98 
Psyche 
[September 
from these widely separated areas. Such differences could be attributed 
to geographical variation ; however, they also have distinctly different 
host preferences. The Brazilian population has a wide host range 
and is a destructive pest of citrus, while the Mexican population is 
of no economic importance, infesting the rose apple and only occasion- 
ally the guava. 
The case of the compound sex determining mechanism encountered 
in A. serpentina is also interesting as this type of system appears to 
be rare in Diptera. Dobzhansky (1935) reported XiX 2 Y system in 
Drosophila miranda Dobzh. believing it to be an example of deter- 
minate disjunction. Cooper (1946), however, clearly showed that a 
XiYX 2 trivalent was actually formed during meiosis. Boyes (1952) 
found the same type of trivalent formed in Hylemya fugax (Meig.). 
It is possible that A. serpentina may also produce a trivalent, but the 
preparations of gonadal tissue using the squash technique were not 
suitable for establishing the interaction of the three sex chromosomes. 
The different chromosome number of 2n=iO reported by Emmart 
(1935) for A. ludens probably resulted from an incorrect interpreta- 
tion of chromosome morphology in her study of meiosis in pupal and 
adult testes. In the present study, larvae as well as adults were 
studied from Cuernavaca, Morelos, the same locality from which 
Emmart collected most of her material. A diploid number of 12 
was always recorded. Meiotic figures in the testes without exception 
had a characteristic haploid number of 6. 
Little can be said about the other Anastrepha species at this time. 
It is likely that a more thorough investigation of the karyotypes 
within this genus will uncover many interesting phylogenetic rela- 
tionships which can now only be hinted at on the basis of the present 
study. 
Spermatogonial metaphase plates can be put to good use in eval- 
uating the chromosome morphology of those species of tephritids whose 
larvae are unknown, as in the case of A. spatulata, or whose larvae 
cannot be readily maintained in the laboratory. Such determinations 
can also be used to obtain tentative identification of larvae collected 
for the first time, and whose chromosome morphology is known only 
from previously captured adults. 
From the cytological data thus far accumulated for the family 
Tephritidae, it appears that the variation between karyotypes is suf- 
ficient to warrant more attention from the taxonomists of this group. 
New methods of handling animal chromosomes, such as the many 
pre-treatments now available, followed by simplified squash tech- 
