ZOOLOGY: J. WEB 
61 
old, was found whose gonads were macroscopically and microscopically well 
developed ovaries. The next frog was again a male. Although the possibility 
of an error in method seemed excluded the writer did not wish to publish the 
fact that both sexes appear in parthenogenetic frogs without having checked 
the result by a new series of experiments. 
These experiments were started in February, 1917. The same precautions 
as in the older experiments were used. Copulating females which had not yet 
laid any eggs were separated from their males and kept separated for at least 
twenty-four hours. The females were repeatedly washed with water during 
the time of isolation, and directly before the experiment were submerged in 
90% alcohol and left there to die. They were taken out, their abdominal 
cavity was opened with sterilized instruments and the oviduct laid bare. The 
eggs were taken out from the oviduct with sterilized instruments, and precau- 
tions were taken that the eggs did not come in contact with the hands of the 
experimenter or with the skin or outside of the frog. Alternate lots of about 
50 to 100 eggs were punctured or kept untreated as controls. None of these 
non-treated eggs ever developed. From the punctured, unfertilized eggs ten 
developed into frogs, of which nine are still alive. The tenth was killed 
December 21 and the microscopic examination of its gonads showed that it 
was a female. This leaves then no doubt that both sexes can be produced 
from the unfertilized eggs of the frog. We have thus far obtained seven 
male frogs and two females, while the determination in two was missed by 
accident. 
How can we account for the production of both sexes? The diploid num- 
ber of chromosomes in the frog seems to be 26, according to Swingle, 3 and, 
therefore, the haploid number 13. The question then arises: Do we find the 
diploid or haploid number of chromosomes in the cells of the parthenogenetic 
frog? Brachet 4 found the diploid number in the somatic cells of a partheno- 
genetic tadpole eighteen days old, but, of course, it was out of the question 
to ascertain the sex of the tadpole. 
The gap can be filled by counting the chromosomes in the fully developed 
parthenogenetic frogs. Thus far the sections of the testes of only one of the 
writer's parthenogenetic frogs have been examined cytologically. This male 
was seventeen months old, had reached the full size of the adult, and had large 
testes with ripe spermatozoa. Prof. R. Goldschmidt, who was good enough 
to examine some of the sections, counted over 20 chromosomes, and there 
can be no doubt that this parthenogenetic male frog possessed the diploid 
number of chromosomes. The writer has not yet been able to ascertain 
whether the nuclei of the female frogs have the haploid or diploid number. 
It is not known whether the female or male is homozygous for sex in the 
frog. If the female were homozygous it would mean that the haploid num- 
ber of chromosomes would be 12 + x and the diploid 24 + 2 x. In this 
case only a female could have the diploid number since 2 x would determine 
a female. Since we find the diploid number of the male parthenogenetic 
