November 2, 1899] 



NA TURH 



23 



n the south-western coast of Persia, where the thermometer 

 has been known not to fall lower than ioo% night or day, for 



orty consecutive days during July and August, and often to 

 reach 128° in the afternoon. Among the highest shade temper- 

 atures we may mention one at night during the Italian occupa- 

 tion of Massowah, when the thermometer is said to have 

 recorded 122°. Temperatures above 120° are occasionally met 

 with in India; I2i°'5 was recorded at Dera-Ishmail-Khan (lat. 

 32° N. ) in 1882, and 126° o at Bhag (lat. 29° N.) in 1859. At 

 Wilcannia on the Darling River, New South Wales, shade 

 temperatures varying from 107° to 129° were recorded on each 

 day from January i to 24 in 1896. Among the low temper- 

 atures (in addition to the extremes mentioned above) we may 

 quote -63°'l at Poplar River, North America, in January 

 1885. During the intense frost in Scotland on December 4, 

 1879, -16° was reported from Kelso and -23° from Black- 

 adder, in Berwickshire. The extremes in or near London for 

 104 years were 97°*i in July 1881, and 4° in December 1796 

 and January 1 84 1. 



The Journal of the Royal Microscopical Society for October 

 contains a short paper (with plate) by Mr. James Yate Johnson 

 on some sponges belonging to the Clionidge obtained at Madeira, 

 in which three new genera are established, named Acca, Nisella 

 and Scantilla. In the section on microscopy is a description 

 of an old compass microscope taken from a German work on 

 the microscopejiby Martin Frobenius Ledermiiller (1763), called 

 Russwurm's " Universal Microscope," which appears to have 

 been a combination of compass and tube microscope in an 

 unusual number of forms ; also a description of " Adams' Com- 

 pendious Pocket Microscope" (1771), which more nearly 

 conforms to the microscopes of the present day than any of 

 those which preceded it. In the section on technique several 

 new pigments are described, also two new methods for orienting 

 small objects. 



BoUettino dell a Societa Sismologica Italiana, vol, v., Nos. 

 z, 3. 1899-1900. — Vertical component microseismograph, de- 

 scription and results, by G. Vicentini and G. Pacher. A 

 reprint of a paper already noticed in Nature —Supplementary 

 considerations with regard to the Umbria-Marches earthquake 

 of December 18, 1897, by A. Issel.— The earthquake in the 

 Parma-Reggio district of the Appennines during the night of 

 March 4-5, 1898, by C. Agamennone. The shock was felt 

 over an acre of about 70,000 sq. km., and was also recorded by 

 horizontal pendulums at Strassburg and Shide ; the velocity of 

 the earth-waves will be considered in another paper. — The 

 Hereford earthquake of December 17, 1896, by C. Davison. 

 A summary (in English) of the writer's report on this earth- 

 quake. — Notices of the earthquakes recorded in Italy, February 

 5 to April 23, 1898 ; the most important being those of Asia 

 Minor on February 5, Cividade (Udine) on February 20 and 

 April 12, Reggio and Parma on March 4, Ferrara on March 9, 

 and distant earthquakes on February 18 and April 15 and 23. 



SOCIETIES AND ACADEMIES. 

 Physical Society, October 27.— Prof. W. E. Ayrton, 

 F.R.S., Vice-President, in the chair.— Dr. S. W. Richardson 

 read a paper on the magnetic properties of the alloys of iron 

 and aluminium. Observations were made upon four alloys con- 

 taining respectively 3-64, 5-44, 9-89 and 18-47 per cent, of 

 aluminium. The alloys were used in the form of anchor rings, 

 and were wound with primary and secondary coils separated by 

 asbestos paper. The temperatures used ranged from - 83° C. 

 to 900" C. The low temperatures were produced by the rapid 

 evaporation of ether surrounded either by ice and salt or by 

 carbon dioxide snow. The high temperatures were obtained 

 either electrically or by gas muffles. In both cases the actual 

 temperatures were deduced from the resistance of the secondary, 

 which was made of platinum wire and wound next the metal. 

