6 o6 
Journal of Agricultural Research 
Vol. XXI, No. 8 
it is probable that weed seed and other impurities in the seed will be 
more satisfactory than the seedling characters as a means of identification. 
The regional strains of common alfalfa may be distinguished from the 
Peruvian, on the one hand, by their more rapid very early growth under 
the long day and from the Grimm, Baltic, Canadian variegated, and 
Turkestan, on the other, by their greater height, erectness, and uniform¬ 
ity under the short day. Table I sets forth briefly the characteristics 
of the seedlings of the various alfalfas under the different lengths of day 
as determined by experiments conducted in the greenhouse at Washing¬ 
ton, D. C. 
Table I. —Characteristics of seedlings exposed to varying periods of light daily. Sown 
in greenhouse fanuary 20; notes taken on March j, 1921 
Varieties. 
Exposed to normal day 
with electric light from 
dusk to 11 p. m. 
Exposed to normal day. 
Exposed to daylight from 
9 a. m. to 4 p. m. 
Peruvian. ...... 
Erect, no branching, 
height 5 inches. 
Erect, no branching, 
height 4 % inches. 
Erect, no branching, 
height 3% inches. 
Erect, no branching, 
height 3 inches. 
Kansas 0 ; .'...... 
Erect, no branching, 
height inches. 
Erect, no branching, 
height 2% inches. 
Grimm.. 
Elect, no branching, 
height 4 4 / 6 inches. 
Low and spreading, 
stipules enlarged, 
basal branches just 
showing, height 2% 
inches. 
Low and spreading, 
rosette-like, stip¬ 
ules enlarged, basal 
branches just show¬ 
ing, height 2 inches. 
Turkestan. 
Erect, no branching, 
height 4 4 / 5 inches. 
Low and spreading, 
stipules enlarged, 
basal shoots just 
showing, height 2 
inches. 
Low and spreading, 
rosette-like, stip¬ 
ules conspicuous, 
basal shoots just 
showing, height 1% 
inches. 
Medicago falcata. 
Erect, slender, no 
branching, height 
5 inches. 
Low and rosette-like, 
stipules conspicu¬ 
ous. height 1X 
inches. 
Low and rosette-like, 
stipules conspicu¬ 
ous. height 1 inch. 
® These measurements were taken too late to show the proper relation between seedlings of the Kansas 
and those of the Peruvian. 
By the use of this method and by the aid of such facts as can frequently 
be had regarding the lots of seed to be identified, it will be found possible 
in many cases to make a reasonably certain varietal identification. The 
method is offered especially to Experiment Station workers and those 
engaged in the administration of alfalfa seed certification activities as 
an aid to and not as a positive means of identifying varieties. It is rec¬ 
ognized that the work herein reported upon is far from complete, but 
investigations are now being conducted to make it more comprehensive 
and useful. 
The method involving the sowing of seed in the field is not recom¬ 
mended where a greenhouse is available. It is less certain and its use is 
limited to sections where the short days of the year are sufficiently warm 
to produce growth. 
