This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project to make the world's books discoverable online.
It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover.
Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the publisher to a library and finally to you.
Usage guidelines
Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to prevent abuse by commercial parties, including placing technical restrictions on automated querying.
We also ask that you:
+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for personal, non-commercial purposes.
+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the use of public domain materials for these purposes and may be able to help.
+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find additional materials through Google Book Search. Please do not remove it.
+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner anywhere in the world. Copyright infringement liability can be quite severe.
About Google Book Search
Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web
atjhttp : //books . qooqle . com/
•-••
Ti
•"• f^fr
nri . mI
./
^-^^y**
s«'-. '
50k
C\53t>r
S € r. 3 2.
v^X ' I
PROCEEDINGS
California Academy of Sciences
THIRD SERIES
ZOOLOO-Y
Vol. I 1897-1899
SAN FRANCISCO 1900
H
COMMITTEE OF PUBLICATION
Charles H. Gilbert, Chairman \ William E. Ritter, G. P. Rixpord.
EDITORS OF ZOOLOGICAL PUBLICATIONS
David S. Jordan, Charles H. Gilbert,
William E. Ritter.
LIBRARY OF THE LELAND SlAM-'n-p JR. UMVFRSITY.
CONTENTS OF VOLUME I.
Title-Page i
Committee of Publication ii
Contents iii
List of Plates iv
Dates of Publication of Separate Articles iv
Index 419
Errata 426
t
Bancroft, Frank Watts.— The Anatomy of Chelyosoma productum
Stimpson. (No. 8, Plate XVIII) 309
Banks, Nathan.— Arachnida from Baja California and Other Parts of
Mexico. (No. 7, Plates XIII-XVII) 205
Calvert, Philip P.— Odonata. from Tepic, Mexico, with Supplemen- tary Notes on those of Baja California. (No. 12, Plate XXV).. . 371
Eisen, Gustav.— Plasmocytes; The Survival of the Centrosomes and Archoplasm of^the Nucleated Erythrocytes, as Free and Inde- pendent Elements in the Blood 'of Batrachoseps attenuates Esch. (No. 1, Plates I, II) 3
Johnson, Herbert P.— A Preliminary Account of the Marine Annelids of the Pacific Coast, with Descriptions of New Species. (No. 5, Plates V-X) 153
Jordan, David Starr.— Description of a Species of Fish (Mitsukurina owstoni) from Japan, the Type of a Distinct Family of Lam- noid Sharks. (No. 6, Plates XI, XII) 199
Miller, Walter.— Scientific Names of Latin and Greek Derivation
(No. 3) "5
Montgomery, Thomas H.— The Gordiacea of Certain American Col- lections, with Particular Reference to the North American • Fauna. (No. 9, Plates XIX, XX) 333
Ritter, William E.— Diemyctylus torosus Esch. The Life-History
and Habits of the Pacific Coast Newt. (No. 2, Plate III) 73
St arks, Edwin Chapin.— The Osteological Characters of the Genus
Sebastolobus. (No. 11, Plates XXII-XXIV) 361
Torrey, Harry Beal.— Observations on Monogenesis in Metridium.
(No. 10, Plate XXI) 345
Wheeler, William Morton.— A Genus of Maritime Dolichopodidae
New to America. (No. 4, Plate IV) 145
COMMITTEE OF PUBLICATION
Charles H. Gilbert, Chairman, William E. Ritter, G. P. Rixpord.
EDITORS OF ZOOLOGICAL PUBLICATIONS
David S. Jordan, Charles H. Gilbert,
William E. Ritter.
LIBRARY OF THE LELAND S7AA'r'n:-p JR. UNIVERSITY.
CONTENTS OF VOLUME I.
Title-Page i
Committee of Publication ii
Contents iii
List of Plates iv
Dates of Publication of Separate Articles iv
Index 419
Errata 426
t
Bancroft, Frank Watts.— The Anatomy of Chelyosoma productum
Stimpson. (No. 8, Plate XVIII) 309
Banks, Nathan.— Arachnida from Baja California and Other Parts of
Mexico. (No. 7, Plates XIII-XVII) 205
Calvert, Philip P.— Odonata. from Tepic, Mexico, with Supplemen- tary Notes on those of Baja California. (No. 12, Plate XXV).. . 371
Eisen, Gustav.— Plasmocytes; The Survival of the Centrosomes and Archoplasm of*the Nucleated Erythrocytes, as Free and Inde- pendent Elements in the Blood of Batrachoseps attenuatus Esch. (No. 1, Plates I, II) 3
Johnson, Herbert P.— A Preliminary Account of the Marine Annelids of the Pacific Coast, with Descriptions of New Species. (No. 5, Plates V-X) 153
Jordan, David Starr.— Description of a Species of Fish (Mitsukurina owstoni) from Japan, the Type of a Distinct Family of Lam- noid Sharks. (No. 6, Plates XI, XII) 199
Miller, Walter.— Scientific Names of Latin and Greek Derivation
(No. 3) 115
Montgomery, Thomas H.— The Gordiacea of Certain American Col- lections, with Particular Reference to the North American • Fauna. (No. 9, Plates XIX, XX) 333
Ritter, William E.— Diemyctylus torosus Esch. The Life-History
and Habits of the Pacific Coast Newt. (No. 2, Plate III) 73
Starks, Edwin Chapin.— The Osteological Characters of the Genus
Sebastolobus. (No. 11, Plates XXII-XXI V) 361
Torrey, Harry Beal.— Observations on Monogenesis in Metridium.
(No. 10, Plate XXI) 345
Wheeler, William Morton.— A Genus of Maritime Dolichopodidae
New to America. (No. 4, Plate IV) 145
COMMITTEE OF PUBLICATION
Charles H. Gilbert, Chairman^ William E. Ritter, G. P. Rixford.
EDITORS OF ZOOLOGICAL PUBLICATIONS
David S. Jordan, Charles H. Gilbert,
William E. Ritter.
LIBRARY OF THE LELAND STANFORD JR. UNIVERSITY.
CONTENTS OF VOLUME I.
Title-Page i
Committee of Publication ii
Contents iii
List of Plates iv
Dates of Publication of Separate Articles iv
Index 419
Errata 426
t
Bancroft, Frank Watts.— The Anatomy of Chelyosoma productum
Stimpson. (No. 8, Plate XVIII) 309
Banks, Nathan.— Arachnida from Baja California and Other Parts of
Mexico. (No. 7, Plates XIII-XVII) 205
Calvert, Philip P.— Odonata. from Tepic, Mexico, with Supplemen- tary Notes on those of Baja California. (No. 12, Plate XXV).. . 371
Eisen, Gustav.— Plasmocytes; The Survival of the Centrosomes and Archoplasm ofthe Nucleated Erythrocytes, as Free and Inde- pendent Elements in the Blood of Batrachoseps attenuatus Esch. (No. 1, Plates I, II) 3
Johnson, Herbert P.— A Preliminary Account of the Marine Annelids of the Pacific Coast, with Descriptions of New Species. (No. 5, Plates V-X) 153
Jordan, David Starr.— Description of a Species of Fish (Mitsukurina owstoni) from Japan, the Type of a Distinct Family of Lam- noidSharks. (No. 6, Plates XI, XII) 199
Miller, Walter.— Scientific Names of Latin and Greek Derivation
(No. 3) "5
Montgomery, Thomas H.— The Gordiacea of Certain American Col- lections, with Particular Reference to the North American • Fauna. (No. 9, Plates XIX, XX) 333
Ritter, William E.— Diemyctylus torosus Esch. The Life-History
and Habits of the Pacific Coast Newt (No. 2, Plate III) 73
St arks, Edwin Chapin.— The Osteological Characters of the Genus
Sebastolobus. (No. 11, Plates XXII-XXIV) 361
Torrey, Harry Beal.— Observations on Monogenesis in Metridium.
(No. 10, Plate XXI) 345
Wheeler, William Morton.— A Genus of Maritime Dolichopodidae
New to America. (No. 4, Plate IV) 145
LIST OF PLATES.
I— II- — Plasmocytes: elements in the blood of Batrachoseps attenu-
atus Esch. III.— Illustrating life-history of Diemyctylus torosus Esch. IV.— Illustrations of Aphrosylus. V-X.— Illustrations of Pacific Coast Marine Annelids. XI-XII. — Mitsukurina owstoni, gen. et sp. nov. XI I I-XVI I.— Illustrations of Arachnida from Baja California.
XVII I. — Illustrations of anatomy of Chelyosoma productutn Stimpson. XIX-XX.— Illustrations of North American Gordiacea. XXL— Illustrations of Metridium fimbriatum Verrill. XXII-XXIV.— Osteological characters of Sebastolobus.
XXV.— Illustrations of Odonata from Tepic, Mexico.
DATES OF PUBLICATION OF SEPARATE ARTICLES.
No. i, April i, 1897; No. 5, Dec. 11, 1897; No. 9, Oct. 12, 1898;
No. 2, Jan. 18, 1897; No. 6, Jan. 18, 1898; No. 10, Oct. 25, 1898;
No. 3, April 10, 1897; No. 7, May 28, 1898; No. 11, Dec. 22, 1898;
No. 4, July 10, 1897; No. 8, Oct. 5, 1898; No. 12, May 22, 1899.
PROCEEDINGS
OP TBB
CALIFORNIA ACADEMY OF SCIENCES
Third Series. Zoology. Vol. I, No. i.
PLASMOCYTES;
The Survival op the Centrosomes and Archoplasm op the
Nucleated Erythrocytes, as Free and Independent
Elements in the Blood op Batrachoseps
attenuatus esch.
BY
Gustav Eisen, Ph. D.,
Curator m the California Acmdemy of Scumcos.
WITH TWO PLATES.
Issued April /, 1897.
SAN FRANCISCO:
Published by the Academy.
1897.
The cost of publication of the present paper has been generously contributed by two members of the Academy of Sciences, the late J. Z. Davis, Director of the Museum of the Academy, and W. S. Keyes, one of the Trustees of the Academy.
PLASMOCYTES;1
Thb Survival of the Cbntrosomes and Archoplasm of thb Nucle- ated Erythrocytes, as Free and Independent Elements in the Blood of Batrachoseps attenuatus Esch.
BY GUSTAV RISEN, PH. D., Curmivr m ike Califtmim Academy ef Sciences.
CONTENTS.
Paob.
Plates I and II.
I. Introductory 4
II. Methods of Investigation 5
General Remarks 5
III. Staining 6
Toluidine 7
Eosin-methyl blue °0" 7
Iron-hctmatoxylin 7
Ehrlich-Biondi and other stains 8
IV. The Blood Elements 8
General Remarks. 8
Non-nucleated Erythrocytes 9
Nucleated Erythrocytes 9
The Fusiform Corpuscles 10
Leucocytes with Polymorphous Nucleus 10
Smaller Mononucleary Leucocytes 12
Leucocytes with Eosinophile Granulation 12
Leucocytes of Various Kinds in Dissolution 12
Plasmocytes 13
V. The Fusiform Elements 14
VI. The Plasmocytoblasts 16
VII. Cytosomb 20
The Plasmosphere 20
Hyalosphere 22
The Granosphere 22
VIII. Archosome 25
The Centrosphere 25
Somosphere and Centrosomes 26
IX. Different Kinds of Protoplasm 27
X. Development of the Plasmocytoblast into Plasmocytes. . 28
XI. The Plasmocytes 29
General Remarks 29
1 Presented for publication May rj, 1896. March 31, 1897.
[3l
4 CALIFORNIA ACADEMY OF SCIENCES. . [*d Ser.,
Paob.
Homology of the Plasmocytes and Plasmocytoblasts 30
Different Kinds 0/ Plasmocytes 31
The Spheres of the Plasmocytes 32
Food Supply in the Somosphere 35
Unequal Staining of the Archosomal Spheres 36
Absence of a Cell Membrane 37
Absence of Nucleus 37
Degeneration of the Plasmocytes 38
Abnormal Plasmocytes 38
Amoeboid Movements 39
Growth and Phagocytosis 41
Duplicity of the Plasmocytoblasts 42
The Ultimate Fate of the Plasmocyte 43
Adhesive Nature of the Cytoplasm 44
XII. The Independence of the Archosome 45
Identification of the Spheres 51
Plasmocyte and Leucocyte 57
XIII. Summary 58
Measurements op Corpuscles 63
Bibliography 64
Explanation of the Figures 66
Stains 72
I. Introductory.
The elements, or corpuscles, of the blood of Batrachoseps attenuates are highly interesting, differing as they do in several important points from the corresponding elements of the blood of all other batrachians of which I have any knowledge. Batrachoseps attenuatus is one of the most common species of the order in this part of California, and material for study may be had at any time of the year and almost anywhere. Not only do the red cells of the blood vary enormously in size and shape, but they differ also from the blood of other batrachians in the fact that very few of them are nucleated. But the most interesting feature of the blood is the presence of a new corpuscle, which I have termed plasmocyte. In this paper I expect to prove that these plasmocytes are the remnants of the extra-nuclear part of fusiform corpuscles; that they consist of the archosome — archoplasm and centrosomes — which has sur- vived, while the nucleus has been destroyed; that this archosome has surrounded itself with various envelopes
Zool.-Vol. I.] EISEN— PLASMOCYTES. 5
of cytoplasm; and that the plasmocytes have thus become free and independent elements of the blood. So far I have only demonstrated the presence of the plasmocytes in Batrachoseps, Phrynosoma, Diemyctylus, and human blood, and it is not improbable that with proper methods they will be found in the blood of other animals. This paper will, however, treat only of the blood of Batrachoseps; but I may be permitted to state that, as regards the human blood, the plasmocytes are so small that without first hav- ing studied the larger ones in the Batrachoseps blood I could never have recognized their structure. In the human blood they have been confounded with blood plates, the structure having been obscured by improper methods of investigation.
Some may, after a perusal of my plates, insist that the blood of Batrachoseps is so called pathological blood, on account of the abnormal form, variation in size, absenccof nuclei, etc.; but I will here hasten to state that this is not the case. Batrachoseps possesses the same form of blood whether young or old, whether examined in the spring, in fall, or in winter. In fact, the blood described here is absolutely normal.
II. Methods of Investigation.
General Remarks. — The delicate plasmocytes can only be studied on cover glass preparations, and even for the other corpuscles this method of investigation was found the most exact and satisfactory. Observation on moist stage and in 0.6 salt solution was also found useful and instructive. The methods generally used for preparing cover glasses with blood of higher animals are useless for batrachian blood. The large corpuscles would roll up and twist, and become so distorted that no minute details could be made out. I have obtained the best results as follows : the covers must be ab- solutely chemically clean and polished. For spreading the blood I use a pair of small forceps with curved prongs of ex- actly the same size and shape. The animal is etherized
Proc. Cal. Acad. Sci., 3D Ser., Zool., Vol. I. October 20, 1896.
6 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
and the head is clipped off just above the heart. Blood is then caught by the curved points of the forceps, which must be closed. The prongs are then quickly passed over the cover glass, always in the same direction and never twice over the same place, as the blood cells would then be disturbed. A zigzag movement over the glass is best when it is desirable to cover the whole surface. The forceps must not be lifted at the margin, but simply be pushed back; the quicker this is done the better the blood will be spread. The blood coagulates with great rapidity, and even if the blood supply would hold out it is hardly possible to procure more than two or three good cover glass preparations from the same animal. With some practice it is not difficult to so spread the blood that the film is only one corpuscle thick, and so that the individual corpuscles are not distorted. Great haste is necessary as a second's delay may result in failure. Furthermore, the corpuscles should be so far apart that the small plasmocytes are entirely free, as, if massed together, they cannot be properly studied.
The cover glasses are then at once placed with the film downwards on clean, dusted, blotting paper, and covered with a bell glass. This is absolutely necessary, as even under well closed bells some dust will penetrate and settle on the upper side of the glass. Afterwards these foreign substances may be mistaken for centrosomes, experience having shown me that through some cause or other these specks of dust frequently settle in just those places where a centrosome is to be expected.
After twelve hours or more of air drying, the cover glass is dropped into a shallow dish containing absolute alcohol, and allowed to remain two hours or longer. It may then be taken out and dried between blotting papers, after which it is ready for staining.
Two points are important to observe: the blotting paper must be smooth and not corrugated; and after the glass has finally become dry it must be brushed off with a fine, clean, soft brush, in order to remove all the dust, which settles with astonishing rapidity, even in a few seconds.
ZOOL.-VOL. I.] EISEN—PLASMOCYTES.
III. Staining.
I have tried a great variety of stains and found only a very few of them useful, while some, like hsematoxylin, proved even injurious. I will mention the stains in order of useful- ness as regards bringing out the details of the plasmocytes. Toluidine. — Watery solution, not quite concentrated. I found this the most useful stain, since it differentiated the various spheres and zones of the plasmocyte with great pre- cision, and without fail. The glass was made to swim in the solution for about three minutes, then washed off with distilled water and dried between pieces of blotting paper. It was then brushed off with a camel's hair brush, and mounted in gum-thus-xylol. The toluidine stains the grano- sphere violet, the other spheres blue, excepting the hyalo- sphere which remains unstained. The centrosomes stand out generally quite black. I tried a number of brands of thionin, but none gave satisfactory results as compared with the toluidine.
Eosin-Methyl Blue "O." — Watery solution of eosin three minutes, washing with water until the stain has receded from the blood serum, leaving only the cells stained. Then watery methyl blue "O" for about ten seconds, washing with water, and mounting as before. This method gave now and then very excellent results, as the eosin has a special affinity for the centrosphere and the hyalosphere, while it leaves the granosphere unstained, the latter being stained by the blue (fig. 49). But this method was never sure, and fre- quently quite unreliable, though when it succeeded it gave results not obtainable in any other way. The eosin demon- strated that the centrosphere is entirely distinct from the granosphere on one side, and from the somosphere on the other. I found methyl blue "O" more satisfactory than any other brand or variety of this stain. It stains quicker and more intensely.
Iron-Hcematoxylin. — Another staining method which I have found of interest and value is the iron-alum-hsematoxylin stain as perfected by M. Heidenhain. The method is the same
8 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
as the one used with sections; the cover glasses are first floated in one liquid, then in the other, and finally washed and mounted in the usual way. By this method the centro- somes in the plasmocytes will stain, but the cytoplasmic sphere will remain unstained. Valuable only as showing the centrosomes.
Ehrlich-Biondi and Others. — Useful for all elements ex- cept for plasmocytes. The latter are diffusely stained, and the respective spheres are seldom differentiated. The effect is to some degree the reverse of toluidine. The hyalosphere is never left clear and is seldom differentiated from the plasmosphere ; the granosphere is frequently left lighter than the centrosphere ; the other blood elements are, however, exquisitely stained. In order to attain the best results the mixture should be acidified with oxalic acid and water, and even the cover glass should finally be washed off with a weak solution of the same. In this way the centrosomal spheres in the leucocytes are brought out strongly and chromatically.
Among other stains I found metanil yellow useful in staining the plasmocytoblast while yet in the erythrocyte. It will now and then, not always, bring out the outlines sharply, but will only give a few details. The method is to first stain for several minutes with an aqueous solution of metanil yellow, wash with water, and double stain with thionin. A second staining with metanil is sometimes nec- essary. By this method I have demonstrated the existence of the plasmocytoblast, as well as the two outer layers of cytoplasm, in perfect, nucleated erythrocytes.
IV. The Blood Elements.
General Remarks. — The respective elements in the blood of Batrachoseps are in short as follows : Nucleated erythro- cytes, non-nucleated erythrocytes, polymorphous leucocytes, lymphocytes with solid round nucleus, fusiform corpuscles, degenerating leucocytes, and finally plasmocytes; the latter now described for the first time. Of the leucocytes there
Zool.-Vol. I.] E1SEN—PLASM0CYTES. 9
are various kinds, the ordinary ones, eosinophile cells, and other strongly granulated cells which do not stain with any of the stains I have so far tried. While it is the fusiform elements and the plasmocytes which will principally occupy our attention, a short description of all the elements is nec- essary. The measurements given later have been calcu- lated by Mr. George Otis Mitchell, whose careful meas- urements of the human blood cells are well known and accepted as standard.
Non-nucleated Erythrocytes. — These constitute by far the great majority of the red blood cells. The proportion be- tween the non-nucleated and the nucleated red blood cells is probably as 99 to 1 at any time, though I have not made a sufficient number of countings to fully ascertain the fact. In some Batrachoseps, especially early in the spring of the year, the nucleated red cells are so scarce that on a well spread cover glass I have found but a single cor- puscle. At other times they are much more numerous, so that in a field viewed under Zeiss A A we may count from 100 to 200 nucleated red blood cells, all the others being non-nucleated. A striking characteristic of all the red blood cells, nucleated and non-nucleated, is their great variation in size. Some are smaller than the human red blood cell, while others surpass it with a diameter seven times as great in every direction; and this variation in size is not confined alone to the non-nucleated red blood cells, but also to the nucleated ones. The smallest nucleated cells besides the nucleus consist of only a very narrow rim of cytoplasm and haemoglobin. In the non-nucleated red blood cells I have never observed any structure that I could at all iden- tify as cytoplasm and centrosomes.
Figs. 1, 2, 3, 4, 5, and 6 represent various non-nucleated cells. There are also numerous cells of the same size and shape as those represented in figs. 7 to 11. The form of the red cells varies considerably, hardly any two being exactly alike; some are round, others oval, while many are oblong and biconcave (fig. 5).
Nucleated Erythrocytes. — To the description already given
IO CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
I can add only a few words as regards the nucleus. The nucleus varies in size considerably, but not so much as the cytoplasmic part. The shape of the nucleus varies more than itssize; thus many nuclei are round, while others are oblong. The former are represented by figs. 10 and 20, the latter by 7, 8, 9, and 11.
I have already stated that with metanil yellow and thionin part of the cytoplasm can be stained enough to show ex- actly the same general structure as the fusiform corpuscles, of which more further on. Of the details of the nucleus I have made no particular study, but I find that it possesses the same polarity as that described by Heidenhain. (See diagrams given in his "Kern und Protoplasma," Taf. ix, fig. 8.)
The Fusiform Corpuscles. — A more detailed description of these will be given further on. Here I will only state that they occur in large numbers and are more numerous than even the nucleated red blood cells. They are found in all stages of degeneration and disintegration.
