December 2, 1897] 



NA TURE 



117 



reason to know that instrument- makers were beginning to get 

 the wire for their resistance-coils also from Germany. He had 

 not enough experience of manganin to say whether that material 

 would stand rough service in the tropics. Prof. Ayrton said 

 the paper had raised the extremely interesting question of the 

 permanency of metals used for resistance-coils. Some time 

 ago he had immersed bare platinoid wires in running water in 

 metal tanks, and the wires all broke into short pieces. He 

 thought, at the time, this might be due to electrolysis. On 

 another occasion he had found that by raising the temperature 

 of platinoid to a dull-red heat in the air, by an electric current, 

 any acquired faults in the wire were corrected, and the original 

 resistance and flexibility were restored. Even when such 

 metals are in good condition, the resistance temperature curve 

 does not return upon itself; it encloses a loop, indicating two 

 distinct values for resistance at each temperature. He had been 

 told by Dr. Muirhead that coils intended for hot climates 

 should be enclosed in air-tight metal cases. English manu- 

 facturers were still dubious in regard to manganin. In 1892 

 he had twenty coils of this material, each of a thousand 

 ohms ; the wire was silk covered. There were 2000 \ olts 

 between the terminals. Their resistance had certainly not 

 changed by i in 1000, although there was some amount of 

 vagueness regarding the fifth figure, which might be due to 

 molecular alteration, for they were heated more than was good 

 for resistance coils. He confes.sed that this manganin had come 

 from Germany. Dr. S. P. Thompson mentioned an alloy that 

 was proposed in Germany under the name of " Constantin." 

 He would like to know whether any information could be ob- 

 tained as to the employment of cast-iron wire. It was a metal 

 that in some respects commended itself. He had observed the 

 failure of some German-silver coils, but he had generally at- 

 tributed it to rough handling. Mr. W. Watson referred to the 

 recent work done at che Reichsanstalt with regard to German- 

 silver and platinoid. It was there found that all alloys con. 

 taining zinc were liable to erratic change of resistance, and 

 were unsuitable for standard coils. Moreover, even the slight 

 amount of zinc introduced into manganin during soldering with 

 soft solder robbed that alloy of its constancy. Silver solder, 

 containing 75 per cent, of silver, should be employed for man- 

 ganin. If Prof. Ayrton's coils were soldered with soft solder, 

 that was sufficient to account for the change in the fifth figure. 

 Shellac varnish was undoubtedly the best protection for coils. 

 Absolute alcohol should be used as the solvent, and the coils 

 should afterwards be heated for some hours at 140° C. If a 

 heating-current was passed through German-silver or platinoid 

 coils immersed in water, the general result was to produce 

 brittleness. — Mr. Watson then described a thermostat which 

 he had contrived for drying the coils after applying the shellac 

 varnish. A hot-air oven contains a thermometer with a platinum 

 contact at the 140° mark, and an 8 candle-power lamp. The 

 thermometer is in circuit with a relay actuating a mercury-key 

 for the 8 candle-power lamp. The key consists of two mercury- 

 cups, and a corresponding U -piece of copper, 'inverted, one 

 limb to each cup. It is important to keep the heating-circuit 

 always made ; for this purpose a 32 candle-power lamp is per- 

 manently connected between the two cups. — Prof. Perry then 

 read a note on a question in thermo-dynamics, arising from 

 correspondence that had taken place between himself, Prof. 

 Ramsay, and Mr. Rose-Innes, with regard to a paper in the 

 Koy. Soc. Trans. Mr. Rose-Innes replied. — The President 

 proposed a vote of thanks to the authors, and the meeting 

 adjourned until December 10. 



Mathematical Society, November 11.— Prof. Elliott, 

 F. R.S., President, in the chair. — The President briefly referred 

 to the recent decease of Lieut. -Colonel J. R. Campbell, who 

 had served on the Council, and rendered signal service to the 

 Society by his liberal gift of 500/. to its funds. — The Treasurer 

 gave an abstract of his report, and, in the course of his remarks, 

 pointed out that the publications of the Society had of recent 

 years grown to such an extent as to tax its utmost resources, 

 and that, were it not for the help of the funds munificently 

 established many years ago by Lord Rayleigh, and more recently 

 iy the late LieuL -Colonel Campbell, the Society would be 

 compelled to materially restrict its sphere of activity. — The ballot 

 having been taken, the President announced that the gentlemen 

 whose names were given in Nature (October 21) had been duly 

 elected to serve on the Council for the ensuing session. — The 

 following communications were made : On an extension of the 



NO. 1466, VOL. 57] 



exponential theorem, by J. E. Campbell; the integral fp"dx, 



and allied forms in Legendre's functions, between arbitrary 

 limits, by R. Hargreaves ; on the Poncelet polygons of a 

 Lima9on, by Prof. F. Morley ; the calculus of equivalent state- 

 ments (No. 7), by H. MacColl ; the character of the general 

 integral of partial diff'erential equations, by Prof. Forsyth ; 

 note on Bessel functions, by H. M. Macdonald. 



