65 



Among the striking peculiarities of sorghum cane, and one which has 

 been noticed throughout the whole time of its culture on this plantation, 

 is its extreme susceptibility to variation of weather or of soil. It has 

 been found that for any variety of sorghum grown here no period of 

 growth or, better, length of time required from germination to matur- 

 ity can be even approximately assigned. For the same variety, grown 

 in two seasons, or planted at different dates in the same season, this 

 period may differ by from three to nearly six weeks. This fact is one 

 which for the present sets at naught any attempt to arrange two or 

 more plantings which shall follow each other with any reasonable reg- 

 ularity in date of ripening. 



Before proceeding with the account of the season's results, the valu- 

 able assistance rendered in the laboratory and otherwise by Mr. O. D. 

 Berwick, jr., of Bayou Sale, La., should be acknowledged. 



EXPERIMENTS OF 1891. 



In the season's experiments two plantings were made at different 

 dates, it being judged unwise to risk the result of past years 7 propa- 

 gation work to the chances of entire loss by drought or excessive rains 

 which might destroy one planting, but would scarcrly continue long 

 enough to endanger two. By this plan, also, it was hoped to distribute 

 the selection of seed over a longer period, avoiding much hurried and 

 therefore unsatisfactory work in the laboratory. 



After study of the field and chemical data collected during the sea- 

 sons of 1889 and 1890 at Calumet, with many varieties of sorghum, it 

 was decided to continue trial of the following ten : Early Orange, Link's 

 Hybrid, Collier's, Ubehlana, Improved Orange, Late Orange, Planter's 

 Friend, Colman Cane, Sterling (plat No. 14 of 1888), Bed Liberian. 



Of these varieties, Colman Cane, Collier's, Planter's Friend, Link's 

 Hybrid, Early, Late, and Improved Orange, were considered to be of 

 best promise, lied Liberian and Sterling (plat No. 14) were retained for 

 a final trial, but with little expectation that they would prove better 

 than in the past. Besides the Calumet-grown stock of Colman Cane, 

 Collier's, and Link's Hybrid, a number of selected seed heads of these 

 varieties, grown at Sterling, Kans., in 1890, were received from Mr. A. 

 A. Denton, superintendent of the Department of Agriculture experi- 

 ment station at that place. From him were also received a few heads 

 labeled " Sterling, lot No. 161," described as a cross of Link's Hybrid 

 with Early Amber, resembling in form the Link's Hybrid type. 



In addition to the above a few special plats were made with seed 

 from certain large sports or crosses produced in several of the 1890 

 plots. The varieties upon which seed selection work was done this 

 season were planted in plats each from a single head, as before noted, 

 and were in most cases duplicated in the two plantings. 



The field in which the 1891 plats were grown was one upon which 

 no sorghum had been previously planted, and was in a crop of ratoon 

 21383 No. 34 5 



