February 20, 1908J 



NA TURE 



Ml 



reply that the latest type of electrostatic separator is the 

 Blake-Morschcr, which was described in a paper read 

 before the Institution of Mining and Metallurgy by E. A. 

 Weinberg (Transactions, 1905, vol. xiv., p. 169). It is of 

 American manufacture, and can probably be obtained from 

 Kraser and Chalmers, of Erith. Earlier forms are 

 exhaustively described in a paper read by H. C. McNeill 

 before the Iron and Steel Institute (Journal, 1S99, vol. 

 Ivi., p. 18). Machinery for the extraction of iron pyrites 

 is made by the German " Humboldt Company," of Kalk. 

 near Cologne. 



The VcrhandhmgcM ilcr deulsclicii physikalisclicn 

 Gesellschaft for December, 1907, contains a coniiiiunica- 

 tion from Prof. E. Wiedemann, in which he directs atten- 

 tion to two Arabic books of the thirteenth and fourteenth 

 centuries, in the former of which the method of magnet- 

 ising a steel needle by rubbing it on a natural lodestone 

 is described, while in the latter instructions are given for 

 mounting a needle so magnetised within a wooden fish, 

 which when placed on water heads always to the north. 

 This appears to be the first known mention of the compass, 

 although the matter is treated as if it were common 

 knowledge at the time. 



\'oi.. iv. of " Conlribulions fro.m the Jefferson Physical 

 Laboratory of Harvard University " contains thirteen 

 memoirs, five of which are from the pen of Prof. B. O. 

 Peirce. Most of these memoirs are reprinted fro.ii vol. 

 .\lii. of the Proceedings of the American .Academy of .'\rts 

 and Sciences, 1906. There is one, on architectural acoustics, 

 by Prof. W. C. Sabine, reprinted from the American 

 Architect for 1900, which well deserves close attention 

 from architects in this country. It is a thoroughly scien- 

 tific attack on the problem of determining the acoustical 

 properties of a room before it is built. The author 

 describes his measurements of the absorbing powers of 

 walls, screens, furniture, and audience, and shows how 

 the constants thus determined can be used in calculating 

 the amount of reverberation to be expected in a large 

 number of cases. In each case direct measurement con- 

 firmed the result of the calculation. 



The report for 1908 of the International Com;nillpe on 

 .Atomic Weights is printed in Xo. ^^^ of ihe Proceedings 

 of the Chemical .Society. From the data here given, and 

 from those cited in previous reports, it is concluded that 

 the entire table of atomic weights is in need of revision. 

 The values assigned to potassium and sodium are too high ; 

 those given to chlorine and sulphur are too low, and 

 these constants affect the delermination of many others. 

 They depend, however, on the atomic weight of silver, 

 which is probabh", but not certainly, as low as 107-88. 

 It is well known that work upon these fundamental 

 constants is now nearing completion in several laboratories, 

 and within a few months it should be possible to enter 

 upon a satisfactory revision of the table, a task which 

 would be unsatisfactory if undertaken now. It is true that 

 the present table contains inconsistencies, but ihey are 

 s.nall in amount, and arc due to inconsistencies in the 

 original data from which the values are derived. .Since 

 issuing the last report Prof. Moissan has died, and has 

 been succeeded on the committee by M. G. Urbain. The 

 report being drawn up in November last does not deal 

 with Ihe striking result obtained by W. Marckxvald in the 

 case of tellurium, which has been published since; this 

 element has long held an abnormal position in the periodic 

 arrangement owing to its appearing to h;ive an atomic 

 weight greater than that of iodine. According to Marck- 



NO. 1999, VOL. 77] 



wald, its correct atomic weight is 12(1-85, that is, 0-12 unit 

 less than the atomic weight of iodine, so that it now falls 

 into line with the rest of the elements. 



-A SECO.SD edition of .Mr. .Arthur Whiting's " Retouch- 

 ing " has been published by Messrs. Dawbarn and Ward, 

 Lid. 



