Walnut St., Boston, Mass.
MASON, Miss IDA M., i Walnut St. Boston, Mass.
MEANS, MR. JAMES HOWARD, 196 Beacon St., Boston, Mass.
MERRIMAN, MRS. DANIEL, Worcester, Mass.
MINNS, Miss SUSAN, 14 Louisburg Square, Boston, Mass.
MINNS, MR. THOMAS, 14 Louisburg Square, Boston, Mass.
MIXTER, Miss M. C., 241 Marlboro St., 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.

NOYES, Miss EVA J., 28 South Willow St., Montclair, N. J.

NUNN, MR. LUCIAN L. Telluride, Colo.

OSBORN, PROF. HENRY F., American Museum of Natural History,
New York.

PHILLIPS, DR. JOHN C., Windy Knob, Newham, Mass.

PHILLIPS, MRS. JOHN C., Windy Knob, Newham, Mass.

PORTER, DR. H. C., University of Pennsylvania, Philadelphia, Pa.

PULSIFER, MR. W. H., Newton Center, Mass.

ROGERS, Miss A. P., 5 Joy St., Boston, Mass.

SEARS, DR. HENRY F., 420 Beacon St., Boston, Mass.

SHEDD, MR. E. A.,

SMITH, MRS. C. C., 286 Marlboro St., Boston, Mass.

STROBELL, Miss E. C., 80 Madison Ave., New York City, N. Y.

THORNDIKE, DR. EDWARD L., Teachers College, Columbia
University, New York City, N. Y.

DIRECTOR'S REPORT. 435

TRELEASE, PROF. WILLIAM, University of Illinois, Champaign, 111.
WARE, Miss MARY L., 41 Brimmer St., Boston, Mass.
WARREN, MRS. S. D., 67 Mt. Vernon St., Boston, Mass.
WHITNEY, MR. HENRY M., Brookline, Mass.
WILLCOX, Miss MARY A., Wellesley College, Wellesley, Mass.
WILMARTH, MRS. H. D., Elliott St., Jamaica Plain, Mass.
WILLIAMS, MRS. ANNA P., 505 Beacon St., Boston, Mass.
WILSON, DR. E. B., Columbia University, New York Ci^y, N. Y.
WILSON, PROF. W. P., Philadelphia Museum, Philadelphia, Pa.

2. MEMBERS, JANUARY, 1916

ABBOTT, PROF. J. F., Washington University, St. Louis, Mo.
ABBOTT, Miss MARGARET B., The Bennett School, Milbrook, N.

Y.
ADDISON, DR. W. H. F., University of Pennsylvania, Medical

School, Philadelphia, Pa.
ADKINS, MR. \V. S., Texas Christian University, Fort Worth,

Texas.

ALLEE, DR. W. C., Lake Forest College, Lake Forest, 111.
ALLEN, PROF. EZRA, 125 Thompson Ave., Ardmore, Pa.
ALLYN, Miss HARRIET M., Hackett Medical College, Canton,

China.
ALSBURG, DR. C. S., U. S. Dept. of Agriculture, Washington,

D. C.
BAITSELL, DR. GEORGE A., Sheffield Scientific School, Yale

University, New Haven, Conn.

BAKER, DR. E. H., 154 W. Randolph St., Chicago, 111.
BANCROFT, DR. F. W., Aloha Farm, Concord, California.
BARDEEN, PROF. C. R., University of Wisconsin, Madison, Wis.
BECKWITH, Miss CORA J., Vassar College, Poughkeepsie, N. Y.
BEHRE, Miss ELINOR H., University of Chicago, Chicago, 111.
BEYER, DR. H. G., Stoneleigh Court, Washington, D. C.
BIGELOW, PROF. M. A., Teachers College, Columbia University,

New York City, N. Y.
BIGELOW, PROF. R. P., Mass. Institute of Technology, Boston,

Mass.
BINFORD, DR. RAYMOND, Earlham College, Richmond, Ind.

436 MARINE BIOLOGICAL LABORATORY.

BINKLEY, Miss LELIA T., University of Texas, Austin, Texas.
BLAKESLEE, PROF. A. F., Carnegie Station, Cold Spring Harbor,

Long Island.

BORING, Miss ALICE M., University of Maine, Orono, Maine.
Box, Miss CORA MAY, University of Cincinnati, Cincinnati,

Ohio.
BRADLEY, DR. HAROLD C., University of Wisconsin, Madison,

Wis,
BROWNE, Miss ETHEL N., Cornell Medical School, New York

City, N. Y.

BUDINGTON, PROF. R. A., Oberlin College, Oberlin, Ohio.
BUMPUS, DR. H. C., Tufts College, Mass.
BYRNES, DR. ESTHER F., 193 Jefferson Ave., Brooklyn, N. Y.
BUCKINGHAM, Miss EDITH N., 342 Marlboro St., Boston, Mass.
CALKINS, PROF. GARY N., Columbia University, New York

City, N. Y.
CALVERT, PROF. PHILIP P., Univ. of Pennsylvania, Philadelphia,

Pa.

CARLSON, PROF. A. J., University of Chicago, Chicago, 111.
CARVER, MR. GAIL L., 307 Adams St., Macon, Georgia.
GARY, DR. L. R., Princeton University, Princeton, N. J.
CASTEEL, DR. D. B., University of Texas, Austin, Texas.
CATTELL, PROF. J. McKEEN, Garrison-on-Hudson, N. Y.
CATTELL, MR. McKEEN, Harvard Medical School, Boston, Mass.
CHAMBERS, DR. ROBERT, JR., Cornell Medical College, New York

City, N. Y.

CHESTER, PROF. WEBSTER, Colby College, Waterville, Maine.
CHIDESTER, DR. F. E., Rutgers College, New Brunswick, N. J.
CHILD, PROF. C. M., University of Chicago, Chicago, 111.
CLAPP, PROF. CORNELIA M., Mount Holyoke College, South

Hadley, Mass.

CLARK, DR. E. R., University of Missouri, Columbia, Mo.
COE, PROF. W. R., Yale University, New Haven, Conn.
COLLEY, DR. R. H., Dartmouth College, Hanover, N. H.
COLTON, DR. H. S., Ardmore, Pa.
COOLIDGE, MR. C. A., Ames Bldg., Boston, Mass.
COPELAND, DR. MANTON, Bowdoin College, Brunswick, Maine.
COWDRY, DR. E. V., Johns Hopkins Medical School, Baltimore,

Md.

DIRECTOR'S REPORT. 437

CRAMPTON, PROF. H. E., Barnard College, Columbia Univer-
sity, New York City, N. Y.
CRANE, MRS. C. R., Woods Hole, Mass.

CURTIS, PROF. W. C., University of Missouri, Columbia, Mo.
DANCHAKOFF, MME. VERA, 451 57tth St., New York City, N. Y.
DAVIS, MR. DONALD W., DePauw University, Greencastle, Ind.
DERICK, PROF. CARRIE M., McGill University, Montreal,

Canada.
DEXTER, DR. J. S., University of Saskatchewan, Saskatoon,

Saskatchewan.

DODDS, PROF. G. S., University of Missouri, Columbia, Mo.
DONALDSON, PROF. H. H., Wistar Institute of Anat. and Biol.,

Philadelphia, Pa.

DORRANCE, Miss ANN, Dorranceton, Pa.
DORRANCE, Miss FRANCES, Dorranceton, Pa.
DREW, PROF. GILMAN A., Marine Biological Laboratory, Woods

Hole, Mass.

