142 



NATURE 



[June ii, 1891 



Physical Society, May 22.— Prof. W. E. Ayrton, F.R.S., 

 President, in the chair.— Mr. C. J. Woodward exhibited Dr. 

 Schobben's form of lantern stereoscope. This instrument con- 

 sists of a dou lie lantern, by which the two pictures of a stereo- 

 scopic slide are projected on a screen. The two pictures are 

 coloured complementary tints by placing pieces of red and green 

 glass in front of the lenses, and each observer views the over- 

 lapping images through spectacles, the eye-glasses of which are 

 also coloured red and green. The stereoscopic effect is very 

 striking. Mr. Boys stated that he had tried to obtain a similar 

 result with the aid of polarized light, by viewing two polarized 

 images through Nicol prisms. No effect was obtained, owing 

 to elliptic polarization produced by the screen, but he thought 

 that if a dead-gold screen had been used instead of an ordinary 

 one, the effect might have been observed. — Prof. Perry, F.R.S., 

 showed a new form of steam-engine indicator. A galvanometer 

 mirror is fixed excentrically to a steel disk, forming one side of a 

 chamber communicating with the cylinder. The pressure of the 

 steam bulges out the disk, and causes the mirror to deflect a ray of 

 light thrown on it in the ordinary way. A rotation of the mirror 

 at right angles to the former is imparted by the movement of the 

 piston-rod. The ray of light traces out the diagram on a screen 

 suitably placed, and the complete figure is continuously visible, 

 owing to the persistence of impressions. This indicator pos- 

 sesses advantages over other forms, in being free from errors due 

 to friction or oscillations of the springs, and the alteration of 

 their elasticity due to temperature changes. The errors of 

 ordinary indicators are considerable at high speeds, owing to 

 the ripples introduced into the indicator diagram. If the natural 

 period of the springs is one-twentieth of the time of a revolu- 

 tion, the diagram is fairly free from ripples, but if it is as much 

 as one-tenth, no amount of friction in the indicator will prevent 

 ripples forming. In the new indicator, the natural period of 

 the disk can be made very short. The steel disks are easily 

 removable, and can be proportioned to suit different pres-sures and 

 speeds. For experimental and teaching purposes it is advant- 

 ageous to see at once the alterations in the diagram caused by 

 changes of load, pressure, &c. Several diagrams were exhibited 

 to the meeting. In reply to Prof. Carus Wilson, Prof. Perry 

 stated that the deflection was proportional to the pressure in the 

 cylinder within the limits any particular disk was intended for. 

 Mr. Addenbrooke thou<iht the instrument an important im- 

 provement on its predecessors, and considered it would prove of 

 great service to electrical engineers. Mr. Swinburne said a 

 peculiar merit of the indicator was that it could be permanently 

 attached to an engine like an ordinary pressure-gauge. He 

 suggested the use of a small telescope instead of the ray method. 

 The President thought that the instrument could be modified so 

 as to be useful for analyzing the shape of the curves representing 

 alternating currents. — On Blakesley's method of measuring 

 power in transformers, by Prof. Perry, F.R. S. This paper 

 refers to the supposed error in Mr, Blakesley's formula due to 

 the fact that transformers show magnetic leakage. The proofs 

 of the formula hitherto given have been obtained by treating the 

 equations in the manner first adopted by Dr. Hopkinson. On 

 this system the reactions of the primary and secondary currents 

 are represented by the equations — 



.o'N 



.^N 



V = RiCi-fP^\ o = R2C2-fS- 



dt lit 



where P and N are respectively the turns on the primary and 

 secondary coils, and N is the magnetic flux between the coils. 

 Here it is assumed that there is no magnetic leakage, and the 

 author thinks that on this account the method is inferior to the 

 original method of Maxwell, in which the induced electromotive 

 forces are expressed in terms of coefficients of self and mutual 

 induction. On the assumption that there are no eddy currents. 

 Maxwell's equations are — 



V = RiCi + LiCi + MCg ; o = R^Cg + MCj + LjCa, 

 in which although Lj, M, and L^ may not be constant, it may be 

 assumed that they are respectively proportional to P'-, PS, and 

 S^, if there is no magnetic leakage ; and if the amount of mag- 

 netic leakage bears a constant proportion to the whole flux, the 

 three quantities may still be assumed proportional to each other, 

 although M2 is less than LjLa. From these equations we obtain 



VC = RjCi^ 



^2p C1C2 



+ kL2 



M\ 



CjCi 



Hopkinson puts down the last term as zero, but owing to the 

 very rapid rate at which Cj changes, the last term is very im- 



NO. II 28, VOL. 44] 



p ort ant , even though M may be but a small percentage less than 

 VLiL,. On integrating this equation, the first two terms on the 

 right-hand side yield Blakesley's formula, and the last term 

 vanishes in the integral, because, however great the magnetic 

 leakage may be, 



(c^C^dt = o, 



when taken over a period because the functions are periodic. 

