HOWARD, HOWARD AND ABDUR RAHMAN 2l5 



As all these organs are practically in contact, self-pollination is easy. By 

 five in the morning, the stigma is generally covered vnth pollen and as insects 

 are not observed till 6 a.m. a great deal of self-pollination is bound to take 

 place. Very frequently, however, the anthers in some flowers do not develop 

 but remain aborted and turn broMTi without shedding any pollen. In these 

 cases the visits of bees easily bring about cross-pollination and it is probable 

 that most of the cross -fertilization which takes place in this crop is due to this 

 cause. Setting readily takes place under muslin and the seed so produced 

 develops normally. In 1910, the seeds of a large number of single plants of 

 til were sown separately at Pusa and the progeny examined. There was no 

 doubt that many of these were heterozygctes and that natural crossing in this 

 crop is considerable. 



Varietal characters. 



During the years 1909 and 1910, a good many sowings of til were made at 

 Pusa, the seed in the first instance being that usually grown by the people in 

 the various parts of India. From the 1909 crop, the seed of single plants, both 

 free flowering and bagged, was saved in a large number of cases for sowino- the 

 next year. These cultures have enabled a preliminary study of this crop to 

 be made. A\Tiile many of the single plant cultures showed that they were 

 raised from heterozygctes, nevertheless some bred true and from these it was 

 observed that the Indian types of til differ in the following characters : — 



1. Colour of seeds. The full development of the colour of the seed coat 

 appears to be a question of development. White and brownish seeds are met 

 with on the same plant while with black seeds, smoky and dark brown indivi- 

 duals occur. Further investigation of this matter is desirable. 



2. Rough and smooth seeds. As a rule, the surface of the seeds is smooth. 

 In the cultures raised from some heterozygotic smooth parents, however some 

 plants Avith rough seeds Avere produced. No homozygotes with rough seeds 

 were isolated. 



3. Colour of corolla. Several different colour types are met with from 

 almost pure white to deep violet. (Plate V.) 



4. Hairiness and smoothness of corolla and capsule. Forms with nu- 

 merous hairs as well as less hairy types occur. 



5. Number of flowers in the leaf axils. At first sight it would appear that 

 in all the varieties of til three flowers in each leaf axil would be usual. As a 

 rule, however, only one flower is produced, the other two bud-like bodies 



