ALFALFA BREEDING FOR DROUGHT RESISTANCE. IS 



yield and quality of the forao;e. All selections have been made with 

 the idea of combining large forage and seed production in the same 

 individual plant, the forage type, however, receiving first considera- 

 tion. A thorough test of the yields of all strains developed is made 

 in Vjroadcast plats and in cultivated rows. It should be said that no 

 proper test of drought resistance has been had in the alfalfa-breeding 

 work up to this time. During the time the work was carried on at 

 Highmore, from 1905 to 1907, inclusive, the annual rainfall was above 

 the average for that station. The season of 1908 at Belief ourche was 

 a dry one, but this was the year when the breeding work was begun 

 there and the plants were too young to afford records of yields under 

 dry conditions. But since the first season's growth of an alfalfa 

 plant is a critical period in its life, and since these selections made a 

 good growth at Belief ourche in the comparatively dry year, 1908, it 

 would seem that they must be at least fairly drought resistant. 



During the season of 1909 the precipitation was again above the 

 average, so that no test of drought resistance was secured that year. 

 It will therefore be necessary to retain all of the progeny rows and 

 plats until a proper test of drought resistance is secured. 



The alfalfa stocks used in the breeding work at Belief ourche con- 

 sisted of selections from six strains which were grown at the High- 

 more (S. Dak.) substation. Two of these strains, South Dakota No. 

 162 and No. 164, are recommended by Prof. W. A. Wheeler in Bulle- 

 tin 101 of the South Dakota Agricultural Experiment Station as the 

 best of the stocks tested at Highmore. The twenty stocks tested 

 there included several hardy stocks imported by the Department of 

 Agriculture previous to the year 1905. The two best varieties, which 

 are described on a later page of tliis bulletin, proved to be perfectly 

 hardy and of good forage and seed producing ability. Four other 

 stocks tested at Highmore, which proved fairly hardy, are also repre- 

 sented in the breeding plats at Bellefourche. In the following dis- 

 cussion each strain is designated by a letter, the selections made from 

 each strain being numbered in consecutive order; as A-1, E- 12, etc. 



Strain A. — This is South Dakota No. 65. The seed was screened 

 from a lot of durum wheat imported from Tashkend, Turkestan, in 

 1902, by the United States Department of Agriculture. It was 

 planted in 1902 on a small plat, about 12 by 50 feet, at Brookings, 

 S. Dak. This plat went through four seasons there (from the si)ring 

 of 1902 to the fall of 1906), and did not suffer any from winterkilling. 



" Seed from this plat [harvested in 19041 was planted at the High- 

 more substation, in 1905, in a selection row. A few of the plants in 

 this row died during the winter of 1905-6, showing that it is not per- 

 fectly hardy under severe test."" The plants now growing at Belle- 



a ^^^^eeler, W. A. Bulletin 101, South Dakota Agricultural Experiment Station, 

 p. 135. 

 196 



