ELISHA MITCHELL SCIENTIFIC SOCIETY. 



43 



izers with different kinds of organic matter in them. The samples were 

 very carefully prepared and were made to pass a 20-mesh sieve. 



1. Contained organic matter, fish-scraps and tobacco dust, 



2. " " tankage," &c. 



3. " cotton seed. 



4. " dried blood, " ammonites," &c. 



I determined total PsO^ in each of these by four methods. 1st, 

 by the Philadelphia method ; 2nd, by dissolving directly in strong, 

 boiling HNO3 with the aid of lumps of KCIO3 thrown in from time 

 to time, — long boiling; 3rd, by fusion with NaaCOg+KNOa and 

 taking up with hot water and HNO3 ; 4th, by burning off organic 

 matter by fire and taking up in strong hot HCI. 



Following are the results obtained : 



TABLE I. 



Sample. 



Fusion with 

 Na2C034- 

 KNO, 



Dissolving in 

 strong, boil- 

 ing HNO3 

 withKClOo 



10.25 

 12.02 

 10.66 

 10.29 



9.26 

 11.92 



9-54 

 9.21 



Phila. Method. 



8.93 



11.67 



991 



9.12 



Destroying organ- 

 ic matter by fire 

 and taking up in 

 strong, hot HCI. 



Per cent. P^Og 

 8.88 



"•93 



10.00 



9.29 



The results from the fusion are seen in every case to be the high- 

 est. The HNO3 and KCIO3 method cannot, I believe, be relied 

 upon in any case, unless it be in case of sample 2. Three of the four 

 cases where the organic matter was destroyed by fire gave higher 

 results than the Philadelphia method. The lowness of the results 

 by this latter method are due, I believe, to imperfect oxidation of 

 the organic matter and to unavoidable loss from so much manipu- 

 lation, despite the fact that the greatest care was taken. 



Four more samples were taken and the fusion method compared 

 with the Philadelphia method alone in these. The results obtained 

 were much nearer together than in the case of the other samples, 

 being larger again in case of the fusion method, but this time 

 not materially so. 



