April 25, 1878] 



NATURE 



505 



Built on wild fancies which thy name surround ? 

 Or doth the story of thy classic ground 

 With the stern facts of Nature's face agree ? 

 What if no tongue may tell ! — thy halo fair 

 Still lingers round the isles which slumber there," 



{" Lyra Devoniensis," p, 135). 



LETTERS TO THE EDITOR 



\The Editor does not hold himself responsible for opinions exf^ressed 

 by his correspondents. Neither can he undertake to return, 

 or to correspond with the writers of, rejected manuscripts. 

 No notice is taken of anonymous communications. 



[ The Editor urgently requests correspondents to keep their letters at 

 short as possible. The pressure on his space is so great that it 

 is impossible otherwise to ensure the appearance even of com- 

 munications containing interesting and novel facts."] 



Indian Rainfall 



As Dr. Hunter has been good enough to mention my name in 

 his letter to Nature (vol. xvii. p. 59) in connection with a 

 comparison of the rainfall of Northern India and the sun-spot 

 period, I may, I trust, be allowed to express my opinion regard- 

 ing the validity of some of the conclusions he draws therefrom. 

 In the first place I would remark that Dr. Hunter's idea of the 

 winter rainfall of Northern India being due to the immediate 

 rebound of the summer monsoon from the Himalayan barrier is 

 at variance with facts in the meteorology of the country. The 

 rebound ought to take place directly the monsoon vapour -current 

 impinges upon the Himalaya, i.e., in the summer. In fact, it is 

 by a succession of oblique rebounds from this impassable barrier 

 that the monsoon is gradually reflected towards the N.W.P. and 

 the Punjab. 



The winter rains, on the contrary, have nothing to do with the 

 monsoon, being, as is well known, due to a branch of the anti- 

 trade, which, descending in the Punjab, is deflected by the 

 Himalayas towards Behar and Bengal, occasionally reaching 

 Calcutta, lat. 22° 35' N. 



Now between the rains of the summer monsoon and those of 

 the anti-trade (or anti-monsoon as it is occasionally called), there 

 is a well-marked interval of bright, clear, settled weather for 

 two or three months throughout Northern India. After this 

 interval the skies again become clouded, and about Christ- 

 mas, or shortly after, the gentle but soaking rain of the cold 

 weather sets in, and is repeated at intervals up to the end 

 of March. It is evident, therefore, that the two currents, 

 monsoon, and anti-trade, are totally unconnected with each 

 other ; and hence arises the desirability, especially in a ques- 

 tion like the present, in which its secular variations are being 

 discussed, of completely separating the rain of the former from 

 that of the latter vapour current. I cannot but think that it is 

 his omission to allow for these vapour currents that has led 

 Dr. Hunter to offer such an erroneous explanation of the re- 

 sults obtained. According to him, copious precipitation 

 should take place in the interval (October to December) between 

 the two seasonal falls, during which clear weather is invariably 

 present. It may be added that the period (December to April) 

 which I took to comprise the winter fall, commenced after this 

 interval. 



The real explanation of the direct variation of the summer, 

 and the inverse variation of the winter rainfall of North India, 

 with the sun-spot period, is the hypothesis which first led to its 

 verification coincidently and independently, by Mr. Hill and 

 myself. 



To enter upon a complete exposition of this hypothesis would 

 occupy too much of your valuable space, but as it has been found 

 to explain most of the anomalies which have hitherto proved such 

 powerful obstacles (especially in extra-tropical regions) to the 

 universal extension of the theory of sun-spot influence (I use the 

 term advisedly) on the different elements of terrestrial meteoro- 

 logy, I will here briefly indicate its general outlines for the 

 benefit of other workers in the same field of investigation. 



The hypothesis, to start with, assumes the solar radiation to 

 vary inversely with the sun-spot frequency. 



It then takes account of the probable effects of such a varia- 

 tion upon the vapour-bearing currents throughout the globe with 

 respect to velocity, direction, season, and latitude. According 

 as trade, anti-trade, monsoon, or anti-monsoon, prevail (i) at 

 different places at the same season, (2) at the same place at dif- 

 ferent seasons, so will specifically distinct effects arise both from 



the amount of vapour brought and its conditions of precipita- 

 tion, to determine which, not only the general conditions intro- 

 duced by latitude and season, but the local and peculiar 

 meteorological functions of the region must be carefully studied. 



