July 6, 19 16} 



NATURE 



389 



establish with some degree of accuracy at what tem- 

 perature working has been carried out and what 

 ranges and durations of annealing have been em- 

 ployed. For this purpose he makes use of measure- 

 ments of grain-size, of a classitication of the degree 

 of "coring" or of " homgenisation " which has been 

 produced, and also of the various indications of cold 

 work or overstrain. Quite apart from its archaeo. 

 logical interest, the paper represents a valuable study 

 of the behaviour of the tin-copper alloys ranging in 

 tin-content from about 2 to 14 per cent, under 

 mechanical deformation and annealing. Less happy 

 are the author's excursions into the domain of theories 

 of plastic strain and of annealing in metals generally ; 

 they burden a lengthy paper with much additional 

 matter scarcely relevant to the subject. 



From the Scientific Materials Co., of Pittsburgh, 

 U.S.A., we have received pamphlets descriptive of the 

 Simatco apparatus for the determination of trans- 

 formation or critical points in iron, steel, or ■ altoys, 

 and of appliances for general metallographic work. 

 While it is difficult to form any real opinion on such 

 appliances without having seen them and tested them 

 in actual use, the fact that special apparatus of this 

 kind is now being placed upon the market in America 

 is significant of the widespread development and ajv 

 plication of metallography. So far as can be gathered 

 from the very clear descriptions and illustrations of 

 the apparatus given in the pamphlets, much of it 

 appears to be highly convenient and ingenious; on 

 the other hand, certain features are obviously open 

 to serious criticism. For instance, the claim is made 

 that a very simple form of well-lagged electrically- 

 wound furnace can by means of a special rheostat 

 be caused to give a uniform rate of rise and fall of 

 temperature over a wide range, and it seems most 

 unlikely that this can be realised. The form of speci- 

 men adopted is also open to objection on the ground 

 that much of the metal is further away from the 

 thermo-couple than is necessary or permissible. The 

 shape adopted arises from the use of a leading- 

 in tube of special shape — in itself very con- 

 venient — by which the wires of the thermo-couples are 

 brought into the specimen. This shape of tube, how- 

 ever, demands a very wide hole, and the effort to 

 compensate for this by a "deep immersion" results 

 in an unsatisfactory shape. Further, for indicating 

 the temperatures of the thermo-junctions, both for in- 

 verse rate and for differential curves, nothing better is 

 provided than a galvanometer with a pointer moving 

 over an ordinary scale. The entire apparatus thus 

 appears to be suitable only for work of the less deli- 

 cate or accurate kind, which, however, is of very con- 

 siderable importance in works practice. 



R' 



PROBLEMS OF CORAL REEFS. 



i ECENT work on coral reefs has established firmly 

 >- the part played by submergence in the produc- 

 tion of encircling and barrier reefs. At the same 

 time, such reefs are shown to be based on extensive 

 platforms, from which there is a further descent to 

 oceanic waters. Mr. T. W. Vaughan points out 

 (Amer. Joiirn. of Science, vol. xli., 1916, p. 134) that 

 the banks off Newfoundland, Nova Scotia, and Cape 

 Cod "would furnish proper habitats for reef-building 

 corals did they not lie outside the life-zone of such 

 organisms," while the corresponding plateaus of 

 Florida and the Central American coast support many 

 reefs. He attributes the general overftowing_ of the 

 marginal land areas in recent geological time to 

 •some diastrophic change in the earth," and is un- 

 willing to accept Glacial control as accounting for all 

 the facts. His paper is an introduction to one on the 



NO. 2436, VOL. 97] 



" Relations of Coral Reefs to Crust Movements in the 

 Fiji Islands," by E. C. Andrews, of Sydney {ibid., 

 p. 135), in which submergence is regarded as essential 

 to the formation of the Great Barrier Reef of Queens- 

 land, while the barrier reefs of the Fijis are reviewed 

 as narrow growths rising from land areas that have 

 been recently submerged. Prof. R. -A. Daly follows 

 {ibid., p. 153) with a paper on " Problems of the 

 Pacific Islands," and emphasises the presence of plat- 

 forms one or two miles to one hundred miles in width 

 as bases for the growth of reefs. He also considers 

 the case of Queensland, and the numerous sections 

 given, drawn to scale, are an important contribution 

 to geography. " The problem of the coral reef," he 

 concludes, "is, in essence, the problem of the plat- 

 form." Mr. T. W. Vaughan, in the Journal of the 

 Washington Academy of Sciences, vol. vi., 1916, p. 53, 

 describes the association of platforms and reefs in 

 the Virgin and Leeward Islands, where the platforms 

 were moulded by marine erosion during Pleistocene 

 time and then submerged, the changes of sea-level 

 thus according with Daly's theory of Glacial control. 

 Readers of Nature will remember a recent considera- 

 tion of this theorv (vol. xcvii., p. 191). 



G. A. J. C. 



SPECTRA IN ELECTRIC FIELDS. 



SHORTLY after Stark's discovery that certain spec- 

 tral lines could be split up into two or more 

 components by an electrical field, an account was given 

 in Nature (May 14, 1914, vol xciii., p. 280), under 

 the title "An Electrical Analogy of the Zeemann 

 Effect," of the experiments of the Italian physicist, 

 Lo Surdo, upon the Balmer series. It was shown by 

 Lo Surdo that the resolution of the four lines in the 

 visible spectrum followed some remarkably simple 

 laws. In a paper, dated December 19, 1915, in the 

 Rendiconti della R. Accademia dei Lincei, C. Sonaglia 

 shows that Lo Surdo's laws hold for the first line in 

 the ultra-violet, i.e. the fifth of the Balmer series. 

 The total number of components into which the line 

 can be resolved is seven, corresponding to the value of 

 the parameter n in the Balmer formula which gives the 

 line, and the number of components the vibrations of 

 which are perpendicular to the field is five, equal to 

 the number which gives the position of the line in the 

 series. 



In the same volume. No. xxiv., are two papers by 

 Rita Brunetti, which detail the results obtained on the 

 helium spectrum by Lo Surdo's method. In the third 

 subsidiary series, in which four lines have been 

 examined, it is found that the number of components 

 into which a line can be resolved is again equal to the 

 value of the parameter n giving the position of the 

 line in the series. For each line there are three un- 

 pjolarised components, while the number of polarised 

 components is equal to (n— 3). In the first subsidiary 

 series only the first member, for which n.=3, possesses 

 any polarised component; for all the lines of this 

 series the number of unpolarised components of any 

 line is (n — 2). It is interesting to notice, when British 

 science is so much under discussion, that the optical 

 apparatus used in all these researches was supplied 

 by an English firm- 



We have also received vc4. xxiv., 96 pp., of Atti 

 della iondazione scieniifica CagnoJa daUa sua Istitu- 

 zione in Pot, containing a report by Prof. G. Vanni 

 on the progress and present position of wireless tele- 

 graphy and telephony. For choice of material, 

 lucidity, and an interesting style this little volume 

 would be difficult to beat. The literature is brought 

 up to about the end of 1914. 



R. S. W. 



