198 



NATURE 



[December 7, 1911 



hook. This date has been verifu!d by Dr. A. WiUt-y, 

 F.R.S. Examination of tht chis(rl shouVd — 



Specific 



G>niposition : 



Si 



S 



.Mn 



Vt gravity. 



traces o'i2 0*003 o'^S nil 99*3% 7 '60 

 Difference being slag and oxide. 



Krdmont shear test showed 16 tons per square inch elastic 

 limit, 26 tons per square inch breaking load. Shock test 

 showed 17 kg. with 85° bend before breaking. Brinell 

 ball test showed hardness numbers 144 and 144 on opposite 

 sides. Scleroscopic hardness, 35. 'Iransverse section 

 shows the specimen to be somewhat carbonised, with 

 carbonised areas on two sides. The presence of martensite 

 and hardenite suggests the important information that the 

 chisel was quenched. The analyses in the paper probably 

 represent the only modern complete determination of the 

 composition of authentic specimens of ancient iron. The 

 percentage of phosphorus, though high, does not greatly 

 differ from modern bar iron. Sulphur is extremely low, 

 showing the employment of a very pure fuel. There is 

 very little silicon, while manganese is entirely absent, 

 which is somewhat remarkable, since nearly all iron con- 

 tains some manganese. From microscopical examination 

 and other tests it results that the specimens represent 

 wrought iron rather than steel. They somewhat resemble 

 puddled iron, and seem to have been made from rather 

 impure ore. The percentage of carbon is low, as is the 

 case for other impurities, with the exception of phosphorus. 

 Slag is present in considerable quantity in a lumpy, 

 irregular form, indicating that the material was not sub- 

 mitted to the amount of forging undergone by modern 

 wrought iron.— Prof. J. S. Townsend : The conductivity 

 of a gas between parallel plate electrodes when the current 

 approaches the maximum value. — Hon. R. J. Strutt and 

 A. Fowler : Spectroscopic investigations in connection 

 with the active modification of nitrogen. II. — Spectra of 

 elements and compounds excited by the nitrogen. The 

 chief results are as follows :— (i) The spectra generated bv 

 the nitrogen afterglow do not differ fundamentally from 

 those which can be produced by other means of excitation. 

 In many cases, however, band spectra are better displayed, 

 and the more refrangible parts of the spectrum are more 

 completely developed. The method therefore adds to our 

 resources for the production of spectra. (2) The spectra of 

 metallic substances approximate to those obtained in the 

 electric arc, or are intermediate between arc and flame 

 spectra. (3) The band spectra given by iodine, chloride of 

 tm, and mercuric iodide are very similar to those obtained 

 froni vacuum tubes. (4) The spectra exhibited by sulphur, 

 sulphuretted hydrogen, and carbon disulphide consist of 

 bands which are quite distinct from those given by sulphur 

 m a vacuum, but resemble the bands of the carbon 

 disulphide flame in air. (5) The cvanogen spectrum which 

 IS developed in the glow by cvanogen and certain other 

 compounds of carbon differs in several respects from that 

 observed m the cyanogen flame or carbon arc. Some of 

 the differences appear to be due to the production of the 

 spectrum at a relatively low pressure in the glow. A new 

 set of bands, occupying positions near the most refrangible 

 edges of the violet groups, occurs in the glow spectrum, 

 and has also been observed during the phosphorescent com- 

 bustion of cyanogen in ozone.— A. Fowler and H. Shaw : 

 Ihe less refrangible spectrum of cvanogen, and its occur- 

 rence in the carbon arc. (i) Revised wave-lengths are 

 given for the bands forming the less refrangible part of 

 the cyanogen spectrum. Numerous bands which have not 

 previously been recorded are included. (2) The heads of 

 the bands can be arranged in regular series similar to 

 those constituting the first positive band spectrum of 

 nitrogen. (3) There are considerable variations in the 

 relative intensities of the various bands, according as the 

 spectrum is obtained from the flame of the burning gas, 

 from a vacuum tube, or from the luminous glow produced 

 by the interaction of certain compounds of carbon with 

 active nitrogen. (4) The complex spectrum of the carbon 

 arc m the red and yellow is almost entirely due to 

 cvanogen. (5) The spectrum of cyanogen in the sun is not 

 of sufficient intensity to give visible indications of the red 

 and yellow bands.— Sir W. Ramsay : Note on the mon- 



NO. 2197, VOL. 88] 



