10 
BULLETIN 1015, U. S. DEPARTMENT OF AGRICULTURE. 
each on the wilt-infested breeding plats at the Arlington Experi- 
mental Farm. After one to three years of continuous selection the 
selected strains were planted beside the resistant varieties from 
which they were selected. 
Table 2. — Effect of continuous selection of varieties of tomatoes on the resist- 
ance to wilt. * 
Variety and year. 
Field. 
Number of plants in 
the experiment. 
Percentage of plants 
showing a trace of 
infection. 
Variety 
under test. 
Selections 
from same. 
Variety 
under test. 
Selections 
from same. 
Norton: 
1917 
3 
1 
1 
2 
2 
1 
2 
1 
15 
16 
48 
16 
16 
16 
48 
48 
48 
175 
32 
16 
112 
64 
16 
25 
33 
50 
52 
19 
69 
19 
31 
65 
63 
1917 
39 
1918 
63 
1918 
19 
1919 
66 
Marvel: 
1918 
s 
1918 
34 
Louisiana Wilt-Resistant: 
1916 
76 
Continuous selection has apparently made no improvement in the 
wilt resistance of these strains. Whenever the real differences be- 
come small, however, as in the present case, they are obscured by 
fluctuations caused by environmental factors. Such a factor is soil 
moisture. On wet spots wilt is much worse than on better drained 
areas. Moreover, water is probably not the only variable environ- 
mental factor that affects the development of wilt. Until the meth- 
ods of reducing these environmental effects are sufficiently refined to 
enable the breeder to distinguish real differences from fluctuations 
it will be impossible to determine whether or not continuous selection 
leads to further improvement. Under the present methods it does 
not seem to do so. The writer has found no difference in this respect 
between tomatoes, which are nearly always self-fertilized, and sugar 
beets, 4 which are normally cross-fertilized. 
DURATION OF RESISTANCE. 
Wilt resistance is apparently as permanent as other characters of 
the tomato. It fluctuates somewhat with changes in soil tempera- 
ture, soil moisture, and other physiological factors, but on the aver- 
age is less variable than tomato fruit characters. Its stability is 
shown by the fact that the varieties developed. by the writer have 
maintained the same degree of resistance under both continued and 
discontinued selection and have remained relatively constant in all 
places tried. This is further supported by the fact that varieties 
I 
* Some recent investigations in sugar-beet breeding. 
465, 51 figs. 1916. 
In Bot. Gaz., v. 62, no. 6, p. 425- 
