Dr. TV. Ramsay on Picoline and its Derivatives. 29 



mercury boils ; they were therefore fractionated in a vacuum 

 produced by a Sprengel's air-pump. After numerous distil- 

 lations, a fraction was separated, which boiled in vacuo at 

 145-155°, and in air at 295-305°. Two determinations of 

 the vapour-density of this oil gave the following numbers : 



I. 81-55: II. 82-80. 



These numbers agree sufficiently well with the vapour-den- 

 sity of dipyridine, 79. 



On analysis it gave the following results : — 



I. Taken, 0-3935 gram. 



C0 2 1-1272 gram, = 78-10 per cent. C. 



H 2 0*2362 gram,= 6*67 per cent. H. 



IE Taken, 0*3003 gram. 



C0 2 . ...... 0-8750 gram, = 79*06 per cent. G. 



H 2 O 0*1778 gram, = 6-57percent.il. 



III. Taken, 0*2830 gram. 



C0 2 0*8181 gram, = 78*81 per cent. C. 



H 4 O 0*1717 gram, = 6*74 per cent. H. 



Dipyridine, C 10 H 10 ]S" 2 contains 75*94 per cent. C. 



6*38 per cent. H. 



Both carbon and hydrogen in every case are too high; and 

 I have been unable to conjecture any reason for the excess. 

 I am inclined to believe, from the vapour-density, as well 

 as from analyses of some compounds of the base, that it really 

 has the formula of dipyridine. 



Isodipyridine is a yellow viscid oil, of 1*08 specific gravity. 

 It boils at 145-155° in a vacuum, and at 295-305° at the 

 ordinary pressure. It does not solidify on exposure to the 

 cold produced by a mixture of snow and salt, nor is crystalli- 

 zation induced by addition of crystals of dipyridine. It 

 dissolves very sparingly in water, and communicates to it its 

 characteristic heavy basic smell : it mixes in all proportions 

 with alcohol and ether. 



Isodipyridine hydrochloride. — TVhen a solution of isodipy- 

 ridine is evaporated to dryness with hydrochloric acid, redis- 

 solved in alcohol, and allowed to remain over sulphuric acid, 

 small, hard, white crystals are deposited. With platinic 

 chloride, the solution of this salt gives a yellow crystalline 

 precipitate of the platinichloride. A determination of platinum 

 was made with the following results : — 



Taken, 0*3903 gram. Loss at 110°, 0*0085 gram, = 2*43 

 per cent. 



Pt 0*1298 gram, = 34*34 per cent. 