atomicity of neon, krypton, and xenon. — H. M. Sudse^^ 



The adherence of flat surfaces. Jhi-* paper deaU aiiI" A- 



variou>i causes prcxluc ing adherenc*- between plan 



which have simply b' • n " wrung " together, a- 



ments are described which were carried out with ,. . .aU>j 



prepared steel gauges. It is shown that the '^cts ol 



atmospheric pressui' ' ; I'.cular atu 



opposing faces are it, and tl: 



chiefly due to the pi - ... -i a mi'""- 



the gauges. Tables arc given con; 



to pull the gauges apart when v 



present, and also comparing the readingi* obtained 



separation occurred in air and in a vacuum. Photoi- 



graphs are shown illustrating the distribution of 



film over the steel, and proving that only a sm.i 



the faces is covered by 'he liquid. The br- •'■ 



of various liquids aro "•-.tiinated from micro- 



ment of the area of cros>.-section under str. 



shown that the tensile strength of water appr 



atmospheres under these conditions. — G. D. VKct.t 



resistance to the motion of a thread of mercur}* in a 



tube. It is shown by theory and experiment that 



mercury index is to l>e moved along a glass tube of smapi 



bore with a velocity v, the difference of pressure P on (Hj 



two ends of the index is given by ^ 



P = 0038/0-1- 8/ijr/o*, 



when a is the radius of the tube and ij the coefficie: 

 viscosity of mercury. — .A. J. Berry : The distillatic 

 binary mixtures of metals in metals in vacuo. PartJ 

 Isolation of a compound of magnesium and zinc. At 

 have been made to isolate definite compounds of 

 metals, one of which, at least, is readily volatile, by 

 filiation of the excess of the more volatile constituf 

 compare Heycock and Neville (Trans. Chem. Soc., 

 1892, p. 914). In the present case it has been shown^ 

 the compound MgZn, discovered by Grube {Zeitsch. 

 Chem., 49, 1906, p. 77) can be prepared by heat 

 mixture of magnesium with an excess of zinc in 

 The excess of zinc distils off, and the residual alloy cc 

 of the intermetallic compound. It has further been 

 that this compound can be distilled without df^comp 

 — F. J. Selby : Analysis of tidal records for Brisbane! 

 the year 1908. This paper gives the results of an analj 

 of one year's tidal records for Brisbane. The analysis 

 made by the method of Sir G. H. Darwin, the tidal aba 

 devised by him being employed. The results are genera 

 in good agreement with those given by Rolin A. Harris 

 Sydney, as deduced from a year's observations of high 

 low waters. Using the constants given by the analysis,; 

 curve was run off on the Indian tide-predicting machine ' 

 the National Physical Laboraton,-, and from a comparis 

 of this with the actual records it was concluded that 

 values of the constants obtained were satisfactory. — H< 

 E. ArmstronflTi E. Frankland Armstrong:, and 

 Horton : Herbage studies. I. —Lotus corniculatus, 

 cyanophoric plant. Early in the summer last year pla 

 of Lotus corniculatus growing on the Thames banks 

 Reading were found to contain a cyanophoric glue 

 but hydrogen cyanid^^ could rarely be detected in 

 from various localities collected later in the season.^ 

 year specimens have been obtained from a great vari« 

 British localities, and these have rarely been tested 

 out obtaining positive results. The climatic condit 

 during the two seasons have been very different, so 

 the differences observed between plants grown in the 

 years would seem to be mainly due to climatic differ 

 This year plants have also been obtained from all 

 Europe — from Norway, France, Holland, Germany, 

 Servia. and Italy. In no case was cyanide found in 

 growing in Norway, where the conditions have been 

 as to favour luxuriance rather than " maturity ' 

 growth. The specimens obtained from other Eur 

 localities in a majority of cases contained cyanide, 

 glucoside is usually accompanied by a corresp 

 enzA-me, but this may occur without the glue 

 Wherever collected, the variety of L. corniculatus 

 as major (uliginosus villosus) has been found to be free 

 from £jlucoside and enzyme : it would therefore seem that 



