129 



The variations in tbe tropical sugar cane are entirely produced in 

 tbat way, as lias already been shown by tbe statements of Professor 

 Morris just quoted. 



Jn Mauritius a sugar cane of the ribbon variety produced two new 

 canes, a green cane and a red cane. This was considered an astonish- 

 ing variation there. Tbe causes of such variations are unknown. It is 

 only known that they do occur, and that valuable new varieties some- 

 times suddenly appear in that way. 



The history of some of the varieties of sorghum would seem to indi- 

 cate, so far as it is possible to obtain accurate information of such mat- 

 ters, that they originated in this way. 



In Indiana, in a field of Cbinese cane, a single cane ripened two weeks 

 earlier than the. other canes. This variation was preserved and named 

 the Early Amber. It is the most widely known of all the varieties of 

 sorghum. In the experimental field of this station there were growing 

 Early Amber canes received from New South Wales, from Cape Town, 

 and from many places, showing its wide distribution. 



In New York, in a field of Early Amber, only one cane ripened before 

 frost. This variation was preserved and named by us Whiting's Early 

 Variety. It matures ten days earlier than the Early Amber. It seems 

 to be a sport from a sport. 



In Tennessee, in a field of Honduras, a single cane ripened two weeks 

 earlier than the other canes. This variation was preserved and was 

 named Link's Hybrid. It is one of the best varieties of sorghum for 

 sugar manufacture. 



It is probable that other cane-growers have seen as valuable vari- 

 ations in their cane fields, and have not recognized their importance. 

 It is worthy of remark that each of these variations was noticed and 

 was preserved merely because it chanced to ripen earlier than the other 

 canes in the same field, and not because its other qualities were recog- 

 nized at the time. 



. In the effort to improve the sorghum plant all such variations from 

 type should be analyzed to determine their value in sugar manufacture. 



WORK AT THE STERLING STATION ON CROSSES OR VARIATIONS. 



It may be said of the work done here in this direction that in the 

 first place it established positively, in the judgment of those in charge, 

 the fact of the very strong tendency of this plant toward variability. This 

 fact has, of course, been frequently noticed and commented upon here- 

 tofore, but as it seems essential that it should be thoroughly and gen- 

 erally understood, we think it advisable to enter into an exposition of 

 the evidence that was obtained to justify us in coming to the very de- 

 cided conclusion we adopted upon this point. The plots which were 

 planted as " crosses " at this station were in every case from single seed- 

 heads, selected by Mr. Den ton, which were very carefully thrashed and 

 cleaned, special precautions being taken to prevent any accidental ad- 

 1405G Bull. 20 9 