Leucocytes with Polymorphous Nucleus. — These are found in varying numbers. In some specimens they are much more numerous than in others. In figs. 14 to 19 I have shown some types, each displaying a pronounced microcentrum which in a general way resembles those described by Heid- enhain, only the microcentrum is surrounded by a small, deep-staining, starlike sphere, which sometimes separates the centrosomes and the centrospheres from each other. A somosphere I have not with certainty observed. Stained with Ehrlich-Biondi, the fine connections between the lobes of the nucleus do not become visible. These fine connec- tions are, however, brought out with toluidine, showing that the various parts of the nucleus are in reality connected. I have never seen entirely isolated parts. These fine connec- tions frequently show one or two minute triangular nodes of very characteristic form, but I have not given them any par- ticular study and wish only to call attention to them. These leucocytes vary but little in size. Two of the figures, igb and igcy represent polymorphous leucocytes stained with
Zool.— Vol. I.] EISEN—PLASMOCYTES. II
toluidine blue. I wish especially to call attention to the star- like pink-colored zone in the center, which appears to me identical with the granosphere. It is brought out only after several hours immersion in toluidine. The centrosomes are rarely stained by the toluidine, and the archoplasmic spheres are much less distinct than when stained with Ehrlich-Biondi. This granosphere in the leucocytes is exceedingly delicate but nevertheless distinct. I have never seen any rays reach to the cell wall; they always stop in the cytoplasm. Rays are frequently seen in the leucocytes extending from the archoplasmic region toward the periphery of the cell, but they consist of two distinct substances: granosphero- plasm, staining pink, and other cytoplasm, staining blue (fig. 19^). The outer part of the ray contains cytoplasmic microsomes, while the middle part of the same ray consists of microsomes of the granospheres. The Ehrlich-Biondi stain is thus misleading, as it does not differentiate the granosphere from the cytoplasm; or, if a slight differ- entiation is made, it appears as though the zonal rays surrounding the microcentrum are all, and throughout, of the same quality, which they are not. For the nuclei of the leucocytes the toluidine stain is entirely unsuitable, as it does not differentiate the various nuclear granulations but stains them all alike.
At times the granosphere, instead of being starlike and consisting of very minute grains of indefinable form, appears to have fallen to pieces, so to say, consisting of a smaller number of larger globular granules, irregularly, scattered about, and not close enough together to form a solid sphere, or zone. The microcentrum, which is generally round in outline, is at other times starlike, irregular, or even broken up into several smaller areas, one adjoining the other as in fig. i8d, which is a toluidine stain. No distinct centrosomes are visible in the leucocyte, as indeed is generally the case with toluidine stains; but there are some shaded portions on the archoplasm which must be considered as corresponding to the somosphere, and which may or may not contain un- stained centrosomes.
12 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
As regards the granulation of the cytoplasm, we can nearly always, in successful preparations, distinguish three distinct kinds besides that of the granosphere. The achromatic granule is strongly refractive and pure white, generally but not always globular, or aggregated into larger globules, around which the two other granulations are sparsely scattered as irregularly formed grains of various sizes. Some of them stain deep blue, while others have a reddish tint, fainter and more bluish than the grains of the granosphere. It appears as if the achromatic granula were of a much greater consistency than the two chromatic ones, as they assume the shape of regularly rounded granules, while the colored grains surrounding them appear compressed or stretched out, accommodating themselves to the greater consistency of the achromatic granule. This refers to the toluidine stain preparations.
Smaller Mononucleary Leucocytes. — These occur in vary- ing numbers, according to the state of the blood. The nu- cleus is very large, round, and compact, while the cytoplasmic part is very small, being reduced to a narrow margin. Frequently this cytoplasmic part stains as if it contained haemoglobin, and this makes me doubtful as to whether these bodies are really leucocytes. Fig. 20 represents one of them. The cytoplasm generally stains much lighter than is figured, but now and then we find a corpuscle intensely stained, as is this one.
Leucocytes with Eosinofhile Granulation. — These are al- ways scarce and vary greatly in size. They show cytoplasm frequently rayed as that of the true leucocytes, but the stain- ing of the parts is reversed. The granulation stains deeply, while the rays, probably corresponding to the grano- sphere, remain pale as in fig. 13. The centrosomes stand out plainly, but the inner spheres do not differentiate. Figs. 19 and 20 represent two of these cells of different sizes. There are some that are yet larger, and these stain less. The smaller the cell the darker it stains.
Leucocytesof Various Kinds in Dissolution. — The cytoplasm and the nucleus appear to disintegrate together. In many
ZOOL.-VOL. I.] EISEX— PLASMOCYTES. 1 3
instances I have seen the inner spheres around the centro- somes stand out sharply, while the nucleus and other parts of the cytoplasm were in the last stages of disintegration ; but I have never seen such a separation of the centrosomes and centrosomal spheres as takes place in the fusiform ele- ments, and, judging from my observations, the microcentrum of the leucocyte does not survive. It is undoubtedly less differentiated and organized than that of the erythrocyte.
Plasmocytes. — I apply this name to a hitherto undescribed element in the blood, first observed by me in the blood of Batrachoseps, and later also in some other batrachians and reptiles, as well as in that of man. These new elements are much smaller than the average erythrocytes, if, indeed, an average can be struck for a corpuscle with such extreme and irregular variations as the erythrocytes in the blood of Batrachoseps. The plasmocytes are only slightly larger than the smallest erythrocytes of the Batrachoseps blood, and similarly, only a little larger than the red blood cells of the human blood ; but even the plasmocytes vary consider- ably, and some are found which are smaller than the human red blood cells. They are generally chiefly characterized by the absence of a cell membrane, and out of about a thous- and plasmocytes only six showed a rounded outline and what I considered a cell membrane. The general form is that of a round or oblong star-shaped body, with more or less frayed or amoeboid projections of the outer layer, while the inte- rior is arranged in varying concentric zones. They occur in large numbers, are more numerous than the fusiform ele- ments, and much more numerous than the nucleated red cells. The object of this paper is to establish the identity of these plasmocytes ; to trace their origin ; to follow their development; and to demonstrate and prove that they are composed of the centrosomes and archoplasm (with part of the cytoplasm) of the nucleated erythrocytes, having dis- engaged themselves from the degenerating and dissolving parts of the fusiform corpuscles, surviving in the blood serum as free and independent elements capable of growth through assimilation of food, and taking their place as blood
14 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
elements, equal in importance to the erythrocytes and leuco- cytes.
V. The Fusiform Elements.
I believe that A. B. MacCallum was the first one to deter- mine satisfactorily that the fusiform elements, or corpuscles, in the blood of batrachians (Necturus) derive their origin di- rectly from the red blood corpuscles; that they constitute, in fact, what remains of the nucleated erythrocyte after the cell wall, haemoglobin, and possibly part of the cytoplasm, have been destroyed or separated. As regards the blood of Ba- trachoseps, this origin of the fusiform corpuscle is so appar- ent that few if any comments are necessary. On my slides I have frequently found nucleated cells that have been injured by pressure, or in which, for some other cause, the cyto- plasmic membrane had been ruptured, thus allowing all of the haemoglobin to escape. Such corpuscles showed the faintly staining cell membrane, with here and there tiny specks of cytoplasm around the edges ; but the nucleus with surrounding cytoplasm was always stained, and in other ways exactly resembled the free fusiform corpuscles. In the yet enclosed fusiform corpuscles I frequently found the same spheres, and the same structure generally, as is seen in the free fusiform elements, with the exception that the nucleus was properly preserved, while that of the latter corpuscle was always in decay; but even in perfect and nucleated erythrocytes, stained with metanil yellow and thionin, I found now and then the cytoplasmic layers brought out in exactly the same way as in the fusiform corpuscles, which leaves no doubt as to the correctness of MacCallum's ob- servations. A further proof is that if a drop of Batrachoseps blood be mixed with a drop of 0.6 salt solution and observed in a moist chamber, we will soon find that the erythrocyte loses its haemoglobin, the cell membrane collapses, and the nucleus with adhering cytoplasm is set free. These re- mains of the erythrocytes closely resemble the fusiform corpuscles, or at least some of them, as it is evident that
Zool.-Vol. I.] EISEX—PLASMOCYTES. 1 5
among the latter we meet with all stages of development and dissolution; development as regards the cytoplasm, dis- solution as regards the nucleus.
A fusiform corpuscle of the blood of Batrachoseps, if stained with toluidine, pure and simple, presents the following structure : A large central nucleus of rather irregular, ob- long form, the two longer sides being always convex, while the two short sides are generally flat, or even concave, each one furnished with a dell. The nucleus itself requires little description, as it is always in a state of rapid dissolution. We find nuclei in all the various stages, some showing a distinct network with fairly well defined chromosomes, others again with only a diffuse mass of ill defined granules. In all the figures given I have, therefore, in no way en- deavored to reproduce a copy of the nuclear structure, but only to show its general form and appearance, the minute details being entirely unimportant.
What attracts us the most in the fusiform corpuscle is the cytoplasmic element which adheres to the nucleus, princi- pally at one, but frequently at both of the poles, some- times, also, as a very thin coating on its long sides. Not only is this cytoplasmic coat thicker at the poles than on the long sides, but the structure of the polar parts is entirely different from that which adheres to the long sides. These sides are covered by a very thin layer of faintly staining cytoplasm. In some corpuscles this layer can be observed without difficulty all around the nucleus, while in others it becomes, at the middle of the long sides, so thin that it is hardly to be observed, and in some instances prob- ably it is entirely absent. The latter appears to be the rule rather than the exception. More rarely this layer is suffi- ciently thick to allow us to define it in two separate layers of about equal thickness, as, for instance, seen in figs. 21 and 27; but at the poles, or at least at one of the poles, this cytoplasm is greatly increased in size, showing that it consists of a number of distinctly staining and differentiating zones or spheres. The toluidine has a marked and distinct affinity for this part of the corpuscle, and stains it in a way
l6 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
that no other stain does. This differentiation is nearly always the same, and we may readily recognize each zone by the coloring alone. Thus it will be seen that while the long sides of the nucleus are covered with two of these cytoplasmic zones or spheres, the poles contain six distinct zones, the two outer being the only ones which continue all around the corpuscle. The polar accumulations must, therefore, be considered as something entirely separate from the bal- ance of the cytoplasm ; they, in fact, give rise to the plas- mocytes, and may, therefore, appropriately be called plas- mocytoblasts, or for the sake of brevity, plasmoblasts.
VI. The Plasmocytoblasts.
In a general way it may be said that plasmocytoblasts are found at each one of the poles of a fusiform cor- puscle; that they stain much darker in the center or at the base, the apex and outside margin being much lighter. Thus with the toluidine stain the margin is always pale blue, while the central or basal part is more violet. At the very base or in the center are generally seen one or more dark dots which are readily identified with the centrosomes of other cells. Instead of remaining in the non-nucleated parts of the erythrocyte when the nucleus is ejected, as is supposed by Heidenhain to be the case in the erythrocyte of the rat, they continue to remain attached to the cyto- plasmic envelope of the nucleus, only later on to separate from it.
As might be expected, we meet with some variation in the form, size, and staining properties of these respective zones, but in the main they are very constant and can nearly always be recognized at once. There are two differently appearing kinds of plasmocytoblasts, but between them are numerous gradations showing that the two extremes cor- respond, one to a stage that is dormant, the other to one that is highly active. In the former, the dark staining part with the centrosomes is situated at the base, close to the nucleus, while in the latter, the dark staining part
Zool.-Vol. I.] EISEN—rLASMOCYTES. 1 7
appears to be in the center of the plasmocytoblast. In the former, the different cytoplasmic layers superpose each other like a series of hollow cones placed one on the top of the other on a level plane, with the centrosomes almost resting on this plane. In the other, or spherical form of plasmocytoblasts, the cytoplasmic layers surround each other as the hulls of seeds. There is here no broad base, but a number of concentric layers of different density, color, and structure. In a general way these zones correspond to the ectoplasm and to the microcentrum (somosphere with cen- trosomes) of some investigators, but as great confusion ex- ists as to names I have considered it best to name each zone or sphere separately, as follows, counting from the exterior to the interior, or from the top to the base: —
A. Cytosome, or cytosomal spheres, ectoplasmatic spheres ; spheres not part of the nucleus and archoplasm —
i. Plasmosphere.
2. Hyalosphere.
3. Granosphere.
B. Archosome, microcentrum, archoplasmic spheres, archoplasm with centrosomes; spheres not part of the nucleus or cytosome —
4. Centrosphere.
5. Somosphere.
6. Centrosome.
As will be seen, and as I expect to demonstrate in the fol- lowing pages, the three outer spheres are purely cytoplasmic spheres, parts of the cell proper, for which, as a whole, I propose to retain the name cytosome, in juxtaposition to the caryosome , or nucleus , or to the three inner spheres , for which , as being of an entirely different nature, I propose the name archosome. This latter corresponds, at least in part, to Heidenhain's microcentrum, and to the archoplasm with centrosomes of some investigators.
The two following diagrams will illustrate this better. The first one gives the shape of the conelike plasmocyto-
i8
CALIFORNIA ACADEMY OF SCIENCES.
[3D Ser.,
blast, while the other one, which really is that of a plasmo- cyte, will give a fair idea of the spherical form of plasmo- cytoblasts.
Cbntrosombs.
■— . PL IHm fttBR E .
HVUOSPHERB.
GRANOSPHBRB.
ROSPHBRB. SO*K>6PHBRB.
Diagram i. Plasmocytoblast at the pole of a fusiform corpuscle from the blood of a Batrachcseps attenuatus.
In this diagram the nucleus is represented as being below the base line, but is not further sketched out, its position only being of interest. Immediately above it is the hyalo- sphere, while on the sides of the nucleus extends the plasmo- sphere. Above the hyalosphere, at the base of the cone, is seen the somosphere containing the centrosomes; surround- ing it is the centrosphere, and above it is the large dotted granosphere. The latter is surrounded by the hyalosphere and by the most exterior of all the spheres, the plasmo- sphere, showing fringed or amoeboid projections.
Zool.— Vol. I.]
EISEN—PLASi\fOC YTES.
19
Somosphbrb Cbntrosomb
Plasmosphere.
Hyalosphere.
Granosphbrb.
Centrosphere.
Diagram 2. Plasmocyte from the blood of Batrachoseps attenuatus.
In this diagram the spheres have the same value as in the former, the only difference being that they have assumed a spherical form. This is a diagram of such a plasmocyto- blast as that represented by figs. 25a, 39, etc., while dia- gram 1 represents such a plasmocytoblast as is seen in fig. 21.
Before I enter into a more minute description of the re- spective spheres, a few words about their general appear- ance may facilitate our understanding of them. In general the effect of the toluidine stain is as follows : The plasmo- sphere stains faintly blue, with darker blue blotches along the outer margin; the hyalosphere remains nearly always pure white, only staining faintly red with eosin, while with rubin it seldom differentiates; the centrosphere ap- pears strongly granulated and stains deeply violet; the centrosphere stains pale blue or deep violet with toluidine, with eosin it stains deep pink; the somosphere stains generally dark, with darker centrosomes. The exceptions to this general rule will be noted further on. Fig. 21 represents a fusiform corpuscle in which the conelike plasmocytoblast is especially prominent at the upper pole, while at the lower pole it is much smaller, and probably
20 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
wanting the microcentrum, in which case it cannot of course lay claim to the name of plasmocytoblast. In the upper one we have no difficulty in recognizing the outer plasmo- sphere with its fringed margin. The transparent hyalo- sphere comes next to it, almost or entirely unstained. The granosphere, which is very large, is stained violet, while the centrosphere is stained even deeper, rather an exception to the rule. Lowest down at the base are seen the somospheres, containing several dark spots, or centro- somes.
The other type of plasmocytoblast, represented by dia- gram 2, is the one that is seen in figs. 321ft, etc. We find that the centrosphere with its enclosures has traveled away from the nucleus and the base of the cone towards its apex, much more so at b than at a. In fig. 35 we find that at the pole b the spheres are arranged according to the conelike type, while at the pole a the spheres have assumed the final spherical shape, the granosphere surrounding the archosome about equally on all sides. The variation between these two extreme types of plasmocytoblasts is such that we seldom find two which are exactly alike, still, the similarity is sufficiently great to allow us without diffi- culty to recognize each respective zone. We will now con- sider these spheres more in detail, beginning with the outer one, or plasmosphere.
VII. Cytosome.
The Plasmosphere. — This, the most exterior sphere of cytoplasm, is especially developed at the poles, where its tendency to assume amoeboid projections is very noticeable. MacCallum and Griesbach have previously described the poles of the fusiform corpuscles as being frayed. The in- creased size of the plasmosphere at the poles may be seen in figs. 32, 33, 34, 35, etc. The plasmosphere may be much larger at one pole of the fusiform corpuscle than at the other, and this inequality in size is also correspondingly shared by the other spheres. In other words, the spheres
Zool.-Vol. I.] EISEN—PLASMOCYTES. 21
increase in size together, and in such a way that the largest plasmocytoblasts are always those which inclose centro- somes and centrospheres. The plasmosphere appears fringed, partly from actually being so, partly, also, on ac- count of a row of dots of dark-staining cytoplasm arranged along the edge. These dots never occur in a continuous line, but run in a zigzag way near the edge. The arrange- ment of these fringed or plasma-projections is like that of the radii in a circle or the rays of a star. The plasmosphere sometimes gives the impression of being at rest, as rep- resented by figs. 21, 27, 34, and others. At other times it appears to have become fixed while in amoeboid activity, as seen in figs. 25a, 290, 32^, 33d, etc. While MacCallum noted the frayed appearance of the plasmosphere and figured it, his method of staining could not bring out any of the de- tails of differentiation, though some of his figures slightly indicate that he had observed some structures corresponding to the inner spheres.
I have already pointed out that in a few plasmocytes I observed a membrane surrounding the plasmosphere, caus- ing the corpuscle to look very much like a real cell. In the plasmocytoblast no such membrane has ever been observed, as all possess a more or less fringed plasmosphere. While in the resting stage the plasmosphere presents an out- line of rounded protuberances, which may be either very small and even, as in fig. 21, or they may be large and un- equal in size, as in fig. 27. When properly stained the cytoplasmic accumulation at the edges is always prominent, and we find it either in the shape of more or less regular globules, or as wedges tapering towards the hyalosphere. The toluidine is the only stain which brings out this cytoplas- mic arrangement, and even a counter stain will prevent them from being observed. These small cytoplasmic masses seldom extend beyond the sharp line of the hyalosphere, and only once did I find them so irregularly scattered that the hyalosphere was obscured, as in fig. 38*.
The question arises as to whether the fringed appearance depends upon amoeboid movements or not. As I will return
Proc. Cal. Acad. Sci., 3D Ser., Zool., Vol. I. (2) Oct. ao, 1896.
22 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
later to this subject, I will only state here that at times it certainly does. When the fusiform corpuscle is finally ejected from the erythrocyte, the cytoplasm is undoubtedly torn from the cell membrane, causing it to assume a star- like appearance, with irregular rays; but later on these rays show forms which can only be explained as the result of amoeboid movements of the plasmosphere. I ascribe amoeboid movements to all the spheres except the hyalo- sphere, which appears always dormant as far as regards change of form.
Hyalosphere. — This sphere extends like an even, narrow, and transparent ring all around the plasmocytoblast. On the three outer sides it is bordered by the plasmosphere, while on the side towards the nucleus it rests against this body. I believe that as a rule the hyalosphere is always found interior to the plasmosphere, though in some instances I have not been able to observe it. In fig. 27, for instance, the hyalosphere is seen below the plasmosphere all around the nucleus, and probably these two spheres always occur to- gether. The hyalosphere appears structureless and hyaline, and is hardly stainable with toluidine. It is always highly refractive. Only by a double stain of eosin and methyl blue "O" has it been possible for me to show with certainty that the hyalosphere is a distinct sphere and not simply a thinner continuation of the plasmosphere. The eosin stains the hy- alosphere pink, while the plasmosphere remains bluish. A characteristic of the hyalosphere is that it is of even size all around, like a transparent highly refractive ring, and that it shows no indication of changing its form by amoeboid move- ments. Until the hyalosphere has closed around the form- ing plasmocyte, this latter, or its counterpart in the plasmo- cytoblast, can only be considered as a fragment of the cell, not yet having resumed that definite form which would char- acterize a finished or fully developed corpuscle.
The Granosphere. — This sphere is the most prominently noticeable of the various zones which compose the plas- mocytoblasts, especially on account of its darker color, but also by its size. When small the shape is always that
Zool.— Vol. I.] EISEV—PLASMOCYTES. 23
of a narrow crescent, as in figs. 21b and 32^; when larger it becomes conelike, and as it recedes from the nu- cleus it assumes a spherical form (fig. 25a, etc.). As the plasmocytoblast grows and tends to separate from the nu- cleus, it carries with it either the whole of the granosphere, as in figs. 25a and 38*, or it leaves behind a narrow crescent of granosphere close to the nucleus, as, for instance, seen in figs. 32*, 39, etc. The part that moves away always con- tains the centrosomes or inner spheres, while the part that is left behind appears homogenous throughout, without trace of a microcentrum, except in case the microcentrum has divided, when a later emigration of a plasmocyte may take place.
While the granosphere generally stains much darker, violet dark in contrast to other spheres which stain blue, this is not always the case, and for some reason or other the staining is inverted even on the same slide. In figs. 27 and 33 the tint it has taken is deep blue and the only differen- tiation is in regard to the intensity of the stain. Figs. 21, 24, 29, 32, 34, and 35 contain normally stained grano- spheres, while figs. 27, 31, and 33 show a granosphere in which we find no trace of the red. This is to a great ex- tent due to the time allowed the toluidine to stain. When too long the differentiation becomes less apparent, as the blue will quickly drive away the pink and violet. An im- mersion in the toluidine for three minutes will generally give the best differentiation, and even five minutes exposure is generally sufficient to destroy the differential effects.
The granosphere is also distinguished by its granulated protoplasm which is always quite prominent. This granu- lation is not even, but irregular, both as regards the size of the granules and their distribution. The periphery of the granosphere is nearly always rather even, pressing as it does against the hyalosphere. The contrast between these two spheres is sharp and striking, as in fig. 21. In size the granosphere is variable. Frequently it is very large, as in figs. 2itf, 22, 35, etc.; at other times it is much smaller, like a thin crescent, as in fig. 23. It frequently happens
24 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
that at one pole of the fusiform corpuscle the granosphere is very large, while at the other it is small or even wanting. This I think depends upon two things: either upon the stage of development of the plasmocytoblast, or upon the absence of or defect in the centrosomes or centrosphere. Grano- spheres which contain no archosome do not appear to increase in size, nor do they separate from the vicinity of the nucleus and become independent. While the plasmo- sphere and hyalosphere often extend all around the nucleus, forming the outer lining of the fusiform corpuscle, the granosphere is always confined to the poles, as seen in figs. 21, 33, etc. Here it exerts a pressure on the nucleus, as it is this sphere which causes the dell in the nucleus, generally found at the poles. Between the granosphere and the nu- cleus there is always a thin rim of hyalosphere, but as this rim is even all around the nucleus it must be the grano- sphere which is the direct cause of the dell. Further on I will refer to this again, and then show that it is the granulated sphere which in other genera of cells also causes a similar dell. The density of protoplasm or the greater tension in the granosphere, which causes this dell, probably could not act in the absence of a cell wall, except for the apparent elasticity and strength of the hyalosphere which prevents the granosphere from escaping. The granosphere is more or less sharply defined from the inner centrosphere. The three outer spheres — plasmosphere, hyalosphere, and grano- sphere— undoubtedly correspond to the ectoplasm of Heid- enhain, a reference to which will be made further on. While I have here referred to the granosphere as being the direct cause of the dells in the nucleus, it is probable that the in- direct cause of the dells is the archosome.
