180 BOTANICAL GAZETTE [SEPTEMBER 
first time. Many of the drawings for these figures were completed 
about two years ago, but my interest in other phases of this work has 
postponed their publication until now. The earlier stages of reduc- 
tion in Oenothera, up to the end of the heterotypic mitosis, have 
already been described in detail (11), so that this paper will deal 
chiefly with the later stages, beginning with the metaphase of the 
heterotypic mitosis. 
The plants from which these studies were made were grown at 
Wood’s Hole, Mass., in 1g05 and 1906, from seeds of DEVRIES. 
The results show that in some cases the number of chromosomes is. 
undoubtedly 21, while in one individual it was 20.2. The number is 
undoubtedly constant in an individual, however, as shown by @ — 
large number of counts, which demonstrated constantly 20 in one 
case and 21 in the other. q 
The external characters were not studied with sufficient care at 
that time to describe them accurately, but from my notes they appeal 
to have been intermediate between O. lata and O. gigas. DEVRIES 
has described in a recent paper (5) the hybrids of O. gigas with other 
forms. I have called attention elsewhere to the fact that the behavior 
of O. gigas in hybridization, as well as its number of chromosomes, 
places it in a different category from the other mutants of O. Lamarcki- 
ana. DeVries finds that O. gigasxO. Lamarckiana forms a Com 
Stant race intermediate between the parents, at least to the second 
hybrid generation. O. gigasxO. Lamarckiana, O. Lamarckiana 
XO. gigas, O. gigasXO. brevistylis, O. gigasxO. rubrinervis, and 
O. rubrinervisXO. gigas, all give constant hybrid races which at 
externally alike in all these crosses. O. lataXO. gigas; howevel 
gives in the F, two types, about 50 per cent. of each; type I inter 
mediate between O. lata and O. gigas, type II intermediate between 
- Lamarckiana and O. gigas. These presumably would all ae 
about 21 chromosomes. There were 13 3 plants in the culture : 
this hybrid in 1907 and a smaller number in 1908. : 
Miss Lutz (24) from a study of 4o individuals of O. lata XO. 88° 
finds more complex conditions in this cross, though apparently * 
* In my first paper (8) this individual was thought to be O. lata XO. Lamarcke* 
soa shed mks rd —— to be derived from fertilization by foreign signs lene 
particular seed pac not having bee when 
were plarited. Be ng been guarded as was supposed 
