Kinsey: The Genus Neuroterus 
23 
and where I have found both insects inhabiting both kinds of 
galls ! This observation is recorded further under exiguus. 
The discovery of these instances of gall polymorphism 
came thru the use of comparative descriptions and the rec- 
ognition of varieties. The phenomenon must now be searched 
for in all primitive Cynipidse, but it very probably must not 
be expected in the higher gall wasps. Gall polymorphism has 
probably been responsible for many misdeterminations where 
too much dependence has been placed on the character of 
the gall, and in some surprising instances has been the reason 
for the creation of complete synonyms among the species of 
Neuroterus. Dr. Patterson has discovered some evidence in 
Cynipidse that male-producing and female-producing insects 
(females) of the agamic generations of a species may live 
in different kinds of galls, and this should be investigated in 
connection with gall dimorphism and with the galls of what 
I have called physiologic species. 
NUMBER OF SPECIES 
There is considerable significance in the rate of discovery 
of the species and varieties of Neuroterus occurring in the 
United States. The record may be summarized by twenty 
year periods as follows: 
By 1863, 3 varieties, 3 species were known. 
By 1883, 9 varieties, 4 species were added. 
By 1903, 17 varieties, 6 species were added. 
By 1923, 34 varieties, 3 species were added. 
The rate of discovery of species has considerably de- 
creased since 1890, and it would appear that we may expect 
a further decrease in the future. Probably not more than 
a couple of dozen species of Neuroterus exist in the United 
States; and Mexico, with its extensive, unexplored oak areas, 
will probably add only moderately to this number, for faunas 
closely related to those of even more southern Mexico are 
to be found in parts of the southern United States. 
The rate of discovery of the varieties has, on the other 
hand, distinctly increased, and promises no retardation for 
a considerable time. We have an apparently reliable means 
of computing the number of varieties which may exist for 
the species already described. Since, as I have shown in 
