68 



NA TURE 



\_Nov. i8, li 



to differ by nearly the maximum possible amount of half au 

 hour, some annoyance is felt, and there is still some opposition ; 

 but it seems quite clear thai, in this country at least, all resist- 

 ance will soon die out. 



As regards the more purely astronomical question of making 

 the astronomical day coincide with the civil day, by beginning at 

 midnight, instead of noon, as it does at present, there is more 

 difference of opinion. For my own part, I am frankly in favour 

 of the change, because I see no use in perpetuating an anomaly 

 which is sometimes annoying and confusing. At the same time 

 the change would, of course, involve some inconvenience to 

 computers and night-observers, and it must be admitted that 

 at present a large number, and possibly a majority, of the most 

 eminent astronomers, in other countries as well as in this, are 

 opposed to it. Those of us whose work falls about as much in 

 the day as in the night, and those, I think, who take a long look 

 ahead, are in favour of the reform ; but those whose work is 

 mainly nocturnal, or is based on observations made chiefly at or 

 near the " witching hour," dread the inconvenience of a change 

 of date in the midst of the record, and the risk of confusion in 

 the interpretation of old obsei-vations. 

 The question, however, seems to me not a very important one. 

 I notice that the visitors of the Royal Observatory have just 

 recommended that the change be introduced into the British 

 Nautical Almanar for 1891. 



Before passing to the moon, a word should be added as to the 

 outcome of the most recent investigations regarding the steadiness 

 of the earth's rotation. Some irregularities in the lunar motions 

 have appeared to justify a suspicion, at least, that they might 

 be caused by irregularities in the length of the day. The re- 

 searches of Newcomb upon ancient eclipses and occultations of 

 stars give results not necessarily inconsistent with this hypo- 

 thesis, perhaps even slightly in its favour, but his careful ex- 

 amination of the past transits of Mercury contra-indicates it 

 pretty decidedly. 



The Moon. — During the past ten years there has been no 

 work upon the lunar theory quite on a level with that of 

 Hansen, Delaunay, Plantamour, and Adams in the years pre- 

 ceding ; but the labours of Nelson, Hill, and Newcomb well 

 deserve mention. The former especially has carried his approxi- 

 mations to a considerably higher point than any of his prede- 

 cessors, though not without making a few numerical mistakes, 

 which have been detected and corrected by Hill. The investi- 

 gation of ancient and mediaeval observations of the moon by 

 Newcomb is also a very important work, as showing clearly that 

 the lunar theory is still incomplete, and that it is impossible by 

 any tables yet made to represent accurately the whole series of 

 observations. A value of the secular acceleration which suits 

 the observations of the last 200 years will not fit the Arabian 

 observations made 1000 years ago, nor will it satisfy the eclipse 

 observations of still more ancient date, unless at least the 

 received interpretation of those ancient eclipses be admitted to 

 be wrong, as Prof Newcomb seems to consider rather probable. 

 From his discussion he derives for the secular acceleration a 

 value of 8" '4, as against the value of I2"'i, deduced by Hansen. 

 It will be remembered probably by every one present that the 

 theoretical value of this quantity is about 6", and that Ferrel, 

 Adams, Delaunay, and others, attributed its apparent increase 

 to 12" to the action of the tides in retarding the earth's rotation 

 and so lengthening the day ; if Newcomb's value is correct, this 

 tidal retardation is cut down from 6" to about 2" 5. 



The study of the moon's surface has been carried on with 

 assiduity, but I do not know that any remarkable results have 

 been reached, though Klein's observation, in 1877, of what he 

 supposed to be a newly-formed crater (Hyginus N.), excited a 

 good deal of interest and discussion for a number of years ; and 

 the most eminent selenographers are still divided in opinion on 

 the question. 



The publication by the German Government of Schmidt's 

 great map of the moon, in 187S, unquestionably marks an epoch 

 in selenography ; and the photographic work of Pritcha'd, and 

 the heliometric determination of the moon's physical libration 

 by Hartwig, must not pass unnoticed. 



Probably, however, the lunar work which has drawn to itself 

 most attention and interest is the investigation of the moon's 

 heat by Lord Rosse and P'rof Langley. 



