October 25, 1895.] 



'SCIENCE. 



535 



ent, which also shows a characteristic spec- 

 trum. The diiference of density was de- 

 tected by the slower rate of diffusion of the 

 true helium, implying higher densitj'. 

 From spectroscopic considerations they are 

 able to separate the composite spectrum 

 into its two constituents, and to conclude 

 that the true helium may have an atomic 

 weight of about 5, and the other of not far 

 from 3. Both of these substances, accord- 

 ing to Kaj'ser, are present in the gas from 

 the AVildbad spring, while according to 

 Bouchard helium from the Pyreneean 

 springs also contains a second ingredient. 

 The lines of both constituents of helium 

 have also been noted in the spectra of sev- 

 eral stars. 



It was formerly asserted that helium and 

 atmospheric argon have a constituent in 

 common, this belief being based on the ex- 

 istence of a certain set of spectral lines 

 common to both. Eecent observations by 

 Lockyer, with an instrument of high dis- 

 persive power, indicate that this coinci- 

 dence is not exact, and hence that the 

 theory of a common constituent is without 

 foundation. 



The only evidence favoring the view that 

 atmospheric argon, as thus far obtained, is 

 pure is found in the apparently constant 

 boiling point of liquid argon, as observed 

 by Olszewski. It may be questioned 

 whether an experiment with only 1-10 c. c. 

 of liquid is sufiBcient to prove the total ab- 

 sence of other substances of unknown boil- 

 ing points. All samples of atmospheric ar- 

 gon naturally have the same composition, 

 being derived ft-om the same source, and 

 argon from other soui-ces has not yet been 

 studied with sufficient accuracy to throw 

 any light on the matter. It is therefore 

 quite likely that the atomic weight of 40 

 will have to be changed to the extent of one 

 or more units. 



From these results it is obvious that the 

 present figures for the atomic weights of 



helium and argon , which are based on density 

 detei'minations with impure material, must 

 be regarded as provisional, but it is evident 

 that one at least of the constituents of crude 

 helium must have an atomic weight lower 

 than the mean and hence must precede 

 lithium. As it is possible that both crude 

 helium and crude argon contain even more 

 than two gases, the problem of their complete 

 separation will doubtless be one of consid- 

 erable difficulty, especially as all seem to 

 possess a high degree of inertness. 



The few experiments thus far made for 

 the purpose of ascertaining if argon is a 

 constituent of living beings have failed to 

 detect it. 



At present there is no evidence that any 

 of these gases are decomposable into simpler 

 constituents, and until such evidence is forth- 

 coming we may continue to regard them 

 as elementary. 



Attempts to produce compounds of argon 

 have given some indications of succeeding, 

 but as yet no substances of definite and con- 

 stant composition have been obtained. Be- 

 sides the compound with benzene vapor de- 

 scribed by Berthelot, this chemist has ob- 

 tained a solid by submitting a m.ixture of 

 argon and carbon disulphide vapor, confined 

 over mercury, to the action of the electric 

 discharge. This substance is said to evolve 

 argon on heating. Ramsay found that, by 

 forming an electric arc between carbon rods 

 in an atmosphere of argon, the spectrum of 

 argon nearly vanished after several hours' 

 action, its place being taken by a brilliant 

 ' channeled ' spectrum. According to 

 Crookes this shows many analogies with the 

 spectra of carbon compounds, and may well 

 proceed from a compound of carbon with 

 argon. If Ramsay's observation that an in- 

 crease of one fifth volume occurs should be 

 confirmed, it would seem to do away with 

 the view that argon molecules are mona- 

 tomic, for a monatomic gas cannot increase 

 in volume on entering into combination. 



