where n' and n' indicate the indices of refraction of crown glass

v

for red and violet rays, and ri and r2 the radii of curvature of crown-glass lenses.

where the corresponding quantities for flint-glass lenses are

indicated by n" and n", P1 and P2. Ρι

T

บ

If the combination of the

two lenses for red and violet possess the same focal distance the two lenses must be such that

With the aid of this equation and the expressions above given for the several focal distances, the radii of curvature which must be given to the two lenses in order to obtain an achromatic system may be calculated with facility.

CHAPTER X.

SPECTRUM ANALYSIS.

65. IF instead of the measurement of indices of refraction the observation and comparison of the spectra proceeding from various sources of light be the subject

of enquiry, the divided circle of the spectrometer may be dispensed with; and the instrument thus simplified constitutes Bunsen's Spectroscope (fig. 111), in which the slit tube A, the prism P, and the observing tube B, are

all arranged just as in the spectrometer. In order, however, to obtain the means of measurement within the limits of the spectrum without a divided circle, a very ingenious apparatus has been introduced. A third tube, C (the scale-tube), has at its outer end, at s, a small photographed scale with transparent divisions, whilst at the inner end is a lens which is placed at about its focal distance from the scale. The scale is illuminated by means of a lamp or candle. The scale-tube is so placed that the rays of light that proceed from the scale and emerge parallel to the axis of the tube are reflected at the anterior surface of the prism in the direction of the observing tube. The observer looking into the telescope sees therefore coincidently with the spectrum of the light F, the image of the scale, which may be used as a measure.

As the rays are deflected from their original direction by the prism, the observing tube in the spectroscope just described must be so placed in regard to the slit-tube as to form an angle which is about equal to the smallest deflection of the middle rays. The source of light to be investigated cannot therefore be looked at directly, a circumstance which renders the arrangement of the instrument difficult and its management somewhat awkward. The direct vision or rectilinear spectroscope (à vision directe) which instead of a single prism contains a combination of prisms, so that there is no deflection, is free from this inconvenience (fig. 104). To this class belongs Hoffman's Spectroscope, and the little (only 3 in. long) pocket Spectroscope of Browning.

66. By means of the spectroscope the spectra of the glowing vapours formerly thrown upon the screen can be very conveniently observed objectively. But

whilst for those researches the dazzling light of the electric flame was requisite, the flame of a Bunsen's burner is now sufficient, at least for the light metals (fig. 1). Instead of the metal itself, some of its chemical combinations, or so-called salts, are usually employed. A small quantity of such a salt is melted at the extremity of a fine platinum wire, and intro duced into the external hottest part of the feebly luminous flame. The salt is decomposed by the heat; the flame