Nov. 26, 1885J 



NA TURE 



79 



shows) they are absolutely concealed by a hedge, while 

 the image of the sloping roof above appears exactly as 

 the roof itself does to the distant spectator who is nearly 

 at the same level ! Such at least must be the case, unless 

 we make the audacious supposition that the more distant 

 parts of this picture represent a flat surface, the drop- 

 scene of a theatre ! ! Let the reader try to put his hand 

 and its images in the aspects shown in Fig. 94 ; or let him 

 try, as in Fig. loi, to see by aH but direct reflection in a 

 concave mirror an object situated far beyond its rim ! 

 On the other hand there are some wierd or Rhada- 

 manthine scenes (as Figs. 134, 135, 13S) ; and a couple at 

 least (Figs. 99, 139) quite Lavaterian in their graphic 

 realisation of human imbecility. 



With such a frame-work or skeleton what could be 

 expected of the book ? Certainly not much ; and it is so 

 far to the credit of Prof. Everett that he has realised a 

 fair amount of that little. But to what class of readers 

 this book can possibly be of use, is one of those un- 

 fathomable questions which only a Mental Philosopher 

 dares to attack. There is not, so far as we have seen, 

 any rcasotiing in the book. Statements merely, and 

 illustrated by pictures ! To each paragraph, when the 

 imprint of its figure-nucleus or c/zV/w'has been exhibited, 

 we might append a slight but important variation of the 

 usual mathematical formula : — 



OUOD ERAT MONSTRANDUM ! 



The Moon, Considered as a Planet, a IVorid, and a Satel- 

 lite. By James Nasniyth, C.E., and James Carpenter, 

 F.R..'\.S., late of the Royal Observatory, Greenwich. 

 With 26 Plates and numerous Woodcuts. (London : 

 Murray, 1885.) 

 This is a third edition of a book which we have already 

 reviewed in our columns. The two previous editions 

 were issued in the quarto form, the present one is in the 

 octavo. It is well known that this work contains the 

 most exquisite illustrations of lunar phenomena extant. 

 They chiefly consist of photographs of models which, 

 when placed in the sun-light, faithfully reproduce the 

 lunar effects of light and shadow. Lovers of astronomy 

 are much indebted to Mr. Nasmyth for his brilliant idea, 

 and it is to be hoped that this re-issue in a cheaper form 

 will bring this admirable volume within the reach of many 

 who have previously been debarred from perusing it. 



LETTERS TO THE EDITOR 



[ The Editor does not hold hi/nsetf responsible/or 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 insure the appearance even 

 of communications containinginteresting and novel facts.^ 

 Weather Forecasts 

 In your number for November 2, 1882, you were kind enough 

 to insert a letter from me on the subject of " Weather Fore- 

 casts." The letter was occasioned by what appeared to me a 

 conspicuous failure in the forecast which was published in the 

 newspapers for October 24. In consequence of communications 

 which followed the publication of my letter and my own con- 

 sideration of the subject, I was led to move in the Huuse o 

 Lords for a Return of the "Storms which have visited the 

 British Islands between Januaiy I, 1874, and December 31 

 1883, and of which no warning has been issued from the Meteoro- 

 logical Office ; with a Notice of the Quarter from which each 

 Unwarned Storm has reached the Coast." This Return, which 

 "■as ordered to be printed August 7, 1884, is in some respects a 

 remarkable document. It contains a recoid of nearly 120 un- 

 warned storms, or an average of nearly 12 in each year. Large 

 as this number appears to be, I was not encouraged by the 



correspondence which I had with several experts to hope that 

 much could be done to improve the system of forecasting, and I 

 have taken no further action. 



One point, however, connected with the Return appears tc 

 me to deserve notice. 



On examining it I found to my surprise that the storm of 

 October 24 was omitted altogether. This seemed to me to be 

 strange ; but my friend, the Rev. F. Redford, a well-known 

 local meteorologist (since deceased), gave me some technical 

 explanation of the omission with which I was compelled to rest 

 content. In the interesting Blue-Book, however, entitled 

 "Principles of Forecasting by means of Weather Charts, by the 

 Hon. Ralph Abercromby," issued by the authority of the 

 Meteorological Council, I find the failure connected with the 

 great storm of October 24, 1882, duly chronicled and recognised. 

 These are Mr. Abercromby's words (p. gi) : — 



" Our last illustration will be that of a kind which fortu:~ateIy 

 rarely occurs, viz. the sudden formation of a cyclone in an unex- 

 pected position, which entirely upsets all forecasting. In Fig. 

 62 [the figure is of course here omitted] we give a chart for 

 6 p.m., October 23, 1882. In it we see the most familar fe.atures 

 of the westerly type of weather, and though the barometer was 

 falling over the Bay of Biscay and rising over Scotland, there 

 was no reason to expect that the ordinary sequence of that kind 

 of weather would be disturbed, that is to say, that west and 

 south-west winds, with rather showery weather, would prevail. 

 Accordingly the following forecasts were issued." 



Then follows the table of forecasts as given in my letter before 

 referred to, and ending with "Warnings, none issued." 



" When we come to look, however, at Fig. 63 [this figure 

 here omitted], the chart for S a.m. the following morning, we 

 find that a small well-defined cyclone had formed during the 

 night over the English Channel, which moved during the day 

 towards north-north-east, and thereby produced continuous rain 

 with complete shifts of the wind through 180° in many parts 

 of the country, so that the forecasts issued were a complete 

 failure." 



Now, observe Mr. Abercromby's practical conclusion : — 



" It may be well to consider how failures of this sort may be 

 guarded against. The answer undoubtedly is, by taking ob- 

 servations at shorter intervals than fourteen hours, as on this 

 occasion. On diis particular night a halo — a sure sign that the 

 cyclone had begun to form — was seen near London at to p.m., 

 and it is therefore certain that if the observations could have been 

 taken at 11 p m., or midnight, such a complete failure vr'ght have 

 been avoided." 



These words seem to imply that something more might be 

 done to give warning of storms. If we undertake to give these 

 warnings, we ought not to let any question of expense stand in 

 the way of making the warnings as complete and as correct as 

 possible. An unwarned storm occurring on an average once a 

 month is a serious consideration. It is not worthy of the 

 greatest maritime nation in the world to be neglectful or 

 niggardly in this matter. H. Carlisle 



Rose Castle, Carlisle, November 20 



P.S. — I forward herewith a copy of the Return of " Unwarned 

 Storms." 



Scandinavian Ice-Flows, 



Referring to the map in CroU's " Climate and Time," which 

 is leproduced in his "Climate and Cosmology" (p. 133), and 

 which traces the ice-flows of the Glacial period from the two 

 sides of the Scandinavian peninsula, it will be seen that the said 

 flow bifurcates in the North Sea, and that the bifurcation is at 

 the position of the Dogger bank. I should be glad to be in- 

 formed whether this fact has been observed or commented upon 

 in any geological work. J. D. Hooker 



Kew, November 23 



