146 ANNUAL REPORT SMITHSONIAN INSTITUTION, 19 38 



maxima occurred midway in the unobserved intervals preceding dis- 

 coveries. By restricting the data to unobserved intervals less than a 

 month, it was possible to use a large number of novae and derive a 

 mean point on the light-curve at about 15 days after maximum, and a 

 short section of the curve in either direction. The mean maximum 

 was then derived by extrapolating this segment of the light-curve on 

 the assumption of a constant slope (of the order of 0.05 magnitude per 

 day). 



Baade, examining the best available photographic light-curves of 

 galactic novae, found that the luminosities faded rapidly during the 

 first few days, then more and more slowly until, by 3 weeks or so after 

 maximum, the slopes were comparable with those observed in M31. 

 He pointed out that the two groups of novae should properly be com- 

 pared in the most reliable section of the mean light-curves rather than 

 at maximum where the data in the spiral are few and unsatisfactory. 

 If the comparison is made at 20 days, the two curves agree fairly well 

 over a short range, and the extrapolation to maximum luminosity can 

 be made on the basis of the well-determined curve for galactic novae. 

 The procedure leads to a mean maximum in the spiral at M = —6.7 to 

 —7, depending on the method of weighting the data, in excellent agree- 

 ment with current values for galactic novae. Moreover, the brighter 

 novae seem to fade more rapidly than the fainter, and such a correla- 

 tion would evidently increase the dispersion about the mean maximum 

 suggested by the incomplete data for the spiral. 



Thus the outstanding discrepancies are removed. We can state 

 with some confidence that the relative luminosities of novae and 

 Cepheids agree with those known in the galactic system, well within 

 the uncertainties of the observations. Novae in the two systems 

 appear to be strictly comparable, and occasional maxima of M— — 9 

 are readily accounted for as random deviations from the statistical 

 mean. 



SUPERNOVAE 



Another type of nova in M31 is represented by the famous star of 

 1885, S Andromeda, which reached a maximum of the order of 

 M= — 15, or twice the total luminosity of the average stellar system. 

 The outburst occurred before the days of the spectrograph, and the 

 many descriptions of the visual observations must be read in the light 

 of information derived from the photographic study of similar stars 

 made during the past year. It now appears that S Andromeda was 

 a typical example of the rare, spectacular supernovae which form a 

 well-defined group of objects quite different from any others known 

 to astronomy. Supernovae represent the sudden release of energy 

 on a tremendous scale, and their unique spectra offer problems whose 

 solution should furnish information of a new kind. 



