90 STUDIES IN GUJARAT COTIONS 
We have, however, 11ow the recent work of Kottur! on this point. He 
worked with various types chiefly of herbaceum cottons, and concluded that 
“all the Indian varieties cross easily and the amount of natural crossing is 
considerable ( up to 6 per cent.) when various varieties are grown in adjoining 
fields.’? Beyond this opinion there is no evidence, so far as we are aware, 
of the degree to which cross-fertilization occurs among herbaceum cottons, or 
of the extent to which the plants now grown are unfixed hybrids of the various 
varieties in cultivation. 
That cross-fertilization does actually occur in the Goghari type of 
herbaceum cotton is clear from the following experiment :— 
Strain No. C 8 Goghari. This is a pure line cultivated without variation 
for several seasons. In 1918=19 the seeds produced by two separate plants of 
1917-18 were sown, and variations in the amount of kapas per boll and in the 
ginning percentage of the kapas were noted which could only have resulted 
from crossing. The rows were planted at least six feet from any other type. 
The actual figures obtained were as under :— 
Number of bolls per pound of kapas.. sie Ae {ne sei 
eee No. 1-44'2 
Ginning percentage oe a “* ** | No. 2-46°6 
Degree to which cross-fertilization occurs. We have started our work by 
isolating pure line cultures of Goghari cotton which have maintained certain of 
their characters for at least two generations. The seeds produced in open 
field cultures in the second generation (the plants being grown no less than 
six feet by three feet apart ) have then been grown, and the proportion of the 
plants which have varied in the next generation in the character studied has 
been determined. The only source from which such variations could arise 
was cross-fertilization from the plants in the field. The following cases 
may be quoted :— 
1. Strain No. B21 Goghari. The character studied was the size and 
shape of the boll. This was absolutely constant in 19-1617 end 1917=18. 
On being grown in open culture in 1918-19, it showed 2°1 per cent. of tLe 
plants bearing bolls of a different character. Again a plant bearing bolls 
of the original shape and size in 1918-19 gave 2°85 per cent. of the plants with 
varying bolls in 1919-20. | 
2. Strain No. C22 Goghart. The same character was studied as in the 
last case, and the boll characters of the strain were constant in 1917-18 and 
1918-19. In 1919-20 this showed plants with varying bolls to the extent of 
I'l per cent. 
1 Mem. Dept. Agrt. India, Botanical Series, X, no. 6, 1920, 
