35S 



NATURE 



[Februakv 7, 189; 



layers of growth, but they were not in any sense tubules. Mr. 

 A. C. Sewaid expressed himself in agreement with the main 

 contention in reference to the occurrence of or;janic tubular 

 structures in the ooliic grains described by the ;iuihor ; hut the 

 expUnalion offered by the author with regard to the radiating [ 

 stnicture sern in certain grains he regarded as unsatisfactory, 

 their general appearance being much more sug^jestive ol 

 secondary changes in the oolitic grains, which had no direct 

 connection with the CiVtvj«^/Aj-tubules. Prof Judd said that as 

 far back as 1S62 the eminent botanist Dr. Ferdinand Cohn 

 had pointed out the important pan played by alR.-c in the for- 

 mation of the Sprudcl>lein of Carlsbad and other calcareous 

 rocks. At a laltr date Bornd, the eminent French algologist, 

 had insisted no less strongly on the work done in perforation 

 and breakinc-up calcareous fragments by other plants. The 

 speaker was inclined to regard snme ol the structures (especi- 

 ally certain of the radial ones) as due to the action of destruc- 

 tive rather than 10 constructive organisms. Mr. H. B. Woodward 

 referred to the mo<lern formation of oolite-grains in waters 

 charged with bicarbonate of lime, whether in proximity to cal- 

 careous spiings or coral reefs. He thought that the slides 

 exhtbitrd by the author did not show connection between the 

 Girvanilla and the concentiic and radiate bands of oolile- 

 grains. Hiof. -Sceley beliexcd that some of the photographs 

 exhil iied a structure which, although imperfectly preserved, 

 might be comi'ared with the sections of tiullipoies which the 

 speaker brought bef)re the British Association at Ba'hin 1S8S, 

 as evidence that some grains of oolite are of organic orijijin. 

 Mr. Kutlcy said that, with regard to the nuclear poitions of 

 some of the sections projected upon the screen, he considered 

 that the author was probably correct in ascribing an organic 

 origin 10 the interlacing tubular structures; but, so lar as the 

 peripheral portion of the oolitic bodies was concerned, he be- 

 lieved that it presented no appieciable difference from that of 

 ordinary calcareous concretions of a purely inorganic origin : 

 similar radiating and concentric structures being also met with 

 in the sih' rulites frequency present in eruptive locks. Prof. 

 .A. II. Green also spolie ; and the auth<ir replied. — On the Lias 

 Ironstone around Banbury, by Edwin A. Walford. The ferru- 

 ginous limestone of the Middle Lias of the Banbury district 

 occurs practically within a ten-mile circle around Banbury. The 

 stone (the Marlsione of the Geological Survey) is an "oolitic" 

 cyprid limestone with much molluscan and crtnoidal dOliris and 

 some quarlz-grains. The author described the lithological 

 charac ers of the rock, and their variations, as traced laterally 

 and vertically, giving a full description of its local devel •pmeni, 

 with a detailed account of the seclims in the principal ex- 

 posures. — Notes on the geology and mineral resources of 

 Anatolia (A-ia Minor), by \V. V. Wilkinson. The route 

 traversed bom northwards to southwards through the city of 

 Broussa lay thmugh a country composed of sedimentary rocks 

 (largely limestones with s^me .shales and conglomerates). In 

 the miiuniains melnmorphic rocks were met with, and also 

 ign- OU.S rocks. The principal igneous rocks noticed are 

 granites and serpentines ; in the latter chrome iron-ore occurs, 

 and IS w'liked. 



Physical Society. January 25.— Mr. Medley concluded the 

 reading III a paper, by Prof. Ayilon and himself, on tests <.f glow- 

 lamps, which was commenced at a former meeting. Wiih ihe 

 newer lamps employed in these tests, it was found thai candle- 

 power, current, and candles j er wait, all rose as Ihe lives of the 

 lamps increxsed. The author-, being surprised at this result, 

 took care 10 satisfy ihcmsrives that the effect observed was due 

 neiiher 10 change in the resistance ol their manganin potcnii- 

 omeier atrip, nor lo uncertainly of contact at the sockets of Ihe 

 lamp>. Starling again with new lamps, they lound that in all 

 cases the ln,ht given out was greater alter the lamps had been 

 glowing for jome lime than it was «hen they were new. In 

 Ihe earlier tests a con iderable falling ofl in candle-power had 

 always Lnkcn place after the lamps hail been rui ning for some 

 time. Further, while the globes of the earlier lamps were 

 alway- mucli blackened, even afiera run of a few humlred hour-, 

 and no became comparatively useless long before ihe filament 

 brole, the Edison-.Swan lamps now examined showed 

 hardly any blackening, even when Ihe filaments lasted over 

 1300 hiinr». The rise in candle-power was always accompanied 

 by a ri»e in current, which »a», however, proportionally much 

 «n alier, so that the con iimpiion of power 1 er candle was 

 actuall) leu alter the lamp had l>een running fifi) hours than it 

 was at the beginning. Among ihe cunclusions drawn by the 



no. 1319, VOL. 51] 



