&(3 REPORT OF THE AGRICULTURAL RESEARCH 



stems. Great variations in staple are found in these 

 plants. Selection experiments have proceeded probably as 

 far as they can be taken and the staple obtained attains the 

 length of -8 inch. It has been proved that the shortest 

 staple is associated with the hardiest and darkest coloured 

 plants. Excepting the characteristic of coloration which 

 is a negligible factor in other parts of India, the above 

 fact is a further confirmation of our experience that hardi- 

 ness and poor staple are companion characters. 



The variety which I have already published under the 

 name of G. oh tusi folium var. hirsutior seemed to be un- 

 suited for cultivation owing to its prolonged season of 

 growth, but it may yet prove useful in cross-breeding. 



The remaining indigenous forms of Punjab cottons are 

 yellow and white flowered forms of neglectum and indicum, 

 all, however, with a strong tendency to show true neglec- 

 tum characters. 



Khaki hirsutum is unsatisfactory from its producing 

 cotton of varying tints and many bolls bear quite white 

 cotton. A good, even, white colour is a desideratum in a 

 cotton produced for sale in the Indian market and buyers 

 are prejudiced against coloured cottons. 



Of the indigenous types the broad lobed sanguineums 

 are by far the most promising and it would be well to draw 

 the attention of the cultivators to these as they are of a 

 particularly desirable type. The outturns of all the plots 

 appear to be excellent. 



The Upland cottons were described to me by an Ameri- 

 can expert as being the best he had seen outside the United 

 States. 



Two varieties come into bearing as early as the indigen- 

 ous cottons, but it is still to be seen whether they are quite 

 early enough to save one watering or not and this is a 

 matter of imjDortant consideration to the zemindar. The 

 successful introduction of these cottons into more general 



