242 T. H. Morgan 
duction of both sexual forms, the male and the sexual female; 
hence the conditions do not determine sex in the sense of producing 
either males or sexual females, but bring to an end parthenoge- 
netic and introduce sexual reproduction. It follows, I think, with 
probability that we are dealing with two different things here, and 
that confusion has resulted from supposing them to be the same. 
These ideas led me to abandon the hope of finding the clue to 
sex determination in the external conditions, however important 
these factors may be in cyclical changes in sex production. In 
recent years the repetition, by several zoologists, of the older 
experiments on tadpoles, caterpillars, etc., that had been accepted 
as demonstrating the influence of external conditions, has shown 
with great probability that those experiments did not establish their 
claim. The discovery at the same time of an internal mechanism 
associated with sex determination has gradually brought convic- 
tion that internal and not external factors determine sex. 
In the group of insects, and especially in the group of Hemip- 
tera to which the aphids belong, sex determination has been 
shown to be associated with an internal factor with which the 
number or the kind of chromosomes is closely linked. Therefore 
we should expect, a priort, to find in the aphids some similar factor, 
if we are to ascribe to it any profound significance. 
The most hopeful field for investigation seemed to be in cases 
where eggs of different sizes exist associated with male and female 
development. In the phylloxerans, near relatives of the aphids, 
these conditions are found. The presence in America of many 
species of Phylloxerans on the hickories gave me the opportunity I 
sought, and a fortunate discovery of two species, in which the 
sexual eggs could be obtained in vast numbers, has made the work- 
ing out of the problem possible, although extremely laborious. 
My first results were published in 1906; since then I have continued 
to study the group, but only during the spring and summer of 
1907 did I obtain the material that has made it possible to work out, 
not only the spermatogenesis, but the entire cycle of cytological 
phenomena. ‘The main facts in regard to the spermatogenesis 
were made out in October and November, 1907, and the briefest 
possible statement of the main facts was given before the Society 
