566 



NATURE 



[October 4, 1900 



The duration of flight is estimated by repeating the letters of 

 the alphabet, minus w, at the rate of five per second, after 

 experience gained from previous practice. 



Among other papers presented to the astronomical depart- 

 ment was one by Mr. C. T. Whilwell, on " The Duration of 

 Totality of the Solar Eclipse of May 28, igcx)." A table which 

 was given illustrated very forcibly the discrepancies between 

 the calculated and observed durations at various observing 

 stations. It was pointed out that to reconcile the observations 

 and calculations by supposing that there were errors in the 

 adopted value of the moon's diameter, or in the position of the 

 observing station, involved the assumption of greater errors 

 than were probable, though each may account in part for 

 the discordance. Another suggestion, due to Mr. Crom- 

 melin, was put forward — namely, that, on account of the 

 irregularities of the moon's limb, the beginning of totality is 

 retarded by an amount corresponding to the movement of the 

 moon required to bring the lowest depressions to the edge of 

 the sun's disc after the assumed geometrical boundary has made 

 contact, while for a similar reason the end of totality would 

 be hastened. A. Fowler. 



CHEMISTR YA T THE BRITISH ASSOCIA TION. 



A LTHOUGH the president of section B, Prof. W. H. 

 ■^^ Perkin, junr., is mainly known as a specialist on poly- 

 methylene compounds, his address upon the teaching of inorganic 

 chemistry proved to be of very general interest and was en- 

 thusiastically received by a large audience. His contention that 

 the present system of examinations would be advantageously 

 superseded by an inspection of the students' laboratory notebooks 

 was favourably commented upon by Sir H. E. Roscoe and Dr. 

 H. E. Armstrong, although it was admitted that the practical 

 difficulties in the way of such a method are very considerable. 

 The presidential address was followed by the report of the com- 

 mittee on the teaching of science in elementary schools, of which 

 Dr J- H. Gladstone is chairman ; the report consisted princi- 

 pally of a discussion of the returns of the Education Department 

 in so far as they concern the teaching of elementary science. 

 The debate which ensued materially assisted the strong case 

 which was subsequently made out in favour of establishing a 

 separate section of the Association for dealing with educational 

 matters. A paper was next read by Dr. Letts and Mr. R. F. 

 Blake on some problems connected with atmospheric carbonic 

 anhydride and on a new and accurate method for determining 

 its amount, suitable for scientific expeditions ; attention was 

 drawn to the variations in the amount of atmospheric carbonic 

 anhydride, and possible explanations of the variations were con- 

 sidered. The authors determine carbonic anhydride in air by 

 absorbing it from about six litres with caustic potash solution, 

 subsequently liberating it by boiling the potash solution 

 with acid in a vacuum and measuring the volume of the 

 carbonic anhydride in a suitable eudiometer. Mr. W. Ackroyd 

 contributed papers on the distribution of chlorine in West 

 Yorkshire and on a limiting standard of acidity for 

 moorland waters. Water from the upper reaches of the West 

 Yorkshire rivers contain from 07 to i 3 parts of chlorine per 

 100,000, but as the sea or a more populous district is approached, 

 the chlorine number becomes much greater. No cases of 

 plumbism have yet been traced to the solvent action upon lead 

 pipes of water of which the acidity is less than the equivalent of 

 0'5 part of sulphuric acid per 100,000; this acidity value 

 is therefore tentatively proposed as a limiting standard for 

 potable waters of moorland origin. Dr. T. W. Hime read a 

 paper on the effects of copper on the human body, in which he 

 sought to show that the agitation against the use of articles of 

 food containing small quantities of copper salts is unjustifiable, 

 because a large number of well-known food stuffs contain 

 copper as a normal constituent and because such articles of 

 food exert no poisonous action at all. Reports were received 

 from the committees on the bibliography of spectroscopy and 

 on the preparation of a new series of wave-length tables of 

 the spectra of the elements. Prof. H. B. Dixon and Mr. F. W. 

 Rixon, in a paper on the specific heat of gases at temperatures 

 up to 400°, showed an apparatus for making such determinations 

 at constant volume in which a steel cylinder containing the gas 

 is heated and dropped into a calorimeter ; the preliminary 

 results obtained with carbonic anhydride were stated. Mr. 

 F. H. Neville communicated a report on the chemical com- 



NO. 



