942 REPORT—1885. 
This interesting observation is given, in terms which are substantially the 
same, in a letter addressed to Oldenburg, and dated Cambridge, April 1672 
(Horsley’s ‘Newton,’ iv. p. 311). In the letter the spectrum of Venus is described 
as a ‘long splendid line.’ 
In conclusion, the author referred to the brief but correct account of Newton’s 
experiments described in the early part of the paper, by Lloyd in his ‘ Undulatory 
Theory,’ and in Verdet’s Works, iii, p. 253. 
15. On the Use of Bisulphide of Carbon Prisms for cases of Hutreme Spectro- 
scopic Dispersion, by Professor C. Piazzi SmytH; and their Results in 
Gaseous Spectra, commented on by Professor ALEXANDER S. HERSCHEL, 
M.A., F.B.S. 
The serious loss of light which occurs by absorption in long trains of dense 
glass prisms was attempted to be diminished by the first author using, in place of 
the usual flint-glass prism between two correcting crown-glass ones, a bisulphide of 
carbon prism with a refracting angle of 104°. A train of such compound fluid 
prisms of 2] inches optic apertures, together with two simple flint-glass prisms, each 
of 64° refracting angle, dispersed the spectrum between A and H over an angle of 
about 60°. The spectrum so formed was examined with an inspecting telescope of 
2:25 inches aperture, 32 inches focal length, and of magnifying power 36. About 
cone-seventh of the whole length of the spectrum could be commanded at one 
time by the range of the micrometer-screw which moved the inspecting telescope 
in angle, without resetting the prism train to minimum deviation for the next 
adjoining portion of the spectrum. 
After overcoming the tendency of the interior glass faces to contract a film 
from the cement used in the prisms, by repeated washings, the well-known - 
difficulties of inequality of temperature in the fluid prisms, disturbing their 
homogeneity and altering the total dispersion of the train, had then to be contended 
with, and were obviated sufficiently for relative measurements by the use of non- 
conducting coverings of the prisms, and by guarding carefully in each set of 
observations against accidental changes of temperature in the room. 
With the spectroscope so arranged the first electric spectrum of a vacuum 
tube examined, was that of oxygen gas. The four distinct lines of the so-called 
compound spectrum of the gas were identified, and three of them proved on 
examination with the instrument’s high dispersive power to be really compound, 
each of the three being resolved into a delicate line-triplet. Three similar line- 
triplets in regular configuration with these in neighbouring parts of the spectrum 
were also found, while a strong oxygen line, in addition to other single lines of the 
spectrum, was detected in the ultra-red at a distance from the ordinary field of 
chromatic light in that direction greater than that of any elementary gas-spectrum 
line yet ocularly measured. 
Of the numerous array of lines produced in hydrogen-tubes which have been 
ascribed to hydrogen, 1,616 lines were well seen and measured, most of them 
extremely sharp and beautifully precise lines. Of these 1,442 are in the space 
where 448 such lines are recorded in the hydrogen line catalogue of this gas’s 
compound spectrum in the Report and tables of the wave-lengths of the elements, 
presented last year by the Spectroscopic Committee of the British Association,’ 
20 are beyond the violet, and 154 are beyond the red limit of the range of the 
same catalogue in that Report. 
As another example of the great resolving and defining powers of the 
bisulphide of carbon prism spectroscope, it may be noted that in a single fluting 
of the compound spectrum of nitrogen near the solar line D, in which twenty 
linelets and haze-bands are figured in the extensive map of that spectrum recently 
communicated to the Imperial Academy of Sciences of St. Petersburg by Dr. 
Hasselberg, 161 sharp lines were seen and mapped: with the train of fluid prisms, 
while only three wave-lengths of the leading edges in the same fluting are noted 
1 British Association Reports, 1884, p. 390. 
—— 
