20 COLORADO AGRICULTURAL EXPERIMENT STATION 
correlation tables might be constructed from the data at hand, this will 
not be done until more data is accumulated. From a cursory eximina- 
tion there would seem to be very little correlation between color and 
leaf habits and ability of the plants to perform. Preliminary counts 
and preliminary computations, however, seem to indicate that there 
may be a relation between the tillering habit and the development which 
actually takes place in the strains and their progeny, when these strains 
prove themselves to be pure lines. 
Some studies have been made on propagating some of these 
strains vegetatively. While we have none of these vegetatively propa¬ 
gated strains in the nursery at the present time, we have had clonal 
varieties which were destroyed as soon as it was shown by experiment 
that this method of propagation was favorable for increased seed 
production of desirable strains and that true clonal varieties could be 
made by such a method. In the case of desirable strains this method 
of propagating may enable us to very quickly increase a desirab'e 
strain for seed increase. Of course, if a strain should happen to be a 
hybrid so that splitting occurred in the F 2 generation, the vegetative 
method of propagation would be of no use, but in the case of pure 
lines it can be utilized to get a larger seed producing capacity of some 
of the pure strains. 
Plate No. 19. 
