76 
Indiana University Studies 
The type insects may include some material of variety 
ojMcus; the leaves containing the pulvinus galls were also 
infested with galls of opacus, and the two were bred together. 
Unfortunately these two insects cannot be separated with any 
certainty, as explained further on, but I have recovered out 
of this lot what would appear to be opacus insects to a total 
of seven per cent, which almost represents the number of 
opacus galls. The paratypes of pidvinus are, then, probably 
almost pure, but should be used with caution in settling any 
fine point. 
If one should ever recognize a physiologic variety, here 
is the instance. The galls of pulvinus show some relation- 
ship to those of other varieties of the species, but in many 
respects are distinct, distinct enough to represent another 
species rather than a variety of verrucarum. The insects 
however do not appear to differ morphologically from those 
of variety opacus, except for vague average differences which 
cannot be utilized with particular individuals. I have spent 
considerable time trying to find distinctive characters which 
would apply to all the individuals of a large series, and altho 
extreme individuals are to be distinguished, no character, 
certainly no set of characters, appeared to me constant enough 
to describe as the differences between opacus and pulvinus. 
Mr. L. 0. Morgan, a graduate student in our laboratories, 
has independently examined this material, and reaches a de- 
cision in accord with my own. Mr. Morgan has been remark- 
ably successful in making distinctions in difficult cases ; since 
he fails to find differences, I question whether other observers 
will. I grant that statistical methods may separate opacus 
and pulvinus insects ; if so, the sharp distinctions of the galls 
will still appear in striking contrast to the close identities 
of the insects. It is reasonable that two taxonomic groups 
should differ either morphologically or physiologically and 
not in both regards. It is however so often true that the 
two sorts of characters differ simultaneously, — or it is so 
often the case that we utilize the morphologic without con- 
sidering the physiologic data, — that we are perhaps unduly 
impressed with a “physiologic’' variety. If galls or other 
physiologic data were not available in such a case as this, 
there would, without doubt, be no one who would distinguish 
the two things. There would follow confusion and outright 
