July 28, 19 10] 



NATURE 



whole of the precipitation takes place during the summer 

 months. Heavy falls at the beginning or end of the year 

 would entirely throw the calculation out, as was shown by 

 Mr. Watt in the issue of April 14. In countries such as 

 England and Scotland, where the rain is fairly evenly dis- 

 tributed throughout the year, the centre of gravity, as 

 determined by Mr. Cook, will nearly always lie between 

 6 and 7, although its true position may be anywhere from 

 o to 12. 



If the monthly rainfalls be plotted round a disc arranged 

 like a clock-face, and then the moments be calculated 

 about rectangular axes passing through the centre of the 

 disc, a series of simple calculations gives the true position of 

 the centre of gravity. 



It is best defined by means of an angle, o, measured 

 clockwise from XII, and a distance, o, expressed as a 

 fraction of the radius. The latter is a measure of the 

 unevenness of the distribution of the rainfall. If equally 

 distributed throughout the year, a = o; if very unequally 

 distributed, a approaches i. If the angle a be divided by 

 30°, a figure is obtained corresponding to Mr. Cook's C.G. 

 But as the true position of the monthly rainfall is at the 

 middle and not the end of the month, o'5 must be deducted 

 from this figure to obtain D, the date in months corre- 

 sponding to the true centre of gravity. Mr. Cook omitted 

 to make this correction. 



In the tables below this method has been applied to two 

 stations in southern India, selected at random : — 



It will be seen that the results are not very different 

 from those obtained by Mr. Cook's method, and the differ- 

 ence is constant, at any, rate for these two stations 

 (o'aS, 0-27). But for English stations very different 

 results would be obtained. The constants have also been 

 calculated for the three imaginarv cases suggested bv Mr. 

 Watt :— 



I . 



II . 



III . 



IV . 

 V . 



VI . 



Vll . 



VIII . 



IX . 



X . 



XI . 



XII . 



Year . 

 "C.G." 

 D 



36 



o 

 0-830 



NO. 2126, VOL. 84] 



It will be seen that D and a, together with the total 

 rainfall for the year, entirely define the distribution, 

 whereas the " C.G." calculated by Mr. Cook's method 

 throws no light upon it. 



This method of specific gravities can, of course, be used 

 for other annual statistics, such as barometric pressures 

 and temperatures. In the latter case, the figures for a 

 would depend upon the zero of temperature selected, and 

 would consequently be different for the Centigrade and 

 Fahrenheit scales. It would perhaps be more satisfactory 

 to take the mean annual temperature of the station as zero. 

 The figures for D would not be affected by the choice of 

 scale. 



The applicability of the method is not confined to 

 meteorology, but may be used for any phenomenon which 

 varies with the time of the year, e.g. vital statistics or 

 railway receipts. .\. Marshall. 



Waverley Cottage, Naini Tal, India, June 14. 



Present Meteoric Displays. 



The Perseid shower appears to have come into play 

 rather earlier than usual this year, for I saw four meteors 

 presumably directed from it on the nights of July 11 to 13. 

 These meteors were of the usual streaking class, and 

 formed a radiant at about 16° -I- 50°, which agrees fairly 

 well with the correct place of radiation at the end of the 

 second week in July. This year I found meteors decidedly 

 rare at the epoch named, but the skies were not very 

 favourable, and twilight very strong. 



By the time these lines appear in print the moon will 

 only slightly interfere with observation, and a clear sky 

 will show many meteors, for at the end of July the 

 .^quarids, as well as Perseids, are generally plentiful ; and 

 there is no danger of confusing the members of the two 

 streams, since their radiants are widely distant from each 

 other. The .\quarids shoot slowly upwards in long flights 

 from a radiant low in the southern sky, while the Perseids 

 are directed in rapid courses from a radiant In the 

 N.N.E. 



On July 29—31 an observer may generally expect to see 

 at least twenty meteors per hour, and especially after 

 midnight, when the number visible usually e.xhibits a very 

 marked increase, the radiants of both the Perseids and 

 -Aquarids taking up a more favourable position for the 

 distribution of their meteors as the night advances. 



It is to be hoped that all the brighter meteors and 

 bolides will be individually recorded this year. The stars 

 of Draco, Cassiopeia, Cepheus, Andromeda, Pegasus, 

 Cygnus, and other constellations afford a ready guide for 

 the accurate registry of meteor-flights, and such data will 

 possess an enduring value as a means of furthering our 

 knowledge. VV. F. Denning. 



Pwdre Ser. 



The following letter, which I received last winter, may 

 possibly throw some light on the questions raised by Prof. 

 Hiighes in his paper on " Pwdre Ser " in Nature of 

 June 23 : — ■ 



" .illegheny, December 4, 1909. 

 " Dear Professor Sciilesinger, — 



" Referring to the falling meteor of which my husband 

 made mention at your lecture last evening, the facts are 