authors were the following : (<i) When a group of Edison-.Swan 

 lamt)s, marked loo-S, are run at 100 volts, and each lamp as 

 its fil.iment breaks is replaced by a new one, it may be ext>ected 

 that the light given out will never subsequently be as small as 

 when all the lamps were new ; {'A an Edison-Swan lamp, 

 marked 100-8, when run at 100 volts will give an average illu- 

 mination- of about 10 candles, and will absorb on an average 

 power of about 4'3 w,atts per candle, so that such a lamp must 

 be regarded as a 43-wati lamp, and not a 30-watt lamp, as is 

 frequently stated ; (<■) the maximum rise of light recorded 

 during the life of any lamp was 45 per cent. ; (li) with lamps 

 of the type examined, there is no point at which it be- 

 comes economical to discard a lamp before its fila- 

 ment actually breaks ; (e) no marked economy can be gained 

 by overrunning such lamps (i.e. by using pressures exceeding 

 100 volts). Prof. Ayrion mentioned that the improvement in 

 glow-lamps after running for some time had been attributed to 

 an improvement in the vacuum. Experiments made on new 

 and used lamps by means of an indue ion coil showed that the 

 more a lamp was u-ed the better the vacuum became, hut he 

 (Prof. Ayrion), though at first inclined 10 adopt this explana- 

 tion, had since found that though in all the lamps examined 

 the progressive improvement of the vacuum was equally 

 marked, the increase in candle-power varied between very wide 

 limits, being very consiierahle in some lamps, and hardly per- 

 ceptible in others. Prof. Kiicker asked if it made any differ- 

 ence to the life of a lamp whether it were kept running con- 

 tinuously until the fibre broke, nr were run for periods of a few 

 hours, alternating with intervals of rest. The latter case would 

 more nearly conespond with the conditions obtaining in prac- 

 tice. Prof Ayrton replied that Ihe lamps were kept running 

 during the night, and were disconnected during the day. — .\ 

 paper, by Prof, .\nderson and Mr. J. A. McClelland, on the 

 temperature of maximum density of water and its coeflicient of 

 expansion in the neighbourhood of this lemicrature, was read 

 by the secretary, Mr. Elder. The dilatometer method was 

 used, but the bulb of the instrument coniained a quantity of 

 mercury, determined by experiment, which for the range of 

 temperature concerned wa- such as to secure the constancy of 

 the remaining internal volume, occupied by water. The ob- 

 served changes were thus the real, and not the apparent changes. 

 The bulb W.1S furnished with a graduated tube of small bore, 

 bent twice at right angles, which served at the same time the 

 purpose of a ground glass stopper. (The joint was made 

 waier-tight by a little Canada balsam.) To determine the 

 coefficient of expansion of the glass, the bulb and tube were 

 filled with mercury at o" C, and healed up to .about 

 97 C, the necessary weighings being afterwards per- 

 formed. The coefficient of expansion of mercury being 

 known, the number of grams o( mercury 10 be kept in the 

 bulb during the experiments on water was calculated. The dilato- 

 meter was next filled with (thoroughly boiled) di-tillcd water 

 at about 8° C, the stopper-end of the graduatetl tube inserted, 

 and the free end dipped under menury, giving at 4' C. a 

 column of mercury whose changes of level could be observed. 

 A thermometer was placed with its bulb close to the middle 

 part of that of the dilatometer ; both being immersed in a water- 

 bath which could be cooled by the addition ol ice-cold water, 

 or healed by radiation from surrounding objects. The ther- 

 mometer used was graduated to tenths of a degree, and was 

 compared with two similarly graduated ones by dillerent 

 makers. The two latter agreed very closely with one another, 

 and one had a Kcw certificate showing no error in the read- 

 ings. Temperatures were written to the fourth decimal place, 

 but accuracy to this extent was not claimed. Three sets of 

 experiments were made, and for each a corresponding curve 

 was drawn. In the first, the water was at atmospheric pressijfc; 

 in the other two, at ijand 2 atmospheres respectively. Coi|- 

 responding to these thice pre-surcs, the temperatures of maxi- 

 mum ilen-ity found were 4°' 1 844 C, 4° 1S23 C., and 4" 1756C. 

 The value 4''l844 C, corresponding to atmospheric pres»ure, 

 is greener than tint generally receiveil. Mr. Rhodes thought 

 that suflicieiil precaution had not been taken to accurately cali- 

 brate the thermometers. He doubled whether temperatures 

 read in the manner dcscribeil could be relied upon to much less 

 than o I. lie iliil not see thnl any real advanage was gained 

 by having mercury inside ihe dibiiooir-ier to compensate for the 

 expansion of the glass. Mr. W. Watson thought that ihe 

 mercury within the vessel would cause further uncertainly by 

 tending' to produce distortion of the glass. lie pointed out 



