870 Report on a re-examination of the 



cleared up, although there remains the best reason for supposing 

 that it chiefly depends upon their capacity to afford carbonic acid. 

 The more alkaline the bases united with these acids in a particu- 

 lar soil, the more favorable are the conditions for vegetation, a 

 fact which is apparently connected with the superior solubility of 

 alkalescent salts. 



It may be doubted whether the steps directed to be taken in 

 the analysis for the determination of the salts of lime are free from 

 all objection. The treatment of the soil after being freed from 

 geine, with dilute hydrochloric acid, must necessarily take into 

 solution aluminium and iron, beside rendering a portion of the 

 silicic acid soluble. On evaporating the fluid to dryness as di- 

 rected, and treating the mass with boiling water, it would there- 

 fore follow that a residuum of alumina and sesquioxide of iron, 

 (owing to the partial decomposition of the chlorides of aluminium 

 and iron from evaporation to dryness,) together with some silicic 

 acid (rendered insoluble by the same treatment,) would go to in- 

 crease the weight assumed to be pure phosphate of lime. 



As might be anticipated therefore, we find the ratio of phos- 

 phate of lime in the soils of Massachusetts exceedingly high, va- 

 rying from 0.5 to 2 per cent. 



Growing out of the same procedure, it appears also, that the 

 proportion of sulphate of lime must generally be rather too high : 

 for if, as we suppose, the hydrochloric acid attacks the aluminium 

 and iron, the aqueous solution regarded in the formula as chlor. 

 calcium only, must contain also the chlorides of iron and alu- 

 minium, as well as some silicic acid. Consequently, we find the 

 sulphate of lime quoted in some of these analyses at 3 p. c, and 

 even higher in a few instances. 



The foregoing inadvertencies (as they strike us) in Dr. Dana's 

 rules of analysis, are not conceived to vitiate in an important man- 

 ner the results contained in the report, nor do we mention them 

 because we imagine they were unperceived by the inventor of 

 the formula or by Prof. H. ; but through a desire to induce these 

 gentlemen to obviate, if possible, the objections urged against it, 

 and still preserve its claims to convenience on the ground of fa- 

 cility of working and accuracy of result. 



The report contains the following remarks respecting the re- 

 sults of these analytic investigations. 



