Thos. L. (Thomas Lincoln) Casey.

Annals of the New York Academy of Sciences online

. (page 6 of 28)
Online LibraryThos. L. (Thomas Lincoln) CaseyAnnals of the New York Academy of Sciences → online text (page 6 of 28)
Font size
QR-code for this ebook

of the ochrea posterior to the lamina or petiole may be called its
ligular portion and is usually supplied by bundles arising tan-
gentially from the main ones.

4. The lateral portions of the primitive leaf, when separated in
greater or less degree, constitute stipules in the usual acceptation
of the term. They are variously modified by subsequent evolu-

*Sir John Lubbock. Jour. Lin. Soc. Lond. 30 : 504. 1894.

Digitized by VjOOQIC

The Nature and Origin of Stipules. 49

tionary changes, by increased development, by basal or total
degeneration, by secondary adnations and various textural modi-
fications. They receive their vascular bundles typically as
branches of the lateral ones of the leaf-trace.

5. The lateral portions of the primitive leaf therefore represent
in potential the ligule, the ochrea, the margins of sheathing peti-
oles and stipules, but they are often incorporated with the other
portions as the wings of petioles and as lateral basal portions
of leaf-blades.

Annals N, Y. Acad. Sci., X, June, 1897. - 4.

Digitized by VjOOQIC

IL— The Ascidian Half-Emhryo.

Read March 8, 1897.

The development of isolated blastomeres of the ascidian egg
has afforded a subject of considerable discussion on the part of
many theoretical embryologists. Chabry* approached the subject
from the experimental side, and, from the results of his many de-
tailed observations and experiments, was led to the conclusion
that one of the isolated blastomeres of the two-celled stage pro-
duced a strict half-embryo. As it was well known that the first
cleavage-plane divided the egg into right and left halves, this con-
clusion seemed altogether probable and of considerable interest.

A number of writers, however, among them Hertwig,t Driesch,J
Weismann,§ Barfurth|| and Roux,^ were led, on the grounds of
Chabry 's results, to opinions more or less at variance with his.
Barfurth considered Chabry to be in greater part correct. Roux
and Weismann believed that during the later development the
missing part was supplied by the other cells through " postgene-
ration." Hertwig states that, in his opinion, Chabry was in er-
ror ; and Driesch also argued that a tj'pical total development oc-
curred. Finally, Driesch** in 1893 repeated Chabry 's experiments,
upon the eggs of Phallusia mammillala, and by the results
wholly confirmed the theoretical conclusions of his previous paper.

* Chabry L. Contribution & Pembryologie normale et teratologiqne des
asoidies simples. Journ. de Panat. et de la phys. XXIII. 1887.

t Hertwig, R. Urmund und Spina bifida. Arch. f. mikr. Anat. XXXIX.

t Driesch, H. Der Werth der beiden ersten Fnrchangszellen in der Ecbino-
dermentwickelnng. Zeit. f. wias. Zool. LIII.

i Weismann, A. Das Keimplasma. 1882.

II Barfurth, D. Halbbildung oder Ganzbildung von halber Grosse. Anat.
Anz. VIII. 1893.

^Rouz, W. tJber des entwickelungsmechaniscbe Yermogen jeder der bei-
den ersten Furohungszellen des Eies. Verbandl. d. Anat. Ges. Wien. 1892.

** Driesch, H. Von der Entwickelung einzebier Ascidienblastomeren.
Archly filr Entwick. der Organism en. I. 3. 1895.

Digitized by VjOOQIC

The Ascidian Half-Embryo. 51

Although at that time reluctant to admit anywhere the occur-
rence of" partial " development, Driesch has since proved, in con-
nection with Morgan, the existence of a partial early development
in the cteuophore egg.* And in a recent paper by the writer f
it has been shown that the isolated blastomere of the snail pos-
sesses the power of forming only a corresponding portion of an
embryo. In a later paper, Driesch,J developing an idea suggested
by Prof. E. B. Wilson and myself (loc. cit.), recognized the ex-
istence of a series among animal eggs, from the nearly isotropic
eggs of the medusa, Amphioxus, fish, sea-urchin, etc., at one ex-
treme, to forms such as the frog and ctenophore, and finally to
the snail, at the other extreme, where the blastomere possesses
such an organization that but a part of an embryo can be formed
and postgeneration cannot occur.