DUGGAR, PROF. B. M., Missouri Botanical Garden, St. Louis, Mo.
DUNGAY, DR. NEIL S., Carleton College, Northfield, Minn.
DUNN, DR. ELIZABETH, Woods Hole, Mass.
EATON, PROF. E. H., Hobart College, Geneva, N. Y.
EDWARDS, DR. D. J., College of the City of New York, New York

City, N. Y.
EIGENMANN, PROF. C. H., University of Indiana, Bloomington,

Ind.
EWALD, DR. W. F., Kaiserin Augusta Str. 78, Berlin W 10,

Germany.

FARNUM, Miss LOUISE W., 43 Hillhouse Ave., New Haven, Conn.
FERGUSON, PROF. J. S., Cornell Univ. Medical School, New York

City, N. Y.
FIELD, Miss HAZEL E., Randolph-Macon Woman's College,

College Park, Va.

FIELD, PROF. IRVING, Auburn, Mass.
FISH, MR. J. BURTON, Boys' High School, Brooklyn, N. Y.
GAGE, PROF. S. H., Cornell University, Ithaca, N. Y.
GARREY, PROF. W. E., Washington University Medical School,

St. Louis, Mo.
GEE, PROF. WILSON, Emory University, Oxford, Ga.

438 MARINE BIOLOGICAL LABORATORY.

GIES, PROF. W. J., Columbia Univ., Dept. Physiological Chemis-
try, New York City, N. Y.

GLASER, PROF. O. C., University of Michigan, Ann Arbor, Mich.

GLASER, DR. R. W., Bussey Institution, Forest Hills, Mass.

GOLDFARB, PROF. A. J., College of the City of New York, New
York City, N. Y.

GOODRICH, MR. H. B., Union College, Schenectady, New York.

GRAVE, DR. CASWELL, Johns Hopkins University, Baltimore, Md.

GREGORY, DR. LOUISE H., Barnard College, Columbia University,
New York City, N. Y.

GREGORY, DR. EMILY R., 501 South 42nd St., Philadelphia, Pa.

GREENMAN, DR. M. J., Wistar Institute of Anat. and Biol.,
Philadelphia, Pa.

GROSS, Miss BEATRIX H., 457 Convent Ave., New York City,
N. Y.

GUNTHER, Miss MAUD C., Eastern High School, Washington,
D. C.

HAHN, DR. C. W., 567 West i86th St., New York City, N. Y.

HANCE, MR. ROBERT T., University of Pennsylvania, Philadel-
phia, Pa.

HARGITT, DR. C. W., Syracuse University, Syracuse, N. Y.

HARMAN, DR. MARY T., Kansas State Agricultural College,
Manhattan, Kans.

HARPER, PROF. R. A., Columbia University, New York City,
N. Y.

HARRISON, MR. A. C., 660 Drexel Bldg., 5th and Chestnut Sts.,
Philadelphia, Pa.

HARRISON, PROF. Ross G., Yale University, New Haven, Conn.

HARVEY, PROF. B. C. H., University of Chicago, Chicago, 111.

HARVEY, DR. E. N., Princeton University, Princeton, N. J.

HAUGHWOXT, MR. F. G., College of Medicine and Surgery,
Calle Herran, Manila, Philippine Islands.

HAYDEN, Miss MARGARET A., Carnegie Inst. of Technology,
Pittsburgh, Pa.

HAYES, PROF. S. P., Mount Holyoke College, South Hadley,
Mass.

HEATH, PROF. HAROLD, Stanford University, San Francisco, Cal.

HEGNER, PROF. R. W., University of Michigan, Ann Arbor, Mich.

DIRECTOR'S REPORT. 439

HEILBRUNN, Dr. L. V., University of Chicago, Chicago, 111.
HOAR, MR. D. BLAKELY, 161 Devonshire St., Boston, Mass.
HOGUE, DR. MARY J., Wellesley College, Wellesley, Mass.
HOGE, Miss MILDRED A., Univ. of Indiana, Arbutus Apts.,

Bloomington, Ind.

HOLMES, PROF. S. J., University of California, Berkeley, Cal.
ISAACS, MR. RAPHAEL, University of Cincinnati, Cincinnati,

Ohio.

ISELEY, PROF. F. B., Central College, Fayette, Mo.
JACKSON, PROF. C. M., University of Minnesota, Minneapolis,

Minn.

JACOBS, MR. MURKEL H., University of Pennsylvania, Phila-
delphia, Pa.
JENNINGS, PROF. H. S., Johns Hopkins University, Baltimore,

Md.

JENNER, PROF. E. A., Simpson College, Indianola, Iowa.
JEWETT, PROF. J. R., Harvard University, Cambridge, Mass.
JONES, PROF. LYNDS, Oberlin College, Oberlin, Ohio.
JORDAN, PROF. H. E., University of Virginia, Charlottesville, Va.
JUST, PROF. E. E., Howard University, Washington, D. C.
KANDA, DR. SAKYO, University of Minnesota, Minneapolis,

Minn.

KELLICOTT, PROF. W. E., Goucher College, Baltimore, Md.
KELLY, MR. J. P., 2163 Gleason Ave., Unionport, N. Y.
KENNEDY, DR. HARRIS, 286 Warren St., Roxbury, Mass.
KING, DR. HELEN DEAN, Wistar Institute, Philadelphia, Pa.
KINGSBURY, PROF. B. F., Cornell University, Ithaca, N. Y.
KINGSLEY, PROF. J. S., University of Illinois, Urbana, 111.
KIRKHAM, DR. W. B., Yale University, New Haven, Conn.
KITE, Dr. G. L., Henry Phipps Institute, Philadelphia, Pa.
KNIGHT, Miss MARIAN V., 36 Bedford Terrace, Northampton,

Mass.
KNOWER, PROF. H. McE., University of Cincinnati, Cincinnati,

Ohio.

KNOWLTON, PROF. F. P., Syracuse University, Syracuse, N. Y.
KNUDSON, PROF. LEWIS, Cornell University, Ithaca, N. Y.
KRIBS, DR. HERBERT, University of Pennsylvania, Philadelphia,

Pa.

44O MARINE BIOLOGICAL LABORATORY.

LANE, PROF. HENRY H., State University of Oklahoma, Norman,
Okla.

LEE, PROF. F. S., 437 West 59th St., New York, City N. Y.

LEFEVRE, PROF. GEORGE, University of Missouri, Columbia, Mo.

LEWIS, PROF. I. F., University of Virginia, Charlottesville, Va.

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

LILLIE, PROF. FRANK R., University of Chicago, Chicago, 111.

LILLIE, PROF. R. S., Clark University, Worcester, Mass.

LINTON, PROF. EDWIN, Washington and Jefferson College, Wash-
ington, Pa.

LOEB, PROF. JACQUES, Rockefeller Institute for Medical Re-
search, New York City, N. Y.

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

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

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

LYMAN, PROF. GEORGE R., Federal Horticultural Board, Wash-
ington, D. C.

LYNCH, Miss CLARA J., Smith College, Northampton, Mass.

LUND, DR. E. J., University of Pennsylvania, Philadelphia, Pa.

McCLENDON, DR. J. F., University of Minnesota, Minneapolis,
Minn.

McCLUNG, PROF. C. E., University of Pennsylvania, Philadel-
phia, Pa.

McGiLL, DR. CAROLINE, Murray Hospital, Butte, Montana.

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

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

MACKENZIE, PROF. MARY D., Carnegie Institute of Technology,
Pittsburgh, Pa.

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

MACKLIN, DR. CHARLES C., Johns Hopkins University, Balti-
more, Md.

MALL, PROF. F. P., Johns Hopkins University, Baltimore, Md.

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

MARTIN, Miss BERTHA E., Wheaton College, Mass.

DIRECTOR S REPORT. 44!

MARQUETTE, MR. WILLIAM, Columbia University, New York
City, N. Y.

MATHEWS, PROF. A. P., University of Chicago, Chicago, 111.

