438 



NATURE 



\_Sepf. 20, 1877 



the Cyanida commences with an excellent conspectus of 

 the genera of that family, followed by detailed descrip- 

 tions of both genera and species ; two well-executed 

 uncoloured plates accompany the paper. 



In the third memoir Dr. T. Thorell gives an account 

 of the Arnnea: collected in Colorado in 1875 by Dr. 

 Packard ; the descriptions of the species are drawn up in 

 the author's usual careful and exhaustive style, and leave 

 nothing to be desired but illustrations, the absence of 

 which we cannot but deplore ; an appendix by Mr. J. H. 

 Emerton (the welMcnown American arachnologist) de- 

 scribes two additional species of the genera Epeira and 

 Drassus, with which two woodcuts are given. 



A. G. Butler 



LETTERS TO THE EDITOR 

 \The Editor does not hold himself responsible for opinions expressed 

 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 as 

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

 is impossible otherwise to ensure the appearaiue even of com- 

 munications containing interesting and novel facts.] 



Temperature of Moon's Sun'ace 



In a recent number of les Mondes (tome xliv., No. i, Sep- 

 tember 6, 1877), M. I'Abbe F. Raillard puts forward a theory 

 to explain the reddish tinge acquired by the moon during a total 

 eclipse, attributing it possibly to the great elevation of tempe- 

 rature caused Dy the continuous exposure of its surface for many 

 days previous to the solar rays, which he thinks may be adequate 

 so to raise its temperature as to render it self-luminous. 



In support of his theory he refers to my experiments with 

 the thermopile, and states that I have found the lunar surface to 

 acquire under solar radiation a temperature of more than 500° 

 Centigrade. Now in a paper published in the Proceedings of 

 the Royal Society, No. 112 (1869), I estimated the radiation to 

 be equal to that of a kmp-blacked surface 500° Fahrenheit 

 higher in temperature at full moon than at new moon, but on 

 repeating the experiments 1 with more care, 197° Fahrenheit, or 

 *loo° Centigrade, was found to be a far more probable value, a 

 large error having crept into the former result. It is, moreover, 

 shown in my last paper that near the middle of the partial 

 eclipse of 1872, Nov. 14, the radiant heat was only about one- 

 half of what it had been two hours before, having kept pace in 

 its diminution with the light. Observations made during the 

 recent eclipse, so far as they go, fully confirm this result, and I 

 much doubt if five per cent, of the heat acquired since new 

 moon is retained till the middle of a total eclipse ; heat, too, 

 which we have shown from its low mean refrangibility as com> 

 pared with that of the direct heat of the sun to have been truly 

 absorbed by the lunar surface. 



M. I'Abbe Raillard appears to be mistaken in supposing it to 

 be the generally received theory that the red tinge is due to 

 dispersion rather than to simple refraction and preponderant 

 absorption of the more refrangible rays in passing through the 

 earth's atmosphere. 



There appears to be, therefore, no ground for supposing that 

 the difference between the " lumiere cendree " of the unillu- 

 minated surface of the new moon and its reddish hue during a 

 total eclipse is to be ascribed to a difference of temperature, and 

 I think that we must fall back on the usual explanation . 



It may also be expected that independently of any tinge due 

 to unequal absorption by the earth's atmosphere the prepon- 

 derance of blue and green on the terrestrial surface may not be 

 without influence on the colour of the "earth-light " which gives 

 rise to the " lumiere cendree " and may contribute to an appre- 

 ciable degree towards forming a contrast between its hue and 

 that acquired by the moon when totally eclipsed, 



September 15 RosSE 



Rainfall and Sun-Spots in India 

 As Prof Balfour Stewart says the true test of a physical cycle 

 is its repetition, and since he evidently regards the tendency to 

 repetition which he has shown to exist in the rainfall of Madras 



I Phitosophkal Tranmctions, 1872. 



as a favourable indication of the presence of a physical cycle such 

 as that claimed by Dr. Hunter, I may perhaps be allowed to 

 supplement my former statements regarding the tendency of the 

 winter rainfall in many stations of Upper India to vary in a cycle 

 corresponding inversely with the solar spots, by exhibiting a 

 similar tendency to repetition in the rainfall of Calcutla. The 

 following table represents the winter rainfall of Calcutta from 

 1S33 to 1S76. 



The rainfall is taken for the months of January, February, 

 March, and April in each year, together with that for December 

 of the preceding year. The November fall is excluded chiefly 

 because experience and an inspection ol the register show that it 

 properly belongs to the summer monsoon rainfall, occurring 

 almost entirely in those years in which the summer monsoon 

 rains are either very heavy cr prolonged, and in fact being nothing 

 else than the last drop they shed before they take their departure. 

 The real winter rains commence in the Christmas week, so 

 December really includes their actual first appearance. As the 

 summer rains seldom begin before the second week in June, we 

 are well within correct limits in taking the rainfall from December 

 to April inclusive. The following table is arranged after the 

 model of that given by Prof. Stewart in his letter to Nature 

 (vol, xvi. p. 161) : — 



The years of minimum sun-spot occur in the first and second 

 series, and the years of maximum sun-spot in the fifth and sixth 

 series. The series of heaviest average winter rainfall are 9, 10, 

 II, I, 2, and those of lightest average rainfall are 5, 6, 7, 8. 

 Taking the mean of the averages ot the five series of heaviest 

 rainfall we get 6 '14 inches, and taking the mean of the averages 

 of the four series of lightest rainfall the result is 4'l6 inches. 

 The same result is exhibited by each cycle individually, thus : — 



Ma 



Cycle A 

 „ c 



I\lm. group, 

 inches. 



6-19 214 



6-87 4-31 



5'05 427 



6'45 4-85 



The evidence of repetition is thus quite as manifest as in Dr. 

 Hunter's case, the only difference being that in the present case 

 the years of minimum sun-spot are those of heaviest, and maximum 

 sun-spot those of lightest, rainfall. In order to render it still 

 more apparent that the cyclical connection with the sun-spots is 

 not the result of accident I will exhibit the difference between 

 the rainfalls in years of absolute minimum and maximum sun- 

 spot : — 



The same connection is maintained when the fall in April is 

 left out or that in November included, so that it is evidently not 

 due to the effect of any particular month, but as may reasonably 



