36 



ART. 19. N. ICHINOIIE : 



\ve can not see that the range is decreasing with time, in 

 average. The mean value of ratios of the period by the duration 

 of the eclipse also has not any systematic evidence. 



Of course, there might be some brighter stars which really 

 belong to this class of variable stars, while astronomers have 

 as yet failed to detect the nature of the light-variation. But 

 these would be comparatively rare except those having a very 

 small range of variation, and the most fruitful field will be 

 found in the stars whose normal brightness is fainter than the 

 ninth magnitude. Thus, the most part of the remaining region 

 will be cultivated by the photographic explorations. 



(b) SPECTRUM OF ALGOLS. 



Among 93 or 94 stars of the Algol-type variables, we know 

 their spectrum only for 34 cases of them. Thus, of about two 

 thirds of them, Ave have not the important knowledge on these 

 stars and it is a great regret for us when we wish to study the 

 properties of this class of variable stars. Classifying their 

 spectrum according to the Draper Catalogue of the Harvard 

 College Observatory, we have the results given in the table 

 below. 



TABLE VIII. 



