266 T. H. Morgan 
chromosomes are present—the largest one being therefore the 
double accessory (Fig. XIV, U-W; Fig. XVI, P,Q). This differ- 
ence in the behavior of the small accessory has made the study of 
this species very perplexing. It might appear that we are deal- 
ing here with two species, one with three and the other with four 
chromosomes, but this view is negatived by a number of facts. 
A close examination of the sexual males fails to reveal two types, 
and since the sexual individuals of different species are readily 
determined, the evidence is strongly in favor of one species. On 
the tree, at New Bedford, Mass., on which this species was 
found swarming out of the galls only one kind of gall was present, 
and since I have studied the sexual forms of the commoner galls 
in this region it is improbable that two species here exist. ‘The 
fact that all intermediate types between union and separation of 
the chromosomes are found also indicates that they all belong to 
the same species; in some individuals the two chromosomes are 
always separate, in others united by a connecting thread; in others 
the smaller is stuck to the side of the larger, and others show 
only a single larger chromosome. On the assumption that there 
are four chromosomes in the male, the results can be brought 
into harmony with the facts found for the stem-mother’s egg, and 
for the number of chromosomes present in the spindle of the male 
egg. This conclusion also harmonizes with the results on the 
Other species). fallax. | 
In the resting stage following the first division, the nuclei of 
the larger cells contain three or four chromosomes each, and it is 
interesting to note that the type that prevails in each individual 
corresponds with the type found in the stage preceding division as 
is shown when both stages coexist in the same testis. 
The equatorial plates of the second spermatocytes also show one 
or the other type (Fig. XVII, LZ, N, O); but the chromosomes are 
crowded and these stages are difficult to find. I have not studied 
them extensively. As the chromosomes divide it is apparent that 
there is no lagging chromosome (Fig. XVII, D’, E’, F) and a few 
successful cases show clearly that the three or four chromo- 
somes, according to the type, divide and pass tothe poles. ‘These 
are the two cells from which the spermatozoa develop. 
