Mat 28, 1909] 



SCIENCE 



867 



The total exposures ranged, in the case of the 

 Andromeda nebula, up to IS' 11", accumulated 

 during three different nights, yielding four- 

 teen lines; and in that of N. G. 0. Y331 up to 

 22"' 22", accumulated during seven nights and 

 yielding two lines. Of the accuracy of the 

 wave-length determinations Dr. Fath says: 



When measuring known spectra the wave- 

 lengths of the lines usually come within 5 or 6 

 Angstrom units of the correct value, although 

 occasionally a larger deviation is found. Thus the 

 third figure given may usually be considered cor- 

 rect. 



In the identification of the lines, it is pos- 

 sible that Dr. Path was unconsciously influ- 

 enced by Scheiner's emphatic assertion:* 



It is thus proven that the Andromeda Nebula 

 exhibits a spectrum of the class Ila, or further, 

 that the greater part of the stars composing the 

 nucleus of this nebula belong to the second spec- 

 tral class. 



The inference that " the spectrum is of the 

 solar type" will, I think, be found to be too 

 hasty. The absence from Dr. Path's plates of 

 any indication of the bright lines of Huggins 

 was, indeed, quite misleading. Bright lines 

 will, undoubtedly, be photographed, later, in 

 the great nebulse of Andromeda. 



The following table records in the first and 

 second columns the wave-lengths and intensi- 

 ties of the lines in the Andromeda nebula as 

 measured by Dr. Fath, in the third and fourth 

 columns the identifications with the spectrum 

 of radium emanation suggested by myself, and 



* Astrophys. Jour., 9, 150. 



in the last column, the identifications proposed 

 by Dr. Fath. 



The wave-lengths of the lines of the radium 

 emanation are taken from the table of Ruther- 

 ford and Eoyds." The second number in the 

 intensity column of radium emanation records, 

 for line No. (1), the value by "photograph 3," 

 and for line No. (7) the "visual" estimate. 

 It should be borne in mind that for the neb- 

 ular lines " the third figure given may usually 

 be considered correct." Both schemes of iden- 

 tification are suggested by theory. Dr. Fath's 

 is supported by the theory that the Andromeda 

 nebula is a cluster of solar stars. No conclu- 

 sive review has been given of such theory. 

 From his own use of Bohlin's parallax of this, 

 nebula, he infers that " the ' star cluster ' the- 

 ory is not very satisfactory." He, however, 

 necessarily recurs to solar lines and "groups 

 of lines," selected, chiefiy, on the basis of mere 

 matching. 



My own identification assumes the theory 

 that all nebulse whatsoever are electrically 

 luminescent results of radioaction in the asso- 

 ciated " stars." This theory has repeatedly 

 been found to be confirmed in accounting for 

 analogous phenomena. It explains why, now 

 bright lines, now dark lines are observed in 

 this nebula. It suggests the kind of chemical 

 elements whose traits are here to be found. It 

 accounts for the variability of this nebula in 

 1885, by asserting that the " star," variable, 

 through radioaction, holds the secret of cosmic 

 evolution. Finally, it banishes the necessity 

 for imagining a lawless collision of a pair of 

 stars for generating spiral nebulse. 



The general agreement in wave-length of 

 this single radio-element with all but one of 

 the nebular lines must be given weight. There 

 are three lines, Nos. (8), (9) and (10), which, 

 both in the spectrum of the nebula and in 

 that of the emanation, follow consecutively. 

 Moreover, their intensities, all conditions con- 

 sidered, are quite consistent with the hypoth- 

 esis of identity. There is no necessity for 

 reminding spectroscopists that, on account of 

 the wide variation of intensity with changes 

 in excitation, fair agreements in such a case 



'Phil. Mag., Ser. VI., Vol. 16, p. 317, Aug., 1908. 



