12 Proceedings of the Royal Society of Victoria. 



Sidgreaves' work shows that the change from the first to the 

 second type — which probably marked the inception of the 

 nebular type — was pretty sharply associated with a definite 

 level of temperature ; probably, therefore, the assumption of the 

 complete nebular type corresponds to a lower temperature level. 

 Why a nebular spectrum should be associated with relatively 

 low temperature is quite another question ; possibly the two 

 rival theories are not concerned in the answer to it, which may 

 be a matter of chemistry alone. Too much -stress must not be 

 laid on the coincidence in point of time between the periodic 

 variation of light and the appearance of the nebular lines, seeing 

 that in the case of Nova Aurigae the variable period came to an 

 end long before any nebular lines showed themselves ; moreover, 

 in the case of -qArgiis the variable period preceded the maximum 

 epoch, while in that of Nova Cygni there is no evidence for its 

 existence. 



3. The discovery of a nebula surrounding the star lends 

 plausibility, at first sight, to the hypothesis of successive collis- 

 ions, and was naturally claimed in support of it ; but it seems 

 to me that the details of Ritehey's investigation tell in a 

 difierent direction. The rapid brightening (alluded to in ^56) of 

 the patch of nebula in contact with the star might of course 

 have resulted from a new collision, the evidence of which reached 

 us between 20th September and 11th November; but in that 

 ■case we might reasonably expect to find that a rise in luminosity 

 of the star occurred somewhere between these epochs. Nothing 

 of the kind was, as a matter of fact, observed. 



4. We must now consider the unique problem oifered by the 

 observed expansion of the nebula. 



Taking Ritehey's measurements of the distances between the 

 star and the principal nebular condensations, we are led by 

 simple inspection to conclude that the rate of expansion was the 

 same all over the nebula, these distances increasing by about 

 ■one-seventh of their total amount in fifty-four days. This is 

 well shown in Table III., in the first column of which we have 

 the letters used by R^itchey to designate the principal condensa- 

 tions, in the second and fourth the angular distances of these 

 from the star on two dates, expressed in seconds of arc, in the 

 third the quotients obtained on dividing the numbers in the 



