342 



NATURE 



[November 27, igig 



is effected by passing the cooled vapours through colte 

 columns and sulphuric acid drying-towers ; arsenic is 

 thus eliminated, and less than 002 per cent, of other 

 impurity (water and inert dust) retained. Details of 

 efficiency and production costs are given. 



Although salvarsan (dihydroxydiaminoarsenobenzene 

 dihydrochloride) has proved to be an effective remedy 

 for syphilis, its use in medicine is open to the objec- 

 tion that its administration involves the use of a some- 

 what elaborate technique. Various attempts have been 

 made to overcome this difficulty, the most successful 

 of which is probably the substitution of the sodium 

 iV-methylenesulphinate (neo-salvarsan) for the parent 

 compound. Medical opinion on the whole is, liowever, 

 in favour of the view that salvarsan is more powerful 

 and more certain in its action than neo-salvarsan, 

 though the latter is not without its advocates. In 

 continuation of work begun in 1907 by Prof. F. L. 

 Pyman and his collaborators, Messrs. Baxter and 

 Fargher, of the Wellcome Chemical Research Labora- 

 tories, described at the last meeting of the Chemical 

 Society a number of arsenic compounds prepared in 

 the hope that they would Ix; suitable for direct intra- 

 venous injection in simple aqueous solution. These 

 compounds are arsenobenzenes of a new type obtained 

 bv the reduction of benzodiazolearsinic acids, which in j 

 turn are produced by the action of acetic or formic 

 acid on diaminophenylarsinic acid and its homologues. 

 The new arsenobenzenes form dihvdrochlorides which 

 are soluble in water, but, though they exhibit a reduced 

 acidity as compared with salvarsan, they still prove 

 to be too acid for direct intravenous injection. These 

 experiments are, however, of considerable interest, 

 forming as thev do the nucleus of further work on the 

 replacement of amino-groups bv heterocyclic nuclei in 

 arsenobenzenes. 



One of the. latest of the many developments of the , 

 Mallet type of locomotive on American railways is a 

 simple or non-compound engine for goods and banking 

 service, built at the works of the Pennsylvania Rail- 

 road. From an article in the Engineer for Novem- 

 ber 7 we extract some particulars of this locomotive, 

 which weighs 287 tons, or just above 400 tons with 

 the tender. There are four cylinders, 305 in. by 

 32 in. ; the driving-wheels are 62 in. in diameter; the 

 boiler-pressure is 205 lb. per sq. in. ; and with a 

 maximum cut-off of 50 per cent, the maximum trac- 

 tive effort is about 135,000 lb. The size of the boiler 

 is notable; its overall length is 54 ft., including a 

 i4-5-ft. fire-box, ii-5-ft. combustion chamber, 20-ft. 

 barrel, and 8-ft. smoke-box. The barrel diameter is 

 from 825 ft. to 9 ft. Expansion movements in the 

 great length of the firebox and combustion chamber 

 are provided for by a folded connection plate forming 

 a U-shaped pocket. A mechanical stoker is used, and 

 the fire-box has a shaking grate operated by power. 

 The grate area is 112 sq. ft., the heating surface 

 6656 sq. ft., and the area of the superheater surface 

 3136 sq. ft. The short cut-off employed in the 

 Mallet engine as a substitute for compounding has 

 been criticised by writers, who consider that the i 

 svstem does not possess the advantages which it mav 

 realise when applied to the usual type of simple 

 locomotive. 



Messrs. Blackie and Son, Ltd., announce "Triumphs 

 of Invention," C. Hall. The Canibridge Universitv 

 Press will shortly publish "The Foundations of 

 Music," Dr. H. J. Watt. Mes.'srs. Hodder and 

 Stoughton are to publish ".Serial Transport," H. 

 Thomas, and "Applied .Aeronautics," G. P. Thomson. 

 Messrs. Longmans and Co. announce a new edition 

 of Prof. W. Watson's "A Text-book of Physics," 



NO. 2613, VOL. 104] 



revised by H. .Moss. Sir Lsaac Pitman and Sons. 

 Ltd., have nearly ready "Electric Lighting in the 

 Home," L. Gaster and j. S. Dow, and "Compressed- 

 Air Power," A. W. and Z. W. Daw. The Vniversity 

 of London Press, Ltd., promise "Africa and Europe" 

 (being Book iii. of the New Regional Geographies 

 -Series). It will include the British Isles and the new 

 boundaries resulting from the Peace Treaty. The 

 section relating to the British Isles will also be issued 

 separately. 



In the official announcement of the reorganisation 

 of the Board of Agriculture and Fisheries which was 

 published in last week's Nature, it should have been 

 stated that Sir A. Griffith-Boscawen has been ap- 

 pointed deputy chairman, and Sir A. Daniel Hall 

 vice-chairman, of the President's .Administrative 

 Council. 



OUR ASTRONOMICAL COLUMN. 



The Leonid .Meteoric .Shower. — Observations at 

 the middle of the present month proved that a few 

 of the meteors were visible, and that the displax was 

 prolonged beyond its usual duration. On the night of 

 Novemiier 15, in 2J hours, Mr. C. P. .Adamson, 

 watching from Wimborne, Dorset, recorded eleven 

 Leonids radiating from i5i°-|-22°. On November 19 

 he saw five Leonids near their radiant at i4g'^ + 23°. 

 The latter result corroborates an observation in 1876 

 November 19-22 at Bristol by Mr. Denning, who saw 

 five Leonids from 149° -1-22°. These figures would 

 appear to prove that there is no decided motion of 

 the radiant similar to that affecting the centre of the 

 great Perseid stream. .\ brilliant meteor was seen 

 by Mr. .\damson on November 19 last at i ih. 5m. 

 It gave a series of fia.shes near the termination of its 

 course, which was from i2o° + 2i° to 140°+ 16°, 

 traversed in two seconds. 



Two Stars with I^arge Parallaxes. — Prof. F. 

 Schlesinger gives particulars in Astr. Journ. (No. 758) 

 of two stars within 14' of each other that both have 

 large parallaxes and proper motions, and yet are 

 apparently quite independent of each other. The 

 brighter star is B.D. + 4i23°, which was found thirty 

 years ago to have an annual P.M of 1-4". The fol- 

 lowing determinations of parallax have been made : — 



Name Parallax Prob. error 



The other star is of the twelfth magnitude, and 

 was independently found by van Maanen and Wolf to 

 have an annual P.M. of 30". Its place for 1900 is 

 R.A. oh. 43m. 53s., N. decl. 4° 544'. 



The following determinations of parallax have been 

 made : — 



Parallax Prob. error 



Name 



Schlesinger ... 

 van Maanen 



027 

 0244 



O-0I2 

 0008 



There would seem to be a fair presumption that the 

 faint star is considerably nearer than the bright one, 

 and hence that their close juxtaposition in the sky is 

 accidental. 



The second star is one of the twenty stars nearest 

 to the solar system, and is evidently (like the Barnard 

 and Innes stars) in the extreme dwarf stage. It 

 would be of interest to determine its visual magni- 

 tude, which is likelv to be brighter than the photo- 

 graphic one. 



