C. A. bARBER 



165 



Table I. 



Unit observations of varieties and places. 



In comparing the general growth of the cane plant in different places, it 

 is, in the first place, necessary to consider the varieties observed there. Saretha 

 and Baroukha, for instance, are characterized by comparatively enormous 

 length of stems, while Mungo is a dwarf variety with very short joints and 

 canes. As the latter cane is present in some places and not in others, Ave 

 cannot obtain reliable figures of general cane growth for any one place by 

 averaging the observations irrespective of whether Mungo is present or not. 

 It would, again, be difficult to compare the cane growth as indicated by the 

 units observed at Nagpur and Sabour, for there is only one of our seven canes 

 grown in common at these two places. The following method has been 

 adopted to overcome this difficulty. Stations are only compared by averaging 

 the same canes grown on them. The Cane-breeding Station, Taliparamba, 

 and Samalkota were first compared, as all of the^m had the whole series gro^nng. 

 Then the Coimbatore Central Farm was compared with these three, omitting 

 Mungo altogether, which was absent on it. Nagpur was compared with these 

 four, as well as with Partabgarh and Aligarh, only in respect to the measure- 

 ments of Saretha, Chin, and Khari. Sabour was compared with the three 

 localities which had its 'five canes, Pusa with the same three, Partabgarh Avith 

 the first four, Shahjahanpur Avith four, and Aligarh Avith three. This method 

 of course greatly increased the Avork, but it AA^as considered the only safe way 

 in Avhich to compare the general groAvth of canes in different places. 



Another factor here obtruded itself. In comparing Pusa Avith the three 

 places in South India A\liich had the same varieties groAAing, it Avas noted that 

 Mungo does rather well in s(nne respects at Pusa, Avhile it does very badly 



