2 BULLETIN 201, U. S. DEPARTMENT OF AGRICULTURE. 



In the analyses, the methods of the Association of Official Agri- 

 cultural Chemists, as published in Bulletin No. 107 of the Bureau of 

 Chemistry, U. S. Department of Agriculture, have been employed. 

 The analytical work in large measure was done by Mr. C. E. Good- 

 rich, assistant chemist, Cattle-Food and Grain Laboratory. About 

 one-fourth of the determinations were made by various other analysts 

 in the same office. 



The work is confined to the continental United States, and the 

 references cited are mainly from departmental and experiment-, 

 station literature. Discretion has been constantly exercised in the 

 admission of data from all sources. Many analyses found in the 

 literature have been omitted from our compilations on account of 

 incompleteness, doubt as to the identity of the species dealt with, 

 evident errors, and other reasons. To obviate such uncertainties in 

 connection with our own work, museum specimens have been pre- 

 served. These, in all cases, can be located by the aid of the serial 

 collection number of the senior writer of this paper, given in each 

 table as "Our sample No." 



It appears that nothing is to be gained by attempting any scientific 

 arrangement of the species discussed, and few botanical data are given. 

 It is considered that an alphabetical arrangement will be more con- 

 venient, and that the indices to the different parts will be much more 

 serviceable to the one using the paper than an arrangement in con- 

 formity with botanical usage. 



Chemical analyses of feeds are used by all feeders in calculating 

 rations and by investigators in performing digestion experiments. 

 Nearly all experiment stations maintain a laboratory to make 

 analyses of feeds in connection with their experiments on the feeding 

 value of various farm products. 



It has been shown by numerous experiments that a plant varies in 

 composition with age. On this account care has been exercised to 

 indicate definitely the condition of growth of the samples analyzed 

 by us. This fact must be considered in any comparisons made. 

 Failure to record such data is a fertile source of irregularity in much 

 chemical work done on natural feeds. 



While these analyses do not show all that is desirable, they do 

 show relative values, when taken in connection with the notes, and 

 they enable a comparison of the species here enumerated to be made 

 with better known feeds. The compilations of other analyses will 

 simplify the labors of agronomists, agriculturists, and agricultural 

 writers generally who have occasion to employ such data. 



GENERAL CONSIDERATIONS. 



It has been quite conclusively shown that the range question in 

 this country is preeminently one of management. The greater part 



