12 -q Argils and surrounding Nebula. 



Many stars (mostly variables, I believe) give spectra 

 vs^hose dark lines are arranged in bands, leaving more or less 

 brigHt spaces between them. It became, therefore, of im- 

 portance to ascertain whether or not the appearance of the 

 spectrum of •>? Arg-iis was due to such peculiarity. 



Stars of about equal magnitude were therefore examined ; 

 in some spectra, notably in that of the red variable E, Leporis, 

 there appeared to be a condensation in the yellow, but in 

 none was the general phenomenon so sharply marked as in 

 r] Argus. 



Besides this evidence, the fact (as will be seen in the 

 sequel) that the bright lines in t] Argus are readily and rea- 

 sonably accounted for in their proper positions, leaves very 

 little doubt that the majority of them are real. 



Moreover it is quite an open question whether the band 

 character of spectrum spoken of may not be partly due to 

 the condition which produces bright lines. In a Orionis, for 

 instance, it seems at least as plausible to assume the presence 

 of that condition which produces bright hydrogen lines, as to 

 deny the presence of hydrogen altogether. 



The phenomenon is necessarily delicate, but by careful 

 manipulation the lines may be handled in a sufficiently 

 satisfactory manner to determine their refrangibilities within 

 not wide Hmits. 



For the purpose of comparing a known spectrum with 

 that of a star, the spectroscope is furnished with a reflector 

 in front of the slit. A small hole in this reflector permits 

 the passage of the star pencil, and the comparison light may 

 at the same time be reflected through the apparatus. 



In this way five lines have been determined more or less 

 satisfactorily. 



One in the red coincides with C, one in the blue with F, 

 thus indicating the presence of hydrogen. 



A third line in the yellow apparently coincides with D 

 sodium line, two in the green with the chief nitrogen line 

 and b respectively ; a sixth line suspected beyond F, may be 

 the third hydrogen line Hy. 



It should be remarked that with the dispersion used, and 

 the width of slit required to see the lines at all satisfactorily, 

 the limits of error may be sufficiently great to bring in two 

 or more competitors for a particular line ; recourse must then 

 be had to collateral evidence. 



In the cases of the red and blue lines there are no suffi- 

 ciently marked competitors in the immediate neighbourhood, 



