Kinsey: The Genus Neuroterus 37 
do not vary to any degree between varieties. This is in 
contrast to the subgenus Dolichostrophus where the species 
are best distinguished on insect morphology. 
Within the subgenus, the galls show some evolution from 
the polythalamous, inseparable, indefinitely formed vernns 
galls to the monothalamous, separable, definitely formed 
umbilicatus galls. The species between may be arranged in 
mostly one linear series which is, of course, too ideal probably 
to represent the real paths of evoluton. But at the best the 
galls remain simple. These simple galls, the very close sim- 
ilarity of the sexes of a species and of the forms of alternate 
generations, are the best indicators of the relatively primitive 
position of the group, more primitive than either Neuroterus 
or Dolichostrophus, to which in many respects our present 
group is related. Batatus is here treated as a Dolichostrophus, 
but it does show some considerable relationship to Diplobius. 
The males of all these groups are more distinctive than the 
females. The best characters for recognizing Diplobius adults 
are the developed malar furrow, the third antennal segment 
winch is hardly longer than the fourth in either the female 
or the male, the simple tarsal claws, size under 1.5 mm., often 
under 1.0 mm., and the close similarity of the male and the 
female. 
Directly alternating generations are described for only 
two of the varieties, belonging to two species, but other 
agamic and bisexual varieties are known in the species niger 
and vernus, and may be predicted with considerable certainty 
for their 14 described varieties. In these cases the alternating 
insects are practically identical except for their modes of 
reproduction, and occur in galls whch are almost identical 
except where some variation is introduced by occurrence on 
different tissues. Two species, howertoni and verrucarum, 
are known only from an agamic form, while floricola has 
only a bisexual form described. It does not seem that the 
alternates of any of these three have been described; it will 
be interesting to find if they occur, as seems likely, in similar 
galls in the two generations. A remarkable addition to our 
knowledge of reproduction in this genus may come if we 
prove, as now seems to be indicated, that two species, um- 
bilicatus and saltatorius, have all of their generations agamic, 
altho they have two generations yearly. 
