102 ANNUAL OF SCIENTIFIC DISCOVERY. 



above conclusions, and have inverted the bright lines of potassium, sodium, 

 lithium, calcium, strontium, and barium. They promise a further extension 

 of their very beautiful and valuable investigations. Pogy. Ann. June, 1800, 

 et Silliman's Journal, November, 1860. 



ON THE PROBABLE COMPOUND NATURE OF SOME OF THE 

 SO-CALLED " ELEMENTS." 



An interesting and elaborate paper by Gustav Tschermak, published in the 

 Proceedings of the Academy of Science of Vienna, and extracted in an 

 abridged form in Knop's Centralblatt (July 4, 1860), on the subject of the 

 la\v of volumes of liquid chemical compounds, affords a support to the views 

 expressed by Mr. Lea, of Philadelphia, and others, " that those bodies which 

 we have as yet failed to decompose we have not found to be elementary." 

 The author therein shows that many of the substances usually classed as 

 elements comport themselves in the physical properties exhibited by their 

 combinations as compound bodies, and that it is possible from these physical 

 properties to determine (hypothetic-ally) the number of " physical " or absolute 

 atoms which he supposes to be contained in a chemical atom of such body 

 or pseudo-element. He endeavors to show that it is possible to calculate the 

 specific gravity of a liquid from its atomic weight and the number of simple 

 (chemical) atoms in its compound molecule as data, but that the results lead 

 to the immediate inference that each chemical atom contains, with few excep- 

 tions, several physical atoms. 



The particulars of the theory of M. Tschermak, and the results deduced by 

 him, are too technical for presentation in the present volume; but a further 

 reference to them may be found in Silliman's Journal for November, I860, in 

 a paper communicated by M. Carey Lea, of Philadelphia, on the subject. 



ON THE NUMERICAL RELATIONS EXISTING BETWEEN THE ELEMENTS. 



In the Annual of Scientific Discovery for 1800, we published an abstract of 

 a paper, by M. Carey Lea, Esq. (contributed to Silliman's Journal), setting 

 forth some exceedingly curious numerical relations existing between the 

 equivalent numbers of the so-called elementary bodies. 1 In the May num- 

 ber of the same Journal we find an additional paper by the same author, in 

 which a new species of relation between the equivalent numbers of the 

 elements is pointed out, wholly distinct, it is believed, from any hitherto 

 noticed, and which Mr. Lea terms "Geometrical Ratios." 



"The arithmetical relations between the equivalent numbers of the 

 elements," says Mr. Lea, " are surceptible of at least an hypothetical expla- 

 nation, on the supposition that the common difference in a series of elements 

 may represent the equivalent numbers of a substance as yet undetermined, 

 which, by its combinations in varying proportions, gives rise to the bodies 

 constituting the successive terms of the series. The new analogies, on the 

 contrary, are more difficult of explanation, even by hypothesis. Their accu- 

 racy, sometimes absolute, renders improbable the supposition that they are 

 mere casual coincidences. 



' The nature of these relations consists in this, that if we take two substances, 

 and examine the ratio which subsists between the numbers representing 



1 See Aniiual of Scientific Discovery for 1860, pp. 279-283. 



