282 
B. W. WELLS 
over a relatively long period of time (one to four weeks roughly) , the chances 
of catching a nucleus in the act of division are very remote. This of course 
is true whether the division is of the mitotic or of the amitotic type. For 
this reason if for no other it is impossible to determine definitely the nature 
of the nuclear cleavage. 
The evidence for amitosis is of an indirect character, such as the frequent 
occurrence of lobulate nuclei ; the very rare display of a constricted nucleus ; 
the frequent occurrence of two nuclei in contiguity ; and the very irregular 
distribution of the nuclei in the cell. But, as heretofore stated, the ex- 
tremely high infrequency of divisions of any kind makes it quite impossible 
to determine the nature of the divisions definitely. 
Differential Growth between Larva and Gall 
Data should be given concerning the difference in time of occurrence 
of the grand periods of growth of the larva and gall respectively, as this is a 
matter of more than passing interest. This difference is readily brought 
out by the following table: 
Measurements of larva and gall at different stages of their concomitant development 
Width of larva Height of gall 
Stage I 0.20 mm. 0.14 mm. 
Stage 2 0.22 mm. 0.80 mm. 
Stage 3 0.31 mm. 3.55 mm. 
Stage 4 0.40 mm. 3.75 mm. 
Stage 5 1 .69 mm. 5.00 mm. 
The grand period of growth for the gall is that of its early existence, 
while that of the larva comes later when the gall is more than half grown. 
This inhibition of growth in the larva for an extended period, which is 
coincident with gall morphogenesis, is in the mind of the writer a very 
significant phenomenon. Just what it implies in its entirety is quite 
unknown, but we may feel reasonably sure that it is related in some manner 
to the production of the gall-making stimulus. The energy of the larva 
appears at first to be consumed in producing the stimulus, and only when 
gall development is well on its way to maturity is this energy released for 
the anabolic processes of the larva itself. This is of course but an assumed 
generalization which cannot be proved until the problem of the nature of the 
stimulus involved has been cleared up. 
The Stimulus Problem 
In many instances the discovery of special structures has been of much 
value in interpreting function. When we raise the physiological problems 
of the nature of the gall-forming stimulus in these prosoplasmas and look 
for some hints concerning it from structural conditions, it must be confessed 
that none are given. I cannot find a single structural fact produced by 
