Ikeno. — On Hybridization of some Species of Salix. If. 179 
\ 
Table III. 
No. and Nature of Experiment. 
G-type. 
M-type. 
New type. 
Totals. 
1. (G-type ^ x G-type cf) ... . 
187 
32 
219 
2. (M-type <j) x G-type <fi) . . . . 
77 
67 
15 
159 
3. (M-type <j> x M-type cF) .... 
4 
19 
8 
3 i 
Totals 
268 
1 18 
23 
409 
% 
69.4 
30.6 
Deviation from 75 % : 25 % expectation 
-5.6 
+ 5.6 
Plants ranked under the column 4 New type 5 in the above table have 
their catkins very similar to those of M-type, but distinguished from the 
latter by their perfect non-hairiness ( 1 . c., p. 44). In my former paper 
I stated my supposition that they will perhaps breed true in later genera- 
tions. In order to confirm it many seeds of this type produced by fertiliza- 
tion inter se were sown in 1918, and many germinated, but a large 
portion of the seedlings perished in the young stage, and I obtained 
only eleven adult plants. Of the latter two produced flowers in 1920,. 
and proved themselves to be the male and female plant respectively, and 
are quite similar to their parents externally. The remaining nine do not 
bear flowers yet. My supposition was thus confirmed, at least partially. The 
problem, whether they were produced by the so-called 4 loss- mutation ’ or by 
the new combination of certain factors, would not be definitely decided 
without further breeding experiments. 
Since the crosses shown in Table III correspond to DRxDR, we 
expect to get 75 per cent. D and 25 per cent. R, if we are dealing here with 
the 3 : 1 segregation. We have actually 69-4 per cent. G-types and 30-6 
per cent. M -types, i.e. + 5*6 per cent, deviation from this expectation. 
The deviation is positive on the side of M-type progeny, and this fact 
is evidently due to the imperfection of dominance : the F 2 progeny are to be 
represented genotypically by 25 per cent. DD + 50 per cent. DR + 25 per 
cent. RR ; in the case under discussion in 5-6 per cent, out of 50 per cent. 
DR the dominance failed, and the recessive character has made its appear- 
ance, or, in other words : 
25 per cent. DD + 44-4 per cent. DR + 5-6 per cent. (DR) + 25 per cent. 
RR = (25 + 44-4) per cent. G-type + (5-6 + 25) per cent. M-type = 69-4 per 
cent. G-type + 30-6 per cent. M-type. 
(DR) denotes the zygotes where the usual dominance fails. 
In short, the segregation follows the 3 : 1 mode, and the deviation 
is due to the imperfection of dominance of the factor D over R. 
3. Back-crosses. 
The results of back-crossing of F 1 hybrids by either of the two parents, 
i. e. DR xR or DR xD, have fully confirmed those obtained in the F 1 as 
well as the F 2 generation. 
