13^^ 



NATURE 



{June lo, 1880 



tbemselves iii the initial and terminal phases of the normal 

 variation. 4. When one contact is warmer than the other the 

 local duration of the excitatory state is le^s in the warmed than in 

 the unwarmed surface. 5. If the surface near one contact is 

 slightly injured, the local duration at the injured surface is 

 diminished in the same way as when the temperature is increa.sed, 

 but if the injury is of such intensity as to destroy its surface, 

 its most prominent effect is to diminish its electromotive 

 activity. 



In an appendix the authors briefly consider the results of 

 slight inequalities produced by mechanical, chemical, or thermal 

 conditions on the potential of the surface of the ventricle in the 

 resting heart, and the influence of temperature on the excitability 

 of the resting heart. 



A full account of the experiments, the results of which were 

 communicated to the Society, will be published in the Journal 

 of Physiology, 



May 27. — " On some Thermal Effects of Electric Currents," 

 by William Henry Preece, General Post Office. Communicated 

 by Prof. Stokes, Sec. R.S. 



I have been engaged for some time past in experimenting on 

 the thermal effects of electric currents, but the final results of 

 those experiments are not sufficiently ripe at present to justify 

 my bringing them before the Royal Society. I have, however, 

 obtained one result which I believe to be sufficiently novel to 

 justify a short preliminary note. 



The mott striking facts elicited by these experiments are : 



1. The extreme rapidity with which thin wires acquire and 

 lose their increased temperature. 



2. The excessive sensibility to linear expansion which fine wires 

 of high re>istance evince. 



Now as the rate of heating, and therefore of expansion and 

 contracli^n, varies very nearly directly as the increment or decre- 

 ment of the currents when these v-iriations are very small, it 

 occurred to me that if a long wire of small diameter and high 

 resistance were attached to a sounding board or to the centre of 

 a disk (such as one of those used for telephones and phonographs) 

 and it formed part of a circuit conveying telephonic currents, 

 sonorous vibrations ought to be reproduced. 



The sketch shows the arrangement of the apparatus used for 

 the experiment. 



A was a stout base of mahogany, on which a brass support C 

 was attached so that it could slide and be fixed at any distance 

 from D. 



D was at first a disk of thin paper, and then of thin iron. 



P was the wire experimented upon whose loose ends were con- 

 nec'ed to terminals on the wooden ba^e, so as to be inserted in 

 the circuit containing a microphone transmitter M and a battery 

 B of six bichromate of potash cells in another room out of 

 hearing. ^ 



A platinum wire of o'ooj inch diameter and 6 inches long 

 from / to p' was first used, and the sonorous effects were most 

 marked and encouraging when the microphone transmitter M 



was spoken into. The articulation, though muffled, was clear, 

 and words could easily be heard. 



1. Experiments were first made to determine the length which 

 gave the loudest sound and the clearest articulation, and, after 

 repeated trials with every variation of length from I inch to 6 

 feet, it was found that a wire 6 inches long gave the maximum 

 effect. 



2. Experiments were then made to determine the diameter of 

 the w ire that gave the best effect, and after repeated trials with 

 every gauge drawn from 0-0005 inch to 0-005 '"^h, it was found 

 that wire of the diameter O'ooi inch gave the best effect. 



3. Experiments were then tried with wires 6 inches in length 

 and O'OOI inch diameter of different materials, viz., gold, iron, 

 aluminium, silver, copper, palladium, and platinum, and they 

 came out in the following order of merit : — 



Platinum ... ... ... ... Very clear. 



Aluminium ... Very variable. 



Palladium Clear. 



Iron Clear. 



Copper Faint. 



Silver Faint. 



Gold ... ... ... ... Very poor. 



4. Tlie effect of mechanical strain was tried. It was found 

 not to vary the effect. When once the requisite tension, which 

 varied with each metal, was obtained, further tightening up did 

 not vjry the clearness or loudness of articulation. 



Gold would scarcely bear the tension required to reproduce 

 sonorous vibrations, hence its low position. 



5. Very thin carbon pencil, '0625 inch diameter, was tried 

 under compression and under tension, but no effect whatever was 

 experienced unless a bad joint was made, when at once a faint 

 microphonic effect was apparent. 6?2i' 



6. No sibilant sounds whatever could be reproduced. --. ;• 



7. That the effect was due to heating and cooling was show n 

 by the fact that it was possible to increase the current to such a 

 strength as to render the temperature of the wire sensible to the 

 touch, and then to make its elongation and contraction by low 

 sounds evident to the eye. 



It therefore appears from these experiments that wires con- 

 veying those currents of electricity which are required for tele- 

 phonic purposes expand and contract as they are heated and 

 cooled, and as the variations in the strength of the current are 

 small compared with the strength of the current itself, the 

 expansion and contraction vary in the same ratio as the con- 

 densation and rarefaction of the air particles conveying the 

 sonorous vibrations which produced these vibrations. 



The mechanical changes, or molecular vibrations in the wire, 

 due directly or indirectly to telephonic currents, i\hich result in 

 the reproduction of sound, bear a close analogy to the mechanical 

 changes due to the direct transmission of sound, but with this 

 important difference, that while the vibrations due to sound are 

 progressive along the wire, and their velocity is low and easily 



