1934 ] 
Polyploidy in the Hymenoptera 
165 
There is very good evidence that about half of the Angios- 
perms are polyploids. From such studies as those reviewed 
by Darlington (Darlington’s Cytology, p. 212) we see that 
in polyploid series one finds many species with double the 
basic chromosome number, fewer with the number obtained 
by a second doubling and still fewer resulting from further 
doublings of the basic number. Such would seem to be the 
case even in this small sample from the Hymenoptera. A 
second point also demonstrated by the abundant plant ma- 
terial is that in polyploid series there are few species with 
prime numbers as their haploid count. In the Hymenoptera 
there are no species with prime numbers. If the chromo- 
some numbers in the order were entirely fortuitous, one 
should expect several species to show prime numbers as 
their haploid count. 
The rareness of polyploidy in the animal kingdom and 
particularly in dioecious species has been considered to be 
due, at least in part, to the fact that such species are in one 
sex heterozygous for the sex chromosomes. Multiples of 
the chromosomes in the vast majority of animals would thus 
cause an upset of the balance between sex chromosomes and 
the autosomes as was suggested by Muller (Am. Nat. 59) in 
regard to polyploidy. He suggests, however, that polyploidy 
might occur in the Hymenoptera. 
In the Hymenoptera it may have been possible to avoid 
this difficulty since they are almost unique among insects in 
that their males are haploid. Thus if there is any order of 
insects in which polyploidy is possible it would seem to be 
the Hymenoptera. 
The suggestion that the series of chromosome numbers in 
the Hymenoptera may have arisen by polyploidy rests, then, 
on these considerations : 1, all but four of the species have 
chromosome numbers which are 4 or a multiple of four; 2, 
the existence of haploid males may have allowed polyploidy 
to occur. 
