June 28, 1894] 



NATURE 



2 £ ■■ 



not instantly reverse, but a certain time elapses before it again 

 attains its full value, that it reverses at a later time at the centre 

 of the core than near its surface, and that the delay in reversal 

 near the centre is due to the electric currents induced in the 

 iron. The object of the present paper is to investigate these 

 effects. 



The magnet experimented upon had a diameter of 4 inches, 

 and formed a closed magnetic circuit. Through a part of its 

 length the cylinder of 4 inches diameter was formed of an iron 

 core surrounded by two concentric, closely fitting tubes. Ex- 

 ploring coils of fine copper wire were bedded in the iron 

 between the surfaces of the tubes. The currents induced in 

 these exploring coils were observed when the current in the 

 main coil of the magnet was reversed. These currents in some 

 cases last for over half a minute. 



Inferences can be drawn from these results as to the be- 

 haviour of other diameters than 4 inches. Comparing two 

 cylinders of different diameters, similar events occur, but at 

 times proportional to the squares of the diameters of the 

 cylinders. From this consideration anl the experiments, a 

 judgment is formed as to the efifecis of local currents in the 

 cores of transformers and of the armature* of dynamo machines. 



June 14. — " The Effect of Mechanical Stress and of .Magneti- 

 sation on the Physical Properties of Alloys of Iron and Nickel 

 and of Manganese Steel." By Herbert Tomlinson, F.K.S. 



Royal Meteorological Society, June 20. — Mr. R. 

 Inwards, President, in the chair. — Mr. R. H. Scott, F.R. S., 

 read a paper on fogs reported with strong winds during the 

 fifteen years 1876-90 in the British isles. Out of a total of 

 ■35 fogs, loS were associated with cyclonic, and twenty-seven 

 with anticyclonic conditions. The majority of the fogs occurred 

 with south-westerly winds and with temperatures very close to 

 the maximum for the day. — Mr. R. H. Curtis read a paper on 

 ■ line characteristic features of gales and strong winds. After 

 calling attention to the unsatisfactory state of anemometry, and 

 after describing the "bridled" anemometer at Holyhead, Mr. 

 Curtis stated that the greatest force of an individual gust which 

 he had met with was registered in December 1891, and 

 amounted to a rate of It I miles per hour, which with the old 

 factor would be equivalent to a rate of about 160 miles per 

 hour. Gusts at a rate from 90 to 100 miles per hour ha.-e 

 many times been recorded, but the usual limit for gusts may be 

 taken to equal about 80 miles per hour, which on the old scale 

 would be equivalent to about 120 miles per hour. Gales and 

 strong winds diffsr in char.-icter very much; and as the result of 

 a prolonged study of their general features, as recorded by the 

 bridled anemometer, the author has been able to group them 

 into three general classes. He then described those gales which 

 are essentially squally in character, in which the gusts consitute 

 the main feature of th:; gale. In an average gale the ordinary 

 gusts follow each other at intervals of atiout ten to twenty 

 seconds, while the extreme gusts occur at the rate of about one 

 per minute. .Another class of gales are those in which the 

 velocity of the wind is tolerafily steady. In the third class are 

 gales which appear to be made up of two series of rapidly suc- 

 ceeding squalls : the one series at a comparatively low rate of 

 velocity, the other at a much higher one, the wind-for>;e shifting 

 rapidly, and very fretpiently from orre series to the other. Mr. 

 Curtis also stated that, on looking carefully over the anemometer 

 records, he had not vinfrequently found very distincly marked a 

 prolonged pulsation in the wind-force, which recurs again and 

 again with more or less regularity, of perhaps twenty minutes 

 or half an hour in some case-, and in others at longer intervals 

 of aljout an hour, more or less. 



Physical Society, June S.— Prof. A. \V. Riicker, F.R.S., 

 President, in the chair. — Prof. Ramsay, in opening the discussion 

 onexperiments on the relations of pressure, volume, and tempera- 

 ture ^f rarefied gases, by Mr. E. C. C. Bily and himself, re- 

 capitulated the chief points in the paper. Siljesliiiin, ill 1S73, and 

 Mendelcel, in 1S75, he said, had both found that gases become 

 less compressible than Boyle's law would give, as rarefaction 

 proceeds. Amagat, in l88j, examined the subject, and con- 

 cluded it w.as impossible to make measuremenis sulTicienily 

 accurate to decide the question one way or the other. In 1S86 

 Bohr investigated the compressibility of oxygen, and found its 

 behaviour abnormal about 07 in.m. pressure. Van der Wen's 

 experiments (1889) led him to conclusions opposite to those of 

 Siljestron and Mendelcef, and those of Melander (1S92) gave 



NO. I :87, VOL. 50] 