The phenomena of phagocytosis will be referred to in an- other place. Here I will state only that they are frequently observed in the plasmocytoblasts as well as in the plasmo- cytes, though principally in the latter. Both of these bodies very often inclose parts of or whole red blood cells, which they are apparently in the act of digesting. Such inclosures are always found in the granosphere, from which it may be
Zool.-Vol. I.] EISEN—PLASMOCYTES. 25
concluded that this sphere possesses digestive properties and can be considered as the digestive organ of the cell and of the plasmocyte.
VIII. Archosome.
The Centrosphere. — We will now consider a part which I think must be held analogous to the archoplasm of some investigators — the spheres surrounding the centro- somes. The centrosphere is nearly always well defined, and sometimes even separated from the granosphere by a thin but distinct unstained border. The position of this sphere in the granosphere is variable ; it may be situated at the base of the cone, or it may be found in the center of the granosphere, or near one of its borders. The outline of the centrosphere is generally smooth and regular; it may be slightly uneven or cloudlike, but is nearly always very distinct, and I believe it is always present. If we consider the staining quality of this sphere we find that with toluidine it generally stains lighter than the granosphere and that it shows much less granulation. But this staining is not always constant; in fig. 35 the centrosphere at a is darker and star-shaped, while at b it is darker and conelike. In fig. 34 the centrospheres at the respective poles are stained lighter than the granosphere. This is also the case in fig- 23.
There may be from one to four centrospheres in one plasmocytoblast; when more than one is found it is evi- dent that they constitute fragments of the original centro- sphere, each fragment having assumed a more or less spherical form, and each one carrying along with it a sep- arate granosphere, the latter also being a fragment of the original granosphere. Thus in fig. 32 we see a single cen- trosphere at each pole, each surrounded by an envelope of granosphere. In fig. 350 the centrosphere is in a state of division, while at 35^ no activity is apparent. In fig. 36, from a fusiform corpuscle stained with Ehrlich-Biondi, we find three centrospheres at each pole, the lower pole at b
26 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
having spread out to such an extent as to enclose one half of the nucleus. In fig. 28 we see a plasmocytoblast with three centrospheres in different stages of development, one of which, having separated itself almost completely from the vicinity of the nucleus, and carrying with it an envelope of granosphere and centrosomes, is apparently ready to form an independent plasmocyte. In the plasmocytes the centro- sphere frequently assumes a large size and becomes more differentiated, evidently a direct effect of development and growth. In fig. 49, which represents a free plasmocyte, the centrosphere is beautifully differentiated, having as- sumed a deep pink eosin stain. In fig. 380 the centrosphere is very large and rounded, pale blue, and surrounded by a narrow rim of granosphere. The centrosphere frequently assumes a star-shaped or irregular form, which indicates that it possesses independent amoeboid movements.
Somosphere and Centrosomes. — The innermost of the spheres, which incloses one or more centrosomes, I have named somosphere. I have not, however, been able in every instance to demonstrate the presence of this sphere; but in many, perhaps in the majority of corpuscles observed, this sphere is distinct from the centrosomes. The dark granules, or centrosomes, accepting this name as Heidenhain understands it, are nearly always surrounded by this special sphere, which generally stains darker than the centrosphere, but sometimes also appears much lighter. It varies much in size and form, but is less regular than any of the other spheres, and undoubtedly possesses amoeboid activity. In fig. 21 the somosphere is well marked, and in its center are distinctly seen the darker granules, or centrosomes. It would be incorrect to state that the centrosphere always en- closes the somosphere, because frequently the latter is seen to lie at one edge of the centrosphere, as represented in fig. 340; or it may be even entirely separate from it, though this maybe caused by accidental pressure. If we compare figs. 22 and 23, we find that in the former the somosphere is very small, a faint tint, so to say, surrounding the granular cen- trosomes. In fig. 23, again, the somosphere is much larger
Zool.-Vol. I.] EISEN—PLASMOCYTES. 2*]
and lies prominently in the white centrosphere, while on the other hand it encloses two small, separated centrosomes. In fig. 38^ the somosphere appears to be absent, the centro- somes standing out free in the centrosphere. In fig. 49, which is in many respects a very instructive one, the somo- sphere is stained dark blue and starlike in form, inclosing some centrosomes of rather uncertain shape. In the early plasmocytoblast the centrosomes always lie very close to- gether, and can only with difficulty be segregated ; but as the spheres grow the centrosomes separate, each carrying with it some part of one of the inner spheres. I have never found more than four centrosomes together in one plas- mocytoblast, and generally their number does not exceed three. In the plasmocytoblasts the somosphere and centro- somes are too small to be readily studied, the larger plasmo- cytes offering much better facilities in this respect.
The relationship of the three inner spheres — those of the microcentrum — is not by far cleared up, but it seems that the somosphere and centrosomes are much more intimately con- nected than the centrosphere and the somosphere.
IX. Different Kinds of Protoplasm.
The distinct differentiation possessed by the various zones naturally indicates that the protoplasm composing them con- sists of at least as many different kinds as there are zones. The word cytoplasm, as referable to all protoplasm con- tained in the cell outside of the nucleus, would thus not express and define the various kinds of protoplasm found in the inner spheres. If we, for convenience sake and with reason of a physiological difference, speak of cytoplasm as distinct from caryoplasm, we can, with equal propriety and for greater distinctness, refer to the protoplasm of the archosome as being distinct from that of the cell and the nucleus. That the spheres of the archosome must be considered as quite distinct from those of the cell and the nucleus is quite evident from what I have mentioned above, and, moreover, they must be considered as a whole
28 CALIFORNIA ACADEMY OF SCIENCES. [3d Sbr.,
by themselves. This unity must ultimately be ascribed to difference in structure, quality, and organization of the pro- toplasm, entitling it to be considered separately. For the protoplasm of the microcentral spheres I therefore propose the word archosomoplasm, giving it equal value and im- portance with the cytoplasm of the cell and of the cary- oplasm of the nucleus. While we may use these words for convenience sake, we may neither imply that the archo- somoplasm, caryoplasm, and cytoplasm are not further sep- arable into distinct kinds, nor that parts of cytoplasm, for instance, may not at times be found mixed with caryoplasm. How many distinct kinds of caryoplasm and archosomoplasm there really are will probably not soon be definitely decided, but I think we can safely argue that every part of protoplasm which differentiates in staining constitutes a kind of its own, differing in quality and function from the rest. That the centrosphere and somosphere do not always differentiate in the same manner does not prove that they are not always equally distinct from each other. Too long exposure to the stain will always destroy the differentiation, while at times permeation with food granules and liquids will greatly lessen or affect their susceptibility.
X. Development of the Plasmocytoblast into Plas-
MOCYTES.
I have already pointed out that by arranging and compar- ing a series of drawings of plasmocytoblasts it soon becomes evident that they are respectively in different stages of development; not only are some of them much larger than others, but the larger ones show a differen- tiation not found in the others. If we study such of the figures as 28, 33^, 37, etc., we observe that the inner spheres have divided, a division apparently caused by a separation of the centrosomes, which latter have carried with them, each one separately, an envelope of one or two spheres. Thus in fig. 350 the white somosphere is dividing and in each division is found a centrosome; in fig. 33 each
Zool.-Vol. I.] EISEX— PLASMOCYTES. 29
somosphere has carried with it an envelope of centro- sphere ; while in figs. 28 and 36 each centrosphere is fully separated from the other. In fig. 28 a further stage has been reached, as here each centrosphere is surrounded by an envelope of granosphere. At a yet more advanced stage this granosphere is surrounded by an envelope of hyalosphere and plasmosphere, as seen in figs. 37,38, and 39. The next stage consists in an entire separation of the new spherical body from the plasmocytoblast. In other words the plasmo- cytoblast has divided into two or more distinct bodies which have gradually freed themselves from all connection with the nucleus, or rather from the thin layers of cytoplasm yet adhering to the nucleus ; the nucleus has continued to disintegrate, while new plasmocytes have steadily developed until they have become free and independent elements of the blood. That these new bodies, or blood elements, are something entirely distinct from mere fragments of the cyto- plasm is evident from several observable facts. The plas- mocytes increase in size, which again shows an independent growth undoubtedly caused by the taking up of nourishment; they have also moved away from the vicinity of the nucleus, showing independent movement; and finally, they have changed their form from a mere fragment to a finished, symmetrical body. The various spheres or envelopes of the new plasmocyte do not show any great irregularity, but instead exhibit a surprising regularity, especially as regards the two exterior spheres, the closing up of which forms the last step in the formation of the plasmocyte. Figs. 28a, 32^, 37, 380, 38^, 39, etc., represent plasmocytoblasts in the last stages of development, the plasmocytes being almost perfected and ready to separate. In 37 the plasmo- cyte is entirely formed, while in 38a and 39 it is yet con- nected with the old fusiform element by a narrow shaft of plasmosphere.
XI. The Plasmocytes.
General Remarks. — In the foregoing I have endeavored to show how the archosomes, or microcentra, of the plas-
30 CALIFORNIA ACADEMY OF SCIENCES. [3d Ser.,
mocytoblasts, have gradually receded from the immediate vicinity of the nucleus; how they have clothed themselves with envelopes of the outer cytoplasmic layers, or spheres; and finally, how they have entirely separated themselves from the fusiform corpuscles, henceforth existing as plasmo- cytes in the blood serum. These new elements possess properties which must characterize them as independent corpuscles. These properties have already been referred to as follows: Assimilation of food through phagocytosis, or through the blood serum; exhibition of independent movements which have enabled them to separate from their connection with the fusiform corpuscle and to live an inde- pendent life in the blood serum ; and further, they have, from the beginning, been something else than mere frag- ments, each one being surrounded by ringlike spheres or zones of differentiated protoplasm, one exterior to the other, an organization which is not found in fragments.
The plasmocyte is thus characterized by the possession of form; interior and symmetrical organization; independent movement, both as a whole and as regards the separate spheres of the microcentrum ; and growth by means of phagocytosis, which includes the process of digestion in the granosphere and in the somosphere.
All these properties also characterize a cell; but the plasmocytes are not cells; they lack, in fact, two most essential characteristics of a perfect cell: they possess no nucleus, and are not generally surrounded by a cell mem- brane.
Homology of the Plasmocytes and the Plasmocytoblasts. — I have carefully examined about one thousand plasmocytes, more or less, under very many stains, and find that while they vary, they do so only within certain limits, and with the exception of a dozen all told they present the same general structure. Comparing them with the plasmocy- toblasts in various stages of development, we find that the plasmocyte possesses every characteristic found in the advanced plasmocytoblast, while it does not possess a single characteristic which is not also found in some plas-
Zool.-Vol. I.] EISEN-PLASMOCYTES. 3 1
mocytoblasts, viz.: those which are ready to separate them- selves from the fusiform corpuscles. Whether stained one way or the other, I find that the respective spheres behave in the same way, and that those of the plasmocyte are al- most exact copies of those of the plasmocytoblast. From the youngest plasmocytoblast to the oldest plasmocyte it is not difficult to arrange a perfect series of forms showing a gradual gradation of one into another. An examination of figs. 40 to 84 will demonstrate this better.
Different Kinds of Plasmocytes. — While all plasmocytes possess a number of characters in common, they still differ in some degree, sufficiently to be worthy of description. For convenience sake we can segregate them into groups, according as we find in them one, two, three, or possibly four separated spheres, side by side, each containing one or more centrosomes; but between these different types there are gradations showing the variations to be of minor importance. As representative of one of these types I will refer to figs. 40, 44, 47, 49,55, 60, etc., taken at random. The common character of these six corpuscles is that of the centrosome or centrosomes being surrounded by an envelope consisting of all the various spheres described above, concentrically arranged. In another type of plas- mocyte we find that the common envelope contains only two of the spheres, the plasmosphere and the hyalosphere, while there are two or three separate archosomes, each of which has its own separate envelope of granosphere. Such plasmocytes are figured in 68, 69, 70, 71, and 72. Thus the division in the plasmocytoblast has in these not extended to the two other spheres. We have here simply an original plasmocytoblast separated from the nucleus of the fusiform corpuscle and closed up by the two outer spheres before the archosomes separated themselves sufficiently for each one to become the center of a plasmocyte. Still another type is represented by figs. 61, 63, 65, 66, 79, etc., in which we find that the division has not even extended to the granosphere. In these the granosphere is continuous in the same way as the two outer spheres, only the archo-
32 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
somes are found separated. Among these types we find those in which one centrosphere encloses one single somo- sphere and one centrosome, while the other centrosphere encloses two distinct centrosomes, as represented in fig. 61 ; or we find that one centrosphere encloses one centro- some, and that the other centrosphere contains three centro- somes as in fig. 65 ; or the division may be more perfect, and we find three distinct centrospheres, each one with a centrosome. In a word, a very great number of combina- tions may exist, each to be considered as the stage in which the plasmocyte was freed. Whether a further division of the archosome could take place in such a way that each centrosome would form the center of a plasmo- cyte is doubtful, and I must leave this question undecided.
The Spheres of the Plasmocytes. — As the spheres of the plasmocytes resemble those of the plasmocytoblast so very closely, only a few remarks will suffice to point out the more apparent characteristics.
The projections of the plasmosphere vary considerably, and I have frequently observed a striking symmetry in their position, in that they occur principally at the poles, thus giving the plasmocyte the appearance of a star, or of a starlike spindle (figs. 46, 73, 74, 51, etc.)
The hyalosphere is nearly always distinct, narrow, even, and pellucid, giving the impression of being solid, as the other spheres rarely encroach on it. It never becomes coarsely granulated, and if stained with Ehrlich-Biondi it differentiates poorly, while with eosin it sometimes stains faintly pink. I have, however, under favorable conditions observed in this sphere a very fine, regular granulation, consisting of even, rounded globules of exceedingly small size, colorless, and of great transparency.
The granosphere is, of course, the most prominent of the spheres as regards color, granulation, and size, though all of these vary within certain limits. It is this sphere which takes up foreign substances and digests them, thus exhibiting phagocytosis (fig. 79). In it we find granules of various sizes, staining more or less intensely. Sometimes the grano-
Zool.-Vol. I.] E/SEX—PLASMOCYTES. 33
sphere is very large, as in figs. 40, 49, 580, 73, and 77; at other times it is narrow but equally distinct, as in figs. 47 and 60. The darkest granules are either accumulated at the margin near the hyalosphere or near the centrosphere, as seen in figs. 53 and 58a, or they are concentrically dis- tributed as in fig. 55.
That the centrosphere is entirely distinct from the grano- sphere is shown by its different staining quality, by its less pronounced granulation, and by the frequently very sharp margin which separates the two spheres. Thus in fig. 49 we see the centrosphere stained pink, while the granosphere is dark blue. The above figure is from an eosin-methyl blue preparation. In fig. 58a the centrosphere is pale blue and the granosphere is dark blue. The somosphere is here very pale and unstained, while the centrosomes are very sharply defined. The centrosphere, more than any other sphere, exhibits amoeboid movements, as seen in figs. 82, 83, etc. In order not to repeat I will leave the detailed description of the various figures to be given at the end of the paper.
The innermost enclosures of the archosome, the somo- sphere, and the centrosomes, may best be considered to- gether, as they undoubtedly are very closely related and are apparently dependent on each other. Sometimes the centrosomes are not distinct, while at other times the somosphere cannot be distinguished. Again, at times, the distinction is prominent, as for instance in fig. 49, where the somosphere has assumed a deep blue, while the cen- trosomes of both remain dark; or, in fig. 65, where the somosphere is lighter blue ; but this absence of either the centrosomes or somosphere is, I think, only apparent, being due to imperfect staining, caused by either too long or too short exposure to the stain. In all successfully stained slides the somosphere and centrosomes are never absent. I have frequently observed that the somosphere and centro- somes do not always lie in the center of the centrosphere, as in fig. 65, but at one side, as in fig. 69, or even outside of it, as in fig. 73. When there are three centrosomes pres-
34 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
ent they form a triangle (figs. 52, 58a) with a position rela- tive to each other very much like that of the centrosomes of the lymphocytes, according to the diagrams of Heiden- hain (see his " Neue Untersuchungen," figs. 3, 25, etc.), or they may simply lie in a half circle, as in figs. 52 and 65. In one plasmocyte I found four centrosomes lying in a square (fig. 60), connected by a film of somosphere. In others the arrangement was less regular and the centrosomes were placed at different depths, one above the other. At no time have I observed more than four centrosomes in the same microcentrum. The somosphere is either diffuse, spherical, crescent-shaped or ringlike. The diffuse and spherical somospheres are always homogenous or very finely granu- lated, while the crescent-shaped or ringlike somosphere is seen to more or less perfectly enclose one or more highly refractive yellowish bodies. When the somosphere is ring- like it is always found to be wider at one point from which it tapers in both directions towards the opposite sides. In the thickened and crescent-shaped part are found the minute centrosomes. When more than one is present they are al- ways found close together and frequently so approximated that only the most delicate manipulation of the light will show them to be separate from each other and from the somosphere.
Between the crescent -shaped and the ringlike somo- sphere there are numerous intermediate links, the extreme forms being of equal frequency. These forms are undoubt- edly due to the enclosures mentioned above. These are of a rather solid nature, and being always round they cause the somosphere to assume the shape of a crescent or ring; the former if the sphere is small and cannot compass the globule ; the latter if it is larger and can extend all around it. That these globules constitute a food supply, perhaps derived from the granosphere, will presently be mentioned. (See figs. 81, 83, 88.) In corroboration of this is the fact that when the somosphere is crescent or ring-shaped the centrosomes appear in greater activity, actually in the process of budding (fig. 83). By budding I do not necessarily im-
Zool.— Vol. I.] EISEN— PLASMOCYTES. 35
ply that new centrosomes are budded off from the mother centrosome, though this might be the case; but it may be as- sumed that this budding is only a part of an amoeboid pro- cess, an expansion which may later on be succeeded by a corresponding contraction.
The budding of centrosomes has already been ob- served and described by M. Heidenhain in his " Neue Un- tersuchungen," and he ascribes it to a process of centro- somal division or multiplication. That such a multiplication of the centrosomes in the plasmocyte takes place is almost certain, but whether it ultimately leads to a division of the plasmocyte remains yet to be demonstrated.
The final effort of the centrosomes and somosphere is probably to separate themselves in such a way that each centrosome, with its surrounding somospheres, forms the center of an archosome and a plasmocyte ; but I think that this division goes on principally while the microcentrum is yet enclosed in the plasmocytoblast. After the plasmo- cyte has once formed, division into two or more plasmocytes may take place, though I have only very rarely found any indication that this is the case. The impossibility of study- ing the plasmocyte without proper staining and fixing makes the determining of this most important question most diffi- cult. In fixed specimens on slides I have now and then found plasmocytes which appear to be in amitotic division, but until special study has been given it this point cannot be decided.
Food Supply in the Somosphere. — Now and then I have observed highly refractive globules in the somosphere, which must either be parts of the somosphere, of secreted, or of foreign matter. These globules may be two or three in number, although usually there is but one. They are al- ways dull yellowish, but strongly refractive, rounded or ir- regular, with sharp outlines. They are not by any means present in every plasmocyte, nor are they found on every slide, though on some slides I find them in al- most every plasmocyte. Since this paper was finished in ms. I have found plasmocytes in large numbers in human
36 CALIFORNIA ACADEMY OF SCIENCES, [3D Ser.,
blood, and nearly every one of them possessed this same refractive globule at one side in its granosphere, seldom in the center. In the batrachian plasmocyte the somosphere lies outside on the surface of these globules, never in them.
The fact that these globules are not always present, nor consistent as regards form, size, and number, induces me to consider them as food particles which are being digested by the somosphere and which may have been either de- rived directly from the blood serum or secreted by the granosphere. In consistency these globules are quite dense, as may be judged from the appearance of the somo- sphere.
The somosphere would then stand in the same relation to the centrosome as that of the granosphere to the centro- sphere and interior spheres. In other words it constitutes a digestive layer for the nourishment of the centrosomes. It is only reasonable to suppose that such delicate organisms as the centrosomes must have specially prepared nutriment, and that they are unable to directly assimilate food sup- plied by the blood serum and by the granosphere. The process would then be as follows: The nutriment supplied by the blood serum is digested by the granosphere; part of what results from this feeds the various spheres of the plas- mocyte, especially the centrosphere ; and as this nutriment is too coarse for the centrosomes, it must, in order to be assimilated by them, be further manipulated by the somo- sphere.
Unequal Staining of the Archosomal Spheres. — A point of considerable interest is the unequal susceptibility to stains exhibited by separate archosomes. For instance, in plasmocytes which contain two or more archosomes with a granosphere surrounding each, we often find that one granosphere or centrosphere has accepted a very dark stain while the other remains quite pale. Thus in fig. 65 we see that the centrosphere of the upper microcentrum is dark violet, while the lower and larger one is light blue. A similar difference is also seen in figs. 64 and 69. This unequal staining is frequently accompanied by a dif-
Zool.— Vol. I.] EISEN— PLASMOCYTES. 37
ference in size. It is generally the larger sphere which stains the most intensely, and this unequal differentiation may be due either to disintegration and decay or to poor nourishment. It appears as though some of the spheres are stronger and better able to procure nourishment than others.
General Absence of a Cell Membrane. — That the plasmo- cytes are not generally surrounded by any cell membrane has already been stated. The exterior layers show projecting plasmarays entirely without any membranous covering. This might be expected on account of the origin of the plasmocyte from a part of the cell which had lost its cell membrane; but out of the very many plasmocytes inves- tigated I have found some, not more than half a dozen in all, which, as far as concerns their exterior margin and form, present a very different appearance. These plasmo- cytes, if such they are,. present a rounded form with smooth outline covered by a distinct cell membrane; but, unlike most plasmocytes, I could never see in them any differ- entiation between the plasmosphere and the hyalosphere, and even the outline of the granosphere was less well de- fined. I believe, however, that these bodies are real plasmo- cytes, though as to the cause of their structure I can venture no opinion. In fig. 59 I have represented one of them. Be- sides these plasmocytes I have also found in the blood very much smaller bodies, appearing like very small isolated nuclei without cytoplasm. My investigations of these are as yet unsatisfactory.
