

580 THE AMERICAN NATURALIST. [Vor. XXXVI. 
support, and in a few days the proportions of a normal hydra 
were assumed. 
Serious difficulties arose in the microscopic study of the 
prepared sections, which it may be well to mention. Although 
undoubted karyokinetic divisions were seen in both neuro- 
muscular and interstitial cells of the ectoderm, and in th- 
endoderm, it was found almost impossible to determine the 
exact amount of dividing tissue, for the following reasons. 
Throughout the entire period of regeneration there is an 
active production of new nettle cells, which are formed from 
interstitial cells of the ectoderm. This process begins by a 
slight sickle- or moon-shaped thickening of protoplasm along 
a part of the cell wall. In certain sections this thickening 
appears as a small, darkly stained rod which may easily be 
mistaken for a small nucleus in process of division. Again, in 
sectioning, this curved, thickened red may be cut at such an 
angle as to bring distinctly into view only its extremities, pro- 
ducing the appearance of two small groups of darkly stained 
material. These could be distinguished from two separating 
groups of chromosomes only by bringing into focus the con- 
necting line of the rod, which would lie slightly above or below 
the ends. In many cases it was found almost impossible to 
determine whether such a cell was forming a cnidocyst or was 
dividing karyokinetically. These difficulties were increased 
by the small size of many of the interstitial cells, by the fact 
that the wall of such cells is not sharply defined, and by the 
fact that at the time of the formation of nettle cells it is very 
often difficult to make out the nuclei of the changing cells. 
Moreover, many nuclei of both neuro-muscular and inter- 
stitial cells were found to be in a state that distinctly suggested 
activity of some sort, — the chromatin mass large and loose, or 
even in many interstitial cells broken and scattered throughout 
the nucleus. This appearance was very different from that o 
nuclei of typical resting cells in the same section, the latter 
containing one or two well-defined nucleoli. Yet the large 
number of nuclei. with the chromatin scattered as just described, 
in sections containing comparatively few karyokinetic spindles, 
would seem to indicate that all such cells were not necessarily 

