KNOWLEDG E 



[Januaby 1, 1898. 



North American Indian of to-day considers the position 

 of the moon's horns with the same practical interest as 

 the Scottish peasant. 



Over all this, it would appear, science shakes her head 

 doubtfully. Lunar influence may be probable, but it is not 

 proven. Some would even go further. Let us listen to a 

 few authoritative utterances on this point. 



In 1895 I find the head of the United States weather 

 service remarking that " Lunar periods [in weather] . . . 

 have all failed to get a foothold in scientific respect, though 

 much time has been put upon them, and they appear 

 theoretically probable." 



Prof. W. Morris Davis, author of one of the best recent 

 books on meteorology, says : " The control of the weather 

 by the moon has long been a favourite idea, but it has not 

 been found to bear the test of accurate comparisons of 

 weather and lunar phases, except in a very faint and 

 imperfect manner." 



Once more, Sir Robert Ball, in his " Story of the 

 Heavens," says : " Careful comparison between the state 

 of the weather and phases of the moon has quite dis- 



things : a certain definite relation to the moon's phases 

 (speaking roughly, a barometric wave to each new and 

 each full moon) appears from time to time, and persists, 

 perhaps half a year, or more. Then it may disappear 

 (from some cause or other), to reappear later on. 



The half-year closing with November, 1897, is, it so 

 happens, a very good example. In the accompanying 

 diagram the curve is that of the daily barometer at 

 Greenwich from -June to November, smoothed with 

 averages of five ; that is, each day point of the curve 

 represents the average of five daily values (»>.(/., that of the 

 3rd of Jime, the five days, one to five, and so on). 



This curve presents, it will be seen, a series of waves 

 corresponding remarkably with the moon's phases. Are 

 we prepared to affirm that so many coincidences are merely 

 fortuitous ? 



This correspondence still persists at the date of writing 

 fDecember 7th), and readers of Knowledge may be 

 interested to watch further developments. Doubtless, it 

 will be masked or obscured ere long : and it may, of 

 course, be argued that those intervening periods of irre- 



6 It IB Xt^ 30 6 IX IS ■>h 3o 5 // 17 1.3 %) J^ 10 /<5 22- 2*- V ,o li, 32. 2? J q- /T XI xy 

 Curre of Daily Barometer, Greenw-icli, June to Norember, 1897 (smoothed witb Fire-Day Arerages). 



credited the notion that any connection of the kind really 

 exists." 



Nevertheless, further study is being given, and will 

 doubtless continue to be given, to this interesting question. 

 Of recent work upon it, may be mentioned that by 

 M. Garrigou-Lagrange, described in a series of papers 

 to the Paris Academy. He attributes to the moon's 

 influence certain periodical oscillations of the pressure and 

 gradients between the Pole and the Equator observed in 

 the meridian of Paris. These are superposed on others 

 which he considers due to the sim ; and the effect is 

 different according as the moon is in a northerly or 

 southerly position. 



A simple and direct way of seeking light on the subject 

 of lunar influence is to plot a number of curves of daily 

 barometric pressure, and see whether any extensive 

 correspondence with the moon's phases can be made out. 

 Having recently done this with the Greenwich data, I 

 would invite attention to some facts which appear to me 

 to be highly suggestive. We seem to find this state of 



gularity (or, in some cases, a different kind of regularity) 

 suttice to overthrow the evidence of casual connection in 

 periods like that here considered. 



Going back as far as 1879, curves of the same type as 

 that here given, and of similar extent, will be found in 

 1883, 1881, 1889, 1893, and 18ii4. AYhy the corre- 

 spondence should come out more clearly at these dates I 

 am unable to say. Perhaps some astronomical cause can 

 be assigned. 



It is easy to see how an experience of long and regular 

 recurrences in weather like that of the years indicated 

 may have given rise to a popular conviction that the moon 

 influences weather ; and, on the other hand, the fact 

 of irregularity subsisting and alternating with regularity 

 might account for the negative results often arrived at by 

 meteorologists when they have superposed the weather 

 data for a long series of limations. 



The presumption of continuance in the type of weather 

 indicated, which the above facts appear to warrant, in 

 a measure might afford some useful help in forecasting. 



