42 REPORTS ON THE STATE OF SCIENCE. — 1915. 



sunset for one or more consecutive days prior to rain. The following 

 instances may be quoted : — 



Feb. 13, 1915 . . Strays strength 2a from 6 a.m. to 6 p.m. 

 Feb. 14, 1915 . . „ 3a from 6 a.m. to 10 a.m. and 



4a from then till 5 P.M. 

 Feb. 15, 1915 . . „ 3a from 6 a.m. to 1 p.m. and 



3c from 1 p.m. to 7 p.m. 



On February 15, 16, 17 and 18, 167 points of rain were recorded, of 

 which only 3 points fell on the 15th. On February 23, 1915, and 

 strays were of strength 3a from 6 a.m. to 6 p.m. On February 24 

 129 points of rain fell. 



This conclusion is borne out in other ways by some of the records 

 forwarded to the Committee. There is evidence that north-west winds 

 on our Atlantic coasts, especially in the winter, are associated with 

 strong strays at Irish stations and at sea. The atmosphei-ic convection 

 produced by the land may be sufficient to account for this. Mr. H. 

 Eicci, of the Marconi Company, who has made reports on two trips 

 I'ound the world, during which he made especially careful daily observa- 

 tions, states that in mid-ocean strays are, as a rule, very few and feeble 

 both in the day and in the night ; but that when the edge of a 

 mountainous continent is approached strong, and even continuous, 

 strays are normal. In this something must depend on the direction of 

 the prevailing wind relative to the land — a matter that will be inquired 

 into later. 



It may be mentioned here as veiy significant that the months of 

 the greatest X stonns in the Mediterranean are shown by the records to 

 be September and October, the months of cyclonic weather. 



As a whole the statistics show that there appear to be two kinds 

 of X storm occurring in the day-time : (1) Those produced by convec- 

 tive conditions in the atmosphere within, perhaps, a hundred miles of 

 the station, which may be termed local X storms ; (2) Those originating 

 at a distance. Eegarding the first class, they may occur almost simul- 

 taneously over a whole continent, but this is only because convective 

 conditions happen to be ruling all over that area. Stations not too far 

 outside the boundaiy of such a region also receive many strays, but 

 apparently their distance must be hmited to within 200 miles of 

 the disturbed regions. In general, we may conclude that the ob- 

 servation of strays in the day-time constitutes a method of feeling the 

 fringe of a region of convective weather, and so anticipating thunder 

 and rain a day or two in advance. Of course, this ability to prophesy 

 the advent of thunder weather is well known and is as old as wireless 

 telegraphy itself ; but hitherto it has been thought that the electric dis- 

 charges at a great distance were responsible for the strays heard at the 

 station attempting to prophesy. The present analysis indicates, rather, 

 that at any rate in the day-time the strays are frequently due to very 

 local discharges, often too weak to give noticeable lightning or thunder, 

 but definitely indicative of an approaching period of instability in the 

 atmosphere. 



The second kind of X storm is not of strictly local origin, but is 

 sometimes traceable in the stray observations made hourly at the Malta 