Geological Society, November 17.— Dr. Henry Hicks, 

 F. R.S., President, in the chair. — The geology of Rotuma, 

 by J. Stanley Gardiner. The author described the relation- 

 ship of the island of Rotuma (situated in lat. 12° 30' S., long. 

 177° i' E.) to the adjoining isles. It is almost separated into 

 two parts, which are united by a narrow neck of sand. The 

 interior is composed of volcanoes, which have emitted lavas and 

 fragmental rocks. Around the volcanic rocks are stratified 

 deposits composed of sea-sand with volcanic fragments. These 

 are partly denuded, and are mantled round by coral-reef and 

 beach sand-flats. A remarkable cavern in the lava of Sol Mapii, 

 with lava-stalactites, was described ; there is a similar cavern in 

 Au Huf Huf. An account of the prevalent meteorological con- 

 ditions was also given. In an appendix by Mr. H. Woods, 

 some of the rocks were] described. They consist of olivine- 

 dolerites and basalts and associated fragmental rocks. — A 

 geological survey of the Witwatersrand and other districts in 

 the Southern Transvaal, by Dr. Frederick H. Hatch. After 

 giving an account of the physical characters of the area, the 

 author proceeded to describe the various rocks, referred to 

 (i) the Karoo system ; (2) the Cape system ; (3) the Primary or 

 Archaean system. The Archaean rocks protrude in a few places 

 through the sedimentary beds, which form the greater part of 

 the area, and consist of an igneous complex of rpcks of varied 

 composition. The Cape system is capable of division into five 

 distinct series : — 



,'Magaliesberg and Gatsrand series ; alternating quart- 



zites, shales, and lava-flows. 16,000 to 20,000 feet. 



Upper J Dolomite and cherts, thickly bedded. 6000 to 8000 feet, 



Beds. \ Black reef ; a bed of quartzite and conglomerate, 20 ] 



to 50 feet, and Klipriversberg amygdaloid ; a basic |- 



volcanic rock, 5000 to 6000 feet. J 



! Witwatersrand beds ; sandstones and conglomerate 

 (in part auriferous). 11,000 to 15,000 feet. 

 Hospital Hill series; quartzitesand ferruginous .shales. 

 8000 to 10,000 feet. 



The Karoo formation is represented by the coal-measures of 

 Vereeniging and the district south of Heidelberg, and by the 

 measures of other coal-areas. They have furnished plants which 

 Mr. Seward refers to in a note as being of Permo-Carboniferous 

 age. The age of the Cape system is doubtful. The Upper beds 

 rest unconformably on the Lower ones, and if the latter be of 

 Devonian age, as has been inferred, the former may represent 

 the Lower Carboniferous rocks. In the discussion which fol- 

 lowed the reading of the paper, the President asked whether 

 there was any fossil evidence to show the age of the beds under 

 those of Permo-Carboniferous age. The unconformities were 

 important, and might indicate that there were rocks in the area 

 older than those of Paleozoic age. Would it not be better to 

 avoid using the term " Primary" for crystalline rocks suggested 

 to be of "Archaean" age? MrT C. Dawson pointed out the 

 great lithological resemblance of the specimens on the table from 

 the Hospital Hill series to some of the fluviaiile rocks of the 

 English Wealden group ; particularly to the lower beds of the 

 Hastings sands and Ashburnham group. The Cape series, of 

 course, vvas a much older formation and had suffered change, 

 but the rocks of the Hospital Hill series appeared to have 

 suffered least. Prof. Le Neve Foster said that he considered 

 that all persons interested in gold-mining were much indebted to 

 the author for his very valuable contribution to our knowledge 

 of the geology of a great gold-bearing region. He desired in- 

 formation upon two points : (l) whether the conglomerate-beds 

 on the southern side of the synclinal are as rich in gold as those 

 which have been so largely worked on the north side ; and (2) 

 what surface-indications guide the prospector in searching for the 

 outcrop of the beds. Is their existence indicated by quartz- 

 pebbles on the surface, or by ferruginous outcrops, or what ? In 

 conclusion, he complained of the use by the author of the ex- 

 pression "mineralisation of the conglomerate" ; he wished in 

 season and out of season to protest against this use of the word 



