6 oo 
Journal of Agricultural Research 
Vol. XXI, No. 8 
sowings were made at North Ridgeville, Arlington Farm, and Bard, Calif. 
Early September sowings gave practically the same results at North 
Ridgeville as at Arlington Farm. The late fall and early winter sowings 
at Bard produced seedlings of much the same appearance as those from 
early September sowings at North Ridgeville and Arlington. This was 
also true of December sowings in the greenhouse at Washington. May 
and June sowings at Arlington and North Ridgeville showed practically 
no difference in the rapidity and habit of growth of the varieties tested. 
A range of field sowings from August 26 to September 16 at Arlington 
showed that five to six weeks of growing weather at that time of the year 
are required to produce plants of a satisfactory size for studying the 
varietal characters of the seedlings. The late August and early Sep¬ 
tember sowings gave good results, but September 16, in this particular 
case, proved to be rather late. 
From these tests it was evident that by sowing seed in late summer or 
early fall in the latitude of Arlington, and northward, it is easily possible 
to determine whether a given lot of alfalfa is of the Peruvian variety, 
the common variety (such as the various regional strains) or the varie¬ 
gated (such as the Grimm, Baltic, and Canadian variegated). From 
spring or summer sowings, especially in the North, the varietal differ¬ 
ences exhibited are hardly great enough to be regarded as sufficient for 
the untrained person to recognize with certainty. Thus far in the 
investigations this method is limited in its application to field sowings 
in the fall in the North and to winter sowings in the extreme South. 
It is also limited to distinguishing between groups of alfalfas, such as the 
common or the variegated, rather than between the strains in these 
groups, as for example the Kansas-grown common strain and the Dakota- 
grown common strain or the Grimm, Baltic, and Canadian variegated. 
But the method makes it easily possible to distinguish between the 
Peruvian and the other alfalfas. 
COMPARATIVE CHAR ACT ERISTICS OF SEEDLINGS 
The differences that were most noticeable in the seedlings resulting 
from fall sowings at Arlington were those of height, branching, and 
compactness. The Peruvian produced erect, sparsely branched seed¬ 
lings of greater height than those of the other varieties. They were also 
more nearly uniform. The seedlings of the Kansas variety very closely 
resembled those of the Peruvian variety in uniformity and general habit 
of growth but were not nearly as tall and showed more of a tendency to 
send out basal branches. The Grimm seedlings were much shorter, 
more branched, and less erect than those of the Peruvian or Kansas 
alfalfas. They also showed more variation than these varieties. The 
seedlings of Medicago falcata were uniform, very short, and exhibited 
a rosette-like habit. The Grimm seedlings were nearly intermediate 
between the Kansas and M . falcata . The difference between the seedlings 
