November 9, 1905] 



NA TURE 



;- 



spells or periods which at times are so strongly pro- 

 nounced that the mean temperature falls, instead of 

 rising, for several days together, and these irregularities, 

 generally speaking, occur everywhere at the same time. 

 (2) That this phenomenon is apparently similarly exhibited 

 over the whole globe. (3) That probably other elements 

 than temperature, even those not generally reckoned as 

 meteorological, exhibit the same peculiarity. (4) That it 

 is not improbable that these occurrences are connected 

 with the sun's activity. In the second portion of the 

 work the author endeavours to show that there is great 

 probability that the same phenomenon really occurs in the 

 southern hemisphere, but the data available do not at 

 present allow of positive conclusions. The magnitude of 

 the work undertaken may be gauged from the fad that 

 the observations of some 3636 years have been discussed. 

 In this portion of the discussion Dr. van Rijckevorsel has 

 departed from the usual method of treating the means of 

 observations from any particular locality as a separate unit, 

 but has thrown all the observations for any year together, 

 irrespective of place or date, as he considers that this 

 method gives better data for the object in view. 



The Proceedings of the Mathematical Society of Edin- 

 burgh for the session 1904-5 open with a systematic 

 paper on the properties of the envelope of the Wallace 

 or Simson line by M. Collignon, including not only geo- 

 metrical, but also kinematical considerations. Dr. 

 Mackay also publishes a bibliographical note intended to 

 accompany Collignon 's memoir. Mr. R. F. Muirhead gives 

 new proofs of Newton's theorem on sums of powers of 

 roots, and also of Waring's expression for the sum of 

 the powers in terms of the coefficients. Mr. E. B. Ross 

 contributes a neat discussion of the degree of contact 

 between a curve and its envelope; and in a paper on polar 

 loci Mr. D. G. Taylor, in order to get rid of the confusion 

 due to multiple values of r for what seems to be 

 graphically one value of B, imagines an infinite number 

 il parallel planes one above the other slit up from o to 00 

 along the initial line and joined together so as to form 

 a kind of helical surface. Dr. Muir communicates a note 

 1 in the condensation of continuants, and Prof. Bromwich 

 gives a useful method for distinguishing the ambiguous 

 1 .ims ill the solution of spherical triangles. 



Is the Transactions of the Faraday Society (vol. i., part 

 iii.1 Mr. Sherard Cowper-Coles gives an interesting account 

 of the various processes which have been suggested for 

 increasing the rate of deposition of electrolytic copper on 

 a commercial scale. It is claimed that the centrifugal 

 process is at least ten times as rapid as any other process. 

 When the mandrel which constitutes the kathode is rotated 

 with sufficient rapidity, smooth, thick deposits of copper 

 in the form of tubes are obtained which show no trace of 

 lamination. The paper is illustrated by numerous plates, 

 in which the influence of the rate of rotation on the 

 character of the electrolytic copper is clearly evident. 



The supposition that radium is a disintegration product 

 ol uranium has received considerable support from the 

 investigations of Strutt, McCoy, and Boltwood. The 

 question whether the production of radium from a pure 

 uranium compound can be experimentally detected would, 

 however, seem to be answered in the negative by recent 

 experiments of Mr. Bertram B. Boltwood, published in the 

 Inn ill on Journal of Science, vol. XX., 1905. Observ- 

 ations on a solution containing 50 grams of uranium, which 

 extended over a period of 390 days, indicate that the 

 quantity of radium formed is less than 1.7x10-" gram. 



no. 1880, vol. 7$] 



This is less than one sixteen-hundredth of the quantity 

 which would be expected from the disintegration theory, 

 and the author concludes that one or more products of a 

 slow rate of change intervene between uranium and radium. 



The Revue ginirale des Sciences For September 15 con- 

 tains a reprint of a lecture which was delivered by Prof. 

 P. A. Guye before the Chemical Societ) oi Paris on new 

 researches on the atomic weight of nitrogen. The author 

 reviews (lie results already obtained, and concludes that 

 sufficient differences exist to render fresh determinations 

 necessary. The classic gravimetric methods are not con- 

 sidered sufficiently accurate, and a description is given of 

 new methods of determining the atomic weight of the 

 element based 011 the analysis of nitrous oxide. A spiral 

 of iron wire is heated electrically in a known weight of 

 the gas and the increase in weight found. In another 

 series of experiments, an iron spiral is similarly heated 

 in a known volume of nitrous oxide. The mean value 

 assigned for the atomic weight is 14-009. An article is 

 also contributed to the same number by MM. J. de 

 Kowalski and J. Dalemont on the teaching of applied 

 science at Fribourg University. 



A\ important paper by Mr. H. v. Stcinwehr on the 

 influence of the size of crystals ..I mercurous sulphate on 

 its relations to electromotive force (" Yorlaufige Mitteilung 

 uber den Einfiuss der Korngrosse auf das electromotorische 

 Verhalten des Merkurosulfats ") has lately been published 

 in the Zeitschrift fitr Instrumentenkunde , 1905, Heft. vn. 

 The paper deals with a subject of great interest, and 

 one which at the present time is occupying a great 

 deal of attention in all countries in connection with 

 the preparation of standard cells. The author experi- 

 mented on samples of mercurous sulphate obtained from 

 different makers, and found that they gave a differ- 

 ence of electromotive force equal to 5x10-' volts; it was 

 also found that they varied in solubility. On examining 

 them under a microscope a difference in the size of crystals 

 was observed, the smaller crystals having a higher solu- 

 l.ilily and higher F.M.F. The subject was pursued both 

 in the direction of reducing the crystals by grinding and 

 of increasing them by crystallisation, and the same result 

 was obtained, viz. the larger the crystals the lower the 

 solubility and the smaller the F..M.F. The author thinks 

 it highly probable that the size of crystals is the chief, 

 if not the only, cause of the differences observed in 

 different samples of mercurous sulphate. He further 

 , utilises the conclusions of Hulett that it is the presence 

 of basic salt that affects the result. In conclusion, he dis- 

 cusses the electrolytic method of preparation suggested by 

 Hulett and Fr. A. Wolff and recommended by Carhart for 

 standard cells, stating that it is bound to lead to the pro- 

 duction of crystals of very varying size, and the device 

 used by them of continuing the stirring after the circuit 

 is broken cannot have the desired effect, as the crystal- 

 lisation of mercurous sulphate is very slow. We await with 

 interest the further communication by the author, and 

 hope he will then have some suggestion as to an improved 

 method of preparation. 



We have received part i. of a book on leather dressing, 

 including dyeing, staining, and finishing, by Mr. M. C. 

 Lamb, director of the leather dyeing department, Herold's 

 Institute, Bermondsey. This portion, containing thirty- 

 pages, is the first of twelve monthly parts of which the 

 book will consist; it deals with sorting, splitting and 

 shaving, and is well printed and illustrated, the working 

 parts of the machines described being explained by 