614, VOL. 62] 



pounds contained in alloys of which the following is a brief 

 abstract. Intermetallic coriipounds may be compared with the 

 unstable compounds of the halogens with each other and with 

 sulphur ; they often bear a great superficial resemblance to their 

 constituent elements and appear to show marked dissociation, 

 or to form systems in true equilibrium with the liquid mixture 

 of their components. 



The intermetallic compounds may be isolated from an alloy 

 (i) by filtration, (2) by volatilisation of excess of a volatile metal, 

 o'' (3) by removing the excess of metal by means of a suitable 

 solvent. Method (i) has been used by Heycock and Neville, 

 who, on filtering a partially solidified solution of gold and 

 cadmium in tin, obtained a crystalline residue having the com- 

 position AuCd ; method (2) was applied to the preparation of 

 the same compound by distilling the excess of cadmium from an 

 alloy of gold and cadmium. Lebeau prepared the compounds 

 SbNa.j, BiNag and SnNa4 by distilling the excess of sodium 

 from alloys in ammonia and nitrogen gas. Debray isolated the 

 compounds PtSn4, RhSn^and RuSn.j, and Le Chatelier obtained 

 CugSn by the action of dilute hydrochloric acid on alloys con- 

 taining excess of tin ; by methods of a similar nature Heycock 

 prepared PtAlj and Stead isolated the crystalline substances 

 Au4Pb, AusPbj, SnSb and SnjAsg. There is, in the application 

 of this method, considerable risk of the solvent attacking the 

 crystals, and Stead has found that the formation of crystals 

 having a core differing in composition from the outside consti- 

 tutes a serious drawback to the method of partial solution 

 regarded as an independent method of investigation. In the 

 systematic study of intermetallic compounds may be placed first 

 that of the chemical equilibrium of the binary system ; this is 

 generally expressed by the freezing-point curve, and has been 

 mainly investigated by Roozeboom and Le Chatelier. Next, 

 and perhaps of equal importance, is placed the microscopic ex- 

 amination of the solid alloys ; whilst thirdly, and more limited 

 in applicability, comes the determination of the difference of 

 electrical potential existing between a metal and its alloys. On 

 determining the freezing-points of a series of mixtures of 

 two metals and plotting the freezing-points as ordinates 

 against the compositions as abscissae, a freezing-point curve 

 is obtained which in its simplest form consists of two 

 branches meeting at an angle— the eutectic angle— lying at a 

 minimum point on the curve. In other cases the freezing-point 

 curve shows a maximum point, but this is not cusped and lies on 

 a gradual change of curvature ; the freezing-point curve may 

 thus consist of a series of branches connected by summits and 

 eutectic depressions. It is pretty generally recognised that the 

 eutectic alloy is merely a conglomerate and not a compound, but 

 it is a remarkable fact that the position of the eutectic point on 

 the curve often corresponds closely with some simple molecular 

 composition ; cases of this have been observed, not only with 

 alloys, but also amongst organic compounds, by Paterno and 

 AmpoUa. The branches of the curve upon which summits lie 

 are caused by the separation of compounds of definite chemical 

 composition from the solidifying magma ; the maximum points 

 lie at positions corresponding to the composition of the com- 

 pound, but Le Chatelier considers that the summit does not 

 necessarily lie exactly at the point indicated by the molecular 

 composition, owing to dissociation occurring in the liquid state. 

 This point, however, needs further investigation. The points 

 upon the freezing-point curve merely denote the temperatures at 

 which solid begins to separate from the magma, but Roozeboom 

 has shown that valuable results may be obtained by plotting, not 

 only the temperatures at which solid begins to separate, but also 

 the temperatures at which complete solidification occurs ; in 

 general, the one curve lies below the other, but they intersect 

 or become one whenever the alloy solidifies as a whole. The 

 microscopic examination of the pattern shown by the polished 

 surface of an alloy which has, if necessary, been etched or 

 heated to produce oxidation colours has been worked at princi- 

 pally by Osmond, Charpy and Stead. The existence of 

 coated crystals is made evident by this method, as in the 

 case of the bronzes rich in tin, in which Stead has 

 shown that the CugSn crystals are coated with CuSn. 

 Le Chatelier has pointed out that in these cases the 

 solid alloy is not in equilibrium, and that annealmg will, 

 in general, cause considerable change. Charpy and Stead 

 also consider that evidence of the existence of series of mixed 

 crystals is obtained by microscopic examination. Rontgen ray 

 photographs of thin sections of alloys which contain one trans- 

 parent metal and one more opaque often give good views of the 



