Professor Apjohn on a new Compound. 167 



precipitated by oxalate of ammonia, and the iodine estimated from the amount of 

 carbonate of lime afforded by the oxalate when calcined at an obscure red heat. 

 An experiment made upon this plan did not give a very satisfactory result ; and, 

 when I considered the great disproportion between the atomic weights of iodine 

 and of lime, I did not feel disposed to repeat the process. 



The iodine may also be taken out of the compound by filings of iron, as well 

 as those of zinc, in the form of iodide of this metal ; and, though the theoretical 

 objection just stated to the process by lime is equally applicable to this method, a 

 single experiment, whose particulars I subjoin, thus conducted, led to a conclu- 

 sion corresponding very closely with that already obtained. 



Three grains of the compound gave 0.72 of peroxide of iron. But this 

 amount of peroxide corresponds to 2.27 of iodine. Hence 8 : 2.27 : : 100 : 28.41, 

 the per centage of iodine associated with the oil, and which exceeds the result, 

 28.14, obtained by the other methods, by a quantity so small, that it may be 

 viewed as affording a corroboration of the correctness of the previous deter- 

 mination. 



Having determined the iodide of potassium, and the iodine in union with the 

 oil, we can now state the composition of the compound, assuming the residue to 

 be oil of cinnamon. 



Iodide of potassium . . . . 12.55 



Iodine 28.14 



Oil of cinnamon 59-30 



99.99 

 That it is the oil itself, and no oxidized or other modification of it, which 

 exists in this compound, I have already assigned reasons for believing ; and as, 

 by the application of such heat as will fuse the compound, no water is set free, 

 it becomes highly probable that the statement above made is a correct represen- 

 tation of its constitution. But the oil of cinnamon has been analyzed, and 

 through the researches of Dumas we are acquainted with its real composition, 

 which he has shown to be represented by the formula c,8 Hg o^. If then the view 

 numerically expressed above be the true one, the 59-30 parts of oil must corres- 

 pond to some integer or at least simple number of atoms. And, reciprocally, if 

 VOL, XVIII. 2 a 



