58 PROCEEDINGS OF THE AMERICAN ACADEMY 



in three natural series, designated, provisionally, as 1st, 2d, and 

 3d series. The first two of these may be conveniently considered 

 as parallel series, since they exist as it were side by side, and in 

 fact extend through and naturally belong to the same portion of 

 the petroleum, and present, moreover, a nearly uniform difference 

 of boiling-point, viz. about 8° C. between any two corresponding 

 or opposite members of the two series, as will appear on inspection 

 of the tables. Both series, however, have in common the boiling- 

 point difference of about 30° between any two contiguous members, 

 corresponding to the common elementary difference of "GH 2 , as also 

 shown in the tables referred to. The members of the 3d series 

 have boiling-points ranging from 0° to 150°, and those of the 2d 

 series from about 8° to 128°. 



In striking contrast with these results, Pelouze and Cahours 

 found in this portion of the petroleum but one series, or only 

 about one half the number of bodies that I obtained. We do not 

 differ, however, as to the general formula (C„H 2n + 2 ) of the bodies 

 belonging to this portion of the petroleum. But in regard to the 

 boiling-point difference for the successive terms of the series, 

 which, as already stated, I have shown to be uniformly about 30° 

 in the £ n H 2n + 2 Series, we find in their results hardly an approxi- 

 mation to uniformity in this respect. On the contrary, their 

 boiling-point differences, corresponding to an elementary differ- 

 ence of -€H 2 , are observed to vary so widely as from 16° or 20° 

 to 38°. 



Yet more remarkable is the circumstance that, above 150° (the 

 highest point reached by either of my first two series), in that 

 part of the petroleum embracing my 3d series, in which I find 

 only bodies of the general formula C n H 2n , they find, on the con- 

 trary, none of these, but continue to obtain bodies which they 

 represent as having the same general formula as those belonging 

 to the more volatile portion of the petroleum.* They maintain that 



* The temperatures here given relate, of course, either to isolated sub- 

 stances, or to sections of the petroleum considered as if composed exclusively 

 of those bodies whose true boiling points have been found to come within the 

 limits of temperature above indicated ; not ignoring the fact, however, that 

 during the process of separation, when applied to a mixture that includes all of 

 the three series, some of the 1st and 2d series come off in reality at tempera- 

 tures considerably above that indicated as the highest limit of these two series, 

 viz. 150° ; nor unmindful of the fact, on the other hand, that some of those 

 of higher boiling-point — members of the 3d series — are carried over with the 



