206 M. A. Delesse's Researches on Granite. 



that not nearly all the sethyle in the salt is obtained in the form 

 of cyanide, and that the process is rendered still more objection- 

 able by the insufferable odour which accompanies it. Messrs. 

 Routledge and Denman have prepared in the following manner 

 several pounds of the cyanide of sethyle and amyle in the labo- 

 ratory of University College without incurring any appreciable 

 loss of the substance, and without any odour being perceptible. 

 Crude iodide of sethyle is mixed with about four times its bulk 

 of spirits of wine, and introduced into a bolthead containing more 

 than an equivalent of powdered cyanide of potassium. The 

 bolthead is connected with a Liebig's condenser, in such a man- 

 ner, that, when it is heated to ebullition in a water-bath, the va- 

 pours will run back into it in proportion as they condense. A 

 drop or two of the distillate is collected from time to time by 

 inclining the apparatus, and tested for iodine by long boiling with 

 alcoholic potash, after which the stench test is applied in the 

 usual manner. When the reaction is complete, the mixture is 

 distilled to dryness, and the residual salts washed out by a little 

 fresh alcohol being distilled off them. The distillates are decom- 

 posed in the usual manner by potash. We have found it advi- 

 sable to use in this operation an apparatus similar to the above- 

 mentioned, with the addition of a tube connecting the upper end 

 of the condenser with one or two Woulfs bottles containing hy- 

 drochloric acid. 



In the preparation of cyanide of amyle, the chloride of that 

 radical was used instead of the iodide, as a small loss of iodine 

 must necessarily occur in the successive actions above described, 

 and in the subsequent reproduction of the elementary body from 

 its potassium salt. It is not unlikely that the chloride of sethyle 

 might also be used with advantage in this reaction, if the vapour 

 were led into the alcohol and cyanide of potassium slowly enough 

 to allow time for the decomposition. 



XXXI. Researches on Granite. By A. Delesse*. 



[TT^ROM the special study of the granitic rocks of the Vosges 

 J- Mountains, the author has made some generalizations of 

 great interest upon the relation of the proportion of silex, and 

 of the nature of the mica, to the age of the mass and to its cir- 

 cumstances of crystallization, also upon the varieties of feldspar.] 

 There are in the Vosges at least two types of granite, distin- 

 guishable by their mineralogical constitution and geological 

 position. 



The first is the granite of the Ballons ; it forms the summits 

 and the central part of the ridge of the Vosges ; its greatest de- 

 * From the Annates des Mines, vol. iii. p. 369. 



