343 



Many attempts were made to obtain sodium- ethyl free from zinc- 

 ethyl, but without success. 



By distillation it was found to be equally impossible either to 



distil off g t from the crystals, or to distil off all ^ ^ I so as 



to leave pure Q jj [ behind. When the crystals are moderately 



heated in a bulb, a singular phenomenon occurs. Gas is evolved, and 

 there remains behind metallic sodium, also metallic zinc, but no car- 

 bonaceous residue. This reduction of a sodium-compound by heat 

 alone is an anomaly in chemistry. 



When the crystals are heated in the water-bath with potassium, 

 a sudden evolution of gas occurs, and there results metallic zinc, with 

 a liquid alloy of potassium and sodium a result likewise peculiar. 



When the crystals are heated in the water-bath with excess of 

 sodium, evolution of gas likewise takes place. 



From these experiments it would seem that the conjoined zinc- 

 ethyl is necessary to the existence of sodium-ethyl ; or more pre- 

 cisely, that some adjunct of a less positive nature than sodium-ethyl 

 is requisite to make the existence of the latter possible. 



Passing on to the other reactions of the crystals 2(ZnC 4 H.) \ 



NaC 4 H 5 } - 



With water there is given pure hydride of ethyl, and hydrated oxides 

 of zinc and sodium. The reaction takes place with great evolution 

 of heat. 



With carbonic acid there is given propionate of soda, which unites 

 with zinc- ethyl forming a double compound, decomposed on the ad- 

 dition of water. To the account of this reaction, published else- 

 where, I have to add that it takes place without evolution of ethyl 

 or any other gas a result which further confirms the formula of 

 sodium-ethyl adopted in this paper. 



With carbonic oxide there is also a reaction, which is in course of 

 examination. 



Cyanogen gas is instantly absorbed, with the formation of a brown 

 solution. 



With ether there seems to be no reaction. For the rest, with 

 oxygen, iodine, &c., I should predict reactions quite analogous to 

 those of zinc-ethyl, but have not specially examined the point. 



