16 



Experiments of this kind are difficult and expensive, and can not well 

 be made by private individuals. The great extent of country interested 

 in, and needing such work renders it highly expedient that the Govern- 

 ment should undertake it. 



It is expected that another station for the trial of new grasses will 

 be located next spring, probably in one of the Southern States. 



NOTES ON GRASSES. 



On the south bank of the Arkansas River, at Garden City, in south- 

 western Kansas, there is a long range of sand hills or sand bluffs. The 

 citizens of the place say that a few years ago these bluffs were destitute 

 of vegetation, and the loose, white sand was blown about in clouds by 

 the strong prevailing winds, but that recently they have become cov- 

 ered with grass. I crossed the river to ascertain what were the grasses 

 that had secured a foothold in the shifting sand beds, and found that 

 they are now sparsely covered principally with Adropogon EaJJii and 

 Redfteldia flexuosa. The first grass is a close relative of A. provincialls, 

 but with stouter culms, shorter and more succulent leaves, larger flow- 

 ers, and the whole plant of a glaucous-green color. Its roots are 

 strong and penetrate deeply in the sand to reach moisture. On sandy 

 ridges and prairies it often takes the place of A. provincialis, and is a 

 promising grass for such situations. The other grass, Redfieldia flexu- 

 osa, formerly called Graphephorum flexuosum, has not been collected by 

 botanists for many years, its known localities being very few. It was 

 therefore a great pleasure to rediscover it here. Its roundish, cylin- 

 drical leaves, one and one half to two feet long, were sparsely scattered, 

 and could with difficulty be pulled up from the long, deep, underground 

 root-stocks, which evidently were very efficient in holding the sand in 

 place. 



Unfortunately there were few flowering culms visible, and they were 

 immature and varied from the typical plant in having but one flower 

 developed in the spikelets. Whether this variation is constant in this 

 locality or is dependent upon the unusual dryness of the present sea- 

 son can not now be determined, but is deserving of future investigation. 



Agropyrum glaucum and A. tenerum are both prevalent inthe dry dis- 

 tricts of Kansas, Colorado, and Dakota, particularly where the native 

 sod has been disturbed, as in old, abandoned wagon-roads and in the 

 neighborhood of ditches. Both kinds are conspicuous on roadsides in 

 the city of Denver, near the shallow irrigating ditches. A. tenerum 

 grows in clumps and does not have running root-stocks. It is an early 

 grass, maturing in July, and afterwards is conspicuous by its whitened 

 culms and leaves. 



In several places in Kansas, Colorado, and Dakota an erect, much, 

 branched, bushy form of Muhleribergia glomerata was observed. Its ap- 



