TESTS OF BARLEY VAEIETIES IX AMERICA 
17 
the State number be preserved in this compihition. In practice this 
is not feasible, and it is especially impracticable ^yiill commercial 
varieties. At a given station most commercial varieties were obtained 
from other stations. If a single State nmnber were to be given to a 
variety it should be that of the first State to number it. Often more 
than one State secured the variety from a common source. In many 
instances the variety was obtained by the State more than once 
either from the same or different sources. A serious study vras made 
as to the best method of presenting the material. No other method 
seemed feasible, so a C. I. (Cereal Investigations) nimiber was given 
to all varieties. Later in the bulletin imder the C. I. number addi- 
tional data are given which could not be repeated throughout the 
publication. 
The part played by the various State stations in introducing 
varieties is emphasized in two ways. The table of C. I. numbers is 
intended to include all varieties listed in the tables, with a definite 
record of the origin of all pedigreed varieties. In the discussion of 
varieties all those of promise are treated individually and credited to 
the originators so far as the writers know them. If a variety is 
promising only at the original station it will be so shown in presenting 
the yields for that place. If it has a wider adaptation this is speciii- 
callv considered elsewhere. 
DISCUSSION OF STATION YIELDS 
In discussing the varietal yields obtained in experiments at niun- 
erous stations covering most of the United States and Canada (fig. 
9) it seems desirable to arrange them in three geographic groups, as 
Fig. ' 
-Map 
showing the locntion of agricultural experiment stations in the United States and 
anada from which yields or observations are reported in this bulletin 
33610°— 25 1- 