The ascidian egg, however, remained unexplained by the contra-
dictory results of Chabry and Driesch. From this consideration
the author was led to an examination of the facts in another
ascidian. The results will, it is hoped, clear up the confusion to
some extent, and will show how far the development is a " partial "
one and in what respects it is " total."

The experiments were performed during the past summer at the
Marine Biological Laboratory, Wood's Holl, upon the eggs of
Molgula manhaltensis, which grows very abundantly upon the
piles and wharves at N^ew Bedford, Mass. Artificial cross-fertil-
ization was resorted to, and the eggs at the desired stage were
spurted in a watch-glass by means of a fine spiral pipette.§ Those
eggs presenting isolations were placed separately in watch-glasses,
and camera drawings of successive stages were made at intervals,
using a Zeiss oc. 4, and obj. C.

As to nomenclature, the system proposed by Kofoid 1| and ap-

* Driesch, H., and Morgan T. H. Von der Entwickelnng einzelner Cteno-
phorenblastomeren. Arcbiv fur Entwick. der Organismen. II. 2. 1895.

t Crampton, H. E., Jr. Ezperimental Studies on Gasteropod Development,
with an appendix on Cleavage and Mosaic Work, by £. B. Wilson. Arcbiv
fiir Entwick. der Organismen. III. 1. 1896.

:t Driescb, H. Betraobtung iiber die Organisation des Eies and ibre Genese.
Arcbiv fiir Entwick. der Organismen. IV. 1. 1896.

i As previously described in connection with the gasteropod experiments.

II Kofoid, C. On some laws of Cleavage in Simax. Proc. Amer. Acad.
Arts and Sciences. Vol. XXIX. 1894.

Digitized by VjOOQIC

62 The Ascidian Half-Embryo.

plied by Castle* to the Ciona egg has been used for obvious
reasons. According to this system, now well known, each cell is
designated by a letter referring to the particular quadrant of the
four-cell stage from which it arose ; in addition it receives an ex-
ponent denoting the generation to which it belongs, and a second
exponent denoting its place in that generation, counting from
below upward.

Detailed Description of Cleavage.

A. Normal Cleavage. — The cleavage of the Molgula egg is pre-
cisely the same as that of Giona and other ascidians, as far as it
has been followed. Therefore, it is unnecessary to discuss the
normal phenomena further than to emphasize a few of the facts
which are important in connection with the cleavage of the frag-

The first and second cleavage-planes are meridional, while the
third is equatorial. An eight-cell stage results (fig. 1) which,
seen from the side, consists of two tiers of four cells each. The
upper tier is shifted anteriorly upon the lower, so that the poste,
rior upper cells are in contact with the anterior ventral cells-
This relation is constant, and characteristic of probably all as-
cidian eggs (Castle, loc. cit., p. 228). Passing to the 16-cell
stage, all the eight blastomeres divide. The spindle axes are in-
clined in such a manner that the anterior products of the anterior
cells (fig. 2 : B ^-2, b ^ *) lie slightly below the median products ;
while the posterior products of the posterior cells lie slightly
above the other cells (fig. 2 : C ^•^, c ^'^), When activity is again
resumed, the dorsal cells remain quiescent, while the ventral cells
segment, and a 24-cell stage results (fig. 3). After a period of
rest the dorsal cells pass into the same generation (sixth) with
the ventral cells, and a morula of 32-cells results. Then the ven-
tral cells divide at labout the same time, while the dorsal cells re-
main quiescent, giving a 48-cell stage.

Further details are unnecessary for our purpose. We empha-
size the fact that, beginning with the 16-cell stage, there is a well-
marked alternation of activity between the cells of the upper and
those of the lower hemisphere of the embryo.