support to Van der Wen's results. To decide the question at 



issue the authors took up the suliject, and their results confirm 

 the conclusions of Siljestrom and Mendeleef. They also prove 

 that oxygen behaves abnormally about 075 m.m., as found by 

 Bohr. Prof. I'erry said some of the terms used in the paper 

 required alteration. The word "elasticity " was employed in 

 several senses ; sometimes being used to denote " ;* v," the jiro- 

 duct of pressure and volume, whilst at others its ordinary mean- 

 ing was intended. He did not quite understand the connection 

 between p v and the thermal expansion, to which the authors 

 refer at the end of their paper. Taking Ostwald's equation for 

 gases {p -i- a) v = R/, he proceeded to show that the coefficient 

 of expansion would be constant whether pv was constant or 

 not. Dr. Burton said he had been accustoiiied to think as 

 pressure was reduced gases approached the simple or " perfect " 

 state. It was very desirable that similar experiments be made on 

 other gases, to ascertain if any had constant coefficients of expan- 

 sion. He failed to see why the internal energy should increase 

 as the pressure decreases, unless, under these conditions, energy 

 travels more by radiation than by conduction or convection. The 

 Presideni, speaking of the adhesion of gases to the surface of 

 glass, suggested experiments on the effects produced by varying 

 the ratio of the surface to the volume of the gas. On the subject 

 of distribution of energy he was inclined to agree with Dr. 

 Burton's view, rather than with the author's suggestion. One 

 would not be led by .;/;ji)r/ reasoning to expect that the internal 

 energy would increase with decrease of pressure. Prof. Ramsay, 

 in reply, said that in the experiments on oxygen at about 075 

 m.m. pressure, the greater part of the gas was sometimes found 

 in one McLeod gauge and sometimes in the other. Only after 

 standing seventy-eight hours did the quantities trapped in the 

 two gauges become equal. The only explanation they coul.l 

 think of was that the temperatures of the gauges might not have 

 been absolutely the same. Speaking of the suggested increase 

 of internal energy with decrease of pressure, he said Prof. 

 Uewar's experiments lend to show that there was little con- 

 duction through vacuous spaces. The President thought Dr. 

 Bottomley's researches had shown that radiation also falls off 

 rapidly .is the pressures become very small. — Owing to the 

 absence of Captain Abney, the exhibition of photographs of 

 flames, which had been announced, was postponed. A paper on 

 the isothermals of ether was read by .Mr. Rose Innes. Taking the 

 important linear law / == /' T - a connecting the pressure and 

 temperature of substances (liquid and gaseous) at constant 

 volume, given by Ramsay and Young {Phil. yPag. vol. xxiii. 

 p. 436), and subsequently confirmed by Amagat and Tait, 

 the author has endeavoured to express the constants / and a in 

 terms of the volume ;■. Using the results of Ramsay and 

 Young's experiments on ether, because they extended over a 

 l.irge range of volume {19 to 300), the following formulae were 

 found to give the best approximation, viz. 



V - 0-9173 Vi 



and a — 



23.300. 030 

 V- -f II 05 j»5 



Tables and curves accompany the paper showing that for 

 volumes between 6 and 303 the agreement of calculated and 

 observed values of / do not differ more than can be accounted 

 for by errors of e.'cperiment. For volumes less than 5 the 

 formula lor / gives numbers quite wrong. In searching lor a 

 physical basis for the expressions for /' and a the author puts /* 

 for V in the above formula, / being the side of a cube whose 

 volume is v. The expression for / then becomes 



P = 



p - a- 



Writing the coefficient of T in the form 



R 



P ' I - c 



, the 



author gois on to show that this form might be expected if the 

 molecules had finite dimensions, for the number 01 impacts on 

 the faces of a cube would be increased in the ratio - -, where 



<: is a quantity depending on the size of the molecules. In- 

 genious arguments suggested by Van der Waal's remark that his 

 lormula must not be pushed beyond the point where the actual 

 volume is less than eight times the volume of the m ilecule«, 

 le.id the author to infer that for ether this limiting volume is 

 somewhat above 6. Hence he would not expect his (Mr. Rose 

 Inues) formula to hold below this volume. The formula for .;, 