Absence of Nucleus. — The absence of any nuclear struc- ture in the plasmocyte is readily explained by the well known law that the cell originates only from the cell, and the nucleus only from a previous nucleus. In one or two instances, however, I have found a structure near the center of the plasmocyte which might perhaps be called an incipient nucleus, an effort to form a new nucleus of some existing nuclear fragments which may have en- tered the plasmocyte. In fig. 75 such a pseudonucleus is seen, as it were, in mitosis (see also fig. 74). Such forms are, however, exceedingly rare, and these two are
Proc. Cal. Acad. Sci., 3D Sbr., Zool., Vol. I. (3) Oct. 21 , 1896.
38 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
really the only ones I have observed. During the disinte- gration of the nucleus and before the plasmocyte has sep- arated, nuclear fragments are often seen floating about in the cytoplasmic sphere, especially in the granosphere. If any larger fragment should happen to be enclosed. in the plasmocyte its influence there might be of more than mo- mentary importance. I must, however, distinctly state that I have nowhere found anything even approaching a per- fect nucleus enclosed in a plasmocyte.
Degeneration of the Plasmocytes. — I consider the plas- mocyte shown in fig. 84 as a form resulting from degener- ation. These forms are of quite frequent occurrence and of various sizes, often larger than the perfect plasmocyte. I was for a long time doubtful as to their nature and their con- nection with the plasmocyte, but of late I have found inter- mediate links which tend to prove that they are forms of plasmocytes. They are characterized by a disintegration of the plasmosphere and hyalosphere, in the place of which we only see what looks like a very delicate membrane into which radiates some substance from the granosphere. The latter is always remarkably granulated, the granules being regular and rounded, staining intensely dark. This granu- lation reminds me greatly of the one possessed by eosin- ophile cells, or by other strongly granulated forms of leuco- cytes. The centrosphere in these plasmocytes is always less distinct, being partly covered with the dark granules of the granosphere, but the somosphere and centrosomes stand out plainly, the somosphere generally being ring-shaped.
Abnormal Plasmocytes. — I have already stated that some plasmocytes show an abnormal structure which is not easy to explain. Such plasmocytes are also very rare. The va- riation refers generally to a duplication of certain spheres or to the presence of some sphere not found in the normal plasmocyte. Thus in fig. 47 we find on one side, the right one, a small crescent-shaped, faintly stained body pushed into the substance of the granosphere. In this instance it may be a fragment of the centrosphere. A more diffi- cult appearance to explain is the one seen in fig. 64.
Zool.— Vol. I.] EISEX^PLASMOCYTES. 39
Here we have seven or eight distinct spheres instead of six. Perhaps the simplest way to explain this is to suppose that the centrosphere, which here is pale blue, has accepted some food, or some other unusual substance which has so arranged itself that it cuts the centrosphere in two. How- ever, this explanation is given for what it is worth, without any pretence to correctness. Every cytologist knows that now and then cells of various kinds are met with which pre- sent an abnormal structure not readily explainable. Another abnormal plasmocyte is one shown in fig. 77. Here the centrosphere appears as a broad band across the grano- spheroplasm. The darker margin of the band with the four black dots I consider to be somosphere and centro- somes.
As to the nature of the various black dots in fig. 56 I am undecided; they may be centrosomes or not. I think the three larger ones certainly are centrosomes, each surrounded by somosphere and centrosphere. The centrosphere of the one to the right is very large, extending across the center of the plasmocyte.
Amoeboid Movements. — The amoeboid movements of the frayed ends of the fusiform corpuscles were mentioned by MacCallum, and have probably been observed by some in- vestigators of the large elements of batrachian blood. MacCallum, however, speaks of the slow vibratory motion of the starlike prolongations, and not of regular amoeboid movements or projections, though from the general tone of his arguments (pages 245, 246) it would, I think, appear as though he considered the amoeboid movements as really existing. This movement has been denied by Eberth, who holds that under ordinary circumstances such move- ments do not exist, at least not in the fusiform cor- puscles; but I have frequently observed plasmocytes and have satisfied myself that in them such movements are common, generally very slow, but under certain condi- tions extremely rapid. If a drop of blood is mixed with salt solution of 0.6, we see that the plasmaprojections change rapidly, pushing out with unusual vigor to a greater
40 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
distance than under ordinary conditions, then becoming suddenly paralyzed and unable to again contract. In fig. 76 I have figured such a plasmocyte. It was at first only oblong rounded, with short plasmospheric projections. In a couple of minutes these had reached their present size.
In ordinary plasmocytoblasts and plasmocytes there is much variation in the appearance of the outer edge of the plasmospheres. In some plasmocytes the rays are pointed and very long, in others again they are rounded and scarcely projecting. While the former rays may be explained as being fragments torn from the cell mem- brane of the erythrocyte, the rounded appearance of the latter can only be considered as a direct effect of amoeboid contraction. Another sign of amoeboid movement is the presence of bacteria, foreign bodies of various kinds and size, as well as fragmentary or even whole red blood cells lodged in the granosphere of the plasmocyte. Their pres- ence can only be explained by amoeboid movements of the plasmocyte, the latter having engulfed them in the same way as do leucocytes and other wandering cells.
As regards each one of the inner spheres, separately, it is evident that the peculiar forms frequently possessed by them must be attributed to amoeboid movements. Such undoubted activity is especially seen in the centrosphere, and to a lesser extent in the granosphere and somosphere. In each one of these spheres we can recognize a rest- ing stage and a stage of amoeboid activity. While the resting form of each of these spheres must be con- sidered as approaching a disk, other forms frequently occur which cannot be the effect of accidental pressure or disturbance. As a plasmocyte with a resting archosome, I consider, for instance, the one shown in fig. 47, where both the centrosphere and the somosphere are oval. Figs. 27, 48, 49, 68, 69, 82, and others, show the various spheres of the microcentrum as arrested in the amoeboid stage. That this amoeboid stage is not confined to the plasmocyte as a whole is evident from such figures as 23, 32^, 35a and 82, all of which show signs of a most active movement of the
Zool.— Vol. I.] EISEX— PLASMOCYTES. 41
centrosphere. In fact the whole progressive movement of the archosome, from the base of the bud (plasmocyto- blast) to its center or upper part, must be ascribed to such amoeboid movements as those indicated in the figures. Similar forms indicating amoeboid movements of the micro- centrum have been described by Rawitz in his paper, " Untersuchungen iiber Zelltheilung," and are illustrated principally in figs. '2 and 3 of said paper. In these and other figures in that paper we also find the granosphere clearly delineated and described as " Zellsubstanzhof."
Growth and Phagocytosis. — While a large number of plasmocytes are not any larger than the largest divisions of the plasmocytoblasts, many of them are much larger than any that I have seen while yet enclosed in the plasmocyto- blast. From this I infer that the plasmocytes increase in size — that they actually grow. Between the smallest plas- mocytes and the largest ones there are those which are of all intermediate sizes. I have measured plasmocytes which were as large as the nucleus of the fusiform corpuscle, but the majority are much smaller, as will be seen by the measurements given further on. This growth can hardly be caused by anything but an assimilation of food. The food supply is probably mostly derived from the blood serum, but some of it, at least, is at times attained by a direct process of phagocytosis. Thus I have frequently encountered plas- mocytes which had engulfed small erythrocytes or their frag- ments, some of the latter appearing to be in a state of decomposition. On every cover glass such examples of phagocytosis are often found. In fig. 79 I have represented a phagocyte which has swallowed a very small erythrocyte, the more interesting because this erythrocyte contains a parasitic protozoa, the life history of which I will soon de- scribe. In many instances, however, what appears to be phagocytosis is not really so. We frequently find that a plasmocyte overlaps, or is superposed, on a red corpuscle, in which case it at first appears as if the plasmocyte was in the act of digesting the red cell. All around the out- lines of the plasmocyte there is seen a pale margin, as if
42 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
the haemoglobin in the red cell had been consumed by the plasmocyte. But this is mere illusion, because the pale ring or area is caused by the pressure of the overlying plasmo- cyte which has pushed the haemoglobin away from its immediate vicinity. Again, wherever phagocytosis occurs no such displacement of the haemoglobin takes place, as illustrated by fig. 79. Even in the plasnrocytoblasts such phagocytosis is frequent. I have already mentioned that the enclosures are principally found in the granosphere, which is thus to be considered as the seat of digestive activity in the cell.
Duplicity of the Plasmocytoblasts. — A very noticeable fact is the frequent occurrence of a plasmocytoblast at each of the opposite poles of the fusiform corpuscle, or rather, more strictly speaking, of the degenerating nucleus. In some instances, even when the granosphere is distinct at each pole, I have not been able to discover the inner spheres and centrosomes ; but in the majority of fusiform corpuscles the respective cytoplasmic spheres, with an archosome, are found at each pole. It appears at the beginning of the degeneration and disintegration of the nucleus, or at the moment when the cell membrane was ruptured and the haemoglobin was diffused, that the erythrocyte was not in actual rest, but at the beginning of mitosis. The centro- somes and archoplasm had evidently already separated and moved to opposite poles of the nucleus, where at their rest- ing places they had caused a dell to appear. The destruc- tion of the cell and its transformation into a fusiform corpuscle, is, therefore, not likely to have been caused by any defect in the centrosomal spheres or archosome, but rather from some defect in the nucleus itself. This de- fect, whatever it may have been, prevented the chromosomes from passing through the preliminary stage of mitosis; they were therefore unable to respond to the action of the archosome. The nucleus in which, at this stage, we should expect to find a great activity among the chromosomes had thus died, that is to say it had become disorganized or paralyzed, so that the action of the centrosomes, instead of
Zool.— Vol. I.] EISEN—PLASATOCYTES. 43
exerting itself on the nucleus, caused the cell wall to rupt- ure and the fusiform corpuscle to be set free in the serum. The very fact that the centrosome and plasmospheres sur- vive— nay, grow, develop, and continue an independent life — while the nucleus perishes, points to this conclusion. The nucleus of the fusiform corpuscle is, as compared to the nucleus of the erythrocyte, always in a state of degen- eration. The various kinds of chromosomes are neither distinguishable in shape nor color, and consist of a mere irregular mass of globules of various sizes, more or less con- centrated towards the center. There is an entire absence of those fine and exquisite details found in every perfect nucleus. The nuclear membrane — the caryotheca — is also frequently ruptured, and we can see small parts of the nu- clear plasma diffusing in an irregular way through the cyto- plasm, indicating dissolution and decay.
Dr. A. Dehler, who was the first to demonstrate the micro- centrum in the erythrocytes of the chicken embryo, de- scribes and figures only one set of centrosomes in the resting blood cell. Also M. Heidenhain, who has so studied in detail the nature of the leucocyte, refers only lightly to the micro- centrum of the erythroblast. In figs. 15 and 16 ("Neue Untersuchungen") he figures two nucleated red blood cells with centrosomes partly visible in the cytoplasm. Fig. 16 interests us most because the nucleus with the polar pro- jections bears a strong resemblance to the fusiform elements of Batrachoseps blood. Heidenhain does not enter upon any detailed description of these cells but simply states that ac- cording to his conclusions the ejection of the nucleus is caused by the tension of the aster rays, in conformity with his now well known tension theory. He believes, also, that after the ejection of the nucleus the centrosomes remain in the erythrocyte. Whatever may be the case with the erythro- cytes of the rabbit, I am positive, as regards the fusiform elements of the Batrachoseps blood, that the centrosomes remain in the plasmocytoblast.
The Ultimate Fate of the Plasmocyte. — A remarkably small proportion of plasmocytes show signs of dissolution,
44 CALIFORNIA ACADEMY OF SCIENCES. [3D Sbr.,
probably not ten per cent., and these appear to be full of large and small vacuoles. The spheres become less dis- tinct, finally diffusing one into the other, and no longer responding to the stains. They become more transparent as the plasma evidently becomes less dense. The interior spheres are the first to disappear from view, the centro- somes with them. In the last stages of its existence the plas- mocyte resembles a large diffuse blood plate. Fig. 78 rep- resents a dissolving plasmocyte with both large and small vacuoles stained with Ehrlich-Biondi. Judging from the few which thus decay, I conclude that the life of a plasmo- cyte is fully as long as that of a nucleated blood cell, and probably much longer.
Adhesive Nature of the Cytoplasm. — Every one who has observed the fusiform corpuscles has remarked upon the adhesive nature of the outer cytoplasmic layer. Frequently a number of such corpuscles are seen adhering together, forming irregular discs. The plasmocytes act exactly in the same way. Not only are they found joined or attached to each other but frequently they are also seen adhering to the margin of the plasmocytoblasts. It is not always easy to determine when we have before us a free plasmo- cyte whether it is simply adhering to the cytoplasm of a plasmocytoblast or is separating from it. The continuation of the two outer spheres must be the criterion of this, though in the first stage of the plasmocyte the difference cannot be very great. If there is any large amount of granosphere yet in position at the apex of the nucleus, and if this granosphere is conelike and contains a microcen- trum, we may assume, with great probability, that the ad- hering plasmocyte is really only adhering and not in the act of separating. After all the plasmocytes have separated from the plasmocytoblast there often remains a thin crescent of granosphere close to the nucleus, but this crescent does not contain any parts of an archosome, therefore cannot produce other plasmocytes.
Zoou— Vol. I.] EISEX—PLASMOCYTES. 45
XII. The Independence of the Archosome.
From the foregoing observations it will be seen, at least in the fusiform elements of the blood, that the archo- some shows an independent life history — surviving, grow- ing, and changing long after the other constituents of the cell have disintegrated As far as I can see the plasmocyte occupies a position equal to that of a real cell. It has all the general qualities of the cell as understood by modern cytologists, with perhaps one exception — the power to re- produce itself; at least no instance of unqualified plasmocytic division has come under my observation. The question now arises as to the relationship which the archosome bears to the nucleus and the balance of the cell. As is well known, there are two opposite views on this matter : one which con- siders the centrosome a constituent of the cell, always pres- ent and of paramount importance in directing the mitosis; the other claiming that the centrosome is only an organ in the cell (like the heart in the animal body), either always present or temporarily differentiated at the time of mitosis — a larger microsome, whatever that may mean, but nothing more. According to the latter theory the only function of the centrosome would be to mechanically direct and carry out the complicated stages of the mitosis, and when this was accomplished its work would be ended. This theory suf- ficed as long as centrosomes were not found in cells which had lost the power of division, or which would never again divide by caryokinesis ; but we now know that centrosomes and archoplasm occur in cells in which cell division will never be repeated. It is not my intention to enter upon this subject extensively at present. However, I will here point out that if we concede that the microcentrum, or archo- some, is always present in every cell, and can recognize that the archosomes in some cells, as for instance in the fusiform corpuscles of Batrachoseps, survive all other parts; that they clothe themselves with various envelopes of cyto- plasmic spheres; that they increase in size and assimilate food; that they remain entirely independent of former asso-
46 CALIFORNIA ACADEMY OF SCIEXCES. [3D Sek.,
ciates in the cell ; and that they are always minutely organ- ized in the same manner; — then, I think, we cannot help but conclude that the centrosome is something other than a larger microsome situated at the junction of the rays of the attraction spheres, in which these rays are inserted, and whose temporary function is to direct the chromosomes and in other ways accomplish caryokinesis.
/ hold, therefore, that the archosome, with its spheres and centrosomes, is not a temporary organ of the cell, but is a most important vital center, capable, under favorable con- ditions and when clothed with certain cytoplasmic envelopes, of growth, assimilation of food, and of movement — in fact, existing as an independent element of the blood.
Connected with this question is another of no less inter- est— the theory of symbiosis, of which Watas6 is now a well known champion. Is the nucleus an organ in the cell, or is it an independent organism — a messmate, so to say — which has associated itself with the balance of the cell for mutual benefit — in symbiosis? Watas£ has done full jus- tice to this theory which in many respects is a most plaus- ible one, and is according to my views most probable. I can do no better than to refer to his admirable lecture upon this subject. But strange to say, Watas£, who has so ar- dently advocated a symbiosis of the nucleus with the cell, has also as eagerly endeavored to destroy the very founda- tion upon which this theory must rest. Referring to his paper upon the independence of the centrosome, we find that he has used every argument to show that the centro- some is merely a large microsome, a variable organ in the cell. Now it can be shown, as I believe I have demon- strated, that the centrosome, with the archoplasmic spheres, is something much more than an organ in the cell, that it constitutes in reality the most vital part in the cell, the surviving center of energy after the connection with the nucleus has been dissolved. Is not this fact an additional support to the theory of symbiosis, a symbiosis not strictly between the nucleus and the cell, but between the nucleus and the archosome, or microcentrum? If we should extend
iris* intrri-* TXL-zinr v»s ni^nr :\"ih::lu:i; ma- :t*r:r* m» * — n- mrxsif tii:k. u;a:-.; in; ■:;nn"i«~inj: tsi-ti**? vtrt m:i*!7it!xu- iiir siifmunn*. :nr-iiii7>= 7*»m:Iia- -*: in* iii'::i;*ur am iiii Tiiusnii;- r^T*. j-^r :ir a? v; iin.'V u" in; Trs^t'.ir 11m; in;s« Tiia-im:- r;Ti*jf art nn nnr ni:it?tf:xi:ifm~ : uranisms* vm::i :t: n:r Ti:ss«*tss i mi^tf.uf. an: in; su£";:tt:oi :iiir in* -n:rr-nu::i;:in;-: uxu'-siRirr* -X :ixr lut-it.ai**.: "tils -frsi!Tii:.'ii*; iirfni sf. I mmi- *onr^7 -"iiusrmiir.'fct.
T~;m m* :ir>rii; — 11:1:112= :r I^;:i:i;nmaii :n nt* rmndiiii :r Hi*/ ::snr":*H:iiit*. r fi**;:ns tt": •:?*«»« ma" ins* :•:•:" jrili^vs nit sunt* irv vnci i;:a::n;?f mu: m; nu:;i^.iir :::m«< :m- r-:m i TumifUtf uiu ac : — rcnaim :ii.- —in i 7r-;—i:ms ""irnna-ni. TTnt :snnr;?f<:niii vriau mis a'vi •■; :r^":iii:T* z~:m i 7 '-*■"•- nis ::;nn-;w:mi* an n :r i-:m ar~ mien.*: mi* liar nurir iiiirn»tn t: in Tr-twrrr. jz -;a::: :»; m; ::us**. an: n n*~ :nin- .i:it iJ. miiniamjii* innr ma- vi" in;:i in; 11 if: 1-7 :x r—n- misss ixiusr nt tn*a:^L 2i:fv_;:a: :r 1 i^air- :i m; r-*n- :u:h3» v* vrniiii iu"r; a: i;:a.~ l ™i:r— — : — ii.ivim*. a~::n;-
simi*- mil ::a — :.*:mi; ::hl"i :"mi-i:ur:inr i:'vr^i in; : :ma-
11:11 :r m; 7ttn-.;-.:T :*jl in;*- mi- n:r. 7i;ri;a7>?. z*± :r f.iiua mniri-sni::*. :iir ;:a:a ii :?: nunn-iirrr :::nii-;Tii;:n::; 1: n;aa; ::±:1 in* .xxxnu.siubte vnnnir r. ~; n:*v i:n:»v iiur 11 1* a*:;i- TWinm h 1 ::rmi:)t*n* :i;n- \z 111-*- ::::-:n:r ::l:il:. 7 : vmr t^sssr urt th**h* nia*n»*nia;!ir :r ;:a::i inntr1 ~^iar i: in*:r 7ii7u:ip*mi!ii: iiv'-'^rninitnr.1 I: in; :.;^:T':.\*:»iitr"* c ::«; i.imi* 3amr"t ii* m; ::*nrr\\-vun; mi: yim :?:?: iii*r*;. :r iii"'; v:* *i: :i*uL vni l in-m;r ;-"n.:7i:?a2: :r in; :ir:;.\r~i- :=ui: :r in;
Z uvri Lrt:ar ^i:^1:?*: i::u" in; ;«,mi>":i;i*r* i: in; a:- firniiiin,,*rt mrin :r in; *:"■■::: r.->: mi; i:n: mi" r :'-:■:: a:'" :~LinL:; 3. Hi* {juxii; -*iian:ii v :::;. :•«:* l: :rr^: .nt r— a:i --iiirrrr \: lit* 17*1: n<: mi* :r ".' :nt : *;I a ^.:mzm. "'i-" -■^" -« "• Ti"'':^.:'^ nar. *»^sr"dH*:f: :r ::i* .■•■fhi -..•:?: «:i--*. ::»; a— .::!••>: mr n.i " :»;
:::mn»:s«: :r ^v: ii:.r::n: ::.:'»:;: *,i::::it:. :n* ::*'ir." * I'l-r.-";
mil 111* ::*nir:M--mi*. ;:*ip:: " v::i::: . LiT'M/t*::.' ■.■■■mi; •snretnr jr i*^a:r. nia^^rnii*::- ::i- ?::«'■* . vll n;:-; ::al
u=enni:a 1: in; i:-;-;^* l::«:»: r.:uu* :' ?.i\,m:.: ■•: .:«; .*:n:T'- rir"T L*:nni^:ni:n :r -;i; a* -.::i Jini.'-n. :i a: • >:i::t: :.;:!: villi;
48 CALIFORNIA ACADEMY OF SCIEXCES. [3D Ser..
centrosomes and other structures in the cell remained dor- mant and separate.
This connection of the archoplasms in two adjoining cells can, I think, best be explained as a necessary conjugation — a preparation for the final mitosis of the cell. It is well known that such conjugations of nuclei have been observed in infusoria, and it has been proven. to be an indispensable rejuvenation of the cell. The conjugation of the archo- plasms is probably a similar necessity — a rejuvenating pro- cess without which, perhaps, a degeneration of the archo- some would ensue. In connection with this I will also call attention to the observations of K. v. Kostanecki and A. "Wierzejski ("Ueber das Verhalten d. sog. achromatischen Substanzen," etc.), Arch. f. Mik. Anat., Bd. 47, 1896, Heft. 2). These investigators have shown how in the fertilized ovum of Physa the centrosome (archoplasm and centrosome) separates from the sperm nucleus and traverses the egg cell in a very independent way (figs. 3, 9, 12, 13). This shows according to my judgment that in the above case the archosome is to a great extent independent of the nu- cleus; but even a greater proof of the independence of the archosome is found in a most important treatise on the Spermatogese von Paludina vivipara, by Professor Leo- pold Auerbach (Jenaische Zeitschrift, Bd. 30, 1896, Heft. 4). This admirable memoir arrived just as I was reading the final proof of the present paper, and time will not per- mit of my making more than a review of the summary of results. Paludina vivipara produces two kinds of sperma- tozoa: one possessing the regular form and structure con- sisting of nucleus and cytoplasm, and probably archosome also; the other kind, which has been known as the worm- like form, contains no nucleus, but is composed principally of cytoplasm. It has the value of a Nebenkern. The chromosomes of the original nucleus are being destroyed or ejected, and the Nebenkern, together with part of the cyto- plasm, develops into an independent spermatozoon. In the figures (13a to q) of Professor Auerbach, I find nothing which resembles a centrosome or archoplasm, but I think
ZOOL.-VOL. L] EISEX-FLASMOCYTES. 49
there can be no doubt as to the homology of my archosome with his Nebenkern, and also with his wormlike sperma- tozoon. The absence of any differentiation of spheres in his Nebenkern is probably due to the stains used or to the fixatives, my own experience being that Saurefuchsin does not differentiate the archosomal spheres, but on the contrary covers up their finer structure. I am confident that with proper toluidine staining the wormlike spermatozoon of Paludina will show a structure not brought out by the coarser fuchsin; however, enough is shown in Auerbach's figures to satisfy me that we have here a real case of inde- pendence of the archosome, and I am confident that further investigation will show an undoubted homology of the worm- ' like spermatozoon of Paludina and the plasmocyte of the blood of Batrachoseps.