* Castle, W. E. The Early Embryology of Ciona intestinalis. Balletin of
Mas. of Comp. Zool. Harvard. Jan. 1896.

Digitized by VjOOQIC

The Ascidian Half -Embryo. 53

B. Cleavage of the ^ blaatomere. — As is well known, the isola«
tion of an ascidian blastomere is effected by the death of its
neighboring cell or cells, and not by an actual separation. The
dead cell partially disintegrates and exerts npon the living cell no
modifying influence, such as mechanical obstruction to rounding
during division, etc.

f . At the normal time, viz : at the time of division of control
eggs, the injured blastomere divides about equally (figs. 4 and 13).
Often when the eggs are operated upon when passing into the 4-
cell stage, evidence of division in the dead cell will remain. In
such cases the division plane of the living cell is seen to be meri-
dional and at right angles to the first. Therefore, it corresponds
with the second cleavage-plane of the normal embryo. In all
cases where it is possible to ascertain the facts this relation ob-
tains. Driesch finds in Phallusia that no such constancy of rela-
tion exists.

f. After a normal period of rest the two cells divide at the
same time. There are thus produced four cells which are ar-
ranged in a manner exactly similar to the half of a normal 8-
celled embryo. Seen from the side (figs. 5, 0) the cells lie so that
two are separated, while two are in contact ; these latter are the
posterior dorsal and the anterior ventral cells, as shown by the
succession of the cleavage planes of the fragment. Precisely as
in the normal 8-celled embryo, there is an anterior shifting of
the dorsal cells upon the lower cells. According as this shifting
is to the right or left, in lateral view, one is confronted by a right
or left half-embryo. From a comparison of the figures, it is seen
that the embryo in fig. 5 is the same as the half turned toward the
observer of fig. 1 ; while that shown in fig. 9 is derived from a
right i blastomere. The appearance of the f embryo in end view
is shown in fig. 14, and a characteristic crossing of the spindle
axes is exhibited, which is similar to their crossing in the com-
plete egg (vide Castle for figures). The four-celled fragment,
then, is in nowise a counterpart of the normal four-celled embryo,
but, on the contrary, corresponds in every particular to the half of
an eight-celled embryo.

From Chabry's fig. 106, it appears that a typical | stage occurs
also in Ascidiella,

^. At the next cleavage, all the cells divide (figs. 6, 10).
Exactly as in the origin of corresponding normal cells, (fig. 2)

Digitized by VjOOQIC

64 The Ascidian Half-Embryo,

.the anterior products otth^' anterior cells (fig. 6 : B**^, b*-* ; figs. 10
and 16: A*'^,a^*) lie slightly below the other cells ; and the
posterior products of the posterior cells (fig. 6 : C*-2, c** ; figs
10 and 16: D^-^, d**) lie slightly above the median products.
On a comparison of fig. 6 and fig. 2, it will be seen, however, that
the topographical relations of the cells of the fragment are quite
diflTerent from the normal. For example in fig. 6, the cell c** is
in contact with B*-^ and b^*, while in the normal Qgg it lies at the

^ other end of the embryo. A similar rearrangement is still better
shown in fig. 10, thai of a right ^ embryo, where D^-2 is in con-
tact with A^*^, while d** is in contact with A^*^ and a^*. This
rearrangement is obviously rendered possible by the absence of
the other half of the embryo, so that the cells cohere in a spherical
form just as a corresponding number of soap-bubbles. It cannot
be considered as a " gliding," for the spindle-axes are from the
first accommodated to the changed conditions. That is ( figs. 15, 16),
the anterior end of the anterior spindles, and the posterior ends
of the posterior mitotic figures are swung somewhat toward the
original first cleavage-plane of the embryo.

Chabry's fig. 113 leaves no doubt that the ^ embryo ot Asci-
diella is precisely the same as that described above for Molgula,
From Driesch's fig. 5, there is no doubt that in Fhallusia the
eight cells are arranged as the normal 8 cells.

