MAGANLAL L. PATEL 121 
And the rain in the latter part of June and especially in the early part of July 
at Surat, for example, is such that excessive falls are very frequent ; if the 
plants are on small ridges these latter disappear and the plants are covered 
up ; if they are on the flat they are drowned by being compelled to remain in 
water-logged soil temporarily. Hence every year the mortality among the 
cotton plants is considerable and very often whole fields have to be resown. 
Some strains resist this drowning better than others. Among Broach deshi 
strains, we have one (Selection | A, cylindrical boll type plant ) which resists 
better than any other we know. Similarly one of the strains of Goghari 
previously described ( B21) is equally good in this respect. 
2. The plants must possess the largest number of monopodia, 7.e., the 
vegetative branches formed in the lower part of the main stem, the number 
and vigour of which very largely determine the bushiness of the plant. They- 
give rise to secondary fruiting branches, which bear a considerable proportion 
of the yield of the cotton plant in Gujarat, and as the flowers on these shoots 
generally appear at a time when they are not likely to be spoiled by rain, it 
is very important that they should be encouraged to the greatest possible 
extent. That thereis a probable relationship (other things being equal ) 
between the number of monopodia and yield is shown by the following 
figures for three strains of Broach deshi cotton at Surat in 1920, which differ 
markedly in this respect. 
Strain Number of | Yield of kapas 
monopodia per acre 
lb. 
1 A cylindrical boil type 6 756 
Cl ; 5 693 
1027 ALF13 7 517 
This difference will be less marked in a very dry yearas in 1918-19, but 
in a normal year such as occurs in eight years out of ten, the above figures 
will indicate what will certainly happen. 
The reason of the advantage which the strain with many monopodia 
will show, lies inthe fact that the cotton is planted in June and on account 
of the rain it is not desirable that flowers should appear in large numbers 
before the middle of November. We refer to this matter again under 
item 4 below, 