MAYER, DR. A. G., Maplewood, N. J.

MEIGS, DR. E. B., Dairy Division Experiment Station, Belts-
ville, Md.

MELTZER, DR. S. J., 13 West I2ist Street, New York City, N. Y.

METCALF, PROF. M. M., 128 Forest St., Oberlin, Ohio.

MINOR, Miss MARIE L., Bryn Mawr College, Bryn Mawr, Pa.

MINOURA, MR. TADACHIKA, University of Chicago, Chicago, 111.

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

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

MOORE, MR. CARL H., University of Chicago, Chicago, 111.

MOORE, PROF. J. PERCY, University of Pennsylvania, Philadel-
phia, Pa.

MORGAN, PROF. H. A., Agricultural Experiment Station, Knox-
ville, Tenn.

MORGAN, DR. ANNA H., Mount Holyoke College, South Hadley,
Mass.

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

MORRILL, DR. C. V., 338 East 26th Street, New York City, N. Y.

MORRIS, Miss MARGARET, Yale University, New Haven, Conn.

MURBACH, DR. L., Central High School, Detroit, Mich.

NACHTRIEB, PROF. HENRY F., University of Minnesota, Minne-
apolis, Minn.

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

NEWMAN, PROF. H. H., University of Chicago, Chicago, 111.

NICHOLS, DR. M. LOUISE, 3221 Race St., Philadelphia, Pa.

OLIVER, MR. WADE W., Ohio-Miami Medical College, Cincin-

i

nati, Ohio.

OSBURN, PROF. R. C., Connecticut College, New London, Conn,
OSTERHOUT, PROF. W. J. V., Harvard University, Cambridge,

Mass.
PACKARD, DR. CHARLES, Columbia University, New York City,

N. Y.

PACKARD, DR. W. H., Bradley Polytechnic Institute, Peoria, 111.
PAINTER, MR. T. S., Yale University, New Haven, Conn.

442 MARINE BIOLOGICAL LABORATORY.

PAPPENHEIMER, DR. A. M., Columbia University, Dept. Pathol-
ogy, New York City, New York.

PARKER, PROF. G. H., 16 Berkeley Street, Cambridge, Mass.

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

PATTEN, Miss J. B., Elm Brook, South Natick, Mass.

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

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

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

PEARSE, PROF. A. S., University of Wisconsin, Madison, Wis.

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

PIKE, PROF. FRANK H., 437 West 59th Street, New York City,
N. Y.

PINNEY, Miss MARY E., Bryn Mawr College, Bryn Mawr, Pa.

PRENTISS, Miss HENRIETTA, Normal College, New York City,
N. Y.

PRICE, DR. WESTON A., Research Commission of the National
Dental Association, Cleveland, Ohio.

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

REA, DR. PAUL M., Charleston Museum, Charleston, S. C.

REIGHARD, PROF. JACOB, University of Michigan, Ann Arbor,
Mich.

REINKE, MR. E. E., Vanderbilt University, Nashville, Tenn.

RICE, PROF. EDWARD L., Ohio Wesleyan University, Delaware,
Ohio.

RICHARDS, DR. A., University of Texas, Austin, Texas.

ROBBINS, MR. W. J., Cornell University, Ithaca, N. Y.

ROBERTSON, Miss ALICE, Wellesley College, Wellesley, Mass.

ROBERTSON, PROF. W. R. B., University Club, Lawrence, Kansas.

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

ROSENOW, DR. E. C., Mayo's Clinics, Rochester, Minn.

RUDDIMAN, Miss MARGUERITE, 441 Senator Street, Brooklyn,
N. Y.

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

SANDS, DR. GEORGIANA, 348 N Main St., Port Chester, N. Y.

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

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

SCOTT, PROF. JOHN W., University of Wyoming, Laramie, Wy-
oming.

DIRECTOR S REPORT. 443

SHULL, DR. A. FRANKLIN, University of Michigan, Ann Arbor,

Mich.

SHUMWAY, MR. WALDO, Amherst College, Amherst, Mass.
SMITH, DR. BERTRAM G., State Normal College, Ypsilanti, Mich.
SOLLMAN, DR. TORALD, Western Reserve University, Cleveland,

Ohio.

SPAETH, DR. REYNOLD A., Yale University, New Haven, Conn.
SPAULDING, DR. E. G., Princeton University, Princeton, N. J.
SPENCER, DR. H. J., 8 West i6th St., New York City, N. Y.
STEWART, Miss MARY W., Barnard College, Columbia Univer-
sity, New York City, N. Y.
STOCKARD, PROF. C. R., Cornell Medical College, New York

City, N. Y.
STREETER, DR. GEORGE L., Johns Hopkins Medical School,

Baltimore, Md.

STRONG, DR. O. S., 437 West 59th St., New York City, N. Y.
STRONG, DR. R. M., Vanderbilt University, Nashville, Tenn.
STURTEVANT, MR. A. H., 528 West I23rd Street, New York City,

N. Y.

TASHIRO, DR. SHIRO, University of Chicago, Chicago, 111.
TAYLOR, Miss KATHERINE A., Cascade, Washington Co.,

Maryland.

TENNENT, PROF D. H., Bryn Mawr College, Bryn Mawr, Pa.
THOMAS, DR. ADRIAN, 2012 Hanover Ave., Richmond, Va.
THOMPSON, PROF. CAROLINE B., 195 Weston Road, Wellesley,

Mass.
TINKHAM, Miss FLORENCE L., 71 Ingersoll Grove, Springfield,

Mass.
TOMPKINS, Miss ELIZABETH M., 2015 Bedford Avenue, Brooklyn,

N. Y.

TREADWELL, PROF. A. L., Vassar College, Poughkeepsie, N. Y.
TURNER, MR. C. L., Marquette Univ. School of Medicine,

Milwaukee, Wis.
UHLENHUTH, DR. EDWARD, Rockefeller Institute for Medical

Research, New York City, N. Y.

VAN CLEAVE, DR. H. J., University of Illinois, Urbana, 111.
VAUGHAN, DR. T. W., U. S. Geological Survey, Washington, D. C.
WAITE, PROF. F. C., Western Reserve University Medical School,

Cleveland, Ohio.

444 MARINE BIOLOGICAL LABORATORY.

WALDRON, DR. F. R., 1701 Hill St., Ann Arbor, Mich.
WALKER, DR. GEORGE, Charles and Center Streets, Baltimore,

Md.

WALLACE, PROF. LOUISE B., Mount Holyoke College, South
Hadley, Mass.

WARBASSE, MRS. J. P., 384 Washington Ave., Brooklyn, N. Y.

WARD, PROF. H. B., University of Illinois, Urbana, 111.

WARDWELL, MR. E. H., Bristol, R. I.

WARREN, PROF. HOWARD C., Princeton University, Princeton,
N. J.

WASTENEYS, MR. HARDOLPH, Rockefeller Institute, New York
City, N. Y.

WATSON, MR. FRANK E., Hobart College, Geneva, N. Y.

WERBER, DR. E. I., Yale University, New Haven, Conn.

WESTON, MR. WILLIAM H., Western Reserve University, Cleve-
land, Ohio.

WHEELER, Miss ISABEL, Dana Hall, Wellesley, Mass.

WHEELER, PROF. W. M., Bussey Institution, Forest Hills, Mass.

WHERRY, DR. W. B., Cincinnati Hospital, Cincinnati, Ohio.

WHITE, Miss E. GRACE, Princeton University, Princeton, N. J.

WHITNEY, DR. DAVID D., Wesleyan University, Middletown,
Conn.

WHITING, MR. PHINEAS W., University of Pennsylvania, Phila-
delphia, Pa.

