324 THE BAILIWICK RAINFALL. 



atmosphere northwards being exceptionally clear, the reflection 

 of St. Catherine's light in the Isle of Wight was visible on 

 the sky. 



At the end of November began a full three weeks' long 

 cold snap. This wintry interval, which set in on the 26th and 

 ended on December 20th, inaugurated what will be known as 

 the severe winter of 1916-17 — the first really hard winter since 

 that of 1894-95. This early burst of cold, well marked as it 

 proved to be (on four days the air temperature at Les 

 Blanches dropped to or below 32 deg., and the mean of the 

 week ending December 16th was 6*4 deg. below the normal), 

 was a very mild affair compared with the bitter conditions 

 prevailing during part of the cold weather that set in on 

 January 9th and prevailed until February 1 5th. 



Each year provides peculiarities in the distribution of 

 rainfall over the Bailiwick, and the most interesting instance 

 in 1916 was probably that of December 17th. On that day, 

 with the barometer at 29*7 in. and only a very light easterly 

 wind stirring, just under three quarters of an inch of rain 

 (0*73 in.) fell at Alderney as reported by Mr. Picot, while at 

 Guernsey (Les Blanches) and Sark the day was quite dry. 

 The cause of these peculiarities is probably the existence of 

 some slight local irregularity in pressure which brings about a 

 fall of rain over the area affected. 



A very fine fire ball, moving from south to north, flashed 

 through the sky at 5.30 p.m. on December 19th, and emitted 

 a brilliant blue light. I was not fortunate enough to see it, 

 but it was reported to me by Mr. C. G. de la Mare and others, 

 and some of the English scientific papers recorded its 

 occurrence as observed in England. 



The water in the well at Les Blanches averaged (for the 

 year) two inches higher than in 1915 and twenty inches above 

 the normal of the 15 years 1902-1916. At the end of March, 

 owing to heavy winter rains, the springs were literally over- 

 flowing their banks. Looking back upon the fact and upon 

 the long summer drought that followed, this proved a blessing 

 in disguise, for despite the steady seasonal drop in the water 

 level from April onwards, there was sufficient in st<>re to 

 prevent a real shortage when the springs approached their 

 lowest. At Les Blanches the flow set in about December 

 17th — a later date than usual by several weeks. Owing too 

 to deficient winter rainfall the rise is so far (February) 

 progressing only slowly. Actually the spring is five feet 

 below the level of a twelvemonth ago, and has not been so low 

 in February since 1909 — a year of very low springs. 



