NATURE 



[April 24, 191. 



At the last meeting the report of the Commission 

 on Maritime Meteorology and Storm Warning Signals 

 was considered. The recommendations of the "com- 

 mission regarding day and night signals, drawn up 

 at the meeting held in London in September, 1912, 

 were adopted except for a few points, such as- the 

 night signal for a hurricane," which was found to be 

 likely to be confused with other signals already in 

 use. These recommendations have already been ■ de- 

 scribed in Nature (loc. tit.). A substantial measure 

 of international agreement in the matter of day 

 and_ night storm warning signals has thus been 

 attained. 



The Rome meeting of the committee was the third 

 which has been held since the Conference of the 

 Directors of Meteorological Observatories and Insti- 

 tutes which met at Innsbruck in 1905. In accord- 

 ance with established practice another conference of 

 directors should be held before the committee can hold 

 another meeting, and it was agreed to call together 

 such a conference for the year 1915. Holland was 

 suggested as a suitable country for the meeting. 



M. Palazzo had been at great pains to entertain his 

 visitors and to afford them opportunities of seeing 

 the geodynamical and meteorological observatories 

 near Rome. On the Tuesday the committee was enter- 

 tained at a dinner, at which the Chief Inspector of 

 Mines presided on behalf of the Minister of Agricul- 

 ture, who sent a message regretting his inability to 

 be present in person. On Wednesday the members 

 were received at the International Institute of Agricul- 

 ture by its president, the Marquis de Cappelli." The 

 whole of Thursday was devoted to an excursion which 

 had for its object the seismological observatory at 

 Rocca di Papa, with which was combined a visit to 

 the Lake of Albano and to Frascati. On Friday after- 

 noon the committee was invited to a meeting of the 

 Physical Society at Rome, where it was welcomed 

 by the president, Prof. Blascona, and subsequently 

 listened to a lecture by Prof. Bjerknes on the fields 

 of force. 



On Saturday afternoon, April 12, the military ob- 

 servatory at Bracciano was visited by motor. This 

 observatory has been recently established, and many 

 of the instruments were not yet finally installed. It 

 is fully equipped, not only for ordinary' meteorological 

 work, but also for taking aerial soundings with kites, 

 registering or pilot balloons. A pilot balloon ascent 

 was carried out in the presence of the visitors, who 

 were subsequently entertained by the commandant 

 and his officers. 



NICKEL STEELS IN CLOCK 

 CONSTRUCTION. 

 T N a pamphlet on " Les Aciers au Nickel et leurs 

 Applications a 1'Horologerie " (Paris, Gauthier- 

 Villars), M. Ch-Ed. Guillaume gives in a simple form 

 an account of the properties of nickel steels and of 

 their application to the construction of compensated 

 clocks, chronometers, torsion clocks, and even watches. 

 The well-known peculiarities of the nickel steels as 

 regards dilatation and variation of elastic modulus 

 and other properties with temperature are briefly 

 described and explained on the ground that the pre- 

 sence of nickel depresses the temperature of the allo- 

 tropic modification which occurs in iron at 890 C, 

 and at the same time changes the transformation point 

 of iron into a wide range of transformation tempera- 

 ture in the alloys. It is when they are within this 

 widened transformation range that these steels possess 

 abnormally low coefficients of expansion, &c. 



M. Guillaume's exposition of the applications of 

 these steels shows, however, that although the alloy- 

 NO. 2269, VOL. 91] 



steel known as "invar" can be produced so as to 

 have negligibly low expansion, that is not the result 

 to be desired for horological purposes. In the case of 

 clock pendulums having an invar rod, with bob and 

 suspension of other metal, the compensation principle 

 of Graham, used in the mercury pendulum, is em- 

 ployed, but the use of a nickel steel of low expansion 

 avoids the use of a liquid and makes the attainment 

 of compensation both simpler and more perfect in its 

 results. A steel of zero expansion would be less con- 

 venient. 



More striking still is the application of nickel steel 

 of a desired (low) coefficient of expansion to the 

 balance-wheels of chronometers of high accuracy. 

 Here the use of these special steels has made it pos- 

 sible to eliminate the second-order errors arising from 

 the fact that compensation effected for two definite 

 temperatures did not, with the older materials, avoid 

 serious errors at intermediate temperatures, owing to 

 the fact that the expansion curves of the two com- 

 pensating metals only crossed at two points and lay 

 widely apart at intermediate temperatures. 



The elimination of this secondary error has made it 

 *vorth while to seek other improvements in chrono- 

 meter construction, so that an almost revolutionary 

 improvement in these instruments has been brought 

 about. For watches in which a compensated balance- 

 wheel is excluded on account of cost, the use of a 

 hair-spring of a special nickel steel, to which some 

 chromium has been added in order to raise the natur- 

 ally low elastic limit, has resulted in the evolution of 

 a cheap method of producing compensated watchej. 

 In this case the abnormal manner in which the elastic 

 modulus of these steels varies with temperature has 

 been utilised. 



The similar anomalous variation of the torsion 

 modulus has also been utilised in connection with the 

 construction of clocks with torsion pendulums, and 

 has brought these clocks into the range of reasonably 

 accurate instruments for the measurement of time'. 

 Thev have the advantage of requiring very little 

 driving power, and can therefore run for four hundred 

 days on a single winding. 



M. Guillaume points out that these important 

 developments must all be regarded as resulting from 

 the study of the internal transformations of solutions 

 and of alloys and that they have resulted indirectly 

 from the study of nickel steels for purposes of metro- 

 logy. The gradual and also the transient changes of 

 dimension to which steels of the " invar " type are 

 known to be subject are fortunately too minute to 

 interfere with these applications, provided the steel 

 has been properly aged. W. Rosenhain. 



I'lRFATIONS OF THE SPECTRUM OF 

 TITANIUM IN THE ELECTRIC FURNACE. 

 T7NHANCED lines are taking a more and more 

 *-^ prominent part in the discussion of both 

 terrestrial and celestial spectra, and another valu- 

 able contribution to the subject comes from the 

 researches of Mr. A. S. King, of the Mount 

 Wilson Solar Observatory (Astrophysical Journal, 

 vol. xxxvii., No. 2, March). The investigation which 

 he has in hand deals with variations in the spectrum 

 of titanium by different temperatures of the electric 

 furnace in order, if possible, to fix the place of the 

 enhanced lines on the temperature scale. As enhanced 

 lines are in general difficult to produce in the furnace, 

 he made the attempt of forcing the furnace tempera- 

 tures up in order to make them appear in the spectra. 

 This he has very successfully accomplished, and in 

 the process he has been able to observe several re- 



