BETWEEN THE ATOMIC WEIGHTS. 241 



Indeed, this could not be otherwise, since, as has been already said, the members of 

 each series are isomorphous, while, as a general rule, to which, however, there are 

 many exceptions, no isomorphism can be established between members of different 

 series. These groups, however, have been merely groups of isomorphous elements, 

 and not series of homologues like those in which the elements are here classed. 



These general remarks will suffice to indicate the principles upon which the classi- 

 fication has been made, and the character of the numerical relation between the 

 atomic weights which has been established. The details of the classification can be 

 best studied by referring to the table, so that it will be only necessary to speak of 

 those points which are of special interest, or which may require explanation, or in 

 regard to which there may be doubt. The series I have named from their common 

 differences. The first I have called the Nine Series, the second the Eight Series, &c. 

 Let us examine the doubtful points in each, commencing with the first. 



The last five members of the Eight Series are connected by so many analogies, 

 that they have been invariably grouped together in the elementary books. There can 

 be no doubt, therefore, in regard to the propriety of placing them in the same series, 

 on the ground of general analogies. Fluorine, it is true, presents some striking points 

 of difference from the rest. Fluoride of Calcium is almost insoluble in water, while 

 the Chloride, Bromide, and Iodide of Calcium are all very soluble. We must, how- 

 ever, remember that we have to do with series, and must not therefore expect to find 

 close resemblances except between adjacent members. If, then, we consider that Oxy- 

 gen is one of the series, and that Fluorine stands but one step removed from Oxygen, 

 while it is two steps removed from Chlorine, the discrepancy in a measure vanishes, 

 for Lime CaO is but slightly soluble in water. Nevertheless, the difficulty does not 

 entirely disappear, for CaFl is much less soluble than CaO, although it should be 

 more soluble judging from the law of the series and the fact that CaCl is so much 

 more soluble than CaFl. 



The solubility of a series of homologous elements or compounds in water, may be 

 regarded as a function of one or more variables. In the case of elements there may 

 be but one variable, but it is easy to see that in the case of compounds there must 

 be several. One of these variables is probably the same which determines the com- 

 mon difference of the series to which the elements or compounds belong ; (it will be 

 hereafter shown that the atomic weights of the homologous compounds are related in 

 the same way as those of the elements ;) the other variables are perhaps the atomic 

 forces which determine the hardness, density, &c. of the solid. We may, therefore, 

 with justice, compare the relative solubilities of a series of homologues to a curve 



