717 



elecd'icity on llie position of tiie ice-cryslals, I met among otliers witli 

 tiie difficuUj, that tlie complete development of tiie circnmscribed 

 lialo seemed at variance witii llie explanation proposed. Therefore 

 I liope to come back to this point afterwards. For each explanation 

 however the observations on the ring and its envelope may be wanted. 

 They follow tlierefore as the concluding part of my remarks. 



BRB = lower tangent arc; O.H =:r circumscribed halo; Kering 

 of 22°; l:=left; r = right. The remaining symbols have the same 

 meaning as in the other tables. 



The mean of the 6 observations on (he red of the ring is 21° 54', 

 only 2' more than that found from the measurements on the top. 



The calculated Ab is meant for white light, the observed A„. for 

 red. Leaving apart the 4 very discordant differences for the first 4 

 observations, the mean difference observation minus calculation is — 0.3°, 

 that means exactly the difference in distance for red and white. 

 Hence these observations of the circumscribed halo are in harmony 

 with the calculation for red. In the 15 measurements on the ordinary 

 ring however, on the contrary a very distinct difference of -|-0.3° 

 remains. 



A. Measurements of the upper and lower top (red). 



mean 21.9 21.8 



Mean of all measurements 21° 52'. for red according to 

 Pernter 21° 34' 



