C. A. Barber 215 



in India, but in the past always unsuccessfully. The experiments have 

 not always been conducted very carefully, and on the founding of the 

 new department it was decided to examine the matter afresh and try 

 to determine the cause of failure in the face of the successful results 

 obtained in Java, the West Indies and elsewhere. The solution of the 

 problem turned out to be remarkably simple. Almost all the experi- 

 ments were made in North India and it transpires that the stamens do 

 not mature and pollen is not formed in the cane flowers there. A 

 cursory glance showed that this was not the case in South India, and in 

 the (Tovernment farm opened at Coimbatore in the Madras Presidency 

 some 40,000 seedlings have been raised during the past two years. 



Proble^n 1. During the short time that the cane-breeding station 

 has been in existence, a very important step in advance has been 

 taken and the first problem, that of obtaining seedlings, has been 

 solved. But in considering the ultimate aim of the station more care- 

 fully, a number of difficulties have cropped up and it is the intention in 

 the present paper to detail some of these further problems, in the hope 

 that help may be available from the great body of readers interested 

 in plant-breeding. 



Problem 2. Most, if not all, of the difficulties in procuring suitable 

 cane seedlings have arisen from the fact that the flowering is irregular — ■ 

 in fact, comparatively rare. If a cane flowers and we obtain seedlings 

 from it, we cannot count on these seedlings flowering. We have no 

 means at present of inducing the canes we are most interested in to 

 flower. Control of the flowering is the second difficulty we have 

 encountered. 



Arrowing of the cane, as the production of the inflorescence is 

 called, is comparatively rare in North India, but occasionally it occurs 

 over large areas. It is viewed with alarm in certain regions, and there 

 appears to be some reason for connecting it with the weather and 

 especially with a failure of the normal rains. Flowering of the sugar- 

 cane appears to be commoner in years of drought, but details on this 

 point are not yet available. In Mysore and Coimbatore, typically dry 

 tracts, flowering is common, and on the other Madras farms where sugar- 

 cane is grown, in Malabar, Godavari and South Arcot, the amount of 

 flowering seems to vary inversely with the moistness of the climate. 

 On the other hand, it is a common belief among Coimbatore cultivators 

 that arrowing is most frequent in water-logged conditions of the roots. 

 Experiments are now being conducted on the farm with difterent 

 soils and different amounts of moisture to see if any effect is produced, 



