94 



NATURE 



[May 26, I J 



and to Phil. Mag., vol. xxx. p 58. The authors suggest that 

 in future the comparative sensitiveness of galvanometers should 

 be expressed in terms of the number of millimetre scale- 

 divisions per micro-ampere, when the observed image or "spot" 

 is one metre from the mirror. Unit angular deflection is there- 

 fore 1/2000 of a radian. Further, for the periodic time, i.e. 

 the time between two transits of the " spot " across some fixed 

 point on the scale, in the same direction, the standard should 

 be ten seconds. It is also proposed to reduce the factor of 

 sensitiveness, as regards resistance, to the common basis of one 

 ohm. The assumption is that, for a given galvanometer, the 

 deflection per micro-ampere is proportional to the 2/5 power of 

 the resistance of the windings. Tables accompanying the paper 

 give complete data for a large number of galvanometers con- 

 structed during the past ten years, and it is possible to trace 

 the improvements in sensitiveness throughout that time. The 

 most sensitive galvanometers are the oscillographs, they have 

 very short periods ; the moving parts are small, the controlling 

 fields very strong. They are designed to indicate the character 

 of rapidly-varying currents. An oscillograph, as improved by 

 Mr. Duddell, was exhibited; its period is O'oooi sec, and its 

 factor of sensitiveness, according to the authors' classification, 

 is greater than any yet obtained. A distinction is drawn as to 

 the use of the term "dead-beat." Maxwell applies it to 

 galvanometers in which the motion is " aperiodic," i.e. to those 

 in which the suspended system, before coming to rest, passes 

 only once through the position of equilibrium. This meaning is 

 retained ; it is not to be confused with " quick-moving " or 

 •'short-period." A pendulum illustrating these distinctions 

 was exhibited. As regards insulation of galvanometers and 

 shunt-boxes, the authors now apply the "guard-wire" principle 

 of Mr. W. A. Price. The instrument to be insulated is enclosed 

 in a metal case provided with a terminal, to which one end of the 

 windings is connected. The second end of the windings passes 

 out through an ebonite bush-piece. This arrangement is said 

 to nullify leakage and to prevent electro-static disturbance of 

 the suspended system. In the second section of the paper, the 

 authors calculate the limiting sensitiveness of galvanometers of 

 the "Thomson" type. The investigation is based upon Prof. 

 Schusters B.A., 1894, paper; it takes into account the period 

 of the suspended system, and the specific magnetisation of the 

 needle. Lastly, the authors discuss the relative merits of long 

 and short periods,?.^, the best "control," for galvanometers 

 intended to indicate zero points in potentiometer operations. 

 They conclude that if the control can be readily altered, and if 

 the sensitiveness can be adjusted for the test, then, for rapidity 

 of working, the " control" should be so adjusted that the sen- 

 sitiveness is approximately two or three times greater than is 

 absolutely needed for the desired accuracy. Prof. Threlfall 

 thought the authors' method of comparing galvanometers very 

 misleading. The results obtained in their comparison of the 

 oscillograph (3,310,000), and the suspended-coil galvanometer 

 (27) might be regarded as the redtictio ad absurduiu of the pro- 

 posed system. The absurdity arose from the dissimilarity of the 

 two instruments. Moreover, the proposed system ignored the 

 fact that sensitiveness may be obtained by optical as well as by 

 electro-magnetic means. Optical sensitiveness, owing to its 

 greater stability, was to be preferred to electro-magnetic sensi- 

 tiveness. The fundamental problem in the construction of 

 galvanometers is an optical one ; it is necessary to decide the 

 mass and dimensions of the suspended parts so as to ensure (i) 

 optical accuracy, and (2) electro-magnetic sensitiveness. Thus, 

 to some extent, the weight of the mirror determines the 

 thickness of the suspension. As an instance of what 

 might be done by optical methods Prof. Threlfall re- 

 ferred to work done by himself and Mr. Brearley. 

 {Phil. Mag., 1896), in which it was possible to measure to 

 I -48 X 10"^' amperes, and, with special refinements, to 

 3 X 10-" amperes. He had found that the best diameter for 

 glass mirrors was i"i cms., with a weight just under o'5 grammes. 

 These were used with a scale at 276 cms. , read by a microscope 

 to 0'04 m.m. The course of the light was : lamp, large lens, 

 small scale, mirror, eye-piece. The period was 25 sees., and 

 the resistance 50,000 ohms. Even better results could be ob- 

 tained by using mirrors of quartz or of blood-stone.- Quartz is 

 incomparably to be preferred to glass. Such figures indicated 

 what could be done by optical sensitiveness, the sensitiveness 

 that the authors ignored. It was pointed out by Prof Threlfall 

 that the controlling field for galvanometers of the " Thomson " 

 type should be straight and uniform. This was best secured by 



NO. 1 49 1, VOL. 58] 



using two magnets, one above and one below the needles. Prof. 

