SOME PROBLEMS OF COELENTERATE ONTOGENY 515 
affording no appreciable advantage over that used in the former 
instance. At this time the chromatin appears only in the form 
of extremely irregular, flocculent patches, scattered here and there 
through the cytoplasmic cell-like aggregates. The same elongated, 
clavicular, or dumb-bell shaped nuclei previously described are 
found in the newer material treated by latest methods. Under 
ordinary treatment the chromatin stainsso intensely as seriously to 
obscure details of structure. Only by prolonged destaining and 
clearing, and by more delicate staining with picro-hematoxylin 
has it been possible to reduce somewhat the flocculent aspects. 
When this is done one may distinguish a granular chromatin con- 
stitution, but the chromosomes have defied all attempts to render 
them distinctive either inform orinnumber. This relates to con- 
ditions in early cleavage, as was pointed out in the previous paper, 
aspects of mitosis become fairly clear in later cleavage. Beckwith 
describes mitoses in early cleavage but makes no reference to the 
anomalous conditions here described. 
With all the pains taken in preparation of my material it must 
be allowed that these conditions are not artifacts, but facts enti- 
tled to the same consideration as others of similar treatment. It 
must be admitted however, that, even at best, our latest refine- 
ments as to staining technique must be accepted as only tenta- 
tively trustworthy. In other words, it becomes more evident 
every day that in protoplasmic and nuclear metabolism there are 
such incessant and intricate variations of chemical conditions that 
one may not assert that a given stain or fixing agent affords any 
certain test of a given state at a given time. On the contrary, it 
will not be denied that a given stain may act in one manner on 
one cell and on another very differently; or indeed, that it may 
in another case fail utterly to yield any results whatsoever in 
differentiation. Under the aspects of chromatin organization, 
or perhaps better, lack of organization, as here portrayed, it has 
not been possible to obtain any definite information as to the 
number or character of the chromosomes. But it may be said, 
as before mentioned, that in Clava, as in Pennaria and Hydraec- 
tinia, there is an enormous fragmentation and dispersal of chro- 
matin at the time of maturation, most of which must be utterly 
