140 



• KNOWLEDGE ♦ 



[Dbc. 16, 1881. 



nifty not lill thr correnpondinj^ expansions of tlio hniin-cnso. Ho 

 ixpliea to yiipi-y 27, llint tlio groiitor niimbcr of doiitlm any 

 unninrricJ mm may tip (iiio to evil living, wliirh ihr nil 

 liBvinj; liomo comforts, avoid ; also, in part, to m ^;l.i U'd nilinonts. 

 The quOHlion really of interest is, however, wlirihcr the married 

 life is in it-self a prophylactic. Wo want liettir siafisties than we 

 have. The pfrcater death-mto amonjf widnwi is scoqib to support 

 the views of Uertillon and Stark ; but, miinif.stly, a cause may net 

 in their cuso which affects the statistics npart from any influencx 

 the mnrriod life may have. KiC'irABn A. rr.ofToR. 



.lUPITEK— DENXING'S COMET. 

 [122] — 1 trust that some of your re;idera will obtain observations 

 of u remarkably britfht spot situated slightly 8. of the equator of 

 Jupiter, and on the X. edge of the groat S. belt. I have observed 

 til transits of this spot since October, 1880. The tollgwing are 

 iho appro.xiniato times when it comes to the central meridian of 

 Jupiter : — 



Dec. II. H. 



17 10 40 



18 6 20 



19 U 51 



20 7 31 



21 13 1 



22 s -11 



23 1 22 



24 52 



25 5 32 



26 11 2 



27 6 43 



28 13 13 



Its period of rotation is Uh. 50m. 6s.G., or 5^ minutes less than 

 that of the ird spot, which is Oh. 5om. U5s. It is in the same 

 longitude as the middle of the red spot on Dceomber 25, and these 

 conjunctions occur nt intervals of lij days. 



The last time I saw the bright spot it was very jflain. This ^vus 

 on the night of December 7, when it was noted exactly on the 

 central meridi.in, at -Ih. 41m. and llh. S3ra. These double traneits 

 are not imfre(|iiently visible now that Jupiter is above the horizon 

 during the gnvitcr ]iart of the night. 



The new cimiet (of 18S1) was observed here on November 25, at 

 lOh., and again at midnight. It was pretty bright and large, with 

 ill-defined and apparently flashing edges. There were many faint 

 stars in the background, so that I could not be certain as to tho 

 characterof the nucleus. — Yours, &c., ^Y. F. DKNNrNc. 



Ashley-down, Bristol, Dec. 8, 1881. 



THE TRUE SPIRIT OF SCIENCE.— WATER SPHERULES. 



[123] — Let me venture to expi-ess a hope that the concluding 

 portion of tho lines from which you have taken your motto may, as 

 1 doubt not they will, tind ex])re!!sion in your magazine, — 

 " Lot Knowledge grow from more to move, 

 .\nd more of rcrcren<:e in ns dwell." 



The spirit of reverence and of humility, the S])irit of Newton, 

 Herschel, and Faraday, seems gra<iually vanishing in many high 

 scientific circles, and a spirit of sclf-snfticioncy and of arrogance 

 growing in its place. It is sad to see how frequently the student 

 of nature of to-day, in the consciousness of his acquisitions, rofcrrod 

 to yesterday, forgets the absurd littleness of his knowledge referred 

 to tho sum of all things. 



But I must not stay to pursue this thonglit, as my object in 

 writing to you was to refer to a qncstiim asked in the copy 6f 

 Knowledge sent to me, No. 3. p. tJO, where a coiTespondent asks 

 what supports or causes tho spherules of water which are often seen 

 rolling on tho surface of water. The explanation hy])Othcticnlly 

 given in your columns is, as you suspect, wholly imtenablo. Tho 

 probable cause is a rfi^erciKc n/fem^)crafi(ro between the drop and 

 the surface on which it rests. 



Aprnpnn of this, the accompanving short paper may, perhaps, 

 interest some of your readers. It was read at a meeting of the 

 Royal Dublin Society on Doe. l.">, 1877, but, so far as I know, 

 has not been published out of the innceGdings of that Society. — 

 Yours, &c., W. F. Barkftt. 



[The paper shall appear shortly. — En.J 



THE WEATHER FORECASTS. 

 [124] — Taking great interest in the daily weather forecasts ns 

 efforts made by tlie director of tho Meteorological Office and his 

 BtafE to predict tho weather in advance, I have for nboui two years 

 past been in the habit of noting briefly tho charactcristlen ef each 

 day, and comparing them nt night with the predictions as piibiislied 



in tho Daily News. I note occasional discrepancies — not nil tiie 

 inting thorn alike. Tho variations in phraseology an, 

 _ tUiy unimportant, consisting in the elision of tha 

 word "vpxy," or making iiit'in'_' variations in tho direction of the 

 wind. For the period of tl.i I-i six months my notes have lieon 

 fuller than formerly, so lh.it 1 v, ill eonline my observationa to those 

 moro particularly. 



As the main conHnsion, I am Ixmi'l i'< say that they aro about as 

 often wrong as right in the mctropolitu ■ i .irea, to which my attontion 

 has been almost exclusively confined. I'hcre is great difflcalty in 

 drawing a sharp lino between " sucr'ie.jcs " and "failures," for 

 often the tempemturo, force, and direotien of wind have proved 

 right; but tho forecast a« to rain or fair -iv. it her has frequently 

 been altogether nnsucccBSfuI, so that it is it e,-isy to state tho 

 percentage of successes. 



