"Our eyes were opened in 1932. In 1931 we 
had a complaint fron. Manning, S. C., our Cleve- 
wilt Strain 1 was wilting. We investigated. In 
sections of the fielt| 95% of the plants were 
wilted, yet there wait no evidence of nematode. 
We decided it must he a new biologic form of 
fusarium wilt. The worst wilt section of this 
field was marked out and in 1932 a variety test 
was planted. This test included all our wilt 
strains, Dixie and tlyree Dixie Triumph strains. 
Much to our chagrin all died badly, but some 
plants survived. These were selected and put in 
plant to rows in 11133 at Sumter, where Mr. 
Paul Bowman had had a similar experience 
with another highly resistant variety (not ours). 
That year many of these plant progenies com¬ 
ing from plants selected at Manning (all from 
strains previously highly resistant at Harts- 
ville) died at Sumter. 
‘‘In addition, at Sfimter in 1933 and 1934 we 
planted two or more sets of our main wilt and 
new strain test. These included ten Cooks, four 
Dixie Triumphs, Dixie, and many Clevcwilts 
and new hybrid strains. Differences in resist¬ 
ance on the two plots was marked, some died in 
Sumter, and held u[ in Hartsville. others held 
up in Sumter and died at Hartsville.” 
The results of these 1933 and 1934 duplicate tests 
at Sumter and at Hartsville indicated the Sumter 
wilt to be different, a third form—differing from both 
the Hartsville and Manning types. We were unable 
to secure a wilt test plot at Manning in 1934, but 
conducted duplicate tests in the three places during 
1935 and 1936. 
In 1935 this test included 148 wilt-resistant 
varieties and strains with 37 check plots, giving a 
total of 185 plots. Two sets of this were planted 
in each place. The behavior of certain of these 
wilt varieties was distinctly different on the three 
plots. From our 1935 test results, all of our best 
wilt-resistant strains that gave most promise of 
high resistance to all forms were selected for further 
test and increase. 
Our 1936 test included all of these plus our stand¬ 
ard commercial varieties. In addition a few were 
planted as indicators of the various wilts. This test 
included 27 wilt varieties and strains and in addi¬ 
tion, every third plot was planted in a susceptible 
variety as a check. This test was repeated ten 
times in each place to try and offset irregularities 
in wilt infestation. A heavy uniform wilt infesta¬ 
tion is almost impossible to secure and the degree 
of wilt damage will vary in the different sets. How¬ 
ever, the relative differences in susceptibility or 
resistance of certain wilt varieties to the Hartsville, 
Sumter and Manning wilts were the same in 1935 
and 1936 tests. 
Cumulative Evidence 
Thus all cumulative evidence points definitely to 
the existence of different Biological forms of 
Fusarium Vasinfectum. 
That these differ in their effect on different cot¬ 
tons bred for wilt-resistance and can be identified 
by these reactions, is shown by photographs 1-8 
taken of the same groups of cottons in the several 
plots. 
£ Photographs 1. 2 and 3—Show the different reactions 
of Clevewilt 6 and Clevcwilt 35-1 to Hartsville, Sumter 
and Manning wilts. 
Photographs 4 and 5—Show two Clevewilt cottons. Both 
resistant to Hartsville wilt: one susceptible to Sumter wilt and 
the other resistant to Sumter wilt. 
HARTSVILLE WILT 
In photographs 1, 2, and 3 note: 
Clevewilt strain 6 (our newest and highest producing wilt cot¬ 
ton) very resistant to the Hartsville and Sumter wilts, and some¬ 
what susceptible to the Manning wilt, Clevewilt ,15-1, highly 
susceptible to the Hartsville ami Sumter wilts and slightly less 
susceptible to Manning than Clevewilt 6. On plot photographed 
at Manning. Clevewilt o produced <>64 pounds of seed cotton to 
the acre and Clevewilt 35-1 produced 774 pounds to the acre. 
On the entire test of ten sets at Manning. Clevewilt 0 . however, 
averaged 818 pounds and Clevewilt 35-1 averaged 771 pounds 
|>er acre. On the nine sets not photographed, wilt infestation 
was lighter. 
In photographs 4 and 5 note: 
Clevewilt 34-10. highly resistant to Hartsville wilt and suscept¬ 
ible to Sumter wilt. On the Hartsville plot of 10 sets. Clevewilt 
34-10 averaged 1536 pounds of seed cotton per acre and Clevcwilt 
strain 5 nveraged 1556 pounds per acre. On tile ten sets at 
Sumter, Clevcwilt 34-10 averaged 712 pounds and Clevewilt 5 
averaged 830 pounds, and on the plot photographed, Clevewilt 
34-10 produced 505 ixmnits ami Clevewilt 5 produced 757 jiounds. 
Clevewilt 34-10 was planted as a Sumter indicator. 
In photographs 6, 7 and 8 note: 
Catch-All 35-50 (a segregate front a Clevcwilt Dixie Triumph 
cross) resistant to Hartsville. susceptible to Sumter and highly 
Note especially the difference between 35-50 in the Sumter and 
Manning plots. 
At Sumter the stand is broken by wilt and the remaining plants 
are dwarfed, showing every evidence of susceptibility and wilt 
damage. 
At Manning the stand is good, plant growth vigorous and 
shows every evidence of resistance. 
In fact 33-50 shows the highest resistance to the Manning 
will of any cotton tested, and led the test in yield. 
LABORATORY TESTS INADEQUATE 
In 1933 we were fortunate in securing for six 
months the services of Dr. P. H. Hornburg, formerly 
of the University of Tennessee Experiment Station, 
a highly trained plant pathologist. He devoted his 
entire time to making pure cultures from wilt plants 
of the different varieties grown in the various test 
plots near Manning, Sumter, Hartsville and else¬ 
where. We hoped to find some differences in the 
growth and development of the various forms of 
wilt that would enable us to identify them. All the 
different types described by Dr. Geo. H. Armstrong 
of Clemson College were found by Dr. Hornburg in 
the Hartsville plot, on which all of our standard 
wilt varieties of cotton held up practically 100%. 
Therefore, we decided that the biologic forms of 
cotton wilt would have to be identified by their 
different reactions on certain varieties of cotton in 
the field. 
RESISTANCE OR TOLERANCE, NOT IMMUNITY 
In all our experience in breeding and experiment¬ 
ing with wilt cottons we have yet to find a single 
upland variety that is immune to wilt. The old 
system of cutting all plants and counting as wilt- 
resistant only those that showed no discoloration of 
the xylem tissues, is wrong in the opinion of the 
writer. The chances are that the only reason some 
of these did not show a discoloration is that they 
luckily escaped. Late in the season when the crop 
is mature but the plant is still green, in our worst 
wilt-infested plots, many plants, even of our most 
wilt-resistant and productive strains, show this 
discloration. 
We, as breeders, are interested only in the cottons 
that will make the most and best cotton under wilt 
conditions. Whether it be tolerance, resistance or 
immunity that enables it to do so is immaterial 
from a practical standpoint. In our work and in 
our records, if a plant makes a normal growth and 
a full crop on badly infested wilt land, it is counted 
as a resistant. In our worst wilt plots, as shown 
by the accompanying photographs, there is not much 
doubt as to those plants that should be counted as 
susceptible. 
0 £ Photographs '< 7. and 8—Show Catch-All 33-50 growing 
in Hartsville. Sumter and Manning wilt plots. Note resistance 
to Hartsville type, susceptibility to Sumter type and high degree 
of resistance to Manning type. 
MANNING WILT 
SUMTER WILT 
