Febbuabt 10, 1911] 



SCIENCE 



235 



ether and does not hold in its own right. Its be- 

 havior is that of an atom of negative electricity 

 pure and simple. Its form is spherical and not 

 spheroidal. Its size is probably less than one ten- 

 million-millionth of an inch. When revolving 

 briskly enough in an orbit within the atom it 

 gives us colored light of highest purity. When 

 violently jostling irregularly about it gives us 

 white light, without it all light would be im- 

 possible. 



" We believe we have found electricity free from 

 matter but never yet matter free from electricity, 

 i'lnally comes the suggestion that matter no less 

 than life may be undergoing a slow but endless 

 evolution. Some of these things and many others 

 have led physicists to suspect that if all electricity 

 were removed from matter nothing would be left, 

 tnat the material atom is an electrical structure 

 and nothing more. 



" There are, however, many stubborn questions 

 to which answers must somehow be found before 

 the so-called electron theory of matter can be 

 accepted unreservedly. As it stands it is at once 

 a most brilliant and promising hypothesis but has 

 not yet reached the full stature of a theory. 



Should it hold good the material atom with 

 its revolving electrons becomes the epitome of the 

 universe. The architecture of the solar system 

 and of the atom, the very great and the very 

 small, reveals the same marvelous plan, the same 

 exquisite workmanship. The conservation of en- 

 ergy becomes an ethereal law and the ether the 

 abiding place of the universal store of energy. 



" To end as we began, we have matter and elec- 

 tricity which some day we may know to be one, 

 and ether and energy. Of these we hope some 

 time to build, in theory, a reasonable world to 

 match the one we now so little understand. 



" When all the interrelations among matter, 

 ether, electricity are separated out and quantita- 

 tively expressed, we believe our work will be 

 complete. 



" Such then is the confession of faith, the very 

 far distant hope of the modern physicist." 



November 1, 1910 



THE PHILOSOPHICAL SOCrETT OF WASHINGTON 



The 687th meeting of the society was held on 

 January 14, 1911, President Day in the chair. 

 Two papers were read: 



Discrepancies among Recent Wave-length Deter- 

 minations: I. G. Priest, of the Bureau of 

 Standards. 



The speaker, in introducing the subject, re- 

 viewed briefly the history of spectroscopic stan- 

 dards, stated that the accuracy at present desired 

 is about O.OOIA., and gave a brief description of 

 the " method of diameters." 



In regard to the discrepancies among the re- 

 sults of Fabry and Buisson, Pfund and Ever- 

 sheim,^ the following conclusions were presented 

 and supported by tabular data.- 



1. The difference (F. & B.).(,s — P.os is not mark- 

 edly systematic when the wnole range of the 

 spectrum covered is considered. Considering the 

 precision of the measures, the systematic dififer- 

 ence that does appear is perhaps negligible. 



2. Throughout the range of the spectrum from 

 5, 167 A. to 6,495 A., inclusive, the difference 

 ( F. & B. ) .OS — P.os is sensibly systematic, the 

 algebraic mean discrepancy being + 0.00 15 A. 

 Out of the total of twelve differences to be con- 

 sidered in this range, only one is negative, viz., 

 — O.OOIA. for X = 5,167ii., the limit of the range. 



3. Throughout the range of the spectrum from 

 4,282A. to 5,002A., the difference (F. & B.).oe — 

 P.os is not markedly systematic, although there is 

 a slight predominance of negative values, the 

 algebraic mean discrepancy being — 0.00045A. 

 Out of the total of eleven differences to be con- 

 sidered, four are positive, five are negative and 

 two are zero. 



4. Considering the whole range of the spectrum 

 covered in common by the several investigators, 

 the results of Eversheim appear to be systematic- 

 ally higher than those of Pfund and Fabry and 

 Buisson by about O.OOIA. 



5. The differences, {F. & B.) .^ — E,^ and P.os — 

 E.oo, when grouped according to sign are also 

 grouped in certain spectral regions as indicated 

 in Table I. In the differences P.os — S.oo ihe co- 

 incidence of the grouping according to sign and 

 the grouping in spectral regions is pronounced 

 and unmistakable. The spectral grouping of the 

 positive and negative differences (F. & B.).os — 

 B.09 ichile less pronounced than for the differences 

 P.os — -B.os is not consistent with this grouping, 

 and the tendency of the groups in the system 

 {F. & B.) 1^ — .0 09 to coincide in spectral posi- 

 tion with groups of the same sign in the system 

 P.os — B.09 is decided. 



^Astrophys. Jour., 28, 195; J. H. Univ. Cir., 

 Feb., 1910, pp. 33 and 34; Ann. der Phy., 30, 

 pp. 837-838. 



- Initials and subscripts refer to authors and 

 year of publication. See also Phys. Rev., 31, 602. 



