500 TRANSACTIONS OF SECTION A. 



absence of visible clouds; fur where the directiou of drift is tangential to the 

 sun's limb the character of distortion is that of a progressive ' rippling ' along 

 the limb in the direction of the movement of the air, and where the direction of 

 drift is at right angles to the limb of the sun this rippling character of distortion 

 is converted into a ' springing' in and out of the area of the sun's image. These 

 co-ordinated movements suggest that the superficial surfaces of layers of air are 

 normally waved or ' rippled,' the troughs and crests of the waves lying in parallel 

 formation at right angles to the direction of their propagation. If this be the 

 case the problem of ' shadow bands' associated with total solar eclipse may find 

 its solution in waves of this nature being made visible upon the ground con- 

 ditionally upon the circumstance of limited illumination. The reliability of the 

 writer's method of observation has been proved by comparison of her records with 

 contemporaneous official records of the Royal Meteorological Society. Her paper 

 on the subject, presented to that society in April last, is to be found in the 

 society's ' Quarterly Journal,' July 1906. The applicability of the method to the 

 study of the drift and conformation of the more inscrutable regions of the air 

 is obvious. It is also to be employed for detecting variations in those atmospheric 

 conditions that find their expression in phenomena of selective absorption and 

 ditfraction of light. Such evidences as have been obtained show a complete accord 

 with established opinions regarding colours associated with sujvise and sunset, 

 in so far as these arc typical of weather. They consist of variations in the fringe 

 of colour associated with the telescopic image of the sun and appear to bo 

 deserving of attention as a genuine meteorological phenomenon. It was claimed 

 in conclusion that it is pre-eminently to the sun, with its penetrating light and 

 measurable disc, that the meteorologist must turn for news of the highest regions 

 of the atmosphere, and to elucidate, amongst other problems, that of the source of 

 atmospheric electricity, which, gi-anted it is ever generated by friction between 

 cloud-masses, is doubtless generated also by friction between contiguous strata 

 of air, moving as these do in so much the more intimate and effectual contact. 



4. On the Radiation of Heat from the Moon. 

 By the Earl of RossE, ll.P., F.R.S. 



In this paper, after shortly recapitulating the principal points of a communica- 

 tion to Section A at Cape Town last year, in which the difficulties in the way of 

 advancing the subject in an unfavourable climate and without the co-operation of 

 other workers are stated and dwelt upon, the main particulars of some heat 

 determinations made during the lunar eclipses of 1898 and 1903, which have been 

 lately fully reduced, are given. 



As had been anticipated, owing to circumstances adverted to above, these 

 observations are found to be quite inconclusive on the two points on which further 

 information was particularly desired. 



The lagging, however, of the decrease and subsequent increase of the heat 

 behind the light was unmistakably detected. 



.5. York Rainfall Records and their Possible Indication of Relation 

 to Solar Cycles. By J. Edmund Clark. 



Meagre records exist for 1811 to 1824, and unbroken records from 1831. 

 The average rainfall from 1831 to 1900 is 24-766 inches. Decades vary from 

 23'265 in the fifties to 28'036 in the seventies. The mean for the eighty-nine 

 years is 24'57 inches. 



The monthly curve (thirty-days means) ranges from 1'.56 inches in March to 

 2 48 in July, 2"51 in August, 2'17 in September, and, the maximum, 2'54 inches 

 in October. 



The September break is peculiar. This month averaged 3-36 inches in tie 



