M. Coste on the Formation of Cells. 383 
blast, inclosed in the thickness of the parietal membrane, disap- 
pearing in proportion as the cell-contents are introduced into the 
cavity of the latter. 
But when we search for the facts upon which so radically ex- 
clusive a theory is founded, we experience the twofold astonish- 
ment of not meeting, in those authors who originated it, with a 
single example the value of which we cannot seriously contest, 
and of not finding in nature those abundant proofs which cause 
a system to prevail, or at least allow the formula to remain, as the 
most faithful expression of the most numerous category. Thus, 
m examining the proofs cited by Schwann in support of this 
hypothesis, we find that they may be reduced, as M. Vogt has 
remarked*, to a single observation directly made on cartilage ; 
and yet this observation, presented by Schwann himself as very 
doubtful, has been shown to be false by the researches of M. 
Vogt on the cartilage of the accoucheur-toad. In fact, in a very 
large number of cases, the nucleolus, to which the theory attri- 
butes the exclusive privilege of causing the amorphous matter to 
produce the cell-walls; im a very large number of cases, I say, 
the nucleolus never appears free and isolated in the midst of the 
cytoblastema. On the contrary, we always find that this cor- 
puscle, even from the first commencement of its appearance, is 
still inclosed in the cavity of the cell, which is previously formed, 
frequent instances of which we find in the tissues of the embryo 
of osseous fishes ; it is evident that in these cases at least the nu- 
cleolus has taken no part in the formation of the cell, as it was 
not in existence when the latter was produced. In other cases, this 
corpuscle did not appear at any period of the existence of the 
cells, and consequently we should have no motive for making it 
intervene as a determing cause, since it does not leave to the 
theory even the pretext of co-existence. This may be easily veri- 
fied by studying the development of the large cells which form 
the internal expansion of the umbilical vesicle of serpents. 
Hence the tardy appearance of the nucleolus in certain cases, 
* “Tn examining the proofs cited by Schwann in support of his opinion, 
we find,” says M. Vogt, ‘‘ that they resolve themselves into a single obser- 
vation made upon cartilage; and moreover it ust be remarked that M. 
Schwann himself presents it to us as very doubtful. In fact, I believe that 
I have shown in my researches on the accoucheur-toad, that this opinion is 
probably erroneous, and that an old cellular cavity almost closed, or the 
half-absorbed nucleus of an old cell, had been taken for the nucleolus of a 
eell in process of formation. I have then no doubt, from the observations 
which we now possess, that the nucleolus, far from being the primitive ru- 
diment of the cell, is, on the coutrary, nothing more tlian a formation result- 
ing from one of the last metamorphoses which the cells undergo. .. . . These 
various facts could not fail to excite doubts in my mind regarding Schwann’s 
theory, and I concluded by considering that it was based upon few facts only, 
and these were mostly susceptible of another interpretation.” 
