8 
The Colorado Experiment Station 
All the desirable points previously mentioned, have not been 
found associated with the most ideal types for hay, but a wider 
selection may secure the desired end. 
The selections for seed yielding traits, have been very positive 
in their results, and the contrast in the production of seed in the 
different varieties, have been the most marked of all. Most of the 
Turkestan plats producing practically no seed, while others among 
the sixty-four plats in the nursery, have produced the phenomenal 
Plate; IV. Contrast in Lkaetnkss 
yield, of two and one half pounds of clean seed, from a square rod, 
with less than one hundred plants. 
The varieties that have been the heaviest seed yielding ones, 
one year, have also, been the best the following season; single plants 
have produced two ounces of clean seed, and the plats sown with 
the seed of these selections, have in turn been heavy seed producers; 
so that this trait is to some extent an inherent quality. We have 
made over a hundred selections of individual plants showing desir¬ 
able qualities, and Plate No. V, shows the comparison in seed 
yield of a few of the average plants. 
While the seed yielding tendency, may be greatly affected by 
the hereditary traits, it is doubtless more generally influenced by 
