490 



NA TURE 



\Scpt. 2 2, 1881 



Taking 70/. per ton as the price of copper of high corductivity 

 (Icnown as "conductivity copper" in the metal marliet), we 

 have "00007/. as the price of a gramme. Multiplying this by 

 S'9 (the specific gravity of copper), v\e find, as the price of a 

 cubic centimetre, 



Tr; -00062/. ... (6) 



and the assumption of 10/. as the par value of one horse-power 

 day and night for 365 days gives, as the price of an erg', 



io/./(3axio''X74Xio*') = ~_^^^Gf U. (7) 



Supposing the actual price to be at the rate of £> x 10/. for the 

 horse-power year, we have 



, of I/. 



(S) 



23X101* 



Lastly, for the specific resistance of copper we have 



^=1640 .... (9) 

 Using (8) and (9) in (5) we find, 



,^..^63XK,'X.640X/.^^^ /£_ 

 23Xio'*X "00002 I "38 



Suppose, for example, p—'^ (that is, electric work throUL^h 

 the conductor for twelve hours of every day of the year to be 

 provided for), and e— i. These supppositions correspond fair'y 

 \yell to ordinary electric transmission of enerj;y in towns for 

 light, according to present arrangements. We have — 



27-6 5-25 

 That is to say, the sectional area of the w ire in centimetres 

 ought to be about a fiftieth of the strength of the current in 

 webers. Thu', for a powerful arc-light cuiTent of 21 weliers, 

 the sectional area uf the leading wn-e should be "4 of a square 

 centimetre, and therefore its diameter (if it is a solid round wire) 

 should be "71 of a centimetre. If we take e—^,\.-^, which cor- 

 responds to 1900/. a year as the cost of 5250 horse-power ( ee 

 Presidential Address, .Section A), and if we take /= i, that is, 

 reckon for continued night and day electric work through the 

 conductor, « e have — 



^ x/381 'Ws' 

 andif <r = 24, .^ = i"24, which makes the diameter 1-26 ceuti' 

 metres, or half an inch (as stated in the Presidentkal Address). 

 But even at Niagara it is not probable that the cc'St of an erg 

 can be as small as ji^ of what we have taken as the par value 

 for England ; and probably therefore a larger diameter for the 

 wire than \ inch will be better economy if so large a current as 

 240 webers is to be conducted by it. 



llluminat'mg Powers of Incandescent Vacuum Lamps with 

 Measured Potentials and Measured Currents, by Sir William 

 Thoinsou and James T. Boltomley. — The electromotive force 

 useid in these experiments was derived from Faure secondary bat- 

 teries, kindly supplied for the purpose by the Societe la Force 

 et la Lumiere in their London office. Two galvanometers w ere 

 used simultaneously, one (called the /otential galvamviu-tcr) 

 for measuring the ditference of potentials between the two ter- 

 minals of the lamp, the other (called the current galvanometer) 

 for measuring the whole strength of the current through the 

 lamp. The potential galvanometer had for its coil several thou- 

 sand metres of No. 50 (B.W.G.) silk-covered wire (of which 

 the copper vvei_,hs about one-twentieth graraaie per metre, and 

 therefore has resistance of about 3 ohms per metre). Its elec- 

 trodes were applied direct on the platinum terminals of the 

 lamp. The current galvanometer had for its coil a single circle, 

 of about 10 centimetre, diameter, of thick wire placed in the 

 direct circuit of the lamp, by means of electrodes kept clo^e 

 together to a ufficient distance fro Ji the galvanometer to insure 

 no sensible action on the needle except from the circle itself. 

 The directive force on the needle which was produced l;y a large 

 semicircular horseshoe magnet of small sectional area was about 

 2h, c.g.s., or fifteen times the earth's horizontal magnetic force in 

 London. This anangenunt would have been better for the 

 potential galvanometer also than the pkm actually used for it, 

 which need not be described here. The ?cale of each galvano- 

 meter was graduated according to the uatnral tangent of 

 the angle of deflection, o that the strength of the current was 

 simply proportional tu the number re.id on the scale in each case. 

 Three lamps were used, Nos. II. .and III. of a larger size 

 than No. I. The experiment w.as continued with higher and 



higher potentials on each lamp till its carbon broke. The 

 illuminating power was measured in the .simplest and easiest 

 way (which is also the most accurate and trustworthy), by letting 

 the standard light and the lamp to be measured shed their lights 

 nearly in the same direction on a white ground (a piece of white 

 paper was used) ; and coinparing the shadows of a suitable object (a 

 pencil w a^ used) ; and varying the distance of the standard light 

 from the white ground till the illuminations of the two shadows 

 were judged equal. The standard used was a regulation " stand- 

 .ard candle," burning 120 grains of wax in the hour. The 

 burning was not actually tested by weighing, but it was no 

 doulit very nearly right ; nearly enough for our purpose, which 

 was an approximate determination of the illuminating powers of 

 each lamp through a wide range of electric pow er applied to it. 

 The follow ing results were obtained : — 

 Lamp No. I. 



Lamp No. II. 



Some of the irregularities of the results in the preceding tables 

 are very interesting and important, as showing the effect of the 



