WRIGHT— RECENT SPECTROSCOPIC OBSERVATIONS. 529 



3. Class O stars, with no (observed) nebulous surroundings. Tem- 

 perature high. 

 In a single instance a star, previously described as belong to Class 

 O, has been found to be surrounded by a planetary nebula, and it 

 appears likely that other observations of a similar character will be 

 made in the future, but such discoveries are hardly necessary to 

 add to the proof of the connection. The only effect would be to 

 remove one or more objects from group 3 and place them in group 2. 

 In considering this relationship it should be borne in mind that the 

 nebulae and the Class O stars have comparatively few points of 

 spectral correspondence, either in wave-length or in character of 

 line, but the nebular unclei are Class O stars, and while their 

 spectra differ from the spectra of their surrounding nebulosities, we 

 have here the undoubted proof of physical association to bridge the 

 gap in spectral similarity. 



Reverting now to the hypothesis diagram of figure 2, this spectral 

 relationship may be interpreted in various ways : 



1. It may be taken as fortifying the hypothesis diagramed on 

 the left and refuting the other. 



2. The planetary nebulae may be regarded as not standing in 

 the prior relationship to the stars indicated in both diagrams, but 

 as representing a development later in life. 



3. They may be bodies exceptional character, not directly related 

 lo those in the supposed ordinary scheme of development. 



While the evidence appears to favor the first there are arguments 

 for and against all of these interpretations, but there is no occasion 

 to discuss them here. If I have indicated at considerable length 

 the possible bearing of the observations on present notions of stellar 

 evolution it has been to point out the critical nature of nebular rela- 

 tionships rather than to attempt to bolster any particular theory. 

 What is regarded as a definite outcome of the work is that it helps 

 to perfect the proof of an element of stellar classification : the 

 relationship of the planetary nebulae to the Class O stars. 



As bearing on the physical conditions obtaining in the planetary 

 nebulae which we find associated with these extremely hot, or elec- 



