458 



NA TURE 



[Sett. 7, 1882 



matter of satisfaction that Southport contains numerous 

 public meeting-places close to each other. The scat- 

 tered position of the Sections at the present meeting has 

 been a very serious obstacle to members wishing to hear 

 pipers in different Sections on the same day. This has 

 been especially the case in Section C, which, being half a 

 mile from most of the other Sections, seldom obtained a 

 good audience, and indeed was only filled when the 

 popular subject of the Channel Tunnel was brought before 

 the Section by Messrs. Boyd Dawkins and De Ranee. 



REPORTS 



Report of the Committee consisting 0/ Prof. Roscoe, Mr. 

 Lockyer, Prof. Dewar, Prof Lvveing, Prof Schuster, Capt. 

 Abney, and Dr. Marshall Watts, appointed at the York Meeting 

 to prepare a New Series of Wave-lengths TaWcs of the Spectra 

 op' the Elements. — This Committee report that they have lately 

 obtained an instrument for the. more exact performance of the 

 process of graphical interpolation, constructed by Messrs. Cooke 

 and Sons of York. And since this instrument has only been 

 received within the last few weeks they are not in a position to 

 make a detailed report to the Association. 



The Report of the Committee consisting op Prof. Balfour 

 Stewart, Thorpe, and Riicker, appointed at the York Meeting to 

 Retort on Methods of Calibrating Mercurial Thermometers was 

 read by Pr.jf. Riicker, Thermometer tubes are in general of 

 un qual bore hi different part-, and the indications of the instru- 

 ment- will thus be erroneous, unless these irregularities are 

 nl owed for. If a short column of mercury broken off from the 

 mam mass in the bulb and tube be measured in different p irts of 

 tie tube, its length will be greater in the narrower, and le-s in 

 the wider part'. Ry means of such measurements the correction 

 for the inequalities in the bore can be applied in two different 

 ways distinguished as methods of calibration and correction 

 respectively. In the first the length of the column of mercury 

 i • measured in various parts of the tube before the scale is etched 

 on it, and the lengths of the divisions are then so adjusted as to 

 make equal differences of scale readings correspond to equal 

 volumes. In the second the tube is in the fir^t instance furnished 

 with a uniform scale, and a table of corrections is afterwards 

 drawn up, by means of which the same end is attaine i as before. 

 In either ca.e the measure nents have to be made in some sys- 

 tematic mann. r, and a number of different methods of perform- 

 ing the ob ervations and calculations have from time to time 

 been proposed. That in u*e at the Ken Observatory ii the 

 simplest of all, while the more elaborate methods have for the 

 most part been proposed by German writers. The report con- 

 -i-ted of a minute discussion of the relative merits of the?e 

 various methods, the chief of which had been applied hy the 

 Committee to the same thermometer, so that the results could be 

 readily compared. The measurements for this purpose were 

 made in the Physical Laboratory of the Yorkshire College. The 

 methods chiefly investigated were Gay Lussac's, Hallstrom's, 

 'Ihiessen's, Marek's, Rudberg's, and Bessel's, both as modified 

 by von Oettingen, and al.-o with further modifications introduced 

 by Professors Thorpe and Riicker. As the result of a long 

 theoretical a id experimental investigation, the Committee con- 

 clude that labour is saved and equal accuracy secured by the 

 repetition of the simplest method of correction (Gay Lussac's), 

 instead of the employment of more elaborate and theoretically 

 more perfect schemes. 



Report of the Committee, consisting of Professors Odling, 

 Huntington, and Hartley, appointed for the Purpose of inves- 

 tigating, by means of Photography, the Ultra- Violet Spark- 

 Spectra emitted by Metallic Elements and their Combinations 

 under Varying Conditions. — This report was drawn up by Prof. 

 Hartley, and communicated to the Section by Prof. Huntington. 

 The object of this investigation was to give, first, a means of 

 readily identifying the metals by photographs of their line 

 spectra ; secondly, a knowledge of the alterations producible in 

 the spectra of metallic salts by the presence of various non- 

 metallic elements ; thirdly, a know ledge of the alterations caused 

 by the dilution of metallic solutions ; fourthly, a possible means 

 of performing rapid quantitative determination of metallic sub- 

 stances by the aid of photography, and obtaining permanent 

 records of the results. These objects have been more or less 

 completely attained, and the results obtained have been the sub- 

 ject of two communications to the Royal Society, which contain 

 an account of the elu;idation of the following points : — (1) The 



solution of the p ac'ical difficulty of obtaining photographs of 

 spark spectra of metallic salts from their solutions ; (2) the 

 c imparison of spectra yielded by metallic electrodes with those 

 obtained from saline solutions ; (3) the variations in the spectra 

 caused by dilution of saline solutions j (4) the sensitiveness of 

 spectrum reactions under certain conditions ; (5) the variation in 

 the spectra of metals caused by alterations in the intensity of the 

 spark employed. A comparison of the spectra of solution of 

 salts with those of metallic electrodes show that in almost all 

 cases the lines of the metals were produced from the solutions. 