 Mr. Blakesley's formula thus appears to hold, whatever the mag- 

 netic leakage. The paper contains several tables of calculations 

 showing the effect of magnetic leakage on thevalue of the terms in 

 the equation. Mr. Blakesley said he doubted the correctness of 

 the assumption that the value of M was the same in the two equa- 

 tions, and thought that the result arri\ed at must be incorrect. 

 Dr. Sumpner did not doubt that if the coefficients could be con- 

 sidered constant, the formula was true whatever the leakage, 

 but he did not consider that the action of transformers justified 

 such an assumption. If the formula were true, it would also 

 hold if there were eddy currents, as these would merely produce 

 the effect of additional secondary coils. He had analyzed 

 Blakesley's method by using a modification of the Hopkinson 

 equations, and had shown that the power as estimated by 

 Blakesley's formula had to be lessened by the fraction repre- 

 sented by the expression 



JxA^I 



Ksdtdt 



e + \Kfksdt 



where Kj, and A., are the instantaneous values of the primary 

 and secondary currents, and oris such that N^ = Nj(i + -r) where 

 N^ and Nj are the fluxes of magnetism through the primary and 

 secondary coils at the same instant ; ^ is a negligibly small 

 quantity compared with the rest of the denominator. In ob- 

 taining this factor no assumptions whatever had been made, and 

 it was easy to see that if Kp and A^ could be assumed sine 

 functions, and x a constant, the value of the factor X became x 

 simply. In only one case did X reduce to zero, and that was 

 when X was a sine function of the same period as hp and A.r. 

 He believed that in actual transformers x was approximately 

 constant. Mr. Swinburne pointed out that the split dynamo- 

 meter was merely a watt-meter, and stated that he had trans- 

 formers which, owing to magnetic leakage, would indicate an 

 efficiency of over 100 per cent, if tested by Blakesley's method. 

 If this method gives an efficiency of 96 per cent., and leakage 

 causes a drop of 2 percent, in E.M.F., the real efficiency is 

 only about 94 per cent. He thought that the assumption that 

 the currents followed a sine law was equivalent to supposing that 

 there was no loss in hysteresis. The President said that no one 

 would be more glad than himself to find that Mr. Blakesley's 

 method was accurate, but he could not agree with Prof. Perry 

 that Dr. Hopkinson was wrong in abandoning the academical 

 method of Maxwell. Prof. Perry replied to the various points 

 raised in the discussion. — A paper on current and potential 

 difference analogies in the methods of measuring power, by 

 Prof. W. E. Ayrton and Dr. Sumpner, was postponed. 



Royal Microscopical Society, May 20. — Dr. R. Braith- 

 waite. President, in the chair.— The President said he regretted 

 to have to announce the deaths of two of their Honorary Fellows, 

 Dr. Carl von Naegeli, of Munich, and Prof. J. Leidy, of Phila- 

 delphia. — Mr. C. L. Curties exhibited a new form of Mayall's 

 mechanical stage, recently manufactured by Zeiss, which gave 

 upwards of an inch motion each way, and merely required to 

 be clamped on the pillar of the microscope when wanted for 

 use. — Mr. Watson exhibited and described a microscope which 

 he had recently made specially to meet the wants of Dr. Van 

 Heurck, of Antwerp. Mr. Mayall, after criticizing the design, 

 for which, he understood, Dr. Van Heurck was responsible, 

 concluded by expressing his regret that Dr. Van Heurck's 

 specification should have resulted in the production of the 

 microscope exhibited. Mr. E. M. Nelson and the Rev. Dr. 

 Dallinger also criticized the instrument adversely. — Mr. Grenfell 

 exhibited the photograph of a small organism, the nature of 

 which he had been as yet unable to determine ; some zoologists 

 and botanists to whom he had shown it were unable to say 

 whether it was vegetable or animal in its nature. He also wished 

 to mention that in the Botanical Gardens, Regent's Park, there 

 were considerable numbers of a free-swimming infusorian 

 known as Tintidhis. It was remarkable for its chitinous lorica. 



