94 
POPULAR SCIENCE REVIEW. 
fifty centim. where the slit was situated, the polarizing instruments being 
between the lens and the slit, and consisting of a double refracting prism of 
spar with a central extraordinary image for analyser, and a large Nicol for 
polarizer. The quartz plate was between these. The prisms of quartz were 
on the coupled principle already employed by MM. Gouy and Thollon (see 
above). Achromatic lenses were used for collimator and telescope, a linear 
thermopile being in the principal focus of the latter. Tables of results thus 
obtained lead to a fine curve, in which abscissae are wave-lengths, and 
ordinates, values of (n'-ri), or the difference of the indices, which are func- 
tions of the wave-lengths and of the substance of the plate. 
Harmonic Ratios in the Spectra of Gases have been noted by Professor 
Johnstone Stoney for three of the hydrogen lines, and others by Soret and 
Lecoq de Boisbaudran. Mr. Arthur Schaster draws attention in Nature 
to a series which he has found in the Iron-spectrum, of which, however, he 
has only as yet examined a seventh part. He states them in a tabular form, 
giving in a first column the corrected wave-length, according to Angstrom ; 
in a second, a ratio expressed fractionally ; in a third, the effect of multi- 
plying the first column by the second; and in a fourth, the nearest observed 
value of other lines : the last column shows the difference between computed 
and observed values. 
A second table gives a set of iron-lines which can be arranged as Har- 
monics of a fundamental wave-length of given magnitude. The differences 
are all small, never reaching an integer, and in one instance sinking to zero. 
He cannot yet say definitely whether all these coincidences are due to acci- 
dent. The true law of distribution, however, he thinks not to have been 
yet found, though harmonic ratios may take a secondary part in it. 
A Fresh Experimental Determination of the Velocity of Light has been 
undertaken by Mr. Nicholson, of the U. S. Navy. He uses, in the main, the 
same apparatus as was employed by Foucault, namely, a revolving mirror and 
reflected image. But in Foucault’s experiments the deflection was too small 
to be measured with the required accuracy. It amounted to only a fraction 
of a millimetre, whereas in these it exceeded 133 millimetres, being about 
two hundred times that obtained by Foucault. It could be easily increased. 
The distance between the mirrors was nearly 2000 feet, the radius or distance 
of the revolving mirror from the slit was about 30 feet, and the speed of the 
mirror about 256 revolutions per second. The deflection was measured 
within three or four hundredths of a millimetre in each observation, so that 
the result is probably correct in this respect to within a ten thousandth. A 
shed was erected at one end of the line, in one corner of which a heliostat 
reflected the sun’s rays through the slit to the revolving mirror, and thence 
to the distant mirror. The shed was blackened inside. The revolving 
mirror consisted of a cast-iron frame, containing, between hardened steel 
points, an arbor carrying the mirror — a disc of plane glass about 1| in. diam., 
silvered on the hither side. A small turbine in a box on the same axis was 
rotated by the entrance of compressed air striking its vanes, and by the 
reaction of the same air in escaping. A disc above the mirror served to 
bring the centre of gravity into the axis of rotation, by means of a series of 
position-testings. It gave no sound when properly balanced, and needed only 
occasional lubrication. 
