44 SMITHSONIAN MISCELLANEOUS COLLECTIONS 



Position V 



VOL. 52 



On account of the number of observations, the result in Position 

 I is entitled to twice as much weight as that in Position II ; and on 

 account of the larger deflections observed, the results ^n Positions 

 IV and V are regarded as each of twice the weight of those in Posi- 

 tion II. It is not thought that the variations of the ratio of trans- 

 missibility between the several observations just noted are beyond 

 the probable errors of the single determinations, so that without dis- 

 tinguishing separate positions, the weighted mean result for the 

 transmissibility of the inner corona-brightness may be regarded as 

 0.364- For positions I, II and IV, V, taken in pairs, the means are 

 0.366 and 0.362 respectively. 



In order to determine the intrinsic corona-brightness as compared 

 with sun-brightness, we must first multiply the average solar deflec- 

 tions observed before and after the eclipse by the two factors appro- 

 priate to allow for the ratio of size of mirror apertures employed and 

 for the introduction of series resistance in the galvanometer circuit 

 respectively. Performing this reduction and introducing also the 

 data of sky-brightness already given, we obtain the following values 

 based on a sun-brightness of 10,000,000: 



Sun near zenith (Flint Island) 10,000,000 



Sky 20° from sun (Flint Island) 140 



Sky distant from sun (Flint Island) 31 



Sky average (Flint Island) 62 



Sky average (Mt. Wilson) 15 



Corona Positions IV and V 13 



Corona Positions I and II 4 



Moon about 50° zenith distance (Flint Island) 12 (?) 



DivSCUSSION OE THE RESULTS 



When we recall the extreme brightness of the sky within a 

 single degree of the sun as compared with that 20° away, and con- 

 sider also the figures just given, the proposal to observe the carona 

 without an eclipse seems an unpromising one. 



