C. A. BARBER 



157 



I take as an exani])lo of this increasing thickness in late-foiined canes one 

 of the Pansahi group, as it has been demonstrated that the presence of early 

 and late canes is a character of the whole of the cane varieties included in it. 

 It will be remembered that, in the Punjab measurements, Kahu was chosen 

 as showing early and late canes most typically, aii.d this is a local member of 

 the Pansahi group. Recent stool dissections of Pansahi (9 months old) are so 

 instructive that a study of one example is appended below. The development 

 of the branches of succeeding orders are so regular as to be almost mathemati- 

 cally exact. In the diagram, a represents the main shoot arising from the 

 bud on the set. There are two branches of the first order, b' and 6', and the 

 former, larger branch, is the one here analysed, c' ''. d^ '\ e' '' are 

 branches of the second, third, and fourth orders respectively. All the cane.s 

 in the clump were full grown., excepting three es. which formed strong shoots, 

 cane-forming at the base and about five feet Jong. There were no other shoots 

 than those in the diagram, which shows that the branching system was 

 completed for the season. 



Brancbino system of k Pansahi shoot from a SI.NOLE BT'D. 