\VIEMAN, PROF. H. L., University of Cincinnati, Cincinnati, Ohio.

WILCOX, DR. ALICE W., Fairbanks Museum of Natural History,
St. Johnsbury, Vt.

WILDMAN, DR. E., 4331 Osage Avenue, Philadelphia, Pa.

WILLIAMS, DR. ANNA W., 549 Riverside Drive, New York City,
N. Y.

WILSON, PROF. H. V., University of North Carolina, Chapel Hill,
N. C.

WOGLOM, DR. WILLIAM H., Columbia University, New York City,
N. Y.

WOODRUFF, PROF. L. L., Yale University, New Haven, Conn.

WOODWARD, Miss ALVALYN E., Vassar College, Poughkeepsie,
N. Y.

YOUNG, MR. D. B., Hartley Hall, Columbia University, New
York City, N. Y.

AMITOSIS IN CELLS GROWING 'IN VITRO.

C. C. MACKLIN.

(From the Department of Anatomy, Johns Hopkins University, and the Marine
Biological Laboratory, Woods Hole, Mass.)

(3 PLATES, 27 FIGURES.)

CONTENTS

Introduction 445

Method 446

Observations 448

Nuclear division by amitosis 449

Subsequent history of cell containing a directly divided nucleus 453

Discussion 455

Fragmentation 458

Vital stains 458

Summary 459

Literature cited 460

Explanation of Plates 462

INTRODUCTION.

During recent years the conviction among cytologists has
become more and more strongly intrenched that the problem of
amitosis can not be satisfactorily solved by investigations based
alone upon the study of non-living tissue, but that its successful
conquest must rely principally upon the correct interpretation
of the succession of morphological and physiological changes
revealed by prolonged observation of the living cell, under normal
and artificially varied conditions. Yet until the advent of the
tissue culture method, workers in biology might well have de-
spaired of ever being able to attack the question in this way.
It is fortunate, therefore, that the technique of tissue cultivation
has been so perfected that even the minute structural details
of the living cell may be readily observed, and that the inspection
may be continued for hours at a time; fortunate, too, that con-
figurations which may be interpreted as stages in the process of
direct division, such as dumb-bell-shaped nuclei apparently
undergoing constriction, and bipartite nuclei, are not infrequently
found in tissue cultures, and, furthermore, that fixed and stained
preparations, to assist in the interpretation of the appearances

445

446 C. C. MACKLIN.

presented by the living cells, are easily made from such cultures.
These favorable circumstances plainly indicated the course
to be followed in an attempt to gain some knowledge of direct
cell division, and accordingly continuous observations of living
cells, and supplemental studies of fixed and stained cells of the
same type, were carried out.1

METHOD

The cells upon which observations were made were growing
at a temperature of 39° to 40° Cent, from the tissue of embryo
chicks from two to ten days old. Cultures were prepared by the
method of Lewis and Lewis ('15), Locke solution being used as a
medium. Activation of the growth was accomplished by the
addition to the medium of a small quantity of autogenous embryo
extract or bouillon. Thus the medium contained, besides the
various salts, small quantities of dextrose (.25 to I per cent.)
and protein. To offset the concentration due to evaporation
during planting and in the moist chamber it was found to be of
advantage to dilute the medium by adding 20 to 25 per cent, of
freshly distilled sterile water. Heart tissue was most frequently
used, and gave growths which were most serviceable for observa-
tion during the second twenty-four hour period.

By arranging the microscope within the incubator where the
tissues were cultivated it was not necessary to expose them to a
changed temperature during observation. A ray screen of copper
sulphate solution was found to be advantageous when artificial
light was used. Evaporation of the drop, with condensation
about the walls of the moist chamber, was lessened by placing a
small drop of distilled water in the cavity of the depressed slide,
and by eliminating air currents from the vicinity of the culture.

Light seemed to have a deleterious effect upon the living
cultures, so that lengthy continuous observation was not found
to be practicable. Accordingly inspections were made as short

1 The procedure of studying amitosis by the tissue culture method was suggested
by M. R. and W. H. Lewis, and I desire to record my appreciation of their kind
assistance and the use of their large collection of fixed and stained cultures, which
was utilized in the investigation. To Prof. F. R. Lillie I also am indebted for his
courtesy in placing a room at the Marine Biological Laboratory at my disposal,
where some of the work was carried on.

AMITOSIS IN CELLS GROWING IN VITRO. 447

as possible, and the light immediately turned off. The diffi-
culties attending direct and prolonged observation of the living
cell are not inconsiderable, for the eye must become accustomed
to distinguish minute structures through the contrast afforded
by varying grades of refractivity. At first only the most re-
fractive bodies are discernible, standing out as bright points or
lines, but gradually the less obvious structures, as mitochondria,
and amoeboid cell processes, come into view. Migration, in
many of the cells, is quite active, and hence it is necessary to
make observations frequently to prevent the cell from wandering
away from the field of vision and becoming lost. The extreme
sensitivity of the cell to light and heat, and to changes in osmotic
pressure of the media from evaporation within the moist chamber,
is responsible for the untimely termination of many observations,
and this difficulty becomes the more important when it is realized
that the study of a single cell must cover many hours to be com-
plete.

It was at first planned to select a single living normal cell and
observe it at frequent intervals over a long period in the hope
that eventually a cell would be found which would divide by
amitosis. This course, however, did not prove to be practicable
on account of the infrequency with which amitosis, even of the
nucleus alone, is met with. In a series of 20 fixed and stained
growths from the heart of the embryo chick, in which there wras
a total of 41,725 cells,1 only 50 constricted nuclei, which could be
regarded as directly dividing, were found (a ratio of I to 835),
and thus the chances of the occurrence of amitotic nuclear divi-
sion in a cell selected at random were but little better than one
tenth of one per cent. To avoid loss of time, therefore, the plan
was adopted of selecting a cell in which the amitotic process

1 The method adopted in making counts was as follows: 20 good preparations
from cultures of chick heart of various ages and stages of growth which had been
fixed and stained were selected. A small square was ruled with a diamond upon a
piece of glass after the method of Isaacs ('15), and this ruled glass was inserted in
the ocular so that a definitely outlined field was marked off upon the tissue culture
preparation on the stage. By manipulating the mechanical stage successive
fields could easily be brought into view, and the cells contained in them counted;
in this way all the cells in the entire new growth were counted except imperfect
cells and those near the original piece, which were several layers deep and were
1ndistinct.

448 C. C. MACKLIN.

appeared to have already commenced, viz., in which the nucleus
showed elongation and equatorial constriction, and observing
in detail the subsequent changes which it underwent.

Such a cell presents an appearance of the following general
type. The nucleus is somewhat lengthened, and, in the zone
equidistant from its poles there is to be seen, on one or both
sides, an indentation. In this concavity is situated characteris-
tically a body, the centrosphere (fig. I, c) whose refractivity is
somewhat greater than that of the surrounding cytoplasm. Its
outline is indistinct, but the edge seems to be irregular with short
toothlike processes. These change their shape slowly, and give
the impression of being pushed out and drawn in very gradually.
They are intimately related to definite refractive bodies — the
mitochondria — which, often rodlike and sometimes threadlike in
form, radiate from the periphery of the centrosphere. Indeed
the movements of the latter (described by Lewis and Lewis '15)
may be responsible for the apparent movement in the periphery
of the centrosphere.

Preparations fixed with osmic acid vapor and stained with
Heidenhain's iron hematoxylin (which was the method generally
employed) disclosed a minute granular body, generally paired,
within the centrosphere, wThich is recognized as the centrosome
or centriole (Fig. 24).

