44 CROSS-FKRTILISATION I.N THE INDIAN COTTONS. 



seed. For this leasou Type 2 has not been used in tlio expciiments 

 conducted with the object of raising races of improved cotton suit- 

 able to the United Provinces. 



One further type of cotton, G. Stocksii, M. Mast., is found in 

 India and has formed the basis of some experiment. Seed of this 

 type was obtained from the wild plants found on the limestone hills 

 arouiul Karachi. Both G. arboretim and G. herhaceum ha\o l)ccn 

 crossed with this plant and in both cases seed was obtained. This 

 seed germinated and gave rise to plants which have been grown 

 for three seasons. They remained diminutive, however, and liavc 

 died without producing any flowers, though the parent plants have 

 flowered freely and subsequent generations have been laised from 

 their seed. From this condition of complete sterility in the F, gen- 

 eration it is but a short step to those cases in which attempts to pro- 

 duce a cross have resulted in complete failuic 



G. Stocksii, though found wild in India, ditt'ors in a remarkable 

 degree from the other members of the Asiatic (Indian) group of 

 Cottons (subsectio Indica) as defined by Todaro.* By Wattf it is 

 placed in a different section of the genus and the correctness of its 

 inclusion in the Genus Gossypium even has been called in question.| 

 In this connection it is of some interest to find that, as far as experi- 

 ments have been made, this plant forms the single instance of a cross 

 between a member of the Asiatic series and other forms of cottons 

 being produced. 



The determination of tlie degree of sterility which may arise 

 in the progeny of crosses such as have been considered above is a 

 still more complex matter. Numerous cultures, involving several 

 thousand individuals, have been carried by successive selfing as far 

 as the h\ generation. Certain of the plants of this generation dis- 

 play a very marked fertility and unduninished vigour. Others, 

 on the other hand, appear to be nearly sterile though a case of com- 

 plete sterihty has not been observed. The cause of this divergence 



• 0>«cr. Biii CiitoDi, 18i'3 



t Wild iiD.l Cultivated Cotloiu of tlic Woild. 



J Wall I. c. p. .Vi. 