 The non-metallic constituents of salts do not yield any marked 

 series of lines. The spectrum of aluminium, as obtained from 

 pure solutions, is free from a group of short or discontinuous 

 line*, which the author has shown to be due to iron. In esti- 

 mating the relative proportions of the constituents of alloys or 

 minerals, only those methods are to be recommended in which 

 solutions are used, as in this way the non-homogeneity of the 

 substance under investigation can alone be obviated. With 

 regard to the reversal of metallic lines, it is pointed out that 

 over-expo-ure suffices to produce reversal without materially 

 influencing the re-t of the spectrum ; and in order to obviate 

 this remit, it is recommended that comparative exposures should 

 be methodically employed to confirm the accuracy of observations 

 made entirely by the aid of photographic repre entations and of 

 spectra. This is especially the case where gelatine or other dry 

 plates containing organic matter are emplcyed. 



Report of the Committee on the Lunar Disturbance of Gravity, 

 by G. H. Uarwin. — shortly after the reading of the first report 

 last year at York, it was found that the instrument with which 

 he and his brother had been working, had broken down, and 

 this together with a series of unforseen circumstances, had pre- 

 vented their continuing their observations, but he still had 

 some remarks to make on the subject. From a remark made 

 by Signor de Rossi on an ob-erved connecti m between baro- 

 metric storms and the disturbance of the vertical, he had been 

 led to make some investigations on the mechanical effects caused 

 by variations of pressure acting on an elastic surface. When a 

 heavy body rests on the surface of the earth in the neighbour- 

 hood of a pendulum, the direction of the pendulum, or the 

 vertical, appears to change, a change due to two causes : first, 

 an actual change due to the attraction of the heavy body on the 

 bob of the pendulum ; and secondly, an apparent change due to 

 an actual change of level caused by the elastic yielding of the 

 surface. Sir W. Thomson had pointed out to him a very re- 

 markable relation between those two effects. If a heavy mass 

 of any form be placed on the surface of an elastic plate of great 

 thickness the deflecti .n produced en a plumb-line suspended 

 over any point of the plate by the attraction of the mass is pro- 

 portional to the slope produced in the plate at the same point by 

 the elastic yielding to the mass. Applying this to the case of 

 variation of barometric pressure, and supposing the earth to have 

 a rigidity between that of glass and copper, he found that the varia- 

 tion of slope between two places 1 500 miles apart due to a difference 

 of 5 cm. of barometric height would be o""oil7, whilst if the 

 attractio 1 of the air be included, it would amount to o"'oi46. 

 Thus, considering two cases of high pressure to right and left, 

 there would be a difference in the p isition of the plumb-line 

 relatively to the earth's surface of o"-02Q2. The amplitude of 

 oscillation at Cambridge due to lunar dis urbance of gravity, 

 as computed on the hypothesis that the earth i . rigid, was in last 

 year's report shown to be o"o2 16, whilst the instrument was 

 capable of detecting changes of o" - oi. As these quantities 

 were all of the same order of magnitude, he came to the conclu- 

 sion that it was hopeless to expect determinations of the lunar 

 effect by experiment based on the pendulum method. There 

 was another effect due to coange of barometric pressure, viz an 

 alteration in the altitude of the surface. Under the same circum- 

 stances as above the difference in height at the two places would 

 be 9 cms. The same reasoning applied to the tides would show 

 that there would appear to be a greater rise and fall of tides 

 that actually exists. This effect is in the opposite direction to 

 that due to the elastic yielding of the earth on account of the 

 tide-rising forces of the sun and moon. Near a coast line the 

 apparent change of the vertical between high and low tides 

 would be far more considerable than in the ca e of variation of 

 barometric pressure. With a difference of true height of » ater 

 between high and low tide of 40 cm., and with a tidal wave- 

 length of 3900 miles, the change in slope at a distance of I kilo- 

 metre from the water's edge would be o"'076.— Sir W. Thomson 

 pointed out a method by which the effect of the attraction of the 