1^-^. When activity is again resumed, only the four lower
cells are affected, while the dorsal cells remain quiescent. A 12-
celled fragment results (figs. 7 and 11), which is exactly equivalent
to a half of the normal 24-cell stage (fig. 3). The quiescence of the
dorsal cells during the division of the ventral cells is the first in-
dication of the alternation of activity in the rhythm of cleavage,
which was found to be characteristic of this type of segmentation.
As in the preceding stage, when the resting condition is assumed,
there is a passive rearrangement of the cells. For example, the
cells A*^ and A^-^ were segmented along an axis inclined at an
angle of 45° to the axis joining their centres at the resting stage.
Again the cells D^* and D^* have retreated around the posterior
end of the fragment.

^. While the eight cells of the lower hemisphere are resting,
the four dorsal cells likewise i)ass into the sixth generation, and a
^ stage results (figs. 8, 12). Its resemblance to the half of a
normal 32-cell stage is still less marked than that of a ^J embryo

Digitized by VjOOQIC

The Aacidian Half-Embryo. 55

to a half of the normal 24-celled stage. This is so, for the reason
that further passive rearrangements of the cells occur, obscuring
the partial character of their origin, and causing the cell complex
in its solid, or ^' complete," condition to resemble a normal or
" complete " embryo. Nevertheless, the succession of rhythmic
cleavages, relation of successive cleavage-planes, etc., point to the
operation of factors which are counterparts of those operating in
a half of the normal embryo.

Later development. The embryo is now ** complete," and gives
rise to a complete blastula and larva. Although the process of
gastrulation has not been carefully observed, enough* of the later
development has been ascertained to prove that a larva arises
which resembles the normal larva, except as regards its smaller
size and certain minor defects. My results, therefore, are en-
tirely confirmatory of those of Driesch upon Phallusia.

C — Cleavage of the J blastomeres, — One of the isolated blasto-
meres of the four-cell stage, is divided at the next cleavage by a
plane which is seen to be at right angles to both of the preceding
planes. Therefore it corresponds to the third cleavage plane of
the normal embryo. The f stage is shown in fig. 17. A subse-
quent cleavage cuts each of the cells equally, and a -^ stage re-
sults (figs. 18, 19), until this time, one is left in doubt as to the
true nature of the fragment, that is, whether it will segment as
a quarter or as an entire egg. However, from this time on, the
character of cleavage is exactly that of a quadrant of a normal

When division next occurs, only the two cells toward the ob-
server segment (fig. 20), and a stage of six cells results, which is
evidently comparable to a t^ embryo only, and not to any stage
of the normal development. After a normal period, the dorsal cells
(lower in the figure) pass into the sixth generation, and an ^
embryo (fig. 21) is the result. As in the previously described
fragments, passive rearrangements occur when the resting condi-
tion is assumed, and the cells flatten down upon one another
(fig. 22). The cells of the ventral half segment at the next period
of activity (fig. 23), while the dorsal cells remain undivided. The
resulting \^ stage, although solid, is nevertheless derived from
the ^ blastomere through a segmentation of a partial character.
This partial character is expressed chiefly in- the characteristic
rhythm of cleavage.

Digitized by VjOOQIC

66 The Ascidian Half-Embryo.

Concerning the later stages, the results of Driesch are again con-
firmed. The young larvae represented in Figs. 25, 26 of this paper
illustrate one point further, although of minor consequence. It
will be seen that the long axis of the ^ larva in fig. 25, and the
long axes of the ^ larvae derived from the same egg, in fig. 26)
are approximately parallel to the principal dorso-ventral-axis of
the original egg.

Summary and Conclusion.

