2;0 



NATURE 



[January io, i9or 



Vol, iii. is similarly mainly bacteriological. Balfour 

 Stewart shows the wide distribution of the Bacillus 

 enteritidis sporogenes of Klein, MacConkey contributes 

 a paper on the differentiation and isolation of the 

 Bacillus coli and Bacillus typhosus from mixtures by 

 means of media containing bile salts. 



The printing, illustrations and general " get-up " of the 

 volumes are excellent. There is, perhaps, a tendency to 

 needless detail in some places, as, for example, in the 

 list of milk samples, but many of the papers are con- 

 tributions of real scientific value. 



EinfUhrung in die Stochiometrie. Von Joachim Bieh- 

 ringer. Pp. xviii + 498. (Brunswick : F. Vieweg und 

 Sohn, 1900.) 

 This book, which differs greatly from the ordinary text- 

 books of chemistry, has for sub-title " A study of the 

 quantitative composition of substances and the properties 

 connected therewith." So far as subject-matter is con- 

 cerned, it covers much the same ground as Part i. of 

 Ostwald's Allgemeine Chemie, the general arrangement 

 evidently being inspired by that work. The treatment, how- 

 ever, is radically different, theoretical matters, although 

 adequately outlined, being made subservient to their prac- 

 tical applications, which are illustrated by a wealth of 

 numerical examples. The arithmetical exercises are, in 

 fact, the chief feature of the book. There are altogether 

 three hundred of them, each of which is provided with a 

 fully-worked solution, the method of calculation being 

 carefully explained. The problems are well-selected, 

 unpedantic, and of real value in their application to 

 laboratory or technical work. Some are, perhaps, a little 

 far-fetched, but even these are instructive, and almost 

 always possess some human interest. For example : 

 " It is desired to convert into sparkling wine 1000 litres 

 of new wine containing 10 per cent, by volume of alcohol 

 and o'5 per cent, of unfermented sugar. How much cane- 

 sugar must be added in order that after fermentation the 

 wine may have a pressure of 5 atmospheres, the tempera- 

 ture of the cellar being 12^?" The student will be 

 especially grateful for the numerous examples of the 

 calculation of molecular weights, and of the results of 

 analysis, both gravimetric and volumetric. Altogether 

 no better guide could be desired to chemical and 

 elementary physico-chemical calculations. 



Travail des Meiaux derives du Fer. Par L. Gages, 

 Capitaine d'Artillerie. " Encyclopedie Scientifique des 

 Aide-Memoire." Pp. 202 ; 40 illustrations. (Paris : 

 Gauthier-Villars, 1900.) Price 2 fr. 50 centimes. 

 This new member of the excellent aide-memoire series 

 worthily upholds the reputation built up by its prede- 

 cessors. It gives in clear, terse language a short summary 

 of the mechanical and thermal treatment to which steel 

 is subjected to prepare it for use in the industries, and 

 the greater part of the remainder of the book is given 

 up to considerations of the theory of hardening and 

 tempering. The solution theory of the constitution of 

 steel is so firmly established in France that it is now 

 taken as an authoritative explanation of the facts ; but 

 some of the thermal and micrographic evidence is given, 

 on which the theory is based. In dealing with "steels" 

 which owe their distinctive properties to the admixture 

 with iron of elements other than carbon, Captain Gages 

 makes a new departure in classifying them in accordance 

 with a law enunciated by Sir William Roberts- Austen, 

 which sets forth the fact that elements having a lower 

 atomic volume than iron tend to harden steel, and those 

 with a higher atomic volume to soften it and make it 

 malleable. This classification of a large class of materials 

 which are comparatively new to engineers seems useful 

 and businesslike, and, indeed. Captain Gages, by bringing 

 into a brief and readable form the results of the labours 

 of Osmond, Werth and many others, has done much to 

 popularise the whole subject. 



NO. 1628, VOL. 6-^ 



LETTERS TO THE EDITOR. 



