lio4» Experiments in the Analjzation of AlarL 47^ 



geftfci in weak nitrous acid. It remained undlfTolved, and 

 weighed 2^. 



7. From the acid in which the iron was digefted, were fepa- 

 ratcd, by foda, 2 graitiS of argil. 



The compofition of this marl, then, according to the above cx- 

 ;amination, is as under 



100. 



The fubjeQ of the above analyfis was found among the keep- 

 ings of my deceafed friend Dr Kennedy, who, in endeavour- 

 ing to teach me the art of analyfu, fometimes employed me with 

 fome fuch fimple minerals ; and when I was difficulted, he ufed 

 to give me directions how to proceed. The marl now examined, 

 I have reafon to believe, was given to Dr Kennedy either by Mr 

 Scott of Harden, or Mr Dcmpfter. If either of thefe gentlemen 

 have any anaiyfis of Dr Kennedy's, I believe they will find his 

 procefs pretty exadly followed, and defcribed, perhaps, nearly 

 by the fame words •, and probably the refult may be nearly the 

 fame. I cannot find, among the papers of Dr Kennedy, in my 

 poiTelhon, any analyies of marl, although I have the produces of 

 fome proceffes which he had employed •, except a few notes, by 

 which it appears, that one fpecimen of Mr Dempi].cr*s marl coiji- 

 tained g4 per cent, of carbonate of lime, 3 of inflammable matter, 

 argil and oxide of iron 2i. In this he appears to have loft only 

 'I per cent. y which Ihows his extreme nicety. Another marl ap- 

 pears to havo contained 551 pt'r cent, of carbon'.\te of lime ; and 

 another only 15!. 



In making his remarks on the above, A. S. may think he has 

 obtained a victory, as it fo happened, that all the other mattery 

 contained in the marl befides the carbonate of lime, were left in 

 the fediment. In this cafe, I yield to him, provided he ufcs pure 

 ^cid ; but only till he ftudies and pra^lifes the art of anaiyfis, and 

 knows the difficulties and unexpecled occurrences which he mult 

 meet with frequently. 



As I have never ufcd lime, and not much marl, in my farm, I 



S 3 cannot 



