L JUN'84 1921 
PROCEEDINGS 
OF THE 
NATIONAL ACADEMY OF SCIENCES 
Volume 6 DECEMBER 15. 1920 Number 12 
PRELIMINARY REPORT ON PTERERGATES IN POGONOMYR- 
MEX CALIFORNICUS 
By Harlow Shapley 
Mount Wilson Observatory, Pasadena, Cai^ifornia 
Communicated by William Morton Wheeler, November 6, 1920 
The phylogeny of castes among social insects has a recognized im- 
portance in the question of development of organic forms through con- 
tinuous or saltatory variation. For the Formicidae the castes are already 
well established in the early Tertiary/ among the first fossil records of 
these insects, although in the primitive ancestral v^asps neither the social 
habit nor the polymorphism of the female appears. Differentiation into 
soldier, worker, and fertile forms is also a long-standing condition with 
the termites, probably dating from the Cretaceous. 
In discussing the origin of castes among ants, Wheeler has considered 
the bearing of the subapterous and apterous females (particularly in the 
genus Monomorium) on the problem of variation through mutation.^ 
He concludes that although the sterile worker-females are now almost 
universally wingless, we should not believe that this micronotal wingless 
condition arose through mutants. Rather "this view of the castes, at 
least so far as their origin is concerned, cannot be maintained, because 
all the available evidence points to their being merely the surviving ex- 
tremes of graduated and continuous series of forms, the annectant mem- 
bers of which have suffered phylogenetic suppression or extinction." For 
the termites Thompson and Snyder find few intermediate types, but trace 
a gradation of characteristics throughout the series of five castes (workers, 
soldiers, and three fertile forms), and find a further argument against 
mutational origins in the fact that the different castes are essentially 
constant both in occurrence and in structure. ^ 
It is of interest, in connection with these views on the phylogenetic de- 
velopment of polymorphism, to find in typical worker ants some onto- 
logical evidence of early stages in the origin of castes — to find, in particular, 
that the embryonic vestiges of wings, discovered in workers of Formica 
by Dewitz,^ do not invariably disappear with the passing of larval and 