At the time I was reviewing the above mentioned mono- graph by Professor Auerbach, my attention was called to a most interesting paper on "The Sexual Phases of Myzos- toma" (Mittheilungen a. d. Zoolog. Station z. Neapel., Bd. 12, 1896, Heft. 2), by Dr. Wm. M. Wheeler.1 In this paper Dr. Wheeler figures certain bodies, living free in the body cavity of Myzostoma, which he describes, provision- ally at least, as parasitic amoebae, under the name of Amceba myzostomatis. The body of this amoeba is at various places produced into fine points, one of which is seen to penetrate the cytoplasm of the ovum of the Myzostoma. In other re- spects the parasite is entirely external to the ovum. The fine needlelike point of the amoeba pierces the ovum more or less deeply, and always from the side furthest away from the nucleus of the ovum, at least so it appears in the figures. The region of the cytoplasmic part of the ovum in the im- mediate vicinity of the inserted point exhibits a most re- markable radiation: as Dr. Wheeler says, "not unlike an astrosphere at the pole of a karyokinetic spindle.' ' This
1 1 am under great obligations to my friend Professor Herbert P. Johnson, of the University of California, for having attracted my attention to the remarkable amoeboid bodies described in Dr. Wheeler's paper.
50 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
astrosphere lies very nearly in the position which we should expect to find occupied by the archosome of the ovum. Occasionally amoebae were found which were not in the act of puncturing the ova; while in a few instances a single amoeba was seen which had two points, each entering the body of an adjacent ovum.
As to the structure of the amoeba, Dr. Wheeler says, "Each contained, besides a number of deeply staining ir- regular granules, a pale round body, which I hesitate to in- terpret as a nucleus, although it is certainly remarkable that no other structure comparable to a nucleus could be found in these amoeboid organisms, when they had been treated with such an excellent nuclear stain as Heidenhain's iron haematoxylin." I think that the absence of a nucleus indi- cates that this body is not an amoeba, but something entirely different; and the question now arises, must we not in this so called amoeba recognize a free and independent archo- some? The pale round body would then be interpreted as a somosphere, and the irregular, deep-staining bodies as centrosomes. An objection to this interpretation of the dark, irregular granules might be made on account of their position, situated as they are outside of the somosphere; but this may be only a temporary position such as occurs also in the plasmocyte, where the centrosomes now and then are found outside of the somosphere, being free in the centro- sphere. In the ova figured by Dr. Wheeler we find no trace of any other archosome, but are told that the cyto- plasm of the ova attacked by the amoeba contained large granules which are larger and more numerous than those which occur in the normal ovum at about the same stage. These granules take up the haematoxylin with avidity. Judging from the figures (Taf. 10, fig. 23, and Taf. 12, fig. 56) these granules in the ova are exactly similar to those in the resting amoeba (fig. 56). In the entering amoeba these granules as well as the pale round body are absent. It ap- pears to me as if the pale round body, or somosphere, and the granules, or centrosomes, had been injected into the ovum by the free archosome. If this interpretation of the
Zool-Vol. L] ErSEN—PLASMOCYTES. 5 1
nature of this amoeboid body is proved to be correct by future investigation, then we shall have here the third known instance of a free and independent archosome, the other two being the plasmocyte in the blood, and the wormlike sperma- tozoon of Physa.
If we now consider the budding of the centrosomes as shown by Heidenhain, and that of the plasmocyte as shown by me, does it not demonstrate that the centrosphere and centrosomes are in reality distinct and independent elements, though as yet we cannot in all cases know them to be in- dependent of each other? Does it not appear also possi- ble that these two structures once existed separately but later on joined in a symbiotic existence, long before the archosome as a whole had joined the caryosome and cyto- some to make up the present cell?
Identification of the Spheres. — I believe it will prove of interest to attempt an identification of the cytoplasmic zones of the plasmocytes and plasmocytoblasts with those observed in perfect cells. Such identification is for several reasons by no means easy. Many investigators have not named the respective cytoplasmic zones observed by them, and, in cases where names have been given, they have frequently used descriptions or descriptive names which are not trans- latable from one language to the other. Another obstacle is found in the different stains used to differentiate the respective zones. The various fixatives which are supposed to preserve the elements of the cell in their original appear- ance undoubtedly frequently accomplish the very opposite, at least with certain parts. Every cytologist knows only too well how differently the stains act after different fixatives have been employed. I will not dwell particularly upon the advantages of the methods I have employed in my investi- gations in this case, but will observe only this, that whatever changes the cytoplasmic zones may have undergone, they are not resultant from the use of violent chemicals.
There is no reason to suppose that we should find a similar grouping of cytoplasm in every cell, even when the cells belong to the same cell species, but in related
52 CALIFORNIA ACADEMY OF SCIENCES. [3d Ser.,
cells we must expect to find the same general cytoplasmic characters. While the leucocytes have received much at- tention from a host of investigators, the finer structure of the erythrocytes has hardly been touched upon. This is greatly due to the interference of the haemoglobin, which does not permit the ready staining of the cytoplasm in any way that would permit a study of the finer details. Heidenhain's fig. 16 ("Neue Untersuchungen") demonstrates this. We see polar projections resembling plasmocytoblasts at each end of the nucleus, faintly visible at the upper end, a little more sharply defined below. The two centrosomes at one pole are strongly brought out, but the respective spheres are not to be seen.
Adolf Dehler, who has made a careful study of the centro- somes of the chicken erythroblasts, gives us as little informa- tion on this particular subject as does Heidenhain. His figures show only sharply defined centrosomes surrounded by a light colored circular sphere. MacCallum's figures show few if any exact details, except the mere outlines of the frayed plasmosphere. E. J. Clay pole gives no details of any kind. In our comparisons we must, therefore, turn to other cells, among which there are few which have been more carefully studied than have the giant cells from the bone marrow of the rabbit by M. Heidenhain; and, as re- gards the cytoplasmic parts they show several points of similarity to our plasmocytes and plasmocytoblasts, we will consider them more particularly.
The concentric arrangement of the cytoplasm, so forcibly pointed out by Heidenhain and Lenhoss^k, is equally dis- tinct in our plasmocytes. As is known, Heidenhain recog- nizes three distinct zones of ectoplasm, and, similarly, three distinct zones are found in the ectoplasm of the plasmo- cyte. How far Heidenhain's outer, inner, and middle lay- ers correspond with the three outer spheres of the plasmo- cyte is more difficult to determine.
Judging from the form, situation, and staining quality, I believe it safe to identify the outer zone of Heidenhain's ectoplasm with my plasmosphere. The identity of my hya-
Zool.-Vol. I.] EISEN—PLASAfOCYTES. 53
losphere with Heidenhain's middle zone is, however, not apparent, and is rather improbable. This middle zone stains strongly and is plainly granulated, while my hyalosphere stains poorly and is characteristically homogenous, appear- ing like an even, pellucid ring. When we turn to the grano- sphere we can hardly identify it with Heidenhain's inner zone of ectoplasm, this zone staining faintly, while my granosphere stains deeply. If, again, we consider the po- sition of the inner zone, we find that it surrounds the nucleus, being in actual contact with it, unlike my granosphere but similar to part of my hyalosphere. On account of its stain- ing quality and general appearance I think that my grano- sphere may be more properly identified with Heidenhain's middle zone. If this is so it is probable that my hyalosphere was originally confined to the vicinity of the nucleus but later pushed itself between the plasmosphere and the grano- sphere. Heidenhain's endoplasm cannot be taken into con- sideration as it is only an invagination of the ectoplasm. The innermost spheres in the plasmocyte I can compare only to Heidenhain's microcentrum, though it must be con- ceded that the analogy is not absolutely certain. Heiden- hain understands by his microcentrum, not only the cen- trosomes but also the particular substance which surrounds and converts them, the whole forming a distinct body by itself. This connecting substance — " primary centro- desmose ' ' — must be referred either to my somosphere or centrosphere ; or, what is more probable, sometimes to the one and sometimes to the other. In this paper when refer- ence is made to the microcentrum it must be distinctly understood that I leave this point undecided. By micro- centrum I mean the centrosomes together with the nearest visible sphere surrounding them and connecting them with each other. In Heidenhain's figures the divisions of the microcentrum are less distinct from each other than they are in the plasmocytes. The distinction between the grano- sphere and the centrosphere is always very good, especially in successful stains with eosin, as shown in figs. 49, 59, or even 48; and even with toluidine the differentiation is
P&oc. Cal. Acad. So., 3D Sbr., Zool., Vol. I. (4) Oct. », 1896.
54 CALIFORNIA ACADEMY OF SCIENCES. [3D Sbr.,
often striking, as, for instance, in fig. 65. The somosphere is probably identical with the faintly staining sphere sur- rounding the microsomes in Heidenhain's figs. 37, 39, 48, 49> 53> etc.
If we compare our cytoplasmic spheres with the gang- lionic cells of the frog, as described by Lenhoss^k, we find several points of similarity. Lenhoss£k endeavored to harmonize the cytoplasmic spheres seen by him with those of other investigators, and met I think with some success. He divides the cytoplasm into two main divisions; an outer one, for which he proposes the name plasmosphere or perisphere, and an inner division which he refers to as centrosphere, the latter being sharply defined like a second nucleus. This agrees exactly with the centrosphere of the plasmocyte, and I have for this reason adopted the name proposed by Len- hoss^k for this highly individualized part. The centro- sphere of Lenhoss£k is frequently surrounded by a white ring, a fact which I, too, have observed at times in the plas- mocytes. A study of Lenhoss^k's figs. 5, 6, 7, and 9 shows that the minute centrosomes are surrounded by a dark film which I identify with my somosphere. As regards the plasmosphere of the ganglionic cells the agreement is less apparent, except that the cytoplasm is prominently concen- tric and consists of at least two, probably three, dis- tinctly differentiated zones, which, however, are not strictly localized. Of these the inner zone is granulated, while the middle one (LenhossSk, Tafl. xv, fig. 3), stains even more intensely than the granosphere of the plasmocyte. In the ganglionic cell as well as in the plasmocytoblast it is the "grosse gekornte Protoplasmaschicht ' ' which causes the dell in the nucleus, a fact which when coupled with other similarities warrants my identifying these two zones with each other.
In one of the very latest contributions to cytology, Dr. O. Van der Stricht has described cell structures which bear a marked resemblance to those of the plasmocytoblasts. In his figs. 19, 20, 21 and 22, but especially in 19 and 20, we find a most striking arrangement of the cytoplasm. It be-
Zool.— Vol. I.] EISEK—PLASAfOCYTES. 55
comes at once evident that the dark-staining granulated zone can be identified with my granosphere. It possesses the fol- lowing qualities in common with the granosphere : It stains darkly; it is granulated; it encloses the microcentrum ; and it is crescent-shaped, the cavity of the crescent joining a light colored zone, which latter partly or entirely surrounds the nucleus, as does my hyalosphere. The microcentrum also is in almost every particular similar to the archosome of the plasmocytoblast. In one of the figures (21) the granosphere is seen to cause a dell in the nucleus, a charac- teristic which renders nearly perfect the similarity. We must suppose that in these cells the hyalosphere if pres- ent is confined to the immediate vicinity of the nucleus, where it is recognizable as a faintly stainable ring. In the paper referred to (pages 257, 258) Van der Stricht has undoubtedly recognized the great importance and promi- nence of the granosphere, stating that the first modifications apparent in the microscope take place in the very compact cytoplasm which surrounds the attraction sphere. He further states that this "granulated zone" probably corresponds to the asteroid region surrounding the attraction sphere in the egg, or to Heidenhain's radiating organic fibers, with which I am ready to agree. This zone has also been recognized and figured by Hermann, A. Prenant, Holl, F. Heneguy, Van Bambeke, and H. Mertens, as was pointed out by Van der Stricht. Among other recent investigators who have recognized the importance, or rather the exist- ence, of the archosome, R. von Erlanger takes a prominent place. In his paper on the testes of the earthworm he de- scribes and figures the archosome and granosphere which he designates as Nebenkern. He recognizes the centro- some as being situated in this body and ascribes to it a dis- tinctly cytoplasmic origin. While this Nebenkern and my archosome and granosphere are undoubtedly homologous, it must be remembered that the Nebenkern, or paranucleus of some other investigators, signifies bodies of a very differ- ent nature and sometimes even foreign to the cell. Erlanger compares his Nebenkern in the sperm cells with similar
56 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
structures described by Butschli and v. la Valette, and calls attention to the great resemblance between them. From this we may conclude that of the cytoplasmic spheres de- scribed by me above, the granosphere is the most constant, probably existing in all the cells.
The large granular spheres which have been described by Meves from the achilles tendon of the frog must also be considered identical with some of the spheres of the plasmocyte. Meves describes the large outer sphere — my granosphere — as concentric layers of indistinct granules which might be considered as membranous formations sur- rounded by thin homogenous cytoplasm. There can be lit- tle doubt that the above structure is identical with my granosphere, in which similar concentric layers are fre- quently observed (Meves, Taf. ix, fig. 2, etc.). In his fig. 10 there appears a pale uncolored sphere surrounding the centrosomes (but inside the centrosphere), which probably corresponds to my somosphere. We must remember that Meves used iron stain which does not differentiate as well as the toluidine.
Especially as regards the somosphere, I think that future investigations will demonstrate its presence in the micro- centrum of many cells, and that many structures which have been described as centrosomes will, when subjected to closer examination, be referred to the somosphere; for in- stance, the branched centrosomes in the pigment cells described by Zimmermann. A real somosphere has been observed by Hacker in the winter egg of Sida crystallina (loc. cit., Taf. xxi, fig. 1), which he calls after Strassburger " tingirbare innenzone." The increase in size and growth of the centrosome in Sida is also pointed out by Hacker, a growth which is probably analogous to the growth of the somosphere in the plasmocyte.
Whether there exists any homology between the plasmo- cytes and the paranuclei described by Bremer from the blood of Testudo Carolina must remain undecided. Judg- ing from the figures accompanying Bremer's paper, as well as from his descriptions, the paranuclear bodies are much
Zool.— Vol. I.] EISEN—PLASMOCYTES. 57
less regular than any of the plasmocytoblasts from the blood of Batrachoseps. How much of this is due to the fixing chemicals used in Bremer's preparations cannot be known until comparative studies have been made with non-fixed blood. The peculiarly shattered nuclear structures would, it seems to me, certainly indicate that the paranuclear struc- tures also had been considerably disturbed before finally being fixed. A further study of these paranuclei is cer- tainly of the highest importance. I have seen somewhat similar bodies in the blood cells of Diemyctylus and Chon- drotus, but not in Batrachoseps.
The peculiar structureless bodies described by Rawitz from the lymphatic gland of Macacus (Arch. Mikr. Ana- tomie, Bd. 45, page 592) are possibly of the same nature as my archosomes or plasmocyte. The want of structure may be due to imperfect fixing methods employed. These bodies occupy the same position in the cell as the Neben- kern. The fact that they are set free in the lymph and prob- ably reach the general circulation makes an homology between them and the plasmocytes probable.
Plasmocyte and Leucocyte. — It is appropriate that we should compare the plasmocyte with the leucocyte in the same blood, and search for some similarities as regards the inner spheres. At the outset we find some very marked differences pertaining to their staining qualities. Thus we find that the Ehrlich-Biondi mixture, which is the most suc- cessful stain to bring out the microcentrum of the leucocyte, fails entirely to give any satisfactory images of the corre- sponding parts of the plasmocyte; and, vice versa, the toluidine does not stain the microcentrum of the leucocyte, neither the spheres nor centrosomes. Another striking differ- ence between the two kinds of corpuscles is the absence of cytoplasmic rays in the fusiform corpuscle. Although care- fully looking out for any trace of them I have failed to find them in the plasmocyte. Now and then a star-shaped grano- sphere occurs, both in the plasmocyte and in the plasmo- cytoblast, but they are quite different from the filaments in
$8 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
other cells and do not appear to have any other function than that of locomotion.
The dark-staining, starlike sphere of the leucocyte, as seen in figs. 14, 15, 16, 18, and 19, is probably homologous with the granosphere of the plasmocyte ; it is the sphere which causes the dell in the nucleus of the leucocyte. The inner pale sphere surrounding the centrosomes in the leucocyte would then correspond to the centrosphere of the plasmo- cyte. In leucocytes stained with toluidine the granosphere is only brought out by several hours of exposure to the stain. This exposure will invariably injure the differentia- tion of the plasmocytes and plasmocytoblasts but it will bring out the granosphere most beautifully, as seen in figs. i6£, 18*, and 190. It will be seen that a ray extending from the archoplasm is covered by different microsomes at different points in the cell. Thus the innermost microsomes consist of particles of granosphere, while the outer ones consist of granules of plasmosphere or blue-staining cyto- plasm.
In fig. 18* the archosome has divided into several smaller semiglobular parts, but the toluidine has not differ- entiated any of its zones nor stained the centrosomes.
Unclassified Corpuscles in the Blood. — Among the plasmo- cytes I frequently find spherical or oval bodies as large as the smallest plasmocytes. They resemble small nuclei, and when stained with toluidine are semitransparent, with darker streaks like marbled veins. They contain neither layers, spheres, nor globules, and their nature is doubtful.
Summary.
1. The erythrocytes in the normal blood of Batracho- seps vary greatly in size and shape, much more so than those of any other known animal. They are in this respect entirely unique. The smallest are smaller than the red cor- puscles of the human blood, while the largest are seven times their diameter. This refers not only to the nucleated, but also to the non-nucleated erythrocytes.
Zool.— Vol. I.] EISEN—PLASMOCYTES. 59
2. The vast majority of the erythrocj-tes are not nu- cleated, hardly any being found in the spring of the year. In the summer and autumn they are more numerous than at any other time. In this, also, Batrachoseps stands alone; all other batrachians possessing only nucleated red blood cells, at least according to our present knowledge.
3. A perfect nucleated erythrocyte of the blood of Batrachoseps consists of three distinct and separately or- ganized parts, which, however, are not of equal importance in the general make-up of the cell. These parts are the Cytosome, consisting of three cytoplasmic spheres — plas- raosphere, hyalosphere, and granosphere; the Archosome, consisting of three archoplasmic spheres — centrosphere, somosphere, and centrosomes; and the Caryosome, or nucleus.
4. The observed facts further verify the theory that the fusiform corpuscles are the remains of nucleated erythro- cytes which for some reason have lost their cell wall and haemoglobin. This fusiform corpuscle is thus newly ejected nucleus to which is yet attached most of the cyto- plasm of the cell.
5. At and after the separation of the fusiform element from the rest of the cell the archosome remains in the fusi- form element until the plasmocyte is formed.
6. The origin of the fusiform corpuscle is due to a de- fect in the nucleus, or more particularly in the chromo- somes. This defect has prevented the chromosomes from assuming the preliminary skein stage preparatory to divis- ion. The archosome, which has already entered upon the second stage of mitosis, having divided and moved to oppo- site poles, cannot, therefore, conclude the process, the energy expended effecting only the rupture of the cell mem- brane, thus setting free the fusiform element.
7. The fusiform corpuscle consists of the nucleus and one or two plasmocytoblasts,1 each one of which consists of six cytoplasmic zones, including centrosomes.
1 In the foregoing, and throughout this paper, I have used the word plasmocyto- blast in order that I may be clearly understood; but for the sake of brevity I propose that this word be made simply plasmoblast.
60 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
8. The plasmocytoblast is capable of development and division into two, three, or more plasmocytes, which finally separate from the nuclear vicinity, each one forming a free and independent active plasmocyte in the blood serum.
9. The plasmocyte undoubtedly survives in this state for a considerable length of time and must take rank with the other principal and active corpuscles in the blood, the erythrocytes and leucocytes.
10. The plasmocyte is capable of growth through assim- ilation of food and exhibits phagocytosis. It contains the same cytoplasmic zones as are found in the plasmocyto- blast but has no nucleus.
11. The determining part of the plasmocyte is the arch- osome, or centrosomes with spheres, as without them the plasmocyte is not formed. Imperfect plasmocytoblasts, or such as do not possess an archosome, do not develop plas- mocytes.
12. The non-existence of attraction rays is probably explained by the absence of a cell wall.
13. The nucleus of the fusiform corpuscle goes quickly to decay, no part of it surviving in the plasmocyte.
14. The archosome is not merely an organ in the cell — a large microsome formed for the special purpose of mitosis — but constitutes a most important element of it, the very center of organization, equal in importance to the nucleus itself. The archosome can no more be said to originate in the cytoplasm than it does in the nucleus. The only con- nection the archosome has with the cytoplasm is that of resting in it and of being partly nourished by it. That the centrosome is at times found in the nucleus, as has been shown by Brauer and Lauterborn, does not imply that it originates in the nucleus, but simply that in some in- stances it temporarily resides there. The archosome pos- sesses the same individuality whether residing in the cyto- plasm or in the nucleus.
15. The survival of the archosome, with its phenomena of growth and phagocytosis, and its general independence, suggests a symbiosis between the archosome, the caryo-
Zool.— Vol. IJ EISEN— PLASMOCYTES. 6l
some , and the cytosome; but of course I do not claim that it proves it. The effect of such a symbiosis would be, among other things, to create a cell membrane and cell division by mitosis. I favor this theory because it explains the survival of the archosome (surrounded by cytoplasmic envelopes) as an independent corpuscle.
