CHROMOSOMES, HEREDITY AND SEK. 
493 
cartiila and anaclioreta sets, and the anachoreta 
parent a (haploid) set of anaclioreta. The secondary 
hybrid therefore contained a double (diploid) set of 
anaclioreta, and a haploid set of curtnla chroinosoines. 
In this case, in spermatogenesis of the secondary hybrid, 
normal pairing took place between the anachoreta chromo- 
somes, while, as in the primary hybrid, the cnrtnla chromo- 
somes divided eqnationall}^, witli the result that the gametes 
of the secondary hybrid contained a haploid curtnla a, 
haploid anachoreta set. If, now, the chromosomes are the 
bearers of Meiidelian factors, the fact is explained that the 
primaiy hybrid, when paired back with curtnla, shows in 
Text-fig. 1. 
•:V.* 
• • • • ^ • • 
v\vy;c; 
• . t • » • 
I 2 3 
4. 
1. First spermatocyte equatorial plate of Pygsera anachoreta, 
30 chromosomes. 2. Similar equatorial plate of P. curtnla, 
29 chromosomes. 3. First spermatocyte equatorial plate of 
hybrid P. anachoreta $ X P. curtnla (^ ; 58 chromo- 
somes, so that only two chromosomes have united to form a 
pair. All the others are about half the size of the spermato- 
cyte chromosomes of the pure species. 4. First spermatocjde 
equatorial plate of secondary hybrid, P. anachoreta $ X 
(P. anachoreta $ X P. curtnla (^); 56 chromosomes, of 
which about 30 are large, and consist of two units paired 
together (presumably ana c ho ret a chromosomes) and about 26 
are small, unpaired, presumalily c u r tula chromosomes. (After 
Federley.) 
general no segregation, since in each generation nearly com- 
plete haploid sets of chromosomes of both species are present 
in the gametes. One or two chromosomes, however, pair and 
separate, and correspondingly one character at least was 
observed which showed Mendelian segregation.^ 
1 The present writer has confirmed Federley’s results as to the 
behaviour of the chromosomes, in reciprocal crosses between Biston 
