OF ARTS AND SCIENCES. 71 



strengthened further by the fact that other observers besides Frank- 

 land and myself have found the boiling-point of hydride of amyl 

 30°, or thereabouts; or that it would begin to boil at very nearly 

 this temperature, indicating at least the probability that it has not 

 a higher boiling-point, for the gaseous butyl hydride, the next 

 lower body of the series, would probably have been so completely 

 expelled from the liquid during the long-continued process of frac- 

 tionation that the little which might remain would not sensibly 

 lower the boiling-point of the amyl hydride. C. G. Williams* 

 found that amyl hydride from Boghead naphtha boiled between 

 30° and 40° ; and a preparation of the same from similar material 

 from cannel coal gave Schorlemmer t the boiling-point 39°-40° ; 

 but subsequently he found that a preparation from American petro- 

 leum boiled at 34°, — which comes nearer to Williams's first obser- 

 vation than to. the lower of his own previous observations; and 

 since he states that he obtained a larger quantity of this, which 

 would enable him to continue longer the process of fractionation, 

 and since a preparation from this petroleum would doubtless con- 

 tain less of contaminating substances that might influence the 

 determination, it may be inferred that his later determination is 

 probably the more reliable of the two. But Pelouze and Cahours % 

 make the boiling-point of this body from American petroleum ex- 

 actly 30°, agreeing precisely with my own determination and 

 with that of Dr. Frankland. Furthermore, Wurtz § states that 

 amyl hydride from amyl alcohol boils also at 30°. 



The determinations above quoted having been made previously 

 to the publication of my petroleum series, probably neither of the 

 investigators, either of the coal or petroleum naphtha, had the re- 

 motest idea of the presence of a series isomeric with the hydrides, 

 and therefore made no special effort to separate them; || and since 

 these determinations were evidently made on mixtures containing 

 variable proportions, and in some cases a preponderating quantity, 

 of these isomeric bodies, the disagreement between their results 

 does not seem surprising, nor that the boiling-point should have 



* Quarterly Journal of the Chemical Society, XV. 130. 



t Journal of the Chemical Society, 1862, XV. 419. 



\ Bulletin de la Socie'te' Chimique, 1863, p. 231. 



§ Comptes Rendus, 1862, LIV. 387. 



|| A re-examination by Schorlemmer has since confirmed my discovery of 

 this series. Proceedings of the Royal Society, XIV. 468; Annalen der Chemie, 

 CXXXV 269. 



