Dr. E. J. Mills's Researches on Melting-point. ' 15 



slight evolution of heat; after four hours' contact, 150 cub. 

 cent, water were added. The precipitate, which was solid on 

 the following morning, was washed with warm aqueous sodic 

 carbonate, and then weighed about 8*5 grin., or 112*5 per 

 cent., the theoretical yield being 132*9 per cent. I will call 

 this specimen J dinitrotoluol. It was crystallized four times 

 from naphtha and three times from spirit. (See Table XIII.) 



[Case III.] The whole of the remainder of D nitrotoluol 

 (p. 12) was treated with nitrous hydric nitrate in the cold ; 

 2*5 grm. of the nitrotoluol being dissolved in 39 grm. of the 

 nitrate, with which it remained in contact 1\ hours. The 

 powder was very electric. I will call this specimen L. It 

 was crystallized four times from naphtha and thrice from 

 spirit. 



The following numbers were obtained with Therm. 2: — ■ 



Table XIII a. 











69-28 





•17 





•20 





•17 





•14 





•12 





•12 





•17 





•22 





•17 





•14 





•03 





•22 





•22 





•22 



Mean 



69*17 



Probable error 



•010 

 2 



Thermometer 





[Case IV.] When rather larger quantities are taken, though 

 the same proportions between the reagents and time of action 

 be observed, a different result is obtained. Thus Y was pre- 

 pared from 4*5 grm. M mononitro toluol; the crude product 

 (122 per cent.) was purified by warming with weak aqueous 

 potash containing ammonia, and was crystallized twice from 

 naphtha, eight times from alcohol. Z was made from 20*5 grm. 

 of a new stock of solid mononitrotoluol (which had been dis- 

 tilled in steam, frozen, pressed, and crystallized twice from 

 naphtha and four times from alcohol); the crude product 

 (120 per cent.) was crystallized twice from naphtha, and eight 

 to nine times (Z 8 , Z 9 ) from alcohol. Both samples had a faint 

 yellowish tinge. 