I have noted particularly eight occasions oi f vretsivo rain or 

 thnnder-storms happening within the months ol 1 ■ mmmcr, not 



one of which had been predicted in our daily foi i^ts, although, 



strange to say, clenr warning had sometimes 1.' . n sent from 

 America, as was noticed by a writer in tho TiMt'S oi July 9 last, 

 signing himself "Observer." The dates I now refi r ^o are — May 

 28, when we had a thunderstorm and torrents of m:iyin the after- 

 noon, certainly not predicted in tho forecast, "cloniii:,innt /air and 

 warm ;" June 5, when we had more than half an inrjto^rain — nobj 

 predicted ; Juno 25, eight hours' rain, from 10 a.m. nil Bo'el'ck ia , 

 London — not predicted, but coining after 4 week of L'forionBly bright "' 

 weather. Then came the heavy thunderstorm of .li/y 5-6, £>' whict— 

 I have already referred, bringing down tho temper;iture &att> aboutt^ 

 92'P. to 70'F. inoiie day; adi-turbaneenotforet ij by our Meteoro- 

 logical Othce, althjugh a warning from the A'. ■ I'orfc Herai ? ha«p3 

 aliealy appeared in all the newspapers. July i."^, when tho prodiG-_' 

 tion was "J'xi , but cool in the morning." Right aS regards tcraj'ora.- , 

 turo, but wrong as failing to give notice of the i .rpcntsof rain ^ucK" 

 fell before midnight. August 8, the woather jTijdicted was " clcudy " 

 or dull at first, then clearing arrain, and yite trd warm," insb .: 1 

 of which the weather in London was fiio until 4 p.m., t ■!- 

 lowed by six hours' rain, which brought d iura tho temperatmo 

 so low, that at noon ne.xt day tlie thormonieter only registoi-1 

 55° I'\ August 12, the prediction "cool, jhangeable, showon." 

 did not prepare tho sportsman for contii:aons rain from mom 

 ing till night (si.vteen hours, at least, without ceatinp) ni:l 

 not well described by "changeable " or " .■s'/owery." The dirpcti n 

 of the wind was wrongly predicted, pro\-iag East and Korth-ea-t, , 

 instead of " Westerly to North-westerly," and almost calm in Loii loii, 

 instead of "strong and S'jually." Passing over August 16, whun ' 

 wo had a fine morning, succeeded by a ^vut evening (tho reverse "f 

 tho prediction), and August. 18, a lino c /Im day, not "squally" it 

 " sliowery," ns predicted, we come to Ai,^fust 25, rjiining all d,ay nil 

 over England (see the column of repr>ris in the Daily yeics of tlie 

 following day), and certainly not fiillilfing tho predictions "lino. 

 gcnevnlly," or "gre.atly improved w,»ther" promised for tho 

 English districts Nos. 2 to .■> inclusive , fcut, on tho contrary, erery- 

 where spoiling tlie harvest prospoctt, <^d remaining bad we^herd 

 for the rest of the month. ' •• \- 



Later instances of " bad shots " n rii Sn ray noto-lxx^k, but for the.- 

 purposo of discussion tlie above niiy suffice, iind with every desiro"^ 

 to keep witliiu the bounds of fair cjriticism, I wish only to pointy I 

 attention to the facts, in tho hope ilat the causes of failure irttiy ho ' 



discovered and eventually remedii li* , ilr. K. H. Scott, in his tuticle 

 on " Forecasting the Weather," pii' iiehcd in Good irurrfj; of Jnly and 

 August, admits his "inability to t *m any estimate of the amount j i g 

 of rain which is likely to fall I'uany day," although ooniiuually, M 

 professing to take acOount of rain Ui his daily forecasts. Tin' wliol©3 W 

 subject gained inipoi-tanco from lie question asked last .\n.'iist, in ' 

 tho House of Commnns, by Mr. li,irp, M.I'., as to the fniHties fer 

 the ))riimnlgation of those woat ! ir-forcc.osts which the l'u>t?n;i~ier- 

 General might be inclined to ijvoP and tho soapons and agri- 

 cultural towns have now tho cl.Jjlco of getting iufomiation on very 

 easy terms. It remains to i^'soen whether the Iieal authorities 

 rightly estimate tho efforts •■ the Meteorological Oilieo to jn-ovido 

 this much-needed infovmation ; The end can only le attained after 

 many failures, for, as ProlV^'Sor Huxley says, "nil true science 

 begins mth empiricism." In Wiat directions, then, are we to look 

 for improvement in the ch ii*cter of tho weather-forecasts ? And 

 ■what new means can be il'iVised to give them greater accuracy? 

 These aro questions which :i))(>ear to bo well worth discussion in tho 

 pages of Knowledgf, — ^I anl/yonrs, &c., 



JonX SPItLER. 



^^ 



WHY RATS GN.\# THROUGH A WATER-PIPE. 



[125]. — The moisture of the atmosphere" is condensed on the 

 outside- of the pipe dming a change of temperature (the cold 

 water in the i>i|ie and the moist warm atmosphei-e), which induces 



^Continued on pagt 143. 