The earliest observations of the kind date back now forty 

 years, when Melloni, in 1846, first detected the moon's heat by 

 means of the then newly-invented thermopile. But the first 

 really scientific iiieasureiiioits are only about fifteen years old, 

 due to Lord Rosse, at Parson-town, and to Marie Davy, at Paris ; 



and they seemed to show that at the time of full moon we 

 receive from our satellite, not merely reflected heat, but warmth 

 railiatetl from the moon's surface ; as if this surface were raised 

 to a considerable temperature by the long insolation to which it 

 has been exposed during the preceding fortnight. Lord Rosse 

 estimated the probable temperature of this heated rock to be as 

 high as from 300° to 500° F. 



But within the past four or five years this conclusion has been 

 called in question. Observations at Parsonstown, of the rapid 

 diminution of radiation during a lunar eclipse, seem to favour 

 the newer view that the moon's surface, like that of a lofty 

 mountain-top on the earth, never gets very hot, since the ab- 

 sence of air enables the solar heat to escape nearly as fast as it 

 is received. 



Prof Langley's recent and still progressing work upon this 

 subject far excels in delicacy and elaborateness anything done 

 before. At first it seemed to show that the temperature of 

 freezing water was never reached even at the hottest parts of the 

 lunar surface ; but the later observations throw some doubt on 

 the legitimacy of this inference. It is found that the radiation 

 from the moon iniquestionably contains a considerable percent- 

 age of rays which have a wave-length longer than any of the 

 heat-rays from melting ice ; and this fact has been supposed to 

 make it probable that the moon's surface was colder than the 

 ice. But then, within a few weeks, Prof Langley has found the 

 long-waved rays in the radiation from an electric arc ! So the 

 question still hangs debatable. 



T/ie Sun's Parallax — I think we may say that, during the 

 past ten years, substantial progress has been made with the 

 problem of the solar parallax. The transit of Venus in 1882 

 adds whatever value its results may have to those obtained eight 

 years before ; but, on the whole, so far as can be judged from 

 the reductions thus far completeil and published, it would seem 

 likely that the outcome of the transit observations will be simply 

 to confirm the results obtained by other methods. It may be 

 that the data obtained from the German heliometer measurements 

 will prove more accordant and decisive than those derived from 

 photographs and from the coutact observations ; there are flying 

 rumours that they will, but it will be necessary to await the 

 official publication for certain knowledge on this point. If they 

 do not, we shall be obliged, hereafter, to relegate transit obser- 

 vations to a secondary rank, as a means of determining the sun's 

 distance. From the various observations of the two transits, 

 different computers have deduced values of the parallax all the 

 way from 8" '6 to 8" '95, corresponding to a distance ranging from 

 95,000,000 to 91,500,000 miles. 



The case is quite different with the heliometer observations of 

 the opposition of Mars, in 1877, made by Mr. Gill at Ascension 

 Island. These give, in a most definite and apparently authori- 

 tative manner, a value of 8"783, and are apparently irrecon- 

 cilable with any value much greater than 8"'8i, or less than 

 8" 75. So far as can be judged from the number, nature, and 

 accordance of the observations, I believe we must accept this as 

 the most trustworthy of the geometrical methods yet employed ; 

 though the weight of the result would certainly be increased if 

 it did not depend to such an extent upon the work of a single 

 indiviJual. 



The confidence of astronomers in the correctness of this value 

 is greatly fortified by the fact that the most recent and reliable 

 determinations of the velocity of light, made by Michelson and 

 Newcomb, in 1877, 1S80, 1881, and 1882, when combined with 

 the Pulkowa constant of aberration determined by Nyren from 

 all the data available up to 1882, give a solar parall.ix accordant 

 with the preceding almost to the hundredth of a second — 8"'794 

 as against 8" 7S3. It is true there are possible theoretical objec- 

 tions to the method ; as, for instance, that the result may be 

 slighted affected by the motion of the solar system through 

 space. Enough is not known certainly about the constitution 

 of the medium that transmits light through space, to decide all 

 such questions a priori and authoritatively ; but it is unques- 

 tionable that any correction needed on account of such possible 

 causes of error must be very minute. 



We believe, therefore, "that it is safe to assume pretty coi- 

 fidently that the solar parallax is about 8"'8 (though probably a 

 trifle less), which makes the sun's mean distance 93,000,002 

 miles, with an error not likely much to exceed 150,000 miles. 

 A larger value of the parallax (about S"'S5) still holds its ground 

 in the nautical almanacs, and undeniably is nearer the average 

 of the results given by all the known methods. But none of 

 the other methods seem to us to compete at all in precision 

 with the two whose authority we accept. 