\The Editor does not hold himself responsible for opinions ex- 

 pressed by his correspondents. Neither can he undertake 

 to return, or to correspond 7vith the writers of, rejected 

 manuscripts intended for this or any other part of NATURE. 

 No notice is taken of anonymous communications.^ 



The Stability of a Swarm of Meteorites. 



Referring to iny paper in Nature of October 11, "On 

 the Stability of a Swarm of Meteorites, &c.," Mr.' R. B. Hay- 

 ward, F.R.S., has called my attention to the fact that the 

 supposition, made near the foot of the first column, that 



l = a cos bit, 7) = b sin wt, 



in other words that the relative orbit of the particle under con- 

 sideration is an ellipse of semi-axes a, b, involves an eccentricity, 



s/i - b'/d', 

 for that ellipse of amount 



n s/^/(fJL - «■■') = \/6«/( >Jn^+ l6/t - «), 



so that in strictness the relative orbit cannot be a circle unless 

 n—o. 



This point ought, perhaps, to have been brought out in my 

 paper as illustrating the rudeness of the approximation of the 

 shape of the swarm to a sphere in any actual case, a matter 

 which I touched on in speaking of the inutility of further refine- 

 ments, such as the effect of the ellipticity of the swarm's orbit 

 round the sun. However, for any probable swarm in the solar 

 system, n'/jj. must be a very small fraction, and so a>^ may be 

 very nearly equal to fjt. (if the centre of the swarm is fixed, 

 ju = &)-), while i^/a is nearly unity. 



Mr. Hayward suggests that the same difficulty may have been 

 felt by others, and that therefore the above explanation may be 

 desirable. Andrew Gray. 



The University, Glasgow, December 29, 1900. 



An Artificial Representation of a Total Solar Eclipse. 



In preparing for polarisation experiments on the solar corona, 

 it is extremely desirable to have an artificial corona as nearly as 

 possible resembling the reality for preliminary work. The only 

 device of the kind that has been used to my knowledge was the 

 arrangement described by Wright in his eclipse report, consist- 

 ing of a cardboard funnel, lined with black cloth, with a light 

 at the back. This gives a ring-shaped illuminated area radially 

 polarised. It is believed that the contrivance about to be 

 described will be found far better adapted to work of this sort, 

 for the artificial corona in this case resembles the real so closely 

 as to startle one who has actually witnessed a total solar eclipse. 

 The polarisation is radial, and is produced in the same way as 

 in the sun's surroundings, and the misty gradations of brilliancy 

 are present as well. So perfect was the representation that I 

 added several features of purely aesthetic nature to heighten the 

 effect, and finally succeeded in getting a reproduction of a solar 

 eclipse which could hardly be distinguished from the reality, 

 except that the polar streamers are straight, as drawn by Trou- 

 velot, instead of being curved, as all the recent photographs 

 show them. The curious greenish-blue colour of the sky, and 

 the peculiar pearly lustre and misty appearance are faithfully 

 reproduced. For lecture purposes an artificial eclipse of this 

 sort would be admirably adapted, and I know of no other way 

 in which an audience could be given so vivid an idea of the 

 beauty of the phenomenon. Drawings and photographs are 

 wholly inadequate in giving any notion of the actual appearance 

 of the sun's surroundings, and I feel sure that any one will feel 

 amply repaid for the small amount of trouble necessary in fitting 

 up the arrangement which I shall describe. 



A rectangular glass tank about a foot square on the front and 

 five or six inches wide, and a six candle-power incandescent 

 lamp are all that are necessary. The dimensions of the tank 

 are not of much importance, a small aquarium being admirably 

 adapted to the purpose. The tank should be nearly filled with 

 clean water, and a spoonful or two of an alcoholic solution of 

 mastic added. The mastic is at once thrown down as an 

 exceedingly fine precipitate, giving the water a milky appear- 

 ance. The wires leading to the lamp should be passed through 

 a short glass tube, and the lamp fastened to the end of the tube 

 with sealing wax, taking care to make a tight joint to prevent 