The position of the centrosphere within the nuclear concavity
or cleft has been noted by various authors, including Maximow
('08), who found the centriole-pair thus situated in cells, the
nuclei of which appeared to be dividing directly, in the mesen-
chyme of the embryo rabbit. In the cells of tissue cultures
examined the centrosphere was absent from the cleft in only
two per cent, of cases, and in these exceptions it may have been
originally situated as in the others.

OBSERVATIONS.

A number of extended observations were made upon living
cells containing such elongated and constricted nuclei, and, as a
general rule, the nucleus rounded out again, assuming the usual
form. Thus it was demonstrated that constriction alone does
not indicate that the nucleus will divide directly. Finally,

AMITOSIS IN CELLS GROWING IN VITRO. 449

however, a cell was found in which the nucleus divided directly
while being watched, and the following paragraph, extracted
from the protocol written at the time, is a brief description of the
process as observed. The drawings (Plate I.) were sketched
free-hand from observation at fifteen-minute intervals, and
afterward retouched by reference to fixed and stained cells of
similar morphology.

8.45 P.M. A cell (Fig. i), growing from a 57-hour culture of 5-day chick heart,
presented an elongated nucleus with a concavity upon one side. The outline of this
concavity was indistinct on account of the fact that the centrosphere (c] was situ-
ated very close to it.

9.00 P.M. (Fig. 2). The nucleus is now straight and the indentation is almost
obliterated.

9.15 P.M. (Fig. 3). The general outline of the cell has changed, and this has
been followed by change in shape of the nucleus. At first there were two nucleoli
within the nucleus, the lowermost being paired, but at 9.30 (Fig. 4) the latter appears
as a dumb-bell-shaped body. The nucleus is now rounded.

9.45 P.M. (Fig. 5). The nucleus is still rounded, and the nucleolar substance
consists of two masses close together.

10.00 P.M. (Fig. 6). The nucleus has become elongated again. There is no
apparent cleft, but the left side, against which the centrosphere is resting is not so
distinct as the right. The central nucleolus now appears as a single structure and
some refractive substance, resembling another nucleolus, is seen in the lowermost
pole of the nucleus.

During the next two fifteen-minute intervals (Figs. 7 and 8) a shortening of the
nucleus occurred, a shallow concavity to the left being noted. This concavity is
deeper at 10.45 (Fig. 9) its outlines being somewhat indistinct and the nucleus has
increased in length; across the middle of the nucleus is a refractive line.

n.oo P.M. (Fig. 10). The concavity is seen with difficulty, and the line across
the nucleus persists. Fifteen minutes after this (Fig. n) the elongated nucleus
shows an indentation upon the right side, opposite the one upon the left, and it
appears to be undergoing constriction into two parts of equal size. Between the
two nuclear portions the refractive line is seen as before, and, from the appearance
shown in cells fixed and stained with iron hematoxylin (Fig. 24) this is evidently a
strand of mitochondria.

11.30 P.M. (Fig. 12). The nuclear sacs are apparently quite separate, and
between them is the strand of mitochondria, and also part of the centrosphere,
this body, still undivided, having retained its position with reference to the nuclear
cleft.

Judging from the behavior of this and other elongated, bent
and dumb-bell shaped nuclei it would seem that the nucleus may
return to its original rounded form provided the constriction has
not gone too far, but, if the degree of constriction passes a certain
critical point the nuclear sacs become completely separated.

450 C. C. MACKLIN.

Furthermore, after the critical point is passed the division occurs
very rapidly.

The study of fixed and stained specimens served to throw con-
siderable light upon the process of direct division, for, by searching
the field, transitional forms were encountered, suggesting stages
in the history of the living nucleus just described. In the ter-
minal stage of direct nuclear division, such as that shown in Fig-
24, mitochondria were found characteristically lying across the
slender strand of nuclear membrane which was all that remained
of the connection between the two portions of the nucleus, and
the centrosphere (which is undivided in these transitional forms,
and also in the end product of nuclear amitosis, the binucleate
cells) is situated in the cleft dividing these parts.

This process involves a more or less equal mass division of the
nucleus without chromosome formation. It results in the pro-
duction of a binucleate cell, and, if the process is repeated, of a
trinucleate or multinucleate cell. Direct nuclear division was
not observed beyond the bipartite stage, but it seems rational to
suppose that (excluding the foreign body giant cells of Lambert
(1912 a and b) the giant cells of tissue cultures are formed from
the successive direct splitting of the nuclear fragments which
become larger by normal growth. With this latter there is
apparently associated an increase in the cytoplasm also.

It was not practicable to make a complete study of amitosis
upon the same cell, and the latter stages of this process, succeeding
direct division of the nucleus, were studied by selecting living
binucleate cells and making prolonged observations upon them.
A typical case, from a 24-hour culture of 8-day chick heart, is
given as follows:

11.50 A.M. The cell, which was of the characteristic connective tissue type,
showed two separate nuclear sacs, whose adjacent surfaces were in close contact.
One sac contained three nucleolar fragments, the other one. Fat granules were
fairly abundant, and were principally congregated at the nuclear poles. Mito-
chondria, long and threadlike, stretched between these granules and, where the
latter were abundant, the strands of mitochondria tended to arrange them in rows.
Mitochondria also radiated from the single centrosphere, situated opposite the
area of contact of the two nuclear sacs. The triangularly shaped cell body was
connected with adjacent cells by three main processes.

12. 20 P.M. A narrow interval can be seen between the nuclear parts, showing
that the latter have moved apart and are quite separate. Their position also has

AMITOSIS IN CELLS GROWING IN VITRO. 451

changed. Variation in the nucleoli is noted, there being now three in one nuclear
sac and four in the other. The outline of the cell body is now quadrilateral, and
this shifting of form has perhaps accounted for the rearrangement in relative posi-
tion and relationship of the nuclear parts. These, at 2.30, were again in contact,
but subsequently repeated the process of moving apart and coming together three
times during the observation, which ended at 11.15 p.m. During this time (nj^
hours) the cell was observed continuously, and underwent constant minor changes,
such as that of the outline, shifting about of cytoplasmic structures, and breaking
up, recombination and variation in size and shape of the nucleolar fragments. After
almost twelve hours the cytoplasm showed no indication of dividing.

These observations brought out the fact that what might be
mistaken for a single nucleus divided by a membrane across its
equator is really two nuclear sacs pressed close together, the equal
tension in the two bodies resulting in a flat membrane between
them, made up of the surfaces of contact. Child (191 1) describes
a type of amitosis which is characterized apparently by the growth
through the nucleus of a membrane or plate, the subsequent
splitting of which leads to the production of two nuclear sacs
quite separate one from the other. Nuclear fission of this type
was not found, and appearances suggesting a process of this
kind probably result from the close relationship of separate
nuclear parts, similar to the condition found in the cells of
tissue cultures. Partitions within the nucleus ave also simu-
lated, in these flattened cells, by long nucleoli, mitochondria
or folds in the nuclear membrane.

These observations also illustrate the characteristic behavior
of the nucleolar bodies, which undergo constant changes in size,
shape, number and position. These bodies stain well with gentian
violet when applied to the living culture. They appear as dark
masses after iron hematoxylin, but if differentiation with iron alum
is carried too far what was before a homogeneous mass becomes a
collection of granules (Fig. 27). The nucleolus appears to be a
concentrated gel of varying density, the granules representing
the denser areas.

