16 



SCIENCE. 



[Vol. I., No. 1. 



end of the tube, and the miiTor, being half-way be- 

 tween these, is at a loop. Hence it tends to set 

 itself at right angles to the vibratory motion. This 

 tendency is opposed by the magnetic forces ; but the 

 image on the scale shifts its position through a dis- 

 tance proportional to the intensity of the action. 

 The instrument reveals an enormous disproportion 

 between sounds whieli, when heard consecutively, 

 appear to be of the same order of magnitude. — 

 (Phil, mag., Sept., 18S2.) c. r. c. [4 



Optics. 



Absorption spectra of ozone and pernitric 

 acid. — The places of eleven absorption bauds due to 

 ozone are catalogued by M. J. Chappuis according to 

 wave-length. Of these, by far the most intense are 

 those having the limits X — 609.3 to 593.5 and X — 

 577.0 to 500.0, which are Nos. 2 and .3 of the table ; 

 next in intensity is A = 535.0 to 527.0, which is No. 5 

 of the table. These bands were observed in light 

 which had traversed a tube 4.5 m. long, containing 

 ozonized oxygen jjrepared at the atmospheric press- 

 ure and a temperature of 15° C. Variations of length 

 of tube and pressure of gas were accompanied by a 

 variation in the intensity of the absorption bands, 

 such that the effect produced seemed proportional to 

 the quantity of ozone traversed by the light. A 

 lowering of temperature, however, produced, inde- 

 pendently of change in density, an increased intensity 

 of the bands. M. Chappuis succeeded also in observ- 

 ing the absorption spectrum of the blue liquid which 

 is obtained by compressing a mixture of carbon 

 dioxide and ozone, in which he found the two char- 

 acteristic bands Nos. 2 and 3 near D ; the absence of 

 the others being attributed to the small quantity of 

 the liquid used. 



If the smallest quantity of nitrogen were present in 

 the tube, other bands of a greater intensity appeared, 

 which M. Hautefeuille and the author were led to at- 

 tribute to an oxygen compound of nitrogen richer in 

 oxygen than nitric acid, and to which they gave tlie 

 name pernitric acid. The stronger of the bands were 

 readily seen in a tube no longer than 0. 1 m. The eight 

 bands attributable to tliis substance are tabulated and 

 described. 



In the second part of his paper, the author gives a 

 discussion of the bearing of his discoveries on the 

 telluric li]ies of the solar spectrum, with the convic- 

 tion that the lines 2, 3, and 5 of tlie ozone spectrum 

 are present in the spectrum of the sun when at the 

 horizon. That a part of the cause of the blue color 

 of the sky is the presence of ozone, is also indicated. 

 — (,7ourn. tU plnjs., Nov., 1S82.) o. s. H. [5 



Reflection of ' actinic ' rays. — • M. <le Chardon- 

 net finds that silver alone, of a large number of solid 

 and liquid bodies, exerts an elective absorption on 

 light of short wave-length. Polishing a body does 

 not alter its action. — {.lourn. de phys., Dec, 1882.) 

 c. s. H. [6 



Saccharimeter. — Note by M. 11. Dnfet on M. 

 Laurent's recent modification of his form of saccha- 

 rimeter, by adding an absorbing plate of bichromate 

 of potash, whereby a source of white light may be 

 used. — {Journ. depli7j!<., Dec, ISS2.) c. s. H. [7 



iP/totometri/.) 

 Stellar photometry. — In a discussion of the ac- 

 curacy attainable by the use of a neutral-tint wedge 

 of glass for the determination of stellar magnitudes. 

 Prof. Pritchard finds that careful measures ought 

 not to be in error more than one-thirteenth of a 

 magnitude. He also fiuds reason to believe that the 

 ordinarily assumed law, that the brightness of a star 



is directly proportional to the siinare of the aperture 

 of the observing telescope, may lead to sensible errors. 

 The paper contains a table of differences of magni- 

 tudes, as determined by himself, compared with the 

 same quantities derived from the Harvard observa- 

 tory. — {Month, not. roy. a.itr. soc, Nov., 1882.) 

 c. s. H. [8 



Photometric measurements of the sun, moon, 

 and electric light. — According to the measure- 

 ments of PouiUet, the sun is radiating 7,000 horse- 

 power per square foot of its surface, or 50 horse-power 

 per square inch. Sir William Tliomson states that 

 the normal current through a Swan lamp giving 20 

 candle-power is 1.4 amperes, with a potential of 40 to 



4.3 volts. Hence the actual work is 61.6 ampere-volts, 

 or watts (so-called). Dividing by 746, we find .085 

 horse-power for the electric activity in a Swan lamp. 

 The filament is 3.5 inches long, and .01 inch in 

 diameter: hence the area of the surface is .11 of a 

 square inch, and the activity at the rate of .75 horse- 

 power per square inch. Therefore the activity of the 

 sun's radiation is about 67 times greater than that of 

 a Swan lamj) for an equal area. 



An experiment on sunlight compared with an ob- 

 servation on moonlight made by our author, has led 

 him to conclude that the surface of the moon radi- 

 ates something not enormously different from one- 

 third of the light incident upon it. The moonlight at 

 the time and place of the observation (York, early 

 in September, 1881, about midnight, near the time 

 of full moon) was found to be equal to that of a 

 candle at a distance of 230 centimetres. The lumi- 

 nous intensity of a cloudy sky was found, about 10 

 A.M. in York, during the meeting of the British as- 

 sociation, to be sucii that light from it through an 

 aperture of one inch area is equal to about one 

 candle. 



An experiment on sunlight last December showed, 

 at one o'clock, the sunlight reaching the author's 

 house to be of such brilliancy, that the amount com- 

 ing through a pinhole in a piece of paper .09 of a 

 centimetre diameter produced an illumination equal 

 to that of 126 candles. The area of tlie candle-flame 

 was 2.7 square centimetres, or 420 times the area of 

 the pinhole, and therefore the intensity of the sun's 

 light was equal to 126 X 420, or about 53,000 that of 

 a caudle-flame. — {Eleetr. rem.nw, Dec. 23, 1882.) 



Sir W. Thomson's first calculation showing that a 

 Swan lamp giving out 20 candle-power uses up only 

 ^V the amount of energy of the sun for the same unit 

 of surface is interesting; but, if we include the ques- 

 tioii of the light obtained, quite a different result 

 will be reached. The total area of the carbon fila- 

 ment, as we have seen, is .11 of an inch; but only 

 half of this, or .0.55 inch (equal to .36 centimetre), 

 can be seen at once; ami this gives out 20 candle- 

 power. The area of the pinhole in the last observa- 

 tion was .0063 square centimetre, and gave out 126 

 candle-power. Hence -jji^-j X -'./,? = 359. — the in- 

 trinsic brilliancy of the sun in terms of the Swan 

 light. The sun therefore radiates 07 times the en- 

 ergy, but 359 times the light, of the Swan lamp, or 



5.4 times the light for every horse-power expended. 

 In May, 1879, the writer conducted some observa- 

 tions on this subject {Proc. ^bver. acad., 1880, xv. 

 236), by which it was found on one occasion that the 

 total brilliancy of the sun, when at an altitude of 

 25°, was 64,700 candle-power at one metre's distance ; 

 and another time, when at an altitude of 40°, 82,000 

 candle-power. The apparent area of the sun's disc 

 at this distance would be .68 centimetres ; and as- 

 suming that the area of the candle flame in this in- 

 stance was 2.7 centimetres, ■which could not be verv 



