250 PROCEEDINGS OF THE AMERICAN ACADEMY 



the base was extracted with ether, contains the excess of dimethyla- 

 mine, and the chloride of dibenzyldimethylammonium, the extraction 

 of which will be described later in this paper. 



Properties of the Benzyldimethylamine. — It forms a colorless liquid 

 with a peculiar smell, boiling at 183-184°* with the column entirely 

 in the vapor and the barometer at 76.53 cm. It is insoluble in water, 

 but mixes freely with alcohol or ether. Two attempts to convert it 

 into a nitroso-compound gave negative results. 



The Chloride is deliquescent, and forms white radiating crystals. 



The Nitrate is also deliquescent, so that crystals can be obtained 

 only with difficulty ; it forms slender white needles, with many shorter 

 ones crossing them at right angles. 



The composition of the base was determined by the analysis of its 

 chlorplatinate and acid ferrocyanide. 



Chlorplatinate^ [C.H-(CH3)2N]^,H2PtClg. — The salt was prepared 

 by adding chlorplatinic acid to the free base, purified by crystallization 

 from water, dried at 100°, and analyzed. 



I. 0.2288 gr. of the salt left on ignition 0.0656 gr. of platinum. 



II. 0.3320 gr. left 0.0944 gr. of platinum. 



III. 0.3994 gr. left 0.1148 gr. of platinum. 



IV. 0.4244 gr. left 0.1216 gr. of platinum. 



Calculated for Found. 



[CjHjCCIIsl.Nl.U^PtClo. I. II. III. IV. 



Platinum 28.63 28.67 28.43 28.74 28.65 



Properties. — The appearance of the substance varied a great deal, 

 according to the conditions under which the crystals were obtained; 

 thus we have observed it in thick orange prisms, in long yellow 

 needles, or in pointed crystals shaped like a spear-head, f It is 

 elightly soluble in water, and in alcohol. Water is the best solvent 

 for it, good crystals being easily obtained from the aqueous solution. 



Ferrocyanide, (C.H.(CH3)oN)^H4Fe(CN)y. — This substance was 

 obtained as a white precipitate, when a solution of potassic ferrocyanide 

 was added to a slightly acid solution of the chloride. It contains no 

 water of crystallization, and its composition was determined by the 

 following analyses. 



* This determination of the boiling point was made with too small a quantity 

 of tlie substance, and must be considered as merely approximate. If we had 

 continued tlie researcii, we sliould have repeated it on a more satisfactory 

 scale. 



t The identity of the substance in all these forms was proved by analyses. 