16. The new plasmocyte in the blood of Batrachoseps may then be defined as a corpuscle, generally without a cell wall and always without a nucleus; but consisting of the archosome, which has surrounded itself with the three outer spheres of cytoplasm, and which survives as an independ- ent corpuscle in the blood serum. It is capable of growth and assimilation of food, and to some extent of amoeboid movements. The archosome itself contains three separate spheres, inclusive of the centrosomes. The plasmocyte possesses the following properties: organization, growth, assimilation of food through phagocytosis, motion, both as a whole and by the individual inner spheres, and, finally, sensitiveness, shown in selecting a certain quality of food
" (erythrocyte fragments and bacteria) .
17. The plasmocytes are derived from the red blood cells and not from the leucocytes. The archosome of the leucocytes does not survive, but disintegrates at the same time as the balance of the leucocyte.
18. The granosphere is the seat of phagocytosis and must be considered as the digestive organ of the cell and the plasmocyte, and the storehouse for accumulated food. The somosphere is probably the assimilative organ of the archosome and especially of the centrosomes. The hyaline globules frequently found in the somosphere I consider as food supply. The process of assimilation would then be as follows : The food, derived from the blood serum and through phagocytosis, is digested and assimilated by the grano- sphere for the benefit of the plasmocytes at large. This food supply when accepted by the archosome is further digested by the somosphere for the special benefit of the centrosomes.
From this we may conclude that the granosphere is a con-
Pmoc. Cal. Acad. Sci., 3D Skjl, Zool., Vol. I. (5) March 31, 1897.
62 CALIFORNIA ACADEMY OF SCIENCES. [3D Sbr.,
stantly recurring zone in the cell and that it is quite easily recognized. Wherever it occurs it seems to possess the same characteristics as regards form, granulation, and stain- ing capacity, and it appears to be a most important element of the cell, probably of secretive, digestive, and assimilative function. The archosome has been recognized in a great number of cells and parts of it have been described under several names: first, I believe, as Nebenkern; later on, as microcentrum, etc. The centrosphere has been ob- served in many cells and is variously termed archo plasm, attraction sphere, centrosphere, etc. The somosphere has been figured as surrounding the centrosomes, but previous to this it has not been named nor has any function been assigned it.
Zool.-Vol. IJ EISEN—PLASMOCYTES. 63
MEASUREMENTS OF THE CORPUSCLES AND THEIR SPHERES, CALCULATED BY PROF. GEORGE OTIS MITCHELL.
NUCLEATED ERYTHROCYTES.
Large nucleated erythrocyte 24. 9 mm.
Round nucleus of erythrocyte 18.26 mm.
Large round nucleated erythrocyte 33. 2 mm.
NON-NUCLEATED ERYTHROCYTES.
Long form 46.48 mm.
Average round 16. 6 mm.
Small round 9.96 mm.
Very small 6.40 mm.
FUSIFORM CORPUSCLES.
Nucleus of corpuscles 24.9 mm.
Cytoplasmic projection, or plasmocytoblast :
at upper end 4. 15 mm.
at lower end 2.49 mm.
Very large nucleus, almost square 19.92 mm.
Plasmocytoblast of fusiform corpuscle 9.96 mm.
LEUCOCYTES WITH POLYMORPHOUS NUCLEI.
Large size 33. 2 mm.
ROUND MONONUCLEARY LEUCOCYTES.
Average 12.28 mm.
EOSINOPHILS LEUCOCYTES.
Large, but not largest 19.92 mm.
Average 14.94 mm.
PLASMOCYTES.
Average, with two microcenters 6.64 mm.
Average, with one microcenter 6.60 mm.
Large 8. 3 mm.
A very long one: Plasmosphere 14.94 mm.
Centrosphere 9.96 mm.
Ovoid form: Plasmosphere 8.30 mm.
Centrosphere 4.98 mm.
Small: Plasmosphere 3.32 mm.
Centrosphere 1.66 mm.
Large round Plasmocyte: Centrosphere 4.98 mm.
Outside of plasmosphere 14.11 mm.
Outside of hyalosphere 8.30 mm.
Total 14.11mm.
Large ovoid: Plasmosphere 11.62 mm.
Hyalosphere 8.30 mm.
Centrosphere 4.81 mm.
64 CALIFORNIA ACADEMY OF SCIENCES. [3D See.,
BIBLIOGRAPHY.
1896. Aurrbach, Leopold. Untersuchungen uber die Spermatogenese
von Paludina vivipara. Jenaische Zeit., Bd. XXX, p. 405. 1893. Bambeke, Ch. Van. Contributions a Phistoire de la constitution de
Pceuf. Arch, de Biol., Tome XIII, p. 89, PL VI. Compare figs.
21, 22, 27, etc. 1893. Braurr, August. Zur Kenntniss der Spermatogenese von Ascaris
megalocephala. Arch. Mikr. Anat., Bd. XLII, p. 153. 1895. Bremer, Ludwtg. Ueber das Paranudearkdrperchen der gekernten
Erythrocyten, nebst Bemerkungen liber den Bau der Erythrocyten
im Allgemeinen. Arch. Mikr. A not., Bd. XLV, p. 433. 1871. Butschu, O. Vorlfiufige Mittheilung liber Bau und Entwicklung der
Samenfaden bei Insecten und Crustaceen. Zeit. Wiss. Zool., Bd.
XXI, p. 402. 1893. Claypole, Edith J. An Investigation of the blood of Necturus and
Cryptobranchus. Proc. Amer. Micr. Soc., Vol. XV, pp. 39-71.
1895. Dehler, Adolf. Beitrag zur Kenntniss des feineren Baues der
roten Blutkorperchen beim Hlihnerembryo. Arch. Mikr. Anat., Bd. XLVI, p. 414.
1896. Erlanger, R. v. Zur Kenntniss des feineren Baues des Regenwurm-
hodens und der Hodenzellen. Arch. Mikr. Anat. , Bd. XLVI I, p. 1.
1892. Griesbach, Hermann. Ueber Plasmastructuren der Blutkorperchen
im kreisenden Blute der Amphibien. Fest. z. Siebenzigsten Ge- burt stage Rudolf Leuckarts, Leipzig, p. 215.
1893. Hacker, Valentine. Ueber die Bedeutung der Centrosomen. Arch.
Mikr. Anat., Bd. XLII, p. 311.
1894. Heidenhain, Martin. Neue Untersuchungen tiber die Centralkor-
per und ihre Beziehungen zum Kern- und Zellenprotoplasma. Arch . Mikr. Anat., Bd. XLIII, p. 423.
1892. Ueber Kern und Protoplasma. Leipzig.
1893. Henneguy, L. Fel. Le corps vitellin de Balbiani dans Pceuf des Ver-
tfbres. Journ. de I* Anat. etdela Phys., Ann. XXIX, p. 1. 1 891. Hermann, F. Beitrag zur Lehre von der Entstehung der karyokine-
tischen Spindel. Arch. Mikr. Anat., Bd. XXXVII, p. 569. 1890. Holl, M. Uber die Reifung der Eizelle des Huhn's. Sitz. K. Akad.
der IViss., Bd. XCIX, Abth. Ill, p. 311. Compare fig. 3. 1896. Kostanecki, K. v. and A. Wierzejski. Ueber das Verhalten der
sogen. achromatischen Substanzen im befruchteten Ei. Arch.
Mikr. Anat., Bd. XLVII, p. 309.
1894. Lauterborn, Robert. Ueber Bau und kerntheilung der Diatomeen.
Verh. d. Natur. Med. Ver. zu Heidelberg, N. F., Bd. V, p. 179.
1895. Lenhossbk, M. v. Centrosom und Sphare in den Spinalganglienzel-
len des Frosches. Arch. Mikr. Anat., Bd. XLVI, p. 345. 1864. Mertens, H. Recherches sur la signification du corps vitellin de Balbiani dans 1 'ovule des mammif&res et des oiseaux. Arch, de Biol. , Tome XIII, p. 389.
Zool.-Vol. IJ EISEN—PLASMOCYTES. 65
1895. Meves, Fr. Ueber die Zellen des Sesambeins in der Achillessehne des Frosches (Rana temporaria) und tiber ihre Centralkdrper. Arch. Mikr. Anal., Bd. XLV, p. 133.
1894. Ueber eine Metamorphose der Attractionssphkre in den Sperma-
togonien von Salamandra maculosa. Arch. Mikr. A not., Bd. XLIV, p. 119. Compare figs. 23, 24, 25, 26, 27, 38, 51, 64. 1891. Prenant, A. Le corpuscule central d'Ed. Van Beneden dans les cellules seminal de la scolopendra. C. R. Soc. de Biol.
1895. Stricht, O. Van der. Contribution & l'&ude de la forme, de la struc-
ture et de la division du noyau. Arch, de Biol. , Tome XIV, p. 257.
1895. Rawitz, Brrnhard. Centrosoma und Attractionssphare in der ruh-
enden Zelle des Salamanderhodens. Arch. Mikr. Anal., Bd. XLIV, p. 555. Compare fig. 3, which shows the granosphere.
1896. Untersuchungen fiber Zelltheilung. Arch. Mikr. Anal., Bd.
XLVII, p. 159. 1886. Vallettb St. George, v. ul. Spermatologische Beitrage. II.—
Blatta Germanica. Arch. Mikr. Anal., Bd. XXVII, p. 1. 1890. Watasr, S. On Caryokinesis. Biol. Lee. Marine Biol. Lab. of
Wood's Holl, p. 168.
1893. On the Nature of Cell Organization. Biol. Lee. Marine Biol.
Lab. of Wood's HoU, p. 83.
1894. Origin of the Centrosome. Biol. Lee. Marine Biol. Lab. of
Wood's Holl, p. 273. 1893. ZiMMBRMANN, K. W. Studien tiber Pigmentzellen. Arch. Mikr. Anal., Bd. XLI, p. 367.
66 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
EXPLANATION OF THE FIGURES.
Figures 1-47, 50-78 have been drawn from Zeiss tV hom. im., with oil im- mersion contact between the Abbe condenser and the underside of the slide. Ocular 2. 3. 4. Figures 44, 48, 79-84 have been drawn from Zeiss apochrom. 3 mm. Ocular 12 and 18. Both daylight and gaslight were used. The images were projected on the working table, outlines drawn with camera, details filled in. They are from cover-glass preparations fixed with absolute alcohol.
All the figures are from elements in the blood of Batrachoseps attenuates.
PLATE I.
NON-NUCLEATED ERYTHROCYTES.
i to 6. Six non-nucleated erythrocytes of different forms and sizes. Ehrlich-Biondi. Oc. 2.
NUCLEATED ERYTHROCYTES.
7. Large round erythrocyte with distinct bipolar nucleus. 8-9. Smaller erythrocytes with oblong nuclei.
10. Large erythrocyte with large nucleus. Above four, Ehrlich-Biondi.
11. Round erythrocyte, Ehrlich's neutrophil stain.
EOSINOPHILS LEUCOCYTES.
12. Average size. Ehrlich-Biondi. Oc. 3.
13. Larger. Ehrlich-Biondi. Oc. 3.
POLYMORPHOUS LEUCOCYTES.
All stained with Ehrlich-Biondi and washed with oxalic acid solution in water. (Except 186 and 19, which are toluidine stains.) Oc. 3.
14. The granosphere is starlike and encloses an archosome with two cen- trosomes separated by a light colored bar.
15. The granosphere is less distinct, and the archosomes are entirely sep- arated, one containing two, the other one centrosome.
16. The granosphere is small but distinct, rounded in outline. The two archosomes are separated.
17. Granosphere not distinct. Three archosomes, one of which has two centrosomes.
i&j. The center of a leucocyte; the nucleus is not figured; the granosphere is starlike; the archosome is round, with three centrosomes of unequal size. Ehrlich-Biondi.
18*. The center of a leucocyte; the nucleus not figured; the granosphere is stained pink; the cytoplasm is bluish; the archosome is not entire, but is broken up into several semiglobular zones; centrosomes not stained. Tolui- dine.
i8r. A leucocyte with three separate archosomes. Ehrlich-Biondi.
19a. A large polymorphous leucocyte; granosphere is stained pink; two centrosomes. Toluidine.
19*. A large leucocyte with pink granosphere. Centrosome not stained. Toluidine.
20. A mononucleary leucocyte stained with Ehrlich-Biondi, the deep stain- ing showing presence of haemoglobin.
ZOOL.-VOL. I J EISEN—PLASMOCYTES. 6 J
FUSIFORM CORPUSCLES WITH PLASMOCYTOBLASTS— 21 TO 39.
All are stained with toluidine, excepting fig. 36, which is stained with Ehrlich-Biondi. In the following the plasmocytoblasts will be referred to simply as pcb.
21. The nucleus is rounded and in a fair state of preservation as regards the outline. There is only one pcb. at the upper pole (a), showing the spheres arranged as cones, one above the other. The smallest and lowest down is the somosphere, in which are seen three dark separated centrosomes. All the spheres, except the plasmosphere and hyalosphere, are stained violet, the latter are blue or bluish. At the lower pole (6) is seen a crescent-shaped frag- ment of granospheroplasm, also stained violet The plasmosphere and hyalo- sphere extend around the left side of the nucleus, but cannot be traced along the entire right side. Oc. 3.
22. The nucleus shows advanced degeneration; the plasmosphere and hyalosphere extend all around. At the upper pole the archosome has ad- vanced far upward into the crescent-shaped large granosphere. Here is seen one centrosome. At the lower pole an indistinct archosome is visible. The nucleus shows a distinct polarity, the pole being marked with a white spot Oc 2.
23. The nucleus shows a strong polarity. At the upper pole is a large pcb. in which the granosphere is very narrow, stained violet The whole archosome is far advanced. The centrosphere is almost unstained. The somosphere is large, with two distinct centrosomes. At the lower pole there is a crescent-shaped granosphere without archosome. Oc 2.
24. The nucleus is in an advanced state of degeneration. A plasmocyte. is nearly ready to separate at the upper pole, having left a crescent-shaped residue of granosphere. A similar crescent of granosphere is seen at the lower pole, but without archosome. Oc. 2.
25. The nucleus is in an advanced stage of disintegration. The upper pole contains a pcb. which has changed into a plasmocyte which is almost ready to separate. The lower part of the pcb. has not yet perfected the outlines of the spheres. There is no residue of granosphere, and only one archosome. At the lower pole there is no pcb. nor any residue of grano- sphere.
26. The nucleus is rounded and swollen, with indications of a strong polarity. The upper pole contains one pcb. with distinct centrosomes and a starlike centrosphere. Plasmosphere and hyalosphere indistinct
27. The nucleus is in fair state of preservation, showing a polarity. At the upper pole is a very large pcb., with the archosome far advanced towards the apex. The plasmosphere and hyalosphere are distinct all around the corpuscle. Oc. 3.
28. There are three archosomes in the pcb. at the upper pole, and one indifferently developed pcb. at the lower pole. Oc. 2.
29. There is a pcb. at each pole. The archosome is far advanced, and the spheres are differentiated. Each archosome contains two centrosomes.
30. Very similar to fig. 29, but the centrospheres are stained deep blue, while in the former they were white.
31. There is a pcb. at either pole. The archosome at the upper pole
68 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
is more advanced than the one at the lower pole. The two outer spheres are not well differentiated. Oc. 2.
32. There is a pcb. at each pole. The granosphere is everywhere violet In the upper pcb. it is cone-shaped, the archosome is moving upwards. The centrosphere is pale white and very narrow. The somosphere is deep blue, with darker centrosomes. The lower pcb. is much further advanced than the upper one. The granosphere is starlike, and the centrosphere is more dis- tinct than the one in the upper archosome. The upper archosome contains three centrosomes, the lower one apparently but one.
33. The pcb. are very large, but the inner spheres are not well differen- tiated. In the upper pcb. (a) there is only one archosome, while in the lower (b) there are three archosomes. Oc. 4.
34. Two pcb., one at each pole. The one at the upper pole is more advanced than the one at the lower pole. The centrospheres are white, the somospheres blue, and the centrosomes dark blue. Oc. 4.
35. There are two pcb., one at each pole. The granospheres are light violet, the centrospheres deeper violet, while the somospheres are pale. In the upper somosphere we see two separated centrosomes, while in the lower one (t>) they are too close to be distinguished. Observe the star-shaped cen- trosphere at the upper pole, while the one at the lower pole is conelike. The former is probably an indication of amoeboid movement.
36. There are two distinct pcb., but only the upper one is of normal size. This one contains three separate archosomes, each surrounded by a concentrating granosphere, and each with a single centrosome. The lower pcb. has spread out, enclosing about half the circumference of the nucleus. In this lower pcb., also, the archosomes have separated, each having a single centrosome. The pale spheres are probably centrospheres. The nucleus is in fair preservation, showing a strong polarity. Ehrlich-Biondi. Oc. 3.
37. This is a very large nucleus with a dissolving pcb. at the pole. Probably the two poles have been brought together, at any rate, the appear- ance is a very irregular one. There are three growing plasmocytes in various stages of development. The lowest one is almost ready to separate as a fully developed plasmocyte, containing apparently three archosomes. The young plasmocyte to the left is the least advanced of all; the nucleus is in a high state of disintegration and unusually swollen. Oc. 3.
38a. In the fusiform corpuscle the lower pcb. has almost separated, and the independent plasmocyte is all but ready. It is connected with the fusi- form corpuscle by a long narrow shaft of plasmosphere. It contains two archosomes, the lower one of which is much the larger. The granosphere is narrow, and differentiated deep blue. The centrosphere is very large and pale blue, containing a paler white zone with a few darker rays and a few small central bodies. The large inner white zone is not easily explainable, except as a part of the centrosphere. The dark round spot at the lower margin is also of doubtful nature. The smaller archosome contains a distinct centrosome. There is a residue of crescent-shaped granosphere at the lower pole. The upper pcb. is probably dormant.
ZOOL.-VOL. IJ EISEN—PLASMOCYTES. 6$
PLATE II.
386. The plasmocyte is almost ready to separate. The nucleus is in a far advanced stage of decomposition. The centrosphere contains three very distinct centrosomes; there is no somosphere.
39. The plasmocyte is almost fully developed; there is a residue of cres- cent-shaped granosphere at the pole; only part of the nucleus is indicated. The plasmocyte contains two archosomes, one of which is imperfect, but with a distinct centrosome. In the larger plasmocyte the granosphere is dark violet; the centrosphere pale violet, with a dark somosphere. There are parts of somosphere and centrosomes scattered along the edge of the cen- trosphere, as is sometimes the case. The pale ring around the granosphere is the hyalosphere.
PLASMOCYTES.
All these figures are shown with Oc. 4, excepting 46, 50, 51, 58 and 76. The stain used for most of the figures was toluidine; Ehrlich-Biondi for 54, 62, 67, 78; eosin-methyl blue " O " for 48, 49, 59; and for figures 82, 83 and 84 Zeiss Apochrom.
40. The granosphere is very long, and the hyalosphere is much wider than is generally the case. The pale violet, almond-shaped zone in the darker granosphere is the centrosphere. The centrosomes and somosphere are not segregated. The thin blue fringe along the margin is the plasmosphere.
41 . Showing a long, armlike, amoeboid projection. The hyalosphere is not differentiated; the granosphere is deep violet; the centrosphere is round and white, with a central centrosome. There is a small centrosome with centro- sphere, but without granosphere, in the lower right margin.
42. A plasmocyte, crescent-shaped, with two archosomes, but only one granosphere. The inner spheres are not well differentiated.
43. A starlike plasmocyte with poorly differentiated archosomes. It is probable that these two figures, 42 and 43, as well as 78, represent plasmo- cytes in degeneration, none of the spheres having properly responded to the stains.
44. A very perfect plasmocyte with well differentiated spheres. The grano- sphere is dark violet and contains two archosomes, the upper one of which is very small; the lower one is large, with a very large oval centrosphere, which shows different layers of plasma; in the center is a square somosphere with four centrosomes.
45. A large, stariike plasmocyte, with a dark, round granosphere. The centrosphere, which is differentiated and irregular, is seen at the lower mar- gin; it contains blotches of somosphere and a round, ringlike centrosome.
46. A starlike plasmocyte with dark and narrow granosphere. The large, paler, inner zone is the centrosphere, and the innermost darker zone is the somosphere with centrosomes.
47. A very perfect plasmocyte, with a narrow rim of violet granosphere. The centrosphere is very large, stained pale blue. The inner somosphere is pale whitish, with a few, small centrosomes in the center.
48. This is a somewhat abnormal, but well differentiated plasmocyte. Eosin-methyl blue " O." The hyalosphere is pure white. The large, pale pink zone is of doubtful character, possibly only part of the hyalosphere. In
70 CALIFORNIA ACADEMY OF SCIENCES. [3D Sbr.,
this case the granosphere is a very narrow, deep blue zone, which surrounds the large, paler blue centrosphere. The somosphere is dark blue, with a bright pink and very large food-granule. The explanation of these spheres is, however, only tentative.
49. A very typical plasmocyte. Eosin-methyl blue " O." The hyalosphere is pink; the granosphere deep blue; the centrosphere is large and pink, while the inner somosphere is starlike and blue, with a few dark centrosomes.
50. The principal interest in this figure is the starlike centrosphere, con- taining a few dark centrosomes. Compare fig. 26, where a similar form of centrosphere is seen.
51a and b. Two starlike plasmocytes In a the granosphere is very narrow and the cntrosphere very large. In b the archosome is not well differ- entiated.
52. A round plasmocyte, well differentiated, but not deeply stained. The pale, rather poorly defined centrosphere contains three very distinct centro- somes of different sizes.
53. The granosphere contains two distinct zones, the inner of which is deep violet. The centrosphere is paler violet. The darker center is the somosphere, with a few very indistinct centrosomes.
54. Stained with Ehrlich-Biondi, showing poor differentiation. The cen- trosomes lie at one edge of the oblong centrosphere.
55. A fully grown plasmocyte. The granosphere contains concentric layers of denser cytoplasm. The centrosphere is denser violet. The centrosomes are arranged in a crescent along the margin of the somosphere.
56. An oblong plasmocyte, with at least three archosomes imbedded in the granosphere. There are two small centrospheres to the left and one very large one to the right, the latter extending half across the granosphere. The large round globules must be explained as centrosomes with somospheres; the smaller, dark granules are of doubtful nature.
57. The hyalosphere is unusually large; the granosphere is violet. There is one large centrosphere, with two distinct centrosomes.
58a. In this plasmocyte the three centrosomes are remarkably distinct, and of different sizes. The paler sphere is the centrosphere, there being no distinct somosphere. The granosphere is deep blue, darker at the left side.
586. A plasmocyte of very much the same nature as the last, only much larger. Here also the centrosomes are distinct and the somosphere is not stained.
59. I have referred to this figure in the text as possibly representing a plasmocyte surrounded by a membrane. It is remarkable in showing no frayed plasmosphere and no clearly differentiated hyalosphere. The dark blue sphere is the granosphere. The round pink one is the centro- sphere. The centrosome is seen clearly in the center, as a very small dot.