.\ si-:cOND edition of the useful " Handbook 10 the 

 Vivaria and Fresh-walcr .■\quaria " at the Horni.iian 

 Museum, Forest Hill, .S.E., has been issued by the London 

 County Council. Copies may be obtained through a book- 

 seller, or directly from Messrs. P. S. King and Son, of 

 Westminster ; the price of the catalogue is one penny. 



Messrs. Crosby Lockwood and Sox have published a 

 fifth edition of Dr. Bernard Dyer's " Fertilisers and Eeed- 

 ing Stuffs; their Properties and Uses," which contains 

 also iho full text of the Fertilisers and Feeding Stuffs .-Xct, 

 1906, the regulations and forms of the Board of Agri- 

 culture, and notes on the .Act by Mr. .\. J. David. Ihe 

 nev edition has been revised, and its price is is. net. 



OUR ASTRONOMICAL COLUMN. 



The Recent Spectrum .\\d M.agnitude of Nov.-x Persei 

 No. 2. — The results of Prof. Hartmann's more recent in- 

 vestigations of the spectrum of Nova Persei No. 2 (1901) 

 appear in No. 4232 of the Astronomische Nachriclilcii 

 (p. 113, February b). Finding that when the magnitude 

 of the star became less than 10-0 he was unable to photo- 

 graph the spectrum with the large instrument used m the 

 previous investigation. Prof. Harlniann devised a new 

 spectroscope in which the collimator objective, of 40 mm. 

 aperture and 60 cm. focal length, was made of U.V. 

 glass, and the camera objective was made of quartz, 

 having an aperture of 40 mm. and a focal length of 

 32 cm. ; quartz prisms were employed, and the distance 

 between H/8 and H5 on the plate was 4-6 mm. This 

 spectrograph was used in conjunction with the 80-cm. 

 refractor, and a good spectrum was obtained with 8j hours' 

 exposure on October 15 and 18, 1907, when the Nova's 

 magnitude was 11-4. 'Ihe main feature of this spectrum 

 is its similarity to the spectrum of the Wolf-Rayet star 

 B.D. 35°. 4001. In both spectra the brightest line is at 

 \ 4688, whilst H/3, H7, and HB are more faintly shown. 

 The fairly strong line in the spectrum of the W.-R. 

 star at A. '4618 is comparatively faint in that of the Nova,, 

 whilst the trace of a line in the latter at \ 3890 is not to 

 be found in the Wolf-Rayet spectrum. 'Ihe chief nebula 

 lines at AA 5007 and 4959 are apparently absent from both 

 spectra or are very faint. 



Determinations of the magnitude of the Nova gave the 

 following results: — 1905, November i, 11-2; 1906, 

 November 24, 11-3; and 1907, October 13, 11-44- 



The Helil'.m Line, D^, as a Dark Line in the Solar 

 Spectrum. — In No. 393 of the Observatory (p. 94. 

 Februarv) Mr. A. .A. Buss discusses the article by Father 

 Cortie, which appeared in the January number, anent the 

 presence of the dark, D,, line of helium in the solar spec- 

 trum. From our previous note (No. 1995. P- 281, 

 January 23) it will be remembered that Father Cortie 

 discussed a photograph obtained by .Mr. Nagaraja, on 

 which both the dark and the bright line of helium, D.,, 

 were supposed to be represented, and came to the con- 

 clusion that the identification was, possibly, a mistaken 

 one. Mr. Buss now advances a nuinber of arguments up- 

 holding the original view. In the first place, he points 

 out that anv arguments on this question suffer considerable 

 uncertaintv'owing to different values being given for the 

 principal lines under discussion. Thus Runge and Paschen 

 give 5S75-870 as the wave-length of the laboratory emission 

 line, whilst in Young's revised list the wave-length of the 

 chromosoheric line is given as 5876 ; that the latter, con- 

 pared with the laboratory line, suffers displacement towards 

 the red is indicated by several different observations. Mr. 

 Nagaraja's dark line lies almost exactly mid-way between 

 the two, at A 3875-930, therefore Mr. Buss considers thai 