Similar observations upon other living cells were carried out,
and in no case did any direct fission of the cytoplasm occur; this
finding was supported by the study of fixed preparations. On
the contrary the history of these binucleate cells was the same
as that of mononucleate cells of the same type. It was even
found that mitosis occurred in these cells, for, during the obser-

452 C. C. MACKLIN.

vation of a binucleate cell in a young culture the good fortune
was encountered of witnessing this entire process in it. A graphic
register of the successive changes is afforded by the drawings
(Plate II.) which were made with the aid of a camera lucida at
the times stated. The following is extracted from the protocol
written at the time of examination :

11.55 A.M. A typical connective tissue binucleate cell (Fig. 13) from a 1 9-hour
culture of 7-day chick heart was selected for observation. The two approximately
equal, sharply outlined nuclear sacs are in close contact, causing a flattening of the
apposed surfaces. Each nucleus contains a single, somewhat irregular nucleolus.
The cell is long and narrow, and the long axis of the nucleus is parallel with that of
the cell in which it is contained. A single centrosphere (c) is found opposite the
area of contact of the two nuclear parts. Numerous fat globules and mitochondria,
the latter showing characteristic movements, are seen.

12.40 P.M. (Fig. 14). The nucleus is still double, and the principal change noted
is the appearance of an additional nucleolus in the lower nuclear part.

i. 20 P.M. (Fig. 15). The two parts of the nucleus are distinct. Only one
nucleolus is now seen in each nuclear sac.

1.50 P.M. (Fig. 16). The cell outline has become modified, the cell body being
shorter, and the long axis of the nucleus has changed so that it is now almost at
right angles to its former direction. The nuclear surfaces are in close contact and
the upper end of the membrane formed by their approximation is indefinite in
outline.

3.05 P.M. (Fig. 17). The cell body has become more fusiform, and the long
axis of the nucleus has rotated through 90°. The double membrane (formed by the
areas of nuclear wall in contact) dividing the nuclear portions cannot be made out
except at the left side, and there indistinctly. An indefinite, refractive substance,
granular in character, is scattered along the line where this membrane had been;
the nucleoli are indefinite, and the uppermost one has been joined by an additional,
very small, mass of the same character. The upper part of the nuclear membrane
is not so distinct as heretofore.

It would appear that the change which has occurred in the part of the nuclear
membrane dividing the two nuclear sacs is a part of the general change affecting the
entire nuclear wall and leading to its gradual disappearance, and the process is
similar to that which occurs in the early stages of mitosis.

5.05 P.M. (Fig. 18). The refractive material in the zone formerly separating
the two nuclear sacs is more prominent than before; it seems to be chromatin. The
nuclear membrane is faintly marked.

Half an hour later this cell is seen to become gradually smaller, and to draw in
its processes. At the same time the nuclear parts become smaller. The nucleoli
also undergo diminution in size. Finally the cell takes a rounded and thickened
form — 6.00 P.M. (Fig. 19) — and is much more refractive than the cells surrounding
it: in fact it resembles a cell in the prophase of mitosis. The fat globules and
mitochondria assume a wreath-like appearance about the central clear space, in
which the nucleoli soon disappear. Though the main mass of the cell is almost
circular there are narrow processes attached to each pole. The cell remains ap-
parently unchanged for some time, though undoubtedly important readjustments
are going on within it.

AMITOSIS IN CELLS GROWING IN VITRO. 453

6.50 P.M. (Fig. 20). The cell is even smaller than before, and more rounded.
In the clear area within it is a refractive bar, which proves to be the equatorial
plate of chromosomes. Individual chromosomes cannot be distinguished, so that
it is impossible to count them, but their ends may be seen, as they project towards
the poles of the spindle. The chromosomes show a slow, oscillatory type of move-
ment, slight in extent. The spindle is represented by a fairly clear area, shaped
like two cones base to base, at the extremities of which the centrosomes are situated.
Astral rays cannot be seen. Mitochondria and fat globules encase this central
area containing the spindle like a shell. This becomes evident by focusing up and
down. Though the uppermost part of the spherical cell is on a level with the flat
resting cells, which are to be found about it, the lowermost part is much below this,
due to the fact that the cell is thickened, and projects into the medium.

7.05 P.M. (Fig. 21). The cell has suddenly become elongated and constricted
at the equator: the plate of chromosomes has evidently split, and the anaphase of
mitosis is being witnessed. The constriction about the middle of the cell can be
seen to be increasing, causing streaming of globules toward its extremities. At
the same time small, bubble-like processes emerge from the borders of the cell,
seeming to be forced out by the internal pressure of the cell body. Into these pro-
tuberances granules flow but later return into the main mass of the cytoplasm.
These processes soon become flattened and extended, forming pseudopodia possessed
of hyaline borders with amoeboid movement. The individual chromosomes of the
two masses in the expanded extremities gradually lose their distinctness and
become dispersed.

7.25 P.M. (Fig. 22). The constriction is more marked and the cells are almost
entirely separated. They are also becoming flattened out. In the upper daughter
cell a clear space for the nucleus is appearing. The constricted zone is somewhat
more highly refractive than the surrounding tissue and resembles a short thread.
This probably contains part of the remains of the spindle. Nuclear details are
not yet visible.

As the cell is watched nucleoli become manifest, at 7.35 P.M. two of these being
seen in each daughter nucleus in a clear space, surrounded by a distinct nuclear
membrane.

8.00 P.M. (Fig. 23). The daughter cells are now almost of normal size. Each
nucleus has a concave side, as is usual, and in each concavity is the new centre-
sphere, from which the mitochondria radiate. Two nucleoli appear in each nu-
cleus. The cells are spread out and flattened, and the fat granules are disposed as
in the ordinary cell.

The entire observation covered eight hours. The more active part of mitotic
division occupied about two hours, but if the initial nuclear changes be included
the duration of mitosis is much longer.

The sequence of changes followed in the above cell are in
almost all respects similar to those of mitosis many times ob-
served in the mononucleate cell. The only difference is that in
the cell described there were to begin with two separate nuclear
sacs instead of one. From the presence, in fixed preparations, of
bipartite nuclei each portion of which is in a condition of spireme,
and the absence, from such preparations, of bipartite nuclei in

454 c- C- MACKLIN.

which one portion only contains a skein, it is gathered that the
prophase in these cells is characterized by the spireme appearing
coincidently in the separate nuclear sacs (Fig. 25) ; later there is
formed from the double spireme a single equatorial plate of
chromosomes. Though these could not be counted there is no
reason for believing that their number was more than that normal
for the mononucleate cell.

Not only has mitosis been demonstrated by observation, both
on living and fixed preparations, to occur in binucleate cells, but
it has been found that it occurs relatively as frequently in these
cells as in those with a unipartite nucleus. By making cell
counts in the aforementioned 20 preparations from chick heart
375 binucleate cells were found in a total cell count of 41,725.
Thus binucleate cells made up a percentage of the total of 0.9,
or a ratio of I in in. Among these binucleate cells 2 were
found which were in the prophase of mitosis,1 a percentage of
0.53. Of the mononucleate cells, which numbered 41,106, 47
were in the prophase, or 0.114 per cent. By comparison of these
ratios it is found that mitosis occurs 4.65 times as frequently
among the binucleate as among the mononucleate cells; allowing
for the limited scope of the observation it seems reasonable to
conclude that mitosis is as frequent a phase of the life of binu-
cleate cells as of mononucleate, and it would seem that this is
their normal method of proliferation. If, in addition, they be
considered as dividing by direct fission (for which there is no
evidence either from living or fixed material) they would then
be possessed of an ability to multiply in excess of that of the
mononucleate cells, and there is no reason for supposing this to
be the case.

No evidence was brought to light that the separate parts of
the bipartite or multipartite nucleus ever combine except during
mitosis.

Another interesting observation was made in a fixed prepara-
tion, viz., that the early changes in the chromatin which presage
the onset of mitosis (i. e., the clumping of the chromatin and its

1 Prophases alone were counted, in estimating cells in mitosis, since in this stage
alone is it possible to distinguish the bipartite from the moiiopartite nucleus, on
account of nuclear fusion in stages later than this in the case of mitosis in the
binucleate cell.

