LEAKE AND RAM PRASAD. 39 



the collection of such observations as have a bearing on this sub- 

 ject, made during the course of a series of experiments having as 

 their object the production of an improved cotton suited to the 

 conditions of the United Provinces. As has been stated, they 

 deal with two distinct phenomena, the occurrence of sterility, 

 chiefly as the result of repeated self-fertilisation, and the frequency 

 of crossing in nature, which two phenomena require separate 

 treatment. 



Sterility as the result of repeated self-fertilisation. 

 It is a matter of common knowledge that the extent of cross- 

 fertilisation occurring in different plants in nature is most diver- 

 gent. At one end of the series lie the dioecious species, in which 

 self- fertilisation is impossible and from which, through the various 

 forms in which, by protandry, protogyny, and the various adap- 

 tations, such as dimorphy, self-fertilisation is injurious, it is pos- 

 sible to pass to the other extreme, represented by plants with 

 cleistogamic flowers, where self-fertilisation is the rule. The facts 

 first collected by MuUer* and Darwinf and more recently by 

 KnuthJ are so well known that it is hardly necessary to make 

 further reference to this general aspect of the subject. It is suffi- 

 cient to note that the cotton flower exhibits none of those adapta- 

 tions which would indicate that it is normally either cross, or 

 self-fertilised. There is, therefore, no a priori reason for antici- 

 pating any marked degree of self-sterility, indication of which can 

 only be obtained from direct observation. Difficulty here arises 

 through the absence of any method for directly measuring the 

 degree of self-sterility, for such sterility may take various forms. 

 Thus the fruit of flowers fertilised by their own pollen may fail to 

 set or, having set, may undergo partial development only and fail 

 to ripen. They may, on the other hand, produce seed which, when 

 sown, gives rise to plants which are completely, or partially, ste- 



* The Fertilisation of Flowers, 1873. 

 t Tlie Effects of Crojs and Self-fertilisation, 1876. 

 I Hand-book of Flower Pollination, 



