142 STITDIES IN INDIAN SUGARCANES 



and doubtless many more in parts of the country where as yet the cane varieties 

 have not been collected. These points are merely indicated here to support 

 the idea suggested that the careful study of minute vegetative differences may 

 be of use in trachi.g the course of development in closely aUied 'varieties of 

 cultivated plants. The line of work with seedlings of each variety is more 

 compheated, in that the minute vegetative differences relied on are ruled out, 

 as they do not appear to be always inherited, but a study of the range of 

 variation in each group will also assuredly assist in fixing the probable wild 

 ancestor. Such work is already possible, with the mass of details collected 

 in the descriptions ol different Indian varieties, but it is not likely that there 

 will be time available in the immediate future for working these details out. 

 One point may, however, be mentioned in passing. We have now a certain 

 knowledge of the relative stability of vegetative characters in the groups 

 forming the subject of this paper. But it is unhkely that this will apply to the 

 same vegetative characters in. other groups, even, if they are derived from the 

 same wild form, and this makes it all the more necessary that each group 

 should be separately studied in all its varieties. 



We have thus far ignored the fact, which has come to light gradually in 

 our study, that in. the Saretha group there is a distinct subdivision, into two 

 sections, primarily separable by the colour of the stem, which may be roughly 

 termed the brown and green stenmied sections. Each section, contaiiis a 

 primitive Punjab variety and they have been n.amed after these. The Katha 

 section is characterized by the regularity of brown, marks or coloratiojis on the 

 canes and the Mesangan section is devoid of these. Katha and Memngan are 

 similar canes in many respects, but the Mesangan group, although including 

 a tbii\ cane, consists chiefly of the thicker forms found in other parts of In.dia 

 than in the Punjal) and neighbouring parts of the United Provinces. Saretha, 

 which is the thickest variety in the Katha section, has a form in the Mesan.gaji 

 section also, sometimes distin.guished as Dhaur Saretha. In the detailed 

 study of these two sections, which is interpolated in many of the character 

 studies, we shall see that the Mesangan section is in. many respects intermediate 

 between the Saretha an.d Sunnabile groups, an.d it may therefore be regarded 

 as, on the whole, a more developed series, further from the origin,al wild form. 



In some of the characters used hi separating the varieties and groups, it 

 has been found difficult to state the differences succinctly, especially in. such as 

 involve compheated series of measurements. For the sake of clearness, certain 

 empirical expressions have been introduced, and the rows of figures have been 

 plotted as curves on squared paper. But, for these to have any general value. 