AMITOSIS IN CELLS GROWING IN VITRO. 455

segregation into short rodlike masses and finally into a spireme)
can take place in a nucleus which is undergoing constriction.
Thus it appears that mitosis can proceed as usual in nuclei
partially divided or wholly divided by the amitotic process.

DISCUSSION.

The direct division of the nucleus is associated with certain
changes of the cell as a whole. Elongation of the nucleus seems
to be a prerequisite, and this apparently is secondary to a
lengthening and narrowing of the cell body occasioned by the
pull of its processes. It has been pointed out that the centro-
sphere is situated characteristically in a concavity at one side
of the nucleus, and, when the nucleus lengthens, this body sinks
deeper into its side; at the same time, judging from fixed prep-
arations, and also from the appearance of the living and dividing
nucleus, mitochondria come to lie across this narrow nuclear
isthmus. These bodies, the centrosphere and associated mito-
chondria, seem to play a part in the fission of the nucleus. The
exact manner of their action is not clear, but it may be that
streaming of the nucleoplasm away from the equator of the
nucleus follows upon the mechanical irritation of the nuclear
membrane by their movements, or possibly upon local alteration
of surface tension from their chemical change. Certainly the
position which these bodies take with reference to the constricted
nucleus points to their participation in the process of fission.

The centrosphere does not divide, nor does it encircle the
nucleus as in the form of division described by Meves ('91).
The nuclear membrane remains intact and nothing resembling
an amphiaster is formed.

Some theories of amitosis have attempted to place the re-
sponsibility for the initiation of the divisional stimulus upon the
nucleolus, and it has been found by some investigators that in
certain nuclei the nucleolus is the first body to become divided.
Such a function of the nucleolus does not obtain, however, in the
nuclei studied, for, although in some cases there was a division
of the nucleolus into two parts, one of which became alloted to
each separate nuclear part, yet, sometimes the nucleolus did not
divide, and one of the nuclear fragments was without visible

456 C. C. MACKLIN.

nucleolar material (Fig. 24). Occasionally two fragments would
be found in one nuclear part while the other had none; again
,it was very frequently found — indeed it wTas the rule — that a
nucleus would have two separate nucleolar fragments without
any nuclear division. In short not only was the division of the
nucleolus in most cases not followed by nuclear fission, but the
latter took place in many cases without division of the nucleolus.

Neither before nor after nuclear fission was there discovered an
instance of differential staining of the two halves of a nucleus
about to be divided, or already divided, as shown by Child ('07)
so that there was no evidence of this kind to support the belief
of Child that there may be a physiological independence of the
nuclear areas even before they become amitotically divided,
manifested by a variation in the staining of the two nuclear
halves. The study of living nuclei, too, divulged nothing in
favor of this hypothesis.

The result of this process of nuclear splitting was the formation
of a cell containing one or more separate nuclei of about equal
size, and each of about the same size as the nuclei of mononu-
cleate cells. After nuclear fission the separated nuclear elements
manifested the power of growth, and seemed to have metabolic
independence. The cell protoplasm also of these cells shows an
ability to increase in bulk. This is especially evident in giant
cells which can thus become quite large.

It is believed, furthermore, that binucleate cells and giant
cells in tissue cultures (except foreign body giant cells which
arise by fusion of previously separate cells) do not arise in any
other manner than that above outlined, for there is no other
adequate explanation of their origin. A careful examination of
living and fixed material does not reveal any evidence of fusion
of cytoplasm without fusion of the nuclei, so that there are no
grounds for admitting this as a possible theory of formation.
Although many of the binucleate cells undoubtedly do migrate
as such from the original piece, wrhere they are doubtless formed
in the same manner as they are in the new growth, yet an increase
of over 100 per cent, in their number in the growth of the second
day as compared with that of the first (as ascertained by making
careful counts of the 20 heart specimens aforementioned) can

AMITOSIS IN CELLS GROWING IN VITRO. 457

hardly be explained by the assumption of an increased emigration
of these forms during the second day; it is more probable that
some of these bipartite nuclei have originated in the new
growth, and the observation of this process here confirms this
conclusion.

Mitosis, too, cannot account for the formation of these bi- and
multipartite nuclei, for in all the cases of mitosis followed through
in the living condition the end result has always been two
daughter cells, quite separate except for a narrow connecting
process, and each containing a single centrosphere. The only
theories remaining for consideration are those of nuclear origin
de novo, or from chromidial extrusions, and these suppositions
are too improbable to discuss here. No other process than that
of nuclear amitosis, therefore, can account for the production
of bi- and multinucleate cells in the cultures examined.

Though the amitotically-divided parts of the nucleus seem
to possess metabolic independence, as noted, they do not appear
to have reproductive independence, for the reason that they are
never dissociated one from the other to become the nuclei of
separate cells. Furthermore they show only one type of cell
division, viz., mitosis, in which the process begins coincidently
in the two nuclear parts, manifested by the simultaneous appear-
ance in each part of a similar spireme. Although in amitosis
there is a mass division of the nuclear material there is no
meristic division, and it appears that before a cell containing an
amitotically divided nucleus can divide it is necessary that the
separated chromatin moieties should recombine. This was done
in the specimen examined during life, for the combined product
of the two nuclear sacs formed a single equatorial plate of chro-
mosomes. Such a type of amitosis, therefore, is not incompatible
with the chromosome hypothesis.

Spiremes in bipartite nuclei, and in dumb-bell-shaped nuclei
evidently undergoing amitosis, are not confined to the cells of
tissue cultures, for Maximow ('08) has described them in the
normally-developing mesenchyme cells of the embryo rabbit, and
this author refers to similar cell configurations which Karpow
('04) describes in the leucocytes of urodele amphibia.

458 C. C. MACKLIN.

FRAGMENTATION.

A note may here be made regarding a pathological change which
nuclei subjected to unfavorable conditions undergo, viz., frag-
mentation. This change consists in a breaking up and degenera-
tion of the nuclei. In cells which had grown for six days in
unchanged media, in which food and oxygen had become depleted
and metabolic products had accumulated (Fig. 26) and also in
cells growing in a medium to which a small amount of ethyl
alcohol had been added (Fig. 27), this form of degeneration was
seen. Multilobulation of the nucleus appeared to antecede the
actual breaking away of the parts. These latter were of different
shapes and sizes, often did not contain a nucleolus, and showed
no power of growth. The cytoplasm enveloping them did not
divide and apparently was incapable of increasing.

In preparations containing forms of this kind no mitoses were
seen, and the phenomenon seemed to be quite different from
nuclear amitosis which occurred only in healthy cells. It is
believed, moreover, that the process of fragmentation has been
confused with that of amitosis, and it is possible that it is this
confusion which has accounted for certain well-known views
which regard amitosis as an evidence of degeneration.

VITAL STAINS.

Finally, a word as to certain so-called "vital stains." It was
hoped that gentian violet would prove to be of value in rendering
visible the minutice of the living cell during its vital changes,
since the results following the use of this dye recorded by Church-
man and Russell ('14) and Russell ('14) with cultures of frog
tissue were so encouraging. Unfortunately the dye proved toxic
to the tissues used in a dilution of I in 200,000, and, though the
nucleoli, nuclear membranes, certain granules and the cell borders
were brought into sharp relief this staining was always accom-
panied by cessation of vital phenomena, and the cells speedily
went into degeneration.

Janus green, in a dilution of I in 80,000, was also toxic, and,
while it stained mitochondria specifically, yet these bodies soon
became granular and lost their characteristic appearance. Hence
neither of these dyes could be spoken of as acting "intravitam,"

AMITOSIS IN CELLS GROWING IN VITRO. 459

and they were of value only in obtaining rapidly information as
to the obscure structure of the cell; the cell, however, was thus
sacrificed.