 Perry said the authors had not asserted that a galvanometer with 

 higher figure of merit, according to their classification, was 

 superior to another of lower figure. It must be agreed that the 

 figure they obtain is a very valuable datum for the comparison 

 of instruments designed for similar purposes ; for instance, in 

 classifying those used by Prof. Threlfall. Mr. Duddell was to 

 be congratulated on the extreme sensitiveness and small period 

 of his oscillograph. Prof. Ayrton, referring to Prof Threlfall's 

 reductio ad absiirdum, admitted that the criticism would carry 

 some conviction if the two instruments were of different kinds ; 

 if, for instance, one possessed a suspended needle and the other 

 a suspended coil. But the argument failed, because both instru- 

 ments were of the suspended-coil type. In one of them Mr. 

 Duddell had developed the advantages to l^ gained by reducing 

 the air-gap. To form an opinion of electro-magnetic improve- 

 ments in galvanometers it was necessary to reduce the results of 

 all instruments to some system of classification. There was no 

 objection, after that, to adding a good mirror, and reading by a 

 good microscope. — The President proposed votes of thanks to 

 the authors, and the meeting adjourned until May 27. 



Chemical Society, May 5.— Prof. Dewar, President, in the 

 chair. — The following papers were read: — The action of 

 hydrogen peroxide on carbohydrates in the presence of iron, by 

 C. F. Cross, E. J. Bevan, and C. Smith. The authors show 

 that the presence of iron salts is necessary to cause the oxid- 

 ation of hexoses and cane-sugar by hydrogen peroxide ; the 

 hydrogen peroxide causes a constitutional change in the hexose 

 molecule yielding products containing the group C(OH) : C(OH). 

 Dicarboxylic acids are produced as the result of secondary 

 reactions. — Note on the oxidation of certain acids in presence of 

 iron, by H. J. H. Fenton. — Properties and relationships of 

 dihydroxytartaric acid. Part ii. Metallic salts, by H. J. H. 

 Fenton. The salts of dihydroxytartaric acid with the 

 alkali metals are now described. — The affinily-constants of 

 dihydroxymaleic, dihydroxyfumaric, dihydroxytartaric and tar- 

 tronic acids, by S. Skinner. On comparing the affinity con- 

 stants of maleic, malonic, succinic, fumaric, tartaric, dihydroxy- 

 maleic, dihydroxyfumaric. dihydroxytartaric and tartronic acids, 

 it appears that the affinity constant increases on introducing 

 hydroxyl groups, and is greater for the lower members of the 

 series of dibasic acids ; the unsaturated acids have larger affinity 

 constants than their saturated isologues. — Note on the enolic 

 and ketonic forms of ethylic acetoacetate, by R. S. Morrell and 

 J. M. Crofts. — The resolution of tetrahydropapaverine into its 

 optically active components, by W. J. Pope and S. J. Peachey. 

 The constitution which Goldschmeidt has attributed to papa- 

 verine indicates that tetrahydropapaverine has the constitution 



(OMe)C.CH : C.CHg.CH.CHo.CH : CH.C(OMe) 

 II I I I II 



(OMe)C.CH : C.CH2.NH CH : CH.C(OMe) 



The authors confirm Goldschmeidt's constitution for papaverine 

 by showing that the tetrahydro-derivative is racemic, and have 

 isolated the dextro- and Isevo-isomerides by means of their salts 

 with dextrobromocamphorsulphonic acid. — Molecular weights of 

 permanganates, perchlorates, and periodates in solution, by J. 

 M. Crofts. Molecular weight determinations of permanganates, 

 perchlorates and periodates in fused Glauber's salt indicates that 

 these salts have the molecular composition M'Mn04, M'C104 

 and MTO4. — The action of chlorine on pyridine, by W. J. Sell 

 and F. W. Dootson. Amongst other products an addition 

 compound of pyridine and chlorine is formed during the action 

 of the latter upon the former. — The oxidation of paranitro- 

 toluenesulphonic acid to dinitrostilbenedisulphonic acid and to 

 paranitrobenzaldehydorthosulphonic acid, by R. Herz and W. 

 H. Bentley. — Determination of molecular weights : modification 

 of Landsberger's boiling-point method, by J. Walker and J. S. 

 Lumsden. 



Royal Microscopical Society, April 20. — Mr. E. M. 

 Nelson, President, in the chair. — Mr. Rousselet exhibited and 

 described a metal lamp chimney made by Mr. Pillischer, and 

 having two openings to carry white and tinted glass. — The Pre- 

 sident exhibited and described a new monochromatic light 

 screen trough of American invention. — Mr. H. G. Madan read a 

 paper ' ' On some organic substances of high refractivity available 

 for mounting specimens for examination under the microscope." 

 Mr. White asked if these media were suitable for histological 

 work. Mr. Madan said piperine and quinodine had been found 



