ORIGIN OF ELEMENTARY SUBSTANCES. 125 



pounds ; SO that the evidence of such stratification of the 

 elementary vapours as I have indicated must necessarily 

 be more fragmentary than that of the geological record. 

 The constant association in nature, however, of several 

 elements belonging to the same group, a remarkable 

 example of which is the presence of lithium, potassium, 

 rubidium, and caesium in a single mineral, lepidolite, 

 appears to confirm this view of the primitive arrangement 

 of elementary vapours. 



In the annexed table are arranged all the known ele- 

 ments in natural groups, wherein gaps appear, as in Tables 

 II. and III., Avhich indicate the existence of missing ele- 

 ments. The atomic weights of other elements which have 

 not been sufficiently investigated are also determined. 



If the theory which I have enunciated of the evolution 

 of elementary substances from hydrogen in definite pro- 

 portions be correct, the numbers representing the atomic 

 weights also represent the number of particles of hydrogen 

 from which the elements were formed. Where these 

 numbers do not coincide exactly, as in the case of Cu=62, 

 and its homologue of position, Zn=64, which are each a 

 unit less than the classical numbers, it is not to be sup- 

 posed that these discrepancies are due to errors of experi- 

 ment, but to some unknown cause which prevents their 

 true atomicity from being ascertained. 



Although the ideas of chemists on the classification and 

 quantivalence of elements have greatly changed during 

 recent years, there is no question that the alkaline metals, 

 lithium, sodium, potassium, rubidium, and caesium belong 

 to the group which I have classified under Hw. Chemists 

 are also agreed that silver, notwithstanding the great 

 divergence of some of its characteristics from those of the 

 alkaline metals, also belongs to the same group. Now 

 some of the physical and chemical properties of copper 