SUMMARY.

The observations above described and the interpretations
thereof may be briefly summarized as follows:

Amitosis was found to involve only the nucleus and was not a
method of cell proliferation. It occurred in normal cells and
was characterized by a separation of the nucleus into one or more
parts which possessed no reproductive independence.

The process of nuclear amitosis consisted in a unilateral or
bilateral constriction, manifested by a narrowing of the nucleus
in the region of its equator, and a streaming of its contents
toward the poles, with final separation of the two nuclear por-
tions. This phenomenon seemed to be associated with the action
of the mitochondria and centrosphere upon an elongated nucleus.
There was no amphiaster or spireme formation and no centre-
some fission. Division of the nucleolus was not an essential.
Repetition of this process leads to the formation of a giant cell.

Not all nuclei which show elongation and constriction divide
by direct fission, but many return to their usual rounded or oval
form. When, however, the constriction has passed a critical
point the division goes on to completion, and this final stage is
rapid.

Cells containing nuclei in process of, or the result of, amitosis
divide by mitosis. Mitosis in binucleate cells, which are the
product of nuclear amitosis, is characterized by the simultaneous
appearance in the nuclear parts of a spireme, from which a single
equatorial plate of chromosomes is formed. Furthermore, bi-
nucleate cells divide as frequently by mitosis as do mononucleate
cells, and this was the only form of division found to occur in
them.

Since the parts of an amitotically divided nucleus do not
become separated as reproductive units but divide only by mi-
tosis, in which the chromatin in the parts is recombined, there is
nothing in nuclear amitosis opposed to the chromosome hy-
pothesis.

The type of amitosis in which the nucleus is split by the growth

460 C. C. MACKLIN.

through it from one side to the other of a membrane was not
found. Nuclear figures simulating this proved to be caused by
the close apposition of separate nuclear sacs, or by nucleoli,
mitochondria or folds of the nuclear membrane.

The dyes, janus green and gentian violet, were toxic and their
presence in the cell was incompatible with its continued life.
They were, however, of service in quickly studying structural
details which were not discernible in the living state.

Nuclear fragmentation, which differs in many ways from nu-
clear amitosis, is a pathological condition, and occurs in degener-
ating cultures.

It is believed that the facts brought to light through the tissue
culture method may be applied to the interpretation of the
phenomena of normally developing cells.

LITERATURE CITED.

Child, C. M.

'07 Studies on the Relation between Amitosis and Mitosis. IV. Nuclear

Division in the Somatic Structures of the Proglottids of Moniezia. V.

General Discussion and Conclusions Concerning Amitosis and Mitosis in

Moniezia. BIOL. BULL., Woods Holl, XIII., 165-184.
'ii The Method of Cell-Division in Moniezia. BIOL. BULL., Woods Hole,

XXI., 280-296.

Churchman, J. W., and D. G. Russell.

'14 The Effect of Gentian Violet on Protozoa and on Growing Adult Tissue.
Proc. Soc. Exper. Biol. and Med., N. Y., XL, 120-124.

Isaacs, R.

'15 A Mechanical Device to Simplify Drawing with the Microscope. Ana-
tomical Record, Phila., Vol. 9, 711-713.

Karpow, W.

'04 (Untersuchungen iiber direkte Zellteilung.) Inaug.-Diss. Moskau, 1904.
Refer, in Jahresber. ii. d. Fortschr. d. Anat. (Schwalbe), Jena, n.F.,X., 42.

Lambert, R. A.

'i2a Variations in the Character of Growth in Tissue Cultures. Anat. Rec.

Phila., VI., 91-108.
'i2b The Production of Foreign Body Giant Cells in Vitro. Jour. Exper.

Med., Lancaster, Pa., XV., 510-515.

Lewis, M. R., and W. H. Lewis.

"15 Mitochondria (and Other Cytoplasmic Structures) in Tissue Cultures.

Amer. Jour. Anat., Phila., XVII., 339-401.
Maximow, A.

'08 Ueber Amitose in den embryonalen Geweben bei Saugetieren. Anat.
Anz., Jena, XXXIII., 89-98.

AMITOSIS IN CELLS GROWING IN VITRO. 461

Meves, Fr.

'91 Ueber amitotische Kernteilung in den Spermatogonien des Salamanders
und Verhalten der Attraktionssphare bei derselben. Anat. Anz., Jena,
VI., 626-639.

Russell, D. G.

'14 The Effect of Gentian Violet on Protozoa and on Tissues Growing in vitro,
with Especial Reference to the Nucleus. Jour. Exper. Med., Lancaster,
Pa., XX., 545-553-

462 C. C. MACKLIN.

EXPLANATION OF PLATES.

PLATE I.

FIGS, i TO 12. Successive stages covering a period of 2% hours in the life history
of a connective tissue cell from a 57-hour culture of 5-day chick heart, growing in
Locke solution (0.5 per cent, dextrose with extract of chick embryo). The nucleus
finally divides by direct fission. Small circles represent fat globules and short
threads mitochondria, c. Fig. i, points to the centrosphere. Free-hand drawings
X about 900. (Description in text.)

BIOLOGICAL BULLETIN, VOL xxx.

PLATE I.

8.4-5 p.m. *

S.i 5 p.m.

9.30p.m. j

9.4-5 p.m f

10 p.m.

10.15 pm.

10.30p.m. g

n.15 p.m. * i

C. C. MACKLIN.

464 C. C. MACKLIN.

PLATE II.

FIGS. 13 TO 23. Graphic record of the process of mitosis in a living binucleate
cell, from a ig-hour culture of 7-day chick heart, growing in Locke solution (i per
cent, dextrose with extract of chick embryo), c, Fig. 13, points to the centrosphere.
Camera lucida drawings. X 1,333- (Description in text.)

BIOLOGICAL BULLETIN VOL. XXX.

PLATE II.

C. C MACKLIN.

466 C. C. MACKLIN.

PLATE III.

FIG. 24. Direct nuclear division in a connective-tissue cell, final stage. Cen-
trosphere between the nuclear parts. Across the slender filament joining these is a
strand of mitochondria. Nucleolus has not divided. Camera lucida drawing
from Preparation No. 2. from 5-day culture of y-day chick heart in Locke solution
(i per cent, dextrose); osmic acid vapor and iron hematoxylin. X 915.

FIG. 25. Spireme in a bipartite nucleus. Prophase of mitosis. Nuclear
membrane and nucleoli are disappearing. Camera lucida drawing from Prep. No.
14, 9-1-15 (Lewis collection). Heart from 6-day chick grown in Locke (0.5 per
cent, dextrose) with a little yolk; fixed on third day of growth in Zenker; stained
with iron hematoxylin (this culture was originally stained with Mallory's connective
tissue stain). On account of the method of fixation the cytoplasmic details are not
shown. X 915.

FIG. 26. Nuclei showing fragmentation. Camera lucida drawings from Prep.
No. 23, 12-1-15 (Lewis collection). 5-day chick stomach in Locke (0.5 per cent,
dextrose). Zenker; Mallory connective tissue stain. Culture grown for 6 days in
the same media. X 1,012.

FIG. 27. Nuclei showing fragmentation. Camera lucida drawings from Prep.
No. 23, 24-11-14 (Lewis collection). 6-day chick stomach in Locke (i per cent,
dextrose) to which ethyl alcohol had been added to make approximately i per cent.
3-day culture. Osmic acid vapor and iron hematoxylin. X 1,500.

8IOLOGICAL BULLETIN, VOL. XXX.

PLATE III.

t

25

26

C C. MACKLIN.

MBL WHOI LIBRARY

UH 17K2 -

10 7 j/