536 



SCIENGE. 



[N. S. Vol. II. No. 43. 



It was early observed that the metallic 

 electrodes in the Pliicker tubes used for ob- 

 taining spectra were very effective in re- 

 moving traces of nitrogen. Even helium 

 seemed to be slowly absorbed, its spectrum 

 gradually becoming fainter and ultimately 

 vanishing. Troost and Ouvrard have found 

 that a powerful silent discharge through 

 tubes of argon or helium containing mag- 

 nesium or even platinum causes absorption 

 of the gases. The absorption is exceedingly 

 slow, but is ultimately complete. Nothing 

 is known as to the nature of the products. 



The elementary nature of helium has been 

 verj' generally conceded. Several hypotheses 

 assuming argon to be a compound of known 

 or unknown elements have been suggested, 

 most of which are without the least experi- 

 mental basis, and need not be noticed here. 

 A view which earlier met with considerable 

 approval is that the argon molecule consists 

 of three nitrogen atoms, Nj . This allotropic 

 form of nitrogen has never been obtained, 

 and we have no means of predicting whether 

 it would be more stable than the common 

 form, Nj. If more stable, and if formed in 

 small amounts in the course of ages, it would 

 tend to accumulate in the atmosphere. 

 This view was supported by a not very 

 close coincidence of densities, argon having 

 a density of about 20, while the density of 

 Ng would be 21. N"ot the slightest evi- 

 dence has been found, however, that argon 

 and nitrogen are convertible into each other, 

 even in traces, while such transformations 

 are quite readily effected in all known cases 

 of allotropism. Neither is argon formed 

 when nitrogen is liberated from combina- 

 tion. If it is a more stable form of nitro- 

 gen it should be one, if not the main pro- 

 duct under these circumstances, just as or- 

 dinary oxygen, not ozone, is the chief pro- 

 duct when this element is set free. Until 

 it can be shown that the evidence of the 

 specific heat ratio, which is based on the 

 kinetic theory of gases, is inconclusive, or 



until argon is actually decomposed into 

 simpler or familiar substances, we must re- 

 gard it as a new element. 



The elementary nature of the new gases 

 being conceded, the first question which 

 arises is whether they represent an entirely 

 new order of substances, or whether they 

 can be harmonized with the classification 

 which has been found to apply to all known 

 chemical elements whose properties have 

 been established with any degree of exact- 

 ness, and if so whether this system will 

 admit of their adoption without any essen- 

 tial change, or whether it must be more or 

 less modified and extended. 



Our present classification is known as the 

 Periodic or Natural System, and is based 

 on a consideration of the broad chem- 

 ical properties of the elements, rather than 

 on special similarities or differences such 

 as are considered in a classification for 

 analytical or other practical purposes. If 

 the elements be arranged in the order of 

 increasing atomic weights there is a grad- 

 ual modification of chemical properties as 

 we ascend in the series, but this does not 

 proceed uninterruptedly in the same sense 

 through the entire list. At certain points 

 there is a break, and the following element 

 has properties resembling one several places 

 back. This abrupt change is perhaps most 

 obvious in respect to the metallic and 

 base forming, and the non-metallic and 

 acid forming properties, and is well illus- 

 trated in the following small table, which 

 gives the first part of the series. 



Li Be B C N F 



7.02 9 11 12 14.03 16 19 



Na Mg Al Si P S CI 



23.05 24.3 27 28.4 31 32.06 35.45 



K 



39.11 etc., etc. 



Lithium (7.02), the element of the low- 

 est known atomic weight excepting hj'dro- 

 gen, is a pronounced metal, electropositive, 



