E. W. Hiigard— Objects and Interpretation of Soil Analyses. 187 
of each winged or soil-condition can be ase 
y first trials of ‘the efficacy of this method of investigation 
were made upon the soils of the State of Mississippi, which, 
fortunately, present extreme variations in character in almost 
every direction, and upon-every key, so to speak, of the soil 
scale. But for this fact, I might, like many before me, have 
abandoned in despair, the hope of attaining any definite results, 
Some of the conclusions reached in this work have been given 
in previous papers (this Journal, Dec., 1872, and others). Since 
then, the material has been considerably increased, and quite 
lately, the aac Melis made under the auspices of the census 
office, upon the oils of the cotton States, have greatly added 
thereto, and given a wider scope to the comparisons. The de- 
tailed record and discussion of the facts so gathered will form 
main points involve 
Lhe taking of representative om specimens is, of course, a ma 
ter of first im pope se, and sometimes of no little dificulty. 
of proper localities, the discrimination between soil and sub- 
soil, a record of depth, natural vegetation, behavior iu cultiva- 
tion, etc. As heretofore stated, I find that with such care, it 
is perfectly practicable to obtain samples representing, typi- 
cally, soil areas of many thousands of square miles; especially 
so when the subsoils are taken as the more reliable indices. 
Method of Analysis—In se ore of the solvent for mak- 
ing the soil-extract to be analyzed, I have been guided by the 
consideration, that pails not sensibly attacked by several 
days’ hot digestion with strong hydrochloric acid, are not likely 
to furnish anything of importance to agriculture, within a gen- 
eration or two. If this assumption seems arbitrary, it at least 
commends itself to common sense. e heavy draught made 
era 
vessel the strength of acid to be used, and the time neces- 
sary to secure the solution of the important substances, I have 
