TRANSACTIONS OF SECTION B. 113 



Experiment III. 



This was suggested by Gladstone and Tribe's paper 

 (Journ. Chem. Soc, xxxv., p. 567). It is simply a modifi- 

 cation of II. 



Three grams of magnesium-filings were heated with 0'5 gram 

 of dry precipitated copper in a current of hydrogen, till the 

 magnesium began to blacken the glass. The couple had much 

 the appearance of the original magnesium-filings, and hence 

 differed markedly from the corresponding copper-zinc couple. 

 The flask was now attached to the condenser, and, when cool, 

 17 grams of ethyl iodide were added. It was then heated in a 

 water-bath till the ethyl iodide ceased to run back, which 

 took twenty hours. 



No gas appeared to come off in this experiment, though the 

 mercury valve indicated a slight excess of pressure within the 

 apparatus. This led to the belief that success was at hand. 

 (I have since noticed that, under certain conditions, such a 

 valve will let a slow current of gas pass through it without any 

 oscillation of the mercury, the gas escaping between the mer- 

 cury and the glass. Hence, I prefer now to use a sulphuric- 

 acid valve in experiments of this nature.) 



The condenser was reversed, and, after the free ethyl iodide 

 had come over, the receiver was changed, and the flask heated 

 in an oil-bath up to 250°. As nothing distilled over, a tin-bath 

 was substituted, and the temperature raised to above 400°, with 

 no result. The residue was the same as in experiment II. 



It will thus be seen that the magnesium-copper couple, pre- 

 pared either by the wet or the dry method, has much the same 

 action on ethyl iodide as magnesium alone. There is no advan- 

 tage in the addition of iodine to the ethyl iodide. 



Experiment IV. 



This was a repetition of Cahours's experiment, except that a 

 magnesium-copper couple, prepared as in III., was used instead 

 of pure magnesium. A great deal of gas was evolved on open- 

 ing the tube, which burnt with a luminous flame v/hen ignited. 

 The residue was the same as in experiments II. and III. 



From the results of these experiments it will be seen that 

 I am unable to confirm Cahours's statement. 



Experiment Y. 



Action of Sodium- Magnesimn Allon on Etlnjl Iodide. 

 The reaction looked for was, — 



Mg+ 2Na -f 2C,H5l = 2NaI -HMg(C2H5)o. 

 Six grams of magnesium filings and 10 grams of sodium 

 were heated together in a hard-glass flask in a current of 

 8 