 The author employed Maxwell's null method of measuring 

 mutual induction, increasing the sensitiveness by the intro- 

 duction of a secohmmeter making about three revolutions per 

 second. In order to test the accuracy of the method some of 

 the experiments were repealed with a ballistic galvanometer in 

 the ordinary way, and the agreement obtained between the 

 results in the two cases was well within the limits of experi- 

 mental error. The chief conclusions to be drawn from the 

 experiments may be summed up as follows: (l) The alloys 

 behave magnetically as though they consisted of two distinct 



NO. 1566, VOL. 61] 



media superposed. (2) The general roundness of the curves 

 and their lack of abruptnti^s near the critical point seems to 

 indicate that the alloys are heterogeneous in structure. (3) The 

 permeability decreases with rise of temperature near the critical 

 point until a minimum value is reached, when further rise of 

 temperature produces very slight diminution, if any, in the 

 permeability. (4) The experiments suggest that the maximum 

 value of the permeability for an alloy containing lo per cent, of 

 aluminium is reached at about -90° C. (5) An alloy contain- 

 ing 1 8 '47 per cent, of aluminium has a critical point at about 

 25° C. , and gives no indication of temperature hysteresis. This 

 alloy probably has a maximum permeability much below - 90° C. 

 The author has found that at high temperatures there is a second 

 maximum on the induction curve. This maximum becomes less 

 and less noticeable as the field is increased. — The Secretary read 

 a note from Prof. Barrett on the electric and magnetic properties 

 of aluminium and other steels. The first part of the note dealt 

 with the electrical conductivity of various alloys, and discussed 

 the effect of composition and annealing upon the value of the 

 conductivity. The second part of the note referred to magnetic 

 effects. The most remarkable efifect produced by aluminium on 

 iron is the reduction of the hysteresis loss. The perme- 

 ability of nickel steels is shown to be very much influenced 

 by annealing. It is found that the addition of a small 

 quantity of tungsten to iron hardly affects the maximum 

 induction, yet increases the retentivity and coercive force. The 

 experiments show that the best steel for making permanent 

 magnets is one containing 74 per cent, of tungsten. The 

 magnetometric method was employed throughout. Prof, S. P. 

 Thompson drew attention to the wide range of temperature over 

 which the author had conducted his experiments, and also to 

 the small number of alloys used. He said a very much finer 

 connection between the properties could be obtained from the 

 examination of more alloys, and expressed his interest in the 

 existence of the second maximum on the induction icurve. He 

 would like to know how the percentage composition of the 

 alloys had been determined. Turning to Prof. Barrett's note. 

 Prof. Thompson referred to the difference in the breadths of 

 the hysteresis curves for aluminium and chromium alloys. Mr. 

 Appleyard asked for information upon the permanence of the 

 curves. Dr. Richardson, in replying, said the compositions were 

 determined by analyses made after the experiments had been 

 performed. It was proposed to carry on the research upon a 

 series of aluminium alloys which he had obtained. The Chair- 

 man expressed his special interest in the agreement which the 

 author had obtained l^etween the ballistic method and the null 

 method of Maxwell increased in sensitiveness by the secohm- 

 meter. — Mr. Addenbrooke exhibited a model illustrating a 

 number of the actions in the flow of an electric current. The 

 model consisted of a spiral of steel wire in the form of a closed 

 circuit. Inside the spiral was placed a wire which was .sup- 

 posed to be carrying the current, and which directed the motion 

 of the spiral. A rotational movement given to one part of the 

 spiral was transmitted by the wire, and produced a rotational 

 movement at another part of the spiral. The resiliency of the 

 spring represents capacity, and the torque electromotive force. 

 Self-induction can be represented by weighting the spring. Prof. 

 Everett expressed his interest in the way that the correspondence 

 between the propagation and rotation agreed with that between 

 the direction of a current and the direction of the magnetic 

 force. Prof. S. P. Thompson agreed that many analogies 

 could be worked out by the model, but gave one or two 

 examples to show that erroneous conclusions might be 

 drawn by pushing the analogy too far. — Mr. W. Watson re- 

 peated some experiments with the Wehnelt interrupter devised 

 by Prof. Lecher. The experiments showed in a clear and 

 striking manner the fact that subsequent sparks tend to pass 

 through the portion of air heated by the first one. In the first 

 experiments motion of the heated air was caused by differences 

 in density, and in the later experiments by allowing the sparks 

 to take Iplace in a strong electromagnetic field. The con- 

 tinuous rotation of the spark in a given field proved the uni- 

 directional nature of the discharge. In reply to Mr. Blakesley, 

 Mr. Watson said he used the word " ionised " in his explan- 

 ations to express simply the fact that the air had been rendered 

 a conductor by the passage of the spark. The Chairman 

 referred to one of the first experiments performed. In this 

 experiment the electrodes consisted of two copper wires in a 

 vertical plane, slightly inclined to one another and nearest 

 together at their lowest points. On switching on the current 



