68 THE CONNECTICUT AGRICULTUEAL 



It has, however, done the best it could, under the circumstances, 

 with due regard to its obligations in other directions. 



The following extract from the Report of the Conn. Board of 

 Agriculture for 1881, pp. 87, 88, will perhaps throw some light on 

 this subject: — 



" QuESTiOK. To what expense is a farmer to be if he wishes 

 to send samples of earth to the Experiment Station for chemical 

 analysis ? 



Answer, It will cost nothing but the freight. The answer to 

 that question suggests another. — "What is the use of analyzing 

 a sample of earth ?" We had the idea extensively promulgated 

 some twenty or thirty years ago, that if a sample of soil were 

 analyzed by a competent chemist, the competent chemist could 

 tell exactly what to put upon the field to make anything grow. 

 Well, the competent chemist can generally tell what to put upon 

 the field without making an analysis. Plenty of good manure 

 will help in almost any case ! 



A little calculation will readily show what a chemist cannot do. 

 You know that it has been frequently a matter of experience that 

 a hundred pounds of Peruvian guano, of the old-fashioned sort 

 that we had twenty years ago, would make the diiference between 

 a good crop and a poor crop, when it happened to be applied to 

 the right land, with the right crop and right weathei*.- That hun- 

 dred pounds of Peruvian guano contained about fifteen per cent, 

 of nitrogen, about fifteen per cent, of phosphoric acid, and about 

 three per cent, of potash, to which 33 pounds of ingredients its 

 fertilizing value was alone due. The soil of an acre of land, taken 

 to the depth of one foot, will weigh about four millions of pounds. 

 Thii'ty-three pounds of fertilizer, and four millions of pounds of 

 soil, assuming that the crop got all its nutriment from the first 

 foot of ground, are the two quantities which, put one above the 

 other, the smallest at the top and a line between, make the frac- 

 tion which the chemist must figure down to if he will find out 

 from an analysis of the soil what element of fertility that soil is 

 deficient in, viz : xtw^tott ^^'.tit.Wo- ^^^^, iii fact, if the chem- 

 ist in two analyses of the same sample of soil gets results whicli 

 agree within yo^.^-g-Q he is lucky and his luck does more towards 

 that result than his skill, for usually the tenth of one per cent. 

 or Ywo" ^s about the limit of accuracy in chemical analysis. It 

 may thus easily happen that the chemist cannot by analysis 

 distinguish between two soils, one of which has had a dressing 



