44 BULLETIN 428, U. S. DEPARTMENT OF AGRICULTURE. 
The summary of results at Brookings and Highmore indicates the 
superior hardiness of Medicago falcata at these points, while Table 
V indicates that factors other than low temperature were responsible 
for winterkilling. 
While the number of plants upon which the percentage of sur- 
vival shown in Table IV was based is too small to make the data 
really dependable, the appearance of the plantings at Highmore was 
very convincing. The rows of Turkestan, North Sweden, Cossack, 
and Cherno were so depleted as to give the entire block an extremely 
ragged appearance, while the rows of Medicago falcata adjoining 
had so nearly the full quota of plants that the mass effect remained 
unbroken. (Figs. 18 and 19.) 
Widely spaced hill plantings, such as were made at Highmore, in 
some respects offer very good opportunity for studying winter- 
killing, since on such plantings the effect of the important factors 
is greatly exaggerated. In row plantings and even in broadcast 
stands at Highmore the mortality of the Medicago sativa varieties 
Fic. 18.—Hill plantings of Medicago falcata varieties, MV. sativa, and M. sativa x falcata, 
commercial varieties and selections. Photographed in August, 1910, Highmore, S. Dak. 
was slight, while it was almost negligible in the case of the Wedicago 
falcata strains. At Brookings the only winterkilling of Medicago 
falcata which could be considered of any consequence occurred dur- 
ing the season of 1912-13. The mortality in this case was doubtless 
due to the presence of an ice sheet over a portion of the plats. It is 
well recognized that Brookings and Highmore are not ideal places 
at which to test the comparative hardiness of alfalfas; nevertheless, 
at the latter point winterkilling is a serious factor in common alfalfa 
under field conditions. 
Georgeson (23) reports Medicago falcata to be the hardiest of all 
the alfalfas tested in Alaska and that it survives the winter without 
“inconvenience.” His tests, which included the Grimm and other 
variegated strains, were made under conditions much less severe than 
those normally obtaining in the Dakotas. 
In all the trials which have been reported there is still lack of 
data on the relative hardiness of the various forms of the species. 
It now seems probable that this fact will be determined only in- 
