GROWING ALFALFA SEED. 447 
Wheeler has used the method on an experimental and field scale 
and his results are confirmatory to those presented in this paper. 
The work on which the conclusions here presented are based 
has been conducted at various experiment farms of this bureau 
and on the farms of Lewis Brott, Sextorp, Neb.; E. Bartholo- 
mew, Stockton, and Dr. W. A. Workman, Ashland, Kan. 
Row cultivation for seed growing has been in use for a num- 
ber of years in the vineyard regions of southern Germany, par- 
ticularly in Baden and Bavaria, in the production of seed of 
Alt-Deutsche Frankische luzerne, a _ well-recognized German 
strain. It is said that alfalfa is grown in cultivated rows for 
seed in parts of Russia, where hand cultivators prove an ef- 
fective and practical means of holding the weeds in check and 
of conserving soil moisture. 
The method has been employed for a number of years by Dr. 
L. Trabut, government botanist of Algeria. Fairchild describes 
a method of growing wheat between alfalfa rows in Algeria 
under light rainfall, where it has been found possible to pro- 
duce a crop of wheat between the wide rows of alfalfa in alter- 
nate years. The practical value of this method for the semi- 
arid portions of the United States was indicated in the publica- 
tion mentioned, without, however, making any direct reference 
to the seed-producing possibilities of alfalfa sown in cultivated 
rows under such conditions. 
Principles of Seed Production.—Although alfalfa has been 
grown increasingly in the West since 1854 or 1855 little has 
been done to develop a rational seed industry. It is a matter 
of common observation that even in recognized seed-producing 
sections the seed crop is very uncertain. A study of some of 
the factors that cause success or failure has indicated some of 
the underlying principles affecting the production of profitable 
seed crops. In Bulletin 118 of this bureau attention was directed 
to the fact that cultivated alfalfa is not a homogeneous species, 
but is composed of numerous races, strains, varieties, and even 
sub-species. These vary greatly in many characters, and espe- 
cially in their seed-producing capacity, no pure varieties of 
known high value comparable with those we have of corn, wheat, 
and other crops having as yet been established. It has also 
been noted that the individuals constituting these diverse races, 
elementary species, or whatever they may be called, exhibit great 
variation among themselves. This is particularly true of their 
ability to set seed. To overcome the source of error resulting 
from this diversity in individual plants the method of vegetative 
propagation described by Westgate and Oliver, of the Bureau of 
Plant Industry, has been used in a portion of this work. 
