ALLISON — ON THE METEOROLOGY OF HALIFAX. 181 



taking up weeks, we had much rain, and the sky was invisible on 

 the 1st ; thorough cloud and heavy rain on the 9th, with no clear- 

 ing till near midnight. The 17th a rainy clouded day. Very 

 little break in the clouds on the 24th, and rain in the evemnjr. 

 Again on the 7th, 15th, 22nd, and 29th, the sky was either quite 

 or mostly clear. Proving that at any stage of the moon's circuit, 

 storms may occur, or fine weather be enjoyed. 



There are some theories regarding lunar halos, and coronge — 

 commonly called indiscriminately, " Circles round the moon" — 

 which are interesting to examine, as these popular ideas are not 

 completely false, being based upon truth. These circles are gene- 

 rally believed to herald rain or snow ; and in so far as they betoken 

 a moist atmosphere 'these signs are correct. But moisture need 

 not be precipitated upon the spot where it is formed. First, let us 

 distinguish between the larger and smaller lunar circles — between 

 halos and coronas. Halos themselves are of two sizes ; always 

 either of 90 or 45 degrees, according to the younger Herschel. The 

 two sizes are generally seen separately, but may appear at once. 

 They are more common in winter than in summer, and sometimes 

 a month passes without one being seen. The prismatic colours are 

 more or less distinct ; here generally pale and undefined at their 

 edges, but the red always inside. Halos are formed by the rays 

 reflected and refracted by the icy crj^stals which make up the high 

 and feathery clouds known to meteorologists as cirri. 



Coronas are in fact interference colours arranged in this shape by 

 the moon's light, with the blue nearest the centre ; al times but 

 one, at times several of these circles are observed; but always con- 

 centric and larger or smaller according to the size of the watery 

 globules in the atmosphere. As has been said the light thrown 

 from the moon and broken upon clouds passing causes this phen- 

 omenon, which seems to be a more sure forerunner of rain or snow 

 than the halo, as it indicates an atmosphere nearly saturated far 

 from Earth. Still even in this case precipitation need not immedi- 

 ately follow. I p-r^ceed to the proof. 



In 1869, 12 lunar halos were here observed. In foiu' cases we 

 liad decided rain or snow ; in four cases cloudy weather with some 

 slight precipitation ; and on the remaining four following days 



