Psyche 
[September 
162 
fested with ciid beetles, (3) parasitized by this species of Cephalono- 
mia, and (4) containing females ready for mating must be fairly 
small. Thus the presence of wingless males may assure that most 
females will be mated and capable of producing female offspring, while 
the winged males may represent a mechanism for providing for enough 
outbreeding to offset the serious effects of continued inbreeding. 
These remarks will apply equally well to the females if one substi- 
tutes “finding the host” for “mating.” That is, the presence of winged 
females throughout the range may be a mechanism for spread of the 
species into new areas containing the ciid hosts, although many winged 
females are doubtless wafted into unsuitable locations. Wingless 
females presumably attack ciids occurring in the same or neighboring 
fungi, and such females might eventually exhaust the available hosts; 
at the very least, a given growth of fungus will die out after a period 
of time. Thus the occasional fertilized, winged females which locate 
a new supply of ciid beetles may play an important role in the survival 
of the species, while the wingless females are responsible for localized 
build-ups in the population. 
Having said all this, I must state that I consider it possible that 
genetics is not directly involved in the polymorphism of Cephalonomia 
perpusilla. Salt (1952) studied the ichneumonid wasp Gelis corrup- 
tor, in which the females are always apterous, the males either 
macropterous or micropterous, and found that on a large host the 
males develop into fully winged individuals but on a small host, pro- 
viding meagre nourishment, the males develop into micropterous indi- 
viduals. In an earlier paper, Salt (1937) showed that Trichogramma 
semblidis produced apterous males when reared on one host, alate 
males when reared on certain other hosts of equal size (or actually 
smaller) ; in this case it is the quality of food rather than its quantity 
which effects morphogenesis. The polymorphic forms of the chalcidoid 
wasp Melittobia chalybii have also been shown to be conditioned by 
trophic factors : the larvae produced from the first few eggs laid on the 
host feed on the blood, and develop into a short-lived form with 
much reduced wings; larvae produced from eggs laid after the first 
few days feed on other tissues and develop into “typical” adults 
(Schmieder, 19 33). In this instance the short-winged forms develop 
much more rapidly and permit a rapid population build-up within the 
large host. 
Clearly such mechanisms as these may have strong selective value 
even if their basis is not “genetic” in the usual sense. These morphs 
are comparable to the castes of social insects rather than to genetically 
