Vol.. 6, 1920 
ZOOLOGY: H. SHAPLEY 
689 
centage of affected forms appears temporarily re-established, notwith- 
standing that 320 of the normal ants had been returned to the nest after 
the June collections. Among the ants taken October 18 and 24 were a 
number of callows with well- developed vestigial wings; this also indicates 
that the tendency to produce pterergates still exists. 
2. There is no obvious external reason why this particular nest should 
show^ such frequent reversion to a remotely ancestral condition of the 
worker ant. The intermittent war with Argentine ants {Iridomyrmex 
humilis Mayr), which is apparently destined to end in the elimination of 
'most of the native ants in the California valleys, is no more severe for the 
nest of pterergates than for many other nests of Pogonomyrmex. This 
nest, however, has access to very little wild barley and similar grasses, a 
common food of the species; on the other hand, it is exceptionally well 
provisioned with mixed grains from a nearby feed store whenever the 
Argentine ants permit the nest to be opened for normal harvesting activi- 
ties. 
3. A small colony of the same species, less than ten feet distant from 
the nest containing pterergates, yields workers indistinguishable from the 
normal ants of the affected nest, but no pterergates; in fact, an examination 
of several thousand individuals from some fifty other colonies of this 
species within a radius of two miles has shown only the one pterergate 
listed above. 
4. In all details of thoracic structure — size, sculpture, pilosity — the 
pterergates are identical with normal ergates of this and neighboring 
nests. In excavating and guarding the nest, and in harvesting, the pter- 
ergates and the normal workers participate equally. Hence these ab- 
normal ants are certainly workers rather than modified fertile forms, such 
as pseudogynes, microgynes, and jS-females. 
5. The seventeen young queens and two males, taken from the nest in 
[une, 1920, appear to be normal in every way when compared with queens 
and males from other nests of the same species. 
6. Grouping the pterergates roughly in order of the development of 
vestiges, we have the following enumeration: 
With minute veinless wing-sacs, or with stubs of broken wings 385 
With sacs from 0.5 to 1.0 mm. long and indistinct veining 219 
With transparent, clearly veined winglets from 0.8 to 1.5 mm. long 132 
The gradation of wing vestiges is, however, perfectly continuous, from 
small protuberances (without appendages) on the mesothoracic segment to 
the most developed winglets, with venation approaching that of the wing 
(6 mm. in length) of the mature queen. This continuity in structure 
may be significant for the problem of variation and the origin of castes. 
7. Four ants from the affected nest have vestiges of both posterior 
and anterior wings — a phenomenon not heretofore recorded, as only an- 
terior wings are represented in all other pterergates. 
