The Mississippi Siinry. — Hilgard. 301. 
text, represents my personal field, ofifice and laboratory work 
dviring four years). I undertook to separate as far as possible, 
the purely scientific part from that bearing directly upon prac- 
tical points, in order to render the latter acessible to unsci- 
entific readers as the nature of the case permitted ; while at the 
same time giving scientific discussion full s\\mg in its proper 
place. This was the more necessary, as my predecessor's re- 
ports had been sharply criticised in this respect : and I think the 
result has justified my judgment in the premises. The volume 
is thus divided nearly evenly between a "geological" and '"agri- 
cultural" portion ; the former giving under the special heading 
of "useful materials" the technically important features of each 
formation, after its geological characters have been discussed. 
In the agricultural portion, it seemed needful at the time to 
give, by way of introduction, a brief discussion of the princi- 
ples of agricultural chemistry, then but little understood by the 
general public; and, accordingly, fifty pages are given to this 
subject and are discussed with reference to the agricultural 
practice of the State. In the special or descriptive portion of 
the agricultural report, the State is divided into "regions" char- 
acterized by more or less uniformity of soil and surface fea- 
tures ; and each is considered in detail with respect to all nat- 
ural features bearing on agricultural pursuits ; special atten- 
tion being given to the nature of the soils, as shown by thei^" 
vegetation and analysis. In the latter respect I departed point- 
edly from the then prevailing opinions, by which soil analysis 
was held to be practically useless. My exploration of the State 
had shown me such intimate connection between the natural 
vegetation and the varying chemical nature of the underlying 
strata that have contributed to soil formation, as to greatly en- 
courage the belief that definite results could be eliminated from 
the discussion of a considerable number of analyses, of soils 
carefully observed and classified with respect both to their ori- 
gin and their natural vegetation, and a comparison of these 
data with the results of cultivation ; and that thus it would be- 
come possible, after all, to do what Liebig originally expected 
could be done, viz : to predict measurably the behavior of soils 
in cultivation from their chemical composition. To what ex- 
tent this expectation has been fulfilled, is hardly apparent from 
the very limited number of analyses which my unaided work 