An isolated blastomere of the Molgula egg segments as if still
forming a corresponding part of an entire embryo. The cleavage
phenomena are strictly partial, as regards the origin of cells,
the inclination of cleavage-planes, and especially in respect to
the rhythm of segmentation. The general appearance of the frag-
ment differs materially from that of a half of a complete embryo, for
the reason that rearrangements of the blastomeres occur, which
tend progressively to mask the partial nature of development.
The end result is a larva of less than normal size, and with defects
in certain of its systems. These defects are undoubtedly due to
the fact that but a portion of the normal amount of material is
available for the formation of the larva; that, for instance, the
chorda of a larva derived from a one-half blastomere, receives but
one-half of the normal number of cells, and consequently a chorda
of one row, and not two rows of cells, results.

In conclusion, one is constrained to adopt the view of Roux-
namely, that in Molgula as in the well-known case of the echino,
derms (Driesch, Wilson, and others) the development begins as
a partial one, but that the missing part is gradually supplied by
the cells already present. Driesch is also entirely correct, as far
as the end result, a nearly complete larva, is concerned.

Explanation of Plate IV.

Magnification of figs. 1-3 about 280 diameters ; of all other figures, 250
diameters. The arrows show the direction of cleavage.
Fig. 1, 8-cell 8tai;e of Ciona from Castle (fig. 23), from the left side.
Fig. 2, 16-cell stage of Ciona from Castle (fig. 24), from the left side.
Fig. 3, 24-cell stage of Ciona from Castle (fig. 43), from the right side.
Figs. 4-8, cleavage of the left }4 blastomere of Molgula, from the side.
Fig. 4, J embryo.
Fig. 5, J embryo.

Digitized by VjOOQIC

The Ascidian Half-Embryo. 57

Fig. 6, A embryo.

Fig. 7, paange to i} embryo.

Fig. 8, H embryo.

Figs. 9-12, cleavage of ibe rigbt }4 blastomere, from tbe side.

Fig. 9, } embryo.

Fig- 10» A embryo.

Fig. 11, if embryo.

Fig. 12, JJ embryo.

Figs. 13-16, cleavage of the rigbt }4 blastomere, from the front.

Fig. 13. } embryo.

Fig. 14, I embryo.

Fig. 15, passage to ^ embryo.

Fig. 16, complete ^ embryo.

Explanation of Plate Y.

Figs. 17-24, cleavage of the }^ blastomere, ventral view.
Fig. 17, f embryo.
Fig. 18, passage to ,*,.
Fig. 19, complete ^.
Fig. 20, ^ embiyo.

^f^' 21 f -fi embryo, immediately after division.
Fig. 22, -fj embryo, in resting condition.
Fig. 23, passage to Jf stage.
Fig. 24, complete \i embiyo.

Fig. 25, }4 larva. The arrow indicates the long axis.
Fig. 26, two }i larvae, from same egg. The arrows indicate the principal

Digitized by VjOOQIC

III. — The Rutherfurd Photographic Measures of Sixty-five Stars

near 61 Gygni.

Read May, 1897.

1. It was but natural that Mr. Rutherfurd, in developing the
art of astronomical photography, should try his skill upon that
star which has attracted the attention of so many investigators
ever since Bessel proved by it the possibility of determining
stellar parallax.

Of these photographs of 61 Gygni and its surrounding stars
taken by Mr. Rutherfurd, nineteen, exposed between 1871, Nov.
9, and 1874, June 13, were measured by Miss Ida Martin more
than twenty years ago, but have remained unreduced until re-
cently placed in my hands for that purpose by Professor J. K.
Rees, Director of the Observatory. The present paper contains
the results of measures of position of stars surrounding 61 Gygni,
and will be followed by a second paper containing the results of
an investigation of the Parallax of 61^ Gygni. The methods of
reduction used so far as measures of distance are concerned are
those presented by Dr. Harold Jacoby in earlier Gontributiona
from this Observatory.

2. In Table I are given the general data of exposure of the
plates, including the computed values of the zenith-distance, par-
allactic angle and refraction factor.

3. Table II contains the means of the refractions computed for
the Eastern and Western impressions from the data of Table I by
the formulae

-— = /c[tan2Ccos2(i> — 3)-fi]
TT — p = — }Kco8eci"tan*Csin2(|) — q).