60. An oblong plasmocyte with four distinct controsomes. The grano- sphere is narrow and very dark; the centrosphere is paler and large; the rectangular field in the center is either only a part of the centrosphere or a greatly extended somosphere.
61 . A plasmocyte with two archosomes, one of which contains two cen-
Zool.-Vol. IJ EISEN—PLASMOCYTES. 7 1
trosomes. The thin, narrow, blue zone around the centrosomes is the somo- sphere. Both archosomes lie in a common granosphere.
62. A plasmocyte stained with Ehriich-Biondi. The stain has not been particularly successful, but the figure is interesting on account of the two separate centrosomes, or somospheres.
63. A plasmocyte with two separate archosomes. This plasmocyte ap- pears to be in a state of degeneration.
64. A plasmocyte with two archosomes, or rather with two separating plasmocytes. The lower one is indifferently stained and possibly in degen- eration; the upper one is most intensely differentiated, and possesses two additional spheres; probably the centrosphere stained pale blue, which has been divided up by the addition of some foreign substance.
65. A plasmocyte with two archosomes in one common granosphere. The upper archosome, which contains a single centrosome, is stained differently from the lower one. In this lower one there are three centrosomes in a crescent zone of somosphere.
66. A plasmocyte with three archosomes in a common granosphere. The paler zones are the centrospheres. The lower one of these contains a large somosphere and a centrosome.
67. Stained with Ehriich-Biondi and unusually well differentiated for this stain. There are four archosomes, but the hyalosphere is only differentiated around one, which contains three distinct centrosomes.
68. A star-shaped plasmocyte with three large archosomes, each one sur- rounded by its granosphere. An additional smaller archosome, indifferently stained, is seen at the lower right margin.
69. A large plasmocyte with two archosomes, stained differently. The one to the left is more perfect, with a large irregular centrosphere, at the lower edge of which are seen two large centrosomes. This plasmocyte may be best explained as having been formed of two, joined together by the plas- mosphere, as the hyalospheres are not continuous.
70. A plasmocyte with two archosomes with distinct differentiation. The one to the left contains a centrosome superposed on an erythrocyte.
71. Another plasmocyte with two archosomes, each containing a centro- some, centrosphere, and granosphere.
72. A compound plasmocyte, consisting of a common plasmosphere, and a common hyalosphere; but with three separate granospheres, each with an archosome.
73. A large plasmocyte. In the granosphere we find the centrosphere without centrosomes, but at its left upper margin are seen differently stain- ing granules, probably foreign matter, reminding one of the brown sphere in fig. 64. The other dark patches may be explained as centrosomes and somo- spheres.
74. On one side of the inner white zone are seen a few dark granules of doubtful nature, possibly centrosomes. The other inner spheres are not easily recognizable.
75. In this plasmocyte the central part contains what greatly resembles chromosomes, and it is possible that there is an effort to establish a nucleus from accidental nuclear matter. There are three faintly differentiated archo- somes in the common granosphere.
72 CALIFORNIA ACADEMY OF SCIENCES. [3D Srr.,
76. A plasmocyte observed in 0.6 per cent, salt solution, showing amoeboid projections of cytoplasm.
77. In this interesting plasmocyte the granosphere is divided by the bar- like centrosphere. Along the edge of the latter is seen on either side a somosphere, each with two centrosomes.
78. A plasmocyte in a very advanced stage of disintegration.
79. A plasmocyte showing phagocytosis. In the granosphere we see to the left two archosomes, each with three centrosomes connected by a narrow zone of somosphere. In the center of the granosphere is seen a small ery- throcyte partly involved by granospheroplasm. In the erythrocyte is seen a large, round disk, a parasitic Plasmodium, Hctmapium riedyi, common in the blood of Batrachoseps, the life-history of which I expect to publish soon.
80 and 81. Two plasmocytes with very refractive food globules in the somo- sphere. The centrosome and somosphere, when visible, are always situated close to the margin, never in its central mass. In this plasmocyte the centro- sphere is the oblong, blue zone. Zeiss Apochrom. 3 mm., Oc. 8.
82. A large plasmocyte with three separate amoeboid centrospheres. Two of these possess a single somosphere with centrosomes, while the middle centrosphere possesses two separate somospheres. One of the somospheres is crescent-shaped and the centrosomes are in the budding stage. Zeiss Apochrom. 3 mm., Oc 8.
83. A large plasmocyte with amoeboid centrospheres and ring-shaped somosphere. In the latter are seen three budding centrosomes connected by a thin rod of centrosomal matter. The substance enclosed by the grano- spheral ring is more refractive and probably composed of food supply which is being digested by the somosphere. Zeiss Apochrom. 3 mm., Oc. 12.
84. This is possibly a degenerating, intensely staining form of plasmo- cyte. The two outer spheres are almost disintegrated, while the granosphere is strongly and globularly granulated, staining intensely. There are three somospheres, one with budding centrosomes.
STAINS.
Eosin. James W. Queen & Co., Philadelphia, U. S. A. Already mixed and in solution. Composition unknown.
Thionin. Actien-Gessellschaft f. Anilin-Fabrikation. Berlin (66,711). 1 per cent, watery solution, with 10 per cent, alcohol.
Toluidine, Blue extra. Actien-Gesellschaft f. Anilin-Fabrikation. Berlin (66,711). 1 per cent, watery solution, 10 per cent, alcohol.
Ehrlich-Biondi. From C. C. Riedy, San Francisco. Rubin S. (No. 243); Methyl-Grun-Kryst (No. 99); Orange G. (3884); all from Actien- Gesellschaft f. Anilin-Fabrikation. Berlin (38,242). Solution acid- ified with oxalic acid.
Methylene Blue "O," Patent. Badische Anilin and Soda-Fabrik. Lud- wigshafen, Germany (9038). 1 per cent, watery solution, alcohol 10 per cent.
All the stains were supplied by C. C. Riedy, San Francisco. The numbers in parenthesis refer to manufacturer's order number.
h
i
P:
:
ft
IfiS^fawijp^-*" ^
PROCEEDINGS
OF THB
CALIFORNIA ACADEMY OF SCIENCES
Zoology.
Third Series.
Vol. I, No.
ERRATA.
Page 75, 14th line, for "Tricha" read Taricha. " 81, 3d line, for " carbondioxide " read carbon dioxide. '* 89, last line, for " fig. 2 " read fig. 3. " 94, 22d line, for " urodelos " read urodelous. " 99, 2 2d line, for " T. cristalus " read T. cr (status. " 100, 6th line from bottom, for "Amb/ystonea" read
Ambly stoma. " 101, 6th line, for " object " read objects. " 104, 17th line from bottom, for "Amb/ystomea" read
Ambly stoma. " 106, 8th line, for "Chondrotus lugubris" read CAon-
drotus tenebrosus. " 106, 17th line, for " Cadis " read Caddis.
WITH ONB PLATE.
Issued January 18 ', /897.
SAN FRANCISCO:
Published by the Academy.
1897.
v s
J
PROCEEDINGS
OF THE
CALIFORNIA ACADEMY OF SCIENCES
Third Series. Zoology. Vol. I, No. 2.
DIEMYCTYLUS TOROSUS Esch.
THE LIFE-HISTORY AND HABITS OF THE PACIFIC COAST
NEWT.
BY William E. Ritter,
Assistant Professor of Biology % Unhtrsity of California. WITH ONE PLATB.
Issued January 18 \ 1897.
SAN FRANCISCO:
Published by the Academy.
1897.
THE LIFE- HISTORY AND HABITS OF THE
PACIFIC COAST NEWT (DIEMYCTYLUS
TOROSUS Esch.).
BY WILLIAM E. RITTER, PH. D., Assistant Professor of Biology, University of CaHfomim.
CONTENTS.
Page.
Plate III.
I.— Historical.
1 1. —The Adult.
General Statement 76
1. Habits 77
{a) Respiration 79
(b) Sloughing 83
(c)Food 83
{d) Movements 84
2. Seasonal Changes 85
3. Sexual Differences 92
III.— Breeding.
1. Mating and Fertilization 93
2. Egg-Laying and Eggs 98
IV.— The Larva.
1. General Description 103
2. Food and Feeding 105
3. Movements 106
V.— The Metamorphosis.
Bibliography 112
Explanation of the Figures 114
I.— Historical.
During the last seventy years a number of zoologists have made observations upon and written about the amphibian to which the following pages are devoted.
It was first described by Eschscholtz ('33) under the name of Triton torosus. The author's introductory statement is as follows1: " Es hat dies Thier Hinsichts der Grosse, der allgemeinen Kdrperform, der Farbe und der Beschaffenheit der Hautdecken eine grosse Aenlichkeit mit Triton cris- tatus-y" and he then proceeds to give a quite detailed com-
1 1 ha-re not had access to a copy of Eschscholtx's Atlas, and am greatly indebted to my friend, Dr. T. S. Palmer, National Department of Agriculture, Washington, D. C, for a compute transcription for me of the part relating to this animal. Proc. Cal. Acad Sci., 3D Sbr., Zool., Vol. I. January 15, 1897.
f73]
74 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
parison between the two species. The description of T. torosus which then follows contains various measurements and anatomical details, on the whole accurate and satisfac- tory. In a few particulars, however, I am at a loss to know how to account for his statements. Thus he tells us that the vas deferens is not white but " ganz schwarz." Of the hundreds of specimens which I have examined I have never seen a case in which the vasa deferentia were any other color than white. The point is one of no particular im- portance, but if such a variation occurs it is certainly very rare, and it is a little curious that Eschscholtz should have happened upon it in the few specimens that he seems to have had.
Again, he states that the testis is not divided into two or three parts by " quere Einschniirungen," but is entirely simple and oval in form. This is certainly not true in the great majority of specimens. It is almost always divided into two or three portions, distinctly and usually quite widely separated from one another.
He erred further in concluding that the openings at the summits of the black-tipped papillae of the integument are only deceptive appearances. As a matter of fact such openings do exist, as I shall show later on.
The next naturalist to study the animal was Gray ('39). This author gives a brief description and also a figure of it. It hardly seems credible, however, that the figure (No. 3, Plate XXXI) to which he refers could have been made from a specimen of this newt. If it was, either the artist or the specimen, or both, were very unworthy representa- tives of their kind.
The best general figure of the species that has here- tofore been published is that cGntained in Girard's Atlas, Herpetology, U. S. Exploring Expedition, Plate I, fig. 1. This is from a sketch from life by Jos. Drayton, and on the whole is quite satisfactory for the variational form which it represents. I have, however, never seen an individual in which there was so strong and abrupt a contrast between the orange of the underside of the trunk and the yellow of
Zool.— Vol. I.] FITTER— DIEMYCTYLUS TOROSUS. 75
the throat as this figure indicates. The eyes are made more prominent than they are in reality, and the pupils are represented as circular, while in truth they are elliptical, as my figures show; but this error is rectified by the author in the text.
Girard adopted for the animal the generic name Taricha, proposed by Gray ('50). In his description he is at fault, or at least only partially right, in several points ; but it will be advantageous to leave the correction of these errors to the appropriate heads under which they will fall in the descrip- tion which is to follow.
I may, however, here call attention to one interesting error in connection with the animal's seasonal variations. Baird and Girard ('53) described Tricha Icevis, "allied to T. torosa. Gray, but smoother, having but slight indications of granulations ;" and " Tail very much compressed, with a fringe along the whole upper edge and the posterior half of the lower." The specimens from which this supposed second species was described were collected by Dr. John L. LeConte, at San Francisco, in February, 1850 (Baird and Girard ['53] , p. 300) . As will be apparent from the account which follows of the animal's changes with the seasons of the year, had Dr. LeConte collected his specimens in mid- summer instead of midwinter, Baird and Girard would have had no T. Icevis.
Girard ('58) speaks of lavis as having been " collected and recorded" by Dr. LeConte under this name, thus seeming to infer that it was Dr. LeConte who first distin- guished a second species. He further states that lavis is distinguished from torosa by being " perfectly smooth," and also by its having " proportionally smaller eyes and more elongated toes." If there is an apparent difference in the size of the eyes of the two forms, it is probably due to the puffy condition of the bodies of the smooth winter forms. This may cause the eyes to seem, relatively to the size of the rest of the animal, smaller than in the "granulated " forms. I have not, however, observed such a difference. Concerning the supposed more slender toes of " Icevis" I
Proc. Cal. Acad. Sex., 3D Swt., Zool., Vol. I. (2) Nor. 17, 1896.
76 CALIFORNIA ACADEMY OF SCIENCES. [3d Ser.,
can only say that I am unable to detect such a distinction. In fact, the reverse is more nearly true; i. e., the toes of " Icevis " are, in proportion to their thickness, less elongated. The most complete and accurate description that we have of the species, particularly of its seasonal varia- tions, is by Cope ('89). To this author's work I shall refer several times later on. Brief references, unimportant from any but a classificatory point of view, have been made to the species by Baird ('tf-$o)9 Skilton, and Boulenger ('82).
II.— The Adult.
General. — It would be superfluous to give here a gen- eral description of the animal, since those by Eschscholtz, and by Girard, and particularly that by Cope, are quite adequate, and the last mentioned is readily accessible to most zool- ogists. As concerns the adults, therefore, I have only to deal with those special points, knowledge relating to which could be obtained only by continuous observation of the animals, living and dead, in large numbers, both in con- finement and in their native haunts, and at all seasons of the year.
I do not use the term adult as meaning full-grown, but I speak of the animal as adult as soon as its metamorphosis is complete. Gage ('91) applies the word only to the olive- green, or viridescent form of Diemyctylus viridescens; but since there does not exist, as we shall see later, so trenchant a division of the life of our species into two periods as is the case in Z>. viridescens, and since I do not know defi- nitely about the limit of the period of growth, I see no satisfactory way of using the term here excepting that in- dicated. How long the animal lives, whether or not it grows during its whole life, and at what age it becomes sexually mature, are questions which I am as yet able to answer only partially.
The average length of nine specimens just metamorphosed I found to be 48 mm. The largest one of this age that I
Zool.— Vol. I.] RITTER—DIEMYCYTLUS TOROSUS. *J*J
have seen (not included in the nine above mentioned) was 60 mm. long. As a very large number of old larvae and young adults have come under my observation, I think it safe to assume that the 60 mm. one is about the maximum size attained by the larvae ; in other words about the maxi- mum attained during the first three-quarters of a year of their lives; i. e., from March to October inclusive, which may be taken as the average larval period. The next larger size that I have found in the fall have had a length of about 80 mm. I therefore conclude that these specimens are in their second year — are about one year and eight months old.
As I have never seen specimens of this size in which the external sexual characters were developed, I conclude fur- ther that sexual maturity is not reached till after the second year. How much after I have no satisfactory means of judging; but comparing the smallest sexually active males observed with the specimens supposed to be approaching two years of age, I think it probable that the males ordinarily become capable of doing their part in the propagation of the species when they are three years old.
The points of chief importance in the section now under treatment are: (1) those relating to the habits of life; (2) those relating to seasonal changes in structure; and (3) those relating to secondary sexual differences. Of course it will be neither desirable nor possible to keep these dis- tinct in description and discussion.
1. Habits. — The first point concerning the habits of the adults to which I would direct attention is the fact that the most strictly terrestrial -period of life is that immediately following metamorphosis.
In view of the notably aquatic habit of the animals for a considerable portion of their adult lives, it has been sur- prising to me to find that they are exceedingly particular about getting out of the water as soon as the metamorphosis is complete. In fact, if they are so situated that they can- not get out by the time their larval characters, the gill-stubs excepted, are gone ; or if after they have once left the water they are immediately put into it again and are not
78 CALIFORNIA ACADEMY OF SCIENCES. [3D Sbr.,
permitted to escape, they die, apparently of drowning, in the course of a few hours.
I have tested this many times and in both ways; i. e., by arranging an aquarium so that transforming larvae could not get out of the water at all, and also by putting into such aquaria recently transformed individuals found on land; and the results have been invariably the same. Particularly those captured on land and put into water seem to drown, for, if placed in an aquarium over night for example, in the morning they are found dead, their mouths open and bodies much enlarged. They appear to have imbibed water over the entire surface of the body.
Furthermore, I have frequently noticed that strictly land- dwelling individuals of larger size, as they are often found late in the summer, seem to have a genuine aversion to water. As an illustration of this, one day in the early fall of last year while searching through one of our caftons from which the water had disappeared excepting for an occasional shallow pool, I found many specimens of me- dium sized adults, all with skins at the extreme of papula- tion, and tails as nearly finless and round as they ever be- come. Although water was easily accessible to them all, not a single individual of the dozens seen, as I remember, was in it. In one instance a specimen in making an un- usual effort to escape me ran close along the side of a pool where the rocks in his course became so sheer as to make it more and more difficult for him to cling to them, and letting go meant an inevitable fall into the water. Against this the animal struggled with great desperation, and when at last the catastrophe came, no drowning human being ever made more frantic efforts to escape than did this unfortu- nate "water dog." It could hardly have been fear of the fall that made him cling so tenaciously to the rocks, for ordinarily the creatures are quite heedless of a tumble from much greater heights than this.
Instances of the apparent dread of water by specimens of this size, form, and habit, one sees frequently. The contrast between this and the ipost complete aquatic habit
ZOOL.-VOL. I.] RITTER—DIEMYCTYLUS TOROSUS. 79
is striking indeed. I am convinced that the full-grown males remain in the water during the winter and spring for months together, without ever once leaving it. This I con- clude, chiefly from constant observations throughout the year on the inhabitants of the Alameda Water Company's reservoir at North Berkeley.
The animals live here literally in thousands, and conse- quently a most excellent opportunity is afforded for study- ing them in nature. Not only do they not leave the water during long periods of time, but they are capable of re- maining for considerable intervals beneath it.
I have watched several specimens get along at least for half an hour without coming to the surface ; and one in- dividual remained at the bottom of a shallow pool in Straw- berry Creek a full hour after I began to watch him before he came up " to blow."
(a) Respiration. — There can be little doubt, I think, that pharyngeal respiration takes place here as Gage has shown to be the case in D. viridescens and various other Amphibia.
The movements of the pharyngeal parts are apparently less conspicuous in torosus than in viridescens, but they are none the less certain and constant.
The whole floor of the mouth within the boundary of the lower jaw, and of the neck, is very gradually depressed for a few seconds, then with a sudden twitch, accompanied by a slight but instantaneous opening of the mouth, the de- pressed parts are contracted, and the water which was taken in — through the nostrils ?— during the depression is forced out. The depression of the floor is ordinarily so slight that it may easily be overlooked. If, however, one is able to watch the animal attentively for some time directly from the side of the head, he will see distinctly enough both the downward and the upward movements of the parts. That water is expelled from the mouth when the twitch oc- curs may be ascertained by carefully watching any fine par- ticles of solid matter, such as powdered carmine, that may be suspended in the water in the immediate vicinity of the mouth. The jet, or currentof expelled water, is not directed
80 CAUFORNIA ACADEMY OF SCIENCES, [3D Ser.,
straight forward in a line with the oral commissure, but is projected ventral ward, almost at an angle of 900 with the commissure. This is caused by the vertical wall formed by the premaxillary and maxillary bones which border the mouth cavity from above, and by the fact that when the water is forced out, the mouth is opened so slightly that the entire outgoing current is brought against this wall and con- sequently is deflected downward by it.
The frequency of the respiratory movements varies some- what, as would be expected, with the condition and tem- perature of the water. Under ordinary circumstances one counts on an average about ten expulsions of water per minute.
Apparently the animal does not usually come to the sur- face of the water to get air, but to expel gas from its mouth. When, under natural conditions, the need of discharging gas is felt by the creature, it makes a sudden start for the surface of the water, frequently swimming almost directly upward. On the instant that the surface is touched by the end of the snout, one or a few small bubbles of gas are dis- charged from the mouth with a slight popping noise, and the animal turns with a quickness quite in contrast to the rather deliberate movements with which he executes the rest of the process and swims back toward the bottom, frequent- ly taking as direct a course as that by which he reached the surface. The head is not usually thrust out of the water at all, and almost the only visible indication on the surface of what has taken place is the small bubble which is formed and floats away for a short distance before it bursts. It is these facts, viz., the nature of the contact of the snout with the surface of the water, the discharge of the bubble of gas, and the instantaneousness of the operation, that lead me to believe no air is taken in.
I am well aware that this conclusion does not harmonize with that reached by Wilder ('76 and '77), Gage and Gage ('86), and Mark C90), from their experimental studies on the respiration of Amia and Lepidosteous, soft-shelled turtles {Amy da mutica and Asfidonectes spt'rifer), and Lepidos
Zool.— Vol. I.] RITTER—D1EMYCTYLUS TOROSUS. 8l
teus osseus, respectively, and I confess that it is not obvious why it should be necessary to go to the surface merely for the purpose of getting rid of gas, presumably carbondioxide, particularly since this gas is so readily soluble in water ; and I confess, also, that it is perfectly comprehensible why it would be helpful to respiration for the animal to take in atmospheric air occasionally, even though it depended largely for its supply of oxygen on taking it from the water by means of its pharyngeal movements. There is no doubt that animals confined for a considerable time in a small quantity of water do take air frequently, but their method of doing this is quite different from the process described above. A large part of the head is raised above the surface of the water, the mouth is opened to a considerable width, and a good quantity of air is gulped down. But I have very rarely seen this done by animals in their native waters, whereas the other operation may be witnessed as many times as an observer might desire, where the animals are as numerous as they usually are in the reservoir of which I have previously spoken. But of course the various ques- tions here raised concerning respiration can only be settled by experiment and chemical examination. /^Gage ('90 and '91) has shown that in D. viridescens the < oral epithelium is columnar and ciliated while the animals are strictly land-living, but is pavemented and non-ciliated during aquatic life ; i. e., during the larval, branchiate period, and when the aquatic habit is reverted to in later life by the adults.
While in general my observations on D. torosus confirm Gage's results, I find some striking exceptions.
On March 9th last, I examined the oral epithelium of eleven large, smooth-skinned, wide-tailed males from the reservoir, all of which I suppose had lived constantly in the water for six months at least. At any rate, there was nothing wanting in them of the characteristics that distinguish the fully aquatic forms to lead me to suppose that some of them had been water dwellers for a longer time than had the others. In two of them, however, I found columnar cells
82 CALIFORNIA ACADEMY OF SCIENCES. [3P Sbr.,
and cilia in abundance. All the others were without either.