Now as the principal effect of a secular change in solar 

 radiated heat must be to cause a similar direct secular change in 

 the normal convection currents of the atmosphere, we may 

 expect the tropical trade-wind and monsoon regions to furnish 

 us with some evidence, whether direct or indirect, in favour of 

 the above hypothesis. 



Little direct evidence has at present been adduced besides that 

 given by Mr. Hill from a comparison of wind velocity in the 

 N.W.P. (Nature, vol. xvii. p. 505). A good letri of indirect 

 evidence, however, is furnished in the monsoon regions by the 

 occurrence of abnormal droughts and floods in contiguous dis- 

 tricts (the drought in the N.W.P. and floods in Assam and 

 Burmah last year were good examples of this kind) at the time 

 of minimum sun-spot, when the velocity of the current being 

 increxsed it travels in a more contracted channel, and, by a more 

 equable distribution of rain at the time of maximum sun-spot, 

 when the velocity of the current being decreased, it is more 

 liable to extend laterally. In the trade-wind regions similar 

 evidence is furnished by the fact of a deficiency of rain and 

 cyclones at the time of minimum sun-spot, with a corresponding 

 excess of both at the time of maximum sun-spot. The augmented 

 velocity of the wind currents at the former epoch, preventing the 

 formation of local areas of condensation and precipitation, and 

 therefore (according to Messrs. Blanford and Eliot's theory of 

 cyclone generation) of cyclones and their accompanying down- 

 pours ; while the diminished velocity at the latter epoch favours 

 the same. 



Finally, the anti-trade which in its seasonal shifts north and 

 south traverses the entire temperate zone, in the winter bringing 

 rain to North India, Palestine, Madeira, California, &c., and in 

 the summer to Northern Eurojie and Siberia, should give signs 

 of a secular change in intensity and humidity, corresponding 

 according to the hypothesis inversely with the sun-spots. In the 

 summer, when large continental areas like Europe are more 

 immediately under the direct influence of solar heat, local con- 

 vection currents being set up will tend to disturb and complicate 

 the effect of any general change in the strength of the anti-trade. 

 In the winter, on the other hand, the obliquity of the incidental 

 solar rays leaves the anti-trade in undisputed possession of the 

 field. At this season, therefore, there should be a marked varia- 

 tion in the rainfall of the temperate zone, more particularly in those 

 regions between 25° and 40° N. and S. lat., where the rainfall 

 of this seaion is the chief rainfall of the year, corresponding 

 inversely with the sun-spots. Even in those regions where the 

 rain falls at all seasons, if we pick out the winter from the total 

 annual falls, as was done by Mr. Draper, for New York 

 (Nature, vol. xvii. p. 15) in accordance with Mr. Hill's admir- 

 able suggestion (vol. xvi. p. 505), the results favour the hypothesis. 

 But they do this in a far more marked manner where the rain 

 of the entire year falls during the winter months, as in the 

 Mediterranean and at Jerusalem, which have consequently hitherto 

 been considered by Dr. Jelinek and Mr. Meldrun to afford 

 strong evidence against the theory of a direct connection between 

 rainfall and sun-spots. The inverse variation of the winter rain- 

 fall of Northern India is only another example of the same law, 

 and shows how extremely important it is to analyse the seasonal 

 variations separately before deciding the question by amere cursory 

 glance at the total annual falls. The apparent anomalies which 

 Dr. Hunter finds presented in the North American rainfalls are, I 

 think, due to his having compared the total annual falls. If he 

 and other investigators will only take the hint dropped by Mr. 

 Hill, and which I cordially endorse, of comparing the seasonal 

 falls separately, they will find, I think, that while the summer rain- 

 falls of the temperate zone show either a non-periodic variation, 

 or symptoms of one coinciding directly with the sun-spots, 

 the winter falls will in general show unmistakable signs of a 

 variation coinciding inversely with that of sun-spot frequency and 

 area. E. D. Archibald 



Sun-spots and Rainfall 



I have read with much interest Dr. Meldrum's paper on 

 Sun-spots and Rainfall in Nature (vol. xvii. p. 448), particu- 

 larly that part of it in which Dr. Hunter's method of discussing 

 the rainfall of Madras is criticised, and a method of inquiry in 

 sun-spot researches is proposed. This method is, so far as I am 

 aware, a new one, and as such, is deserving of careful examina- 