The argument for entering this table is p,

4. Table III. — The corrections to the position-angle due to pre-
cession, nutation and aberration will be found in column two.
These were computed by the formulae

Digitized by VjOOQIC

Sixty-five Stars near 61 Cygni. 59

a'^20."o68ina8ec<5 y' = cos a tan (J

/?'= coflflsecrf <5' = sin a tan <J

Ap = ( r— «' — ^«' — -B/3' — C!^' — -D«J'.

The epoch T= 1873.0 has been selected to which to reduce all
the observations. The substitution of the coordinates of 61 ^ Cygni
for this epoch gives :

^i>7i = — 36"+ [1.254] A + [9.9560] 5+ [9.7460] C+ [9.742] D.
^1>75, = — 18 + [1.254] A -f [9.9560] B + [9.7460] 0+ [9.742] D,
^Pn= o 4- [1.254] A + [9.9560] B + [9.7460] C4- [9-742] D,
APt4 = +i8 + [r.254] ^ + [9.9560] B+ [9.7460] C-f [9.742] D.

Where ^p^i denotes the correction to be applied to the posi-
tion-angle for the plates made in 1871, and so on in the other years
as denoted by the subscripts.

6. Precession and nutation have no effect upon the distances ;
but aberration does have, and its amount is given by :

7":=(tanc8in«5-f-8iuacosrf) sin i"
(J"= — cos a 008 «5 sin i"

As =(0/'+2M")«

For 61^ Cygni this becomes

A8=J[4.i4in]C+ [4.433n] -2>J« for all years,

and is additive to the distances to reduce them to 1873.0. This
factor of 8 is given in column three of Table III.

6. The logarithms of the Besselian day-numbers, taken from
the American Ephemeris, are :


log A.







1. 104

1. 178





1. 198




















1. 218






II, 12,











16, 17,

9.41 1




18, 19,






7. In the second portion of Table III. is given the mean of the

Digitized by VjOOQIC

60 Rutherfurd Photographic Measures of

East and West zero-corrections computed for each by the

formula *

t; = }X;2 tan 6 — y-fa?

in which v is the zero-correction to be added to all observed posi-
tion angles of each plate.

In the next column are the special corrections f required by the
position-angles of the Western impressions in consequence of
using the same zero-point in measuring both Eastern and West-
ern impression s4 The sum of these two columns is then given in
column six, which, therefore, contains the final correction as
actually applied in the reductions.

8. In Table IV. is given the tangent correction. This is always
negative and its unit is .oooi divisions of the micrometer. It
has been computed by the formula :

Correction = — Ja^d'sin* i" = [1.7887U ] a*

where s denotes the distance in divisions of the glass scale and d
is the value of one division of the scale in seconds of arc.

Table V.— Measures of Distance.

9. The first column contains the numbers of the stars in order
of right ascension and also in parentheses, for convenience of ref-
erence to the original measures and plates, are the numbers as as-
signed by RuTHEEFUED. The number of the plate is given in col-
umn two, after which follows the observed distances for the Eastern
and Western impressions. The numbers set down are the frac-
tional part of the measured distance expressed in divisions of the
glass scale, the whole number of divisions being ordinarily the
same as that given in th^ final corrected distance. Where there is
a change of .8 or .9 in the observed distance, it is an indication of
a change of a unit in the whole number of divisions in passing
from the observed distance to the corrected mean. In columns
five, six and seven are placed the corrections as applied for refrac-
tion,§ aberration II and scale^f respectively; these, with addition of

♦Annals N. Y. Acad, of Sci., Vol. VI., p. 272.
t Ibid., p. 278.
t Ibid., p. 240.
\ Table II and Paragraph 2.
II Table III and Pai-agrapb 4.

Online LibraryThos. L. (Thomas Lincoln) CaseyAnnals of the New York Academy of Sciences → online text (page 6 of 28)