Again, the specimen which I have already mentioned as having remained under water an hour after I began watch- ing him before he came to the surface, I found to possess an abundantly ciliated oral epithelium. This case seems to be in striking opposition to Gage's conclusion that "in all forms of Amphibia and in all stages after the complete dis- appearance of food yolk, ciliated epithelium is absent from the mouth when the respiration is mostly aquatic, and water is frequently taken into the mouth " (Gage, '90). But I must say that since this specimen's skin was fully papillated and its tail-fin was much reduced, I should not have con- sidered it to be strictly aquatic had it been brought to me without information as to where it was captured. Having, however, seen it remain at the bottom of the stream for so long a time, and having convinced myself by careful ob- servation that it was regularly carrying on the pharyngeal respiration, I can see no sufficient grounds for supposing that it had been anything else than a water dweller for six months or more, at least, i. e., during the whole winter, spring, and summer thus far, viz., to June 29th, the time of writing. And this is the more probable from the fact that the animals do undoubtedly return from the smooth to the papillated condition without leaving the water. I speak of this more fully elsewhere.
I have also found the mouth epithelium ciliated in several females which I have no reason for supposing less aquatic than several others in which such cilia were not present. But it must be borne in mind that absolute certainty, either as to how long or how constantly particular animals have been in water, is possible only upon direct observation, or from having kept them under conditions that would preclude the possibility of their leaving the water.
I find, after having examined a large number of cases, no exception to Gage's conclusion, "that in forms with mostly aerial respiration, when water is rarely taken into the mouth, the mouth is lined with a ciliated epithelium."
Zool.-Vol. I.] RITTER—DIEAfYCTYLUS TOROSUS. 83
I think it highly probable that cutaneous respiration takes place to a considerable extent in animals that have lived long in water. This seems particularly probable as regards those in the extremely smooth, soft-skinned condition which characterizes many males during the late winter. In this state the epidermis is found, when examined in section, to be almost entirely devoid of the external layers of flattened, cornified cells which distinguish the epidermis in its hard, papillated condition.
It becomes, as one might say, stratum mucosum in its entire thickness. This undoubtedly renders it more per- meable to gases. The subepithelial connective tissue also becomes markedly augmented in thickness atthesametime, and what is of more importance still, a striking increase in vascularity occurs, at least in many cases.
I hope at some future time to study the question of cuta- neous respiration in these animals.
(b) Sloughing. — The little that need be said on this subject may be inserted here. There is, so far as I have been able to determine, no particular season of the year at which this occurs. It frequently happens that the outer layer of the epiderm, the layer which alone is shed, is thrown off, in whole or in part, in large or small pieces, almost imme- diately after animals are captured and placed in the aquarium. This is particularly likely to occur if the quantity of water in which they are confined be small. One often finds the skin from the limbs quite entire floating in the water ; also that from the tail ; but from the trunk and head it usually comes off in fragments.
Land-dwelling individuals are not infrequently seen with numerous fragments of dry epidermis clinging to various parts of their bodies. These animals have a very tattered, forlorn look. As soon, however, as the old rags are fully gotten rid of a much brighter, sleeker mien is presented.
The sloughed epidermis is an important article of diet for the animals.
(c) Food. — Like many other long-tailed amphibians this species is far from fastidious. It eats almost anything, almost
84 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
everything, or almost nothing, according to the conditions. As already mentioned, its own cast off epidermis is not re- jected as having fulfilled its whole usefulness merely because it can no longer serve as wearing apparel, but it is utilized as food without hesitation. I have found large quantities of it in the animals' stomachs. But the eating of their own skins is not the only way in which they put in practice the doctrine that home products should be consumed as largely as possible. During the breeding period their own eggs and young form an important food staple, particularly, as it seems, for the old males. One often sees one of these fel- lows taxing his ingenuity and mouth capacity to the utter- most in an effort to get a large egg mass whole into his stomach; and his efforts are frequently successful. I have also seen such males pulling to pieces the jelly of bunches in which the embryos were well developed, apparently for the purpose of extracting the little ones ; I must, however, admit that I have never found young larvae freed from the jelly in the stomach of an adult.
Small snails and slugs, both larvae and adults of numerous species of insects, sow bugs, earth worms, etc., will gen- erally be found ia greater or less quantity when an inventory is taken of the contents of a stomach. But although the animals eat heartily when an abundance of food is at hand, and are not very particular as to the kind of food, they can- not, I think, be regarded as particularly voracious. They certainly can endure for months together with very little to eat, and they never, so far as I see, show by their actions any signs of hunger.
(d) Movements. — In their movements, whether on land or in water, they are very deliberate and clumsy, even for long- tailed amphibians. On occasion they can, particularly in swimming, push themselves along with considerable alac- rity; but they are incapable of the almost lightning-like movements frequently executed by some other salamanders.
They may almost always be easily captured with the hand. They show little signs of fright, or inclination to flee from a human being. Sometimes, however, apparently
Zool.-Vol. IJ RITTER—DIEMYCTYLUS TOROSUS. 85
when the conditions are such that one's foot-falls produce a slight jar, either to the water itself or to the bottom on which the creatures may be resting, they will swim or run away for a short distance as one approaches them.
From various tests and observations I do not believe that the sense of sight is of much use to them in distinguishing objects, excepting for short distances — not much beyond a foot it appears — for objects the size of a man's hat.
They are in no sense given to hiding themselves from the light of day. During their aquatic career it seems to be their chief care to dwell in still waters. Nowhere have I seen them in any such abundance as in the reservoir al- ready mentioned ; and this lies out perfectly free to the full light of day, and there is almost nothing in or about the water that can be used as a hiding place or a screen from the daylight. It seems hardly possible that they are in- duced hither by a particular abundance of food, since great care is taken by the water company to keep the re- servoir as free from life of all kinds as possible. I may mention that the "water dogs " are taken out in great num- bers, as I am told by the Italian workman who lives at the res- ervoir as an employee of the company, and buried in the ground to prevent their getting back to the water, the fear being, of course, that they are in some way injurious to the water. As a matter of fact, however, their presence is rather an advantage, since they undoubtedly act as scaven- gers for the water, so far as animal life is concerned.
When living on land they often make long excursions out into open areas; e. g., they are familiar objects to every- body about the University of California, so frequently are they seen slowly and silently strolling about the grounds and over the sidewalks. At such times they do not mani- fest the least inclination to escape being captured; indeed, I have sometimes imagined them to be presenting them- selves as museum or laboratory offerings.
2. Seasonal Changes. — As a considerable part of the interest attached to the results of my studies on the seasonal characteristics of our Dicmyctylus consists in the compar-
86 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
ison made possible between it and its eastern relative, I can- not do better than to carry the comparison along as I pro- ceed with the narration of my observations. To this end it will be best to quote at the outset Gage's summary of his own studies on D. viridescens. His paragraphs that will concern us here are as follows: —
"4. After the gills are absorbed the animal leaves the water, and the color gradually changes from an olive-green to brownish-red, and finally, during the same season, as- sumes a bright yellowish-red, the vermillion spots remaining and becoming partly surrounded by black pigment."
"6. In the autumn of the third, or the spring of the fourth year after hatching (when two and one-half or three years old), the red changes for a viridescent coloration. This may occur with or without entering the water. If the water is entered the animal changes to an aquatic mode of life/'
" 8. After becoming adult and transforming to the viri- descent coloration, the Diemyctylus always remains of that general color, and never again becomes red, even when kept out of water a whole year, thus showing that the color- ation is dependent neither on food, season, nor environ- ment, but is normal for a given period of life only."
"9. The adult viridescent forms are purely aquatic under favorable conditions, and after once entering the water do not leave it, although they are able to live for several months, and perhaps indefinitely, in moist places, wholly out of water."
The changes of form and color in our species which take place during metamorphosis from the larval state, I give in detail in my description of the larva. On that point I need consequently do no more here than refer to figs. 4, 5, 6 and 7.
I have already shown that immediately after metamorphosis the adults are more strictly aerial than at any other period in their lives, and that at this time they possess the tuber- culated skin and narrow tail in the fullest degree. I have also given reasons for believing that the males, at least,
Zool.-Vol. I.] RITTER—DIEMYCTYLUS TOROSUS. 87
usually arrive at sexual maturity when three years old. Sexual maturity in the males is not reached, so far as I know, until the aquatic habit is assumed for the second time, and the tail has again become much expanded and laterally compressed. A change of color is, however, not essential to the arrival at this state. We are not, therefore, it appears to me, justified in speaking of a second trans- formation in this species as Gage does in D. viridescens. The difference between a young adult male and a sexually mature one is certainly considerable, but it is very much less than that between the gill-breathing larva and the adult. Fur- thermore, the transition to the mature condition is much more gradual than is the true metamorphosis. But what is most important of all, and what constitutes the greatest distinction between the two species, is the fact that, according to Gage, the viridescent, or adult form of the eastern species, having been once assumed, the red form is never again reverted to; our species, on the other hand, may certainly go back from the condition which seems to correspond to the viridescent, or adult form of the eastern species, to that which apparent- ly corresponds to the red, or immature adult stage of that species. The characteristics of the two conditions in our species will be readily understood by an examination of figs. 1, 2 and 3.
Figure 3 represents a full-grown male, as these are frequently found in autumn, crawling about on land, quite remote from any water.
Figure 2 represents another of about the same size, taken from the reservoir at North Berkeley, on the first of January, 1896; i. e., during the height of the season's amours. (I point out elsewhere that the males lead the way by some weeks in love-making, so that egg-laying does not begin for some time after the males are abundantly pre- pared for the occasion.)
In addition to the great width of tail and strength of limbs of the latter as compared with the former, which conditions come out with ample distinctness in the figures, there is in reality a difference almost as striking and characteristic in
88 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
the structure of the skin, although this does not appear quite as clearly in the figures.
I refer, of course, to the papulation and the greater hard- ness of the epidermis of the land-dwelling form. The " granulation " of the skin, as it has been usually designated, pertains, when the condition is at the height of its develop- ment, to almost the entire surface of the animal. It is, however, most pronounced on the back and part way down on the sides of the body. In these regions the papillae are both larger and nearer together than elsewhere. They are least developed about the end of the nose, on the tail, particularly along its ventral side, on the inner surface of the limbs, and on the soles of the hands and feet. They are also always considerably less numerous and lower on the belly than on the back.
Each papilla is crowned by a cap of deeply pigmented cells, in the middle of which is the opening of a gland. Figure 13 represents a surface view of a small piece of sloughed epiderm. It is not my intention to enter into a histological description of the epidermis at present, but I may say that there can be no doubt about the presence of openings at the summits of the papillae. Even a superficial examination of microtomic sections of the skin is sufficient to convince one of the fact.
The papillae make their appearance in all individuals, males as well as females, some time before metamorphosis is com- plete. The females retain them throughout life, so far as I am able to determine, while they disappear almost entirely, in many cases absolutely, from males when the aquatic habit is fully resumed. My belief is, though the proof of this is not complete, that in general the papillae disappear more and more as the animals grow older.
In many specimens, a good example of which is presented by fig. 2, while the epidermis is perfectly smooth, i.e., is wholly without the papillae, there are frequently seen a great number of light colored spots corresponding in size and distribution with the papillae when these are present. They are evidently remnants of the papillae. In still
Zool.— Vol. I.] RITTER-DIEMYCTYLUS TOROSUS. 89
another condition, an instance of which is illustrated by fig. 1, absolutely no trace of papillae is to be found. From the comparative rarity and large size of speci- mens of this kind, I conjecture that these individuals have arrived at about the extreme age to which the species attains.
I must also call attention to the faded out appearance of the two last mentioned specimens. It looks very much as though they were actually sun-bleached. Individuals of this color are by no means rare. In fact, specimens are frequently seen which are more conspicuously light colored as they are seen in the water than those here figured appear to be. Prof. Cope mentions that "in the rough specimens the brown becomes almost black;" and that "in smooth specimens the brown is pale, and has an olive tinge." In general this is true, but it is by no means always the case that the rough ones are almost black. As much as can be said is that the darkest color ever attained is by the rough ones, and that the lightest color reached is by the smooth ones. This color variation takes place in the males alone, the females retaining with considerable constancy their char- acteristic seal brown color.
The decidedly dark color of the side of the body, as compared with other parts, shown in fig. 2, will be no- ticed. This condition heightens the impression above mentioned, that the light color is due to sun bleaching — and I may say that specimens of this kind have been seen more frequently than elsewhere in the North Berkeley reser- voir, which, as already said, is fully exposed to the sun.
I have previously stated that the papilla ted, narrow-tailed condition may be again assumed by the smooth-skinned, broad-tailed specimens. Of this I have positive evidence from experiments with animals kept in confinement. Last winter I placed some two dozen specimens, most of which were males with well developed tails and much reduced pa- pillae, in a terrarium in the laboratory. The most of them are still living (July 3), but they are now all like the one shown in fig. 2. The tails are reduced to the minimum in
90 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
depth, and the epidermis is nearly at the maximum of papu- lation; and I am convinced by numerous observations on animals in nature, particularly by those made during the present spring and summer, that the same reversion takes place normally. I have recently (in midsummer) examined many males, not only at the reservoir but also from several streams in the vicinity, and have not found one which pos- sesses the smooth skin.
Not only this, but it can actually be seen that the papillae are growing out on skins that were previously smooth. The black tips, at first projecting but slightly above the surface, make their appearance while the general groundwork still shows the smoothness and softness which characterize it when the papillae are wholly absent. The feel to the hand of the condition here described is quite different from what it is at any other period during the animal's life.
It will be remembered that Gage's conclusion concerning D. viridescens is "that the coloration is dependent neither on food, season, nor environment, but is normal for a given period of life only," since he found that the red coloration changes to the viridescent whether the animal enters the water or not, and that it is not resumed even though viri- descent specimens are kept out of water for long periods of time. Professor Gage appears to be particular to limit his denial of the potency of environmental change in this case to its influence on color. What he thinks about the cause of the change in the form of the tail, for example, which accompanies the change of color, he does not tell us. Con- cerning the replacement of a ciliated by a non-ciliated oral epithelium, as the animals change from the aerial to the aquatic mode of respiration, he remarks, however, that " the change has something the character and certainty of a simple chemical reaction."
As regards D. torosus, the facts above pointed out, viz., that the repapillation of the skin and reduction of the tail take place even though the animals do not leave the water, might seem to indicate that even structural changes in this species are " normal for a given period of life," and are not
Zool,— Vol. I.] RITTER—DIEAfYCTYLUS TOROSUS. 91
dependent on " food, season, nor environment." But I have seen nothing to indicate that the wide-tailed, smooth-skinned condition of the males is ever assumed without the adop- tion of the aquatic mode of life ; and so obviously and per- fectly is it an adaptation to such a life that one can scarcely believe it to have been produced by any other cause.
To say that the characters distinguishing it are " second- ary sexual characters," and so dismiss the subject with the supposition that an explanation of the facts has been given, is wholly unsatisfactory, even though the males alone present these characters, and that at the time of sexual activity.
It appears to me that instead of regarding the seasonal changes as independent of environment, or as having been produced in response to the needs of the reproductive func- tion, we should come nearer to a satisfying explanation by supposing that originally and immediately they were caused by change in the mode of life of the animals; and that now the characters have become so thoroughly established by heredity that they have acquired a considerable degree of independence of the causes which produced them.
It seems probable that the reason why D. torosus re- verts to the rough-skinned, narrow-tailed state after it has passed a period in the opposite condition, whereas D. viri- descens never leaves the viridescent, aquatic form after hav- ing once assumed it, is to be found in the difference in habitat of the two species. The region inhabited by viri- descens is not particularly drier at one season of the year than another, and the streams and ponds which the animals make their home are perennial; so that, so far as environ- ment is concerned, there is no reason why they should leave the water when once they have betaken themselves to it and become well established in it. With torosus the case is different. Throughout the greater portion of its range very many of the streams and ponds in which it lives during the winter and early summer dry up almost entirely (I doubt considerably if the larvae ever come to metamorphosis in streams that wholly disappear during the summer) ; and as a consequence the animals find it to their advantage, in
Pkoc. Cal. Acad. Sci., 3D Sbr., Zool., Vol. I. (3) Nov. 18, 1896.
92 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser..
their search after food, to spend much of this period on land.
It would, I believe, be about as easy to establish the prop- osition that the annual period of sexual activity is deter- mined by life habits primarily induced through environ- mental influences, of which those pertaining to food and moisture are probably most important, as it would to establish the opposite proposition; viz., that the life habits and their accompanying structural characters are secondary to sexual activity. But this is a very difficult problem, though per- haps not wholly unsolvable.
3. Sexual Differences. — The characters distinctive of the sexes have been mostly adverted to already, though indirectly.
Until sexual maturity is reached there is, so far as I have determined, no way of distinguishing them by superficial inspection. Males and females are alike papillated and narrow-tailed.
As already pointed out, the females never leave this con- dition. Whether this is due to a failure on their part to take on the broad tail and smooth skin, because these are in real- ity male characters, or to the fact that they are less aquatic in their tastes and habits, or to some other cause, I do not know. My observations do, however, lead me to believe that on the whole the males are more fond of the water than are the females.
The great development of the lips of the male cloaca as compared with those of the female during the breeding season is very distinctive. In addition to the great en- largement of these parts, there is a dark band extending down from the general dark ground color of the dorsum and sides of the body on the cloaca almost to the ventral edge of the lips (fig. 2). The female cloaca never becomes enlarged to any extent, and is always without the lateral bands.
During the breeding period there is developed on the in- ner surface of each femur in the males a patch of epider- mis harder and more corrugated than in adjacent parts, and
ZOOL.-VOL. I.] RITTER—DIEAfYCTYLUS TOROSUS. 93
also furnished with a dark, dirty pigment that does not exist here at other times. The glands in this area are peculiar to it also, but whether they are the ordinary epidermal glands modified for the occasion only, or are permanently different from the latter, I do not know. These patches are rather transient and disappear very soon after the culmination of the reproductive activity. As a rule the soles of the hands and feet of the males become black, and the tips of the digits capped with black at this season, but this character is apparently less constant than are the others mentioned. Figure 10 illustrates the several points described. Prof. Cope's unqualified statement that " the epidermis on the ex- tremities of all the digits is horny," is an error, probably induced by the author's having observed the condition here described. Ordinarily the epidermis of the toe tips does not differ from that on other parts of the members, excepting that it is somewhat smoother.
III.— Breeding.
1. Mating and Fertilization. — Gage ('91) states that an autumnal mating, at any rate so far as the male's part in the process is concerned, takes place in D. viridescens. Spermatophores, he says, are emitted precisely as in spring, or the proper breeding season. Copulating pairs of D. torosus are not infrequently seen in the early fall in the streams and ponds of this local- ity. Thus last year I observed such a pair as early as Sep- tember 23d. Instead, however, of there being an autumnal as distinct from a spring mating in this species, we have to say that the season's amours begin thus early in the year, for the process goes on without interruption from its begin- ning to the termination of the breeding time. But here as in D. viridescens it is an affair of the males alone. Sperm masses are discharged but there are no ova for them to fer- tilize till some months later, as we shall see further on. Gage is at a loss to understand the significance of the au- tumnal mating in D. viridescens. From what has been said
94 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
about D. torosus it appears that the autumnal pairing here means merely that the males lead the way by some months in the yearly amours — their sexual products mature earlier by this length of time than do those of the females. If we might suppose that D. viridescens or its immediate ancestors once lived in a climate something like that now prevailing in California, where the winters were not sufficiently rigorous to materially retard their physiological processes, we might then suppose that its autumnal mating was likewise the lead- ing of the way by some months of the males in the sexual activity of the year. Then with a change in climatic con- ditions, either by a migration of the species or by an actual change of climate, the severe cold of winter intervened be- tween the time when the males began their amours and that at which the females were ready to co-operate with them in the reproductive act. Still, from old habit the males began their amorous advances as before ; but by the severity of the cold to which they later became subject, they suffered a check in their activities till warmth returned with the fol- lowing spring ; and such an order of procedure as we now see in the species resulted. But in this connection it is im- portant to bear in mind that in some species of urodelos Amphibians, i. e., Salamandr a maculosa, Knauer('78) and Pfitzner ('80), the same females reproduce regularly twice a year, in spring and in autumn.
That internal fertilization occurs in this species, as has now been amply proved to be the case in numerous other Urodela, there can be no doubt, since females kept in the laboratory isolated from the males have frequently deposited eggs, and these have always developed in all respects like those deposited under natural conditions in the streams and ponds. Unfortunately my observations on this point are incomplete. The animals appear adverse to carrying on their amours in captivity; at least they have thus far not gratified my desire to have them do so. My information is consequently limited to what I have seen them doing in their natural haunts.
While, therefore, I have not been able to determine with
Zool.— Vol. L] RITTER—DIEAfYCTYLUS TOROSUS. 95
certainty all the details of the method of fertilization, I still believe that the sperm by some means reaches and enters the cloaca of the female during the act of copulation. I would not, however, be understood to mean by copulation that an actual uniting of the external reproductive parts of the two sexes is an essential element in it. I simply mean by it the grasping and holding of the female by the male. My reasons for believing this are chiefly two : First, although nothing is more common during the breeding season than to find the females held in the grasp of the males, I have watched for a great deal, but have never seen a suggestion of such processes as are gone through in several other spe- cies; e. g., in Axolotl (Gasco '8i and Zeller '90), Triton (Gasco '80, Zeller '90 and '91), or D. viridescens (Jordan '91 and Gage '91). In all these cases fertilization consists in a preliminary love-making, during which the male discharges one or more spermatophores, not while he holds the female in his embrace (in Axolotl this phase of the proceeding is apparently omitted entirely) , but free upon the floor of the aquarium, where it is afterwards picked up by the cloaca of the female. My second reason for believing fertilization to be more direct than this in our species is the fact that I have captured one copulating pair in which a large quantity of sperm was contained in and protruding from the cloaca of the female. This case might seem to be conclusive on the point, though of course it is not wholly so, for in the first place the actual passage of the sperm mass from the male to the female was not observed, and in the second place it is quite conceivable that this passage may have taken place by some means less direct than that supposed ; or again it is by no means impossible that the sperm mass might have been obtained by the female from some other male than the one with which she was found mating. But, on the whole, when all the facts observed are considered, it seems to me that the belief above expressed is warranted,
The lips of the male cloaca become enormously tumid and enlarged at the height of sexual activity (fig. 2), and during copulation these are made to straddle the dorsum of
$6 CALIFORNIA ACADEMY OF SCIENCES. [3D Ser.,
the base of the tail of the female very nearly over her cloaca. In several instances, a notable one being that mentioned above, where a large quantity of sperm was found in the cloaca of the female, these lips have been so much extended as to reach fully half way down across the tail of the female. One is reminded of a saddle on a horse's back. At such times the rugae on the inner surface of the lips (fig. 10) are very prominent. They project beyond the edge of the lips all around the cloaca in the form of a fringe, and are quite conspicuous in contrast with the dark color of the outer skin, they being a lively pink, owing to their great vascu- larity.