44 
Psyche 
[June 
nishes proof that oculata is not a simple dominant. Further- 
more, in B, C, and E oculata, presumably heterozygous if 
illepida is recessive (because they were the offspring of 
illepida females) produced low percentages of illepida. If 
ocidata were dominant the expected proportions of illepida 
would be 25% in B and C, and 50% in E. 
The high percentage of ocidata offspring in all the crosses 
occurs in a manner showing close approximation to the 
results expected from segregating recessives with a rare 
production of illepida, which could be explained by the 
presence of two or more pairs of recessive genes either 
of which in the homozygous recessive state could produce 
oculata. Thus, if there are two genes any one of which in 
the recessive state would produce oculata, a cross like those 
in D (Table 1) could theoretically produce 25% illepida, 
provided that the illepida parent is heterozygous for both 
pairs of genes. The actual number obtained was 19%. If 
it is assumed that any one of three genes in the recessive 
state might produce oculata, the actual ratios of the five 
crosses will conform more closely with the theoretical. 
Whatever the genetic mechanism may be, it can be 
concluded that illepida is not a simple recessive. If such 
were the case, inbreeding could not possibly produce any- 
thing except illepida, but actually the result is 50% oculata. 
(Table 1, F.) Conversely, the fact that inbreeding of five 
pairs of oculata from an oculata female gives only ocidata 
suggests that this character may be a recessive. There are 
of course the other patterns on the vertex to be considered, 
but no attempt is being made to explain their occurrence. 
Even though the manner in which the variations are 
produced cannot be explained easily and regardless of the 
percentages of the two varieties obtained from the dif- 
ferent crosses, the important point is that the characters 
which have caused the naming of oculata and illepida are 
inherited. Moreover this appears to be true for the other 
varieties previously mentioned, including those with dif- 
ferences in darkness of wing venation. 
An eighth variety recognized by Smith (1932) is separata 
Banks which “is distinguished by the absence of a connec- 
tion between the black loop under the antennae and the 
black genal band.” If the connection is very faint the 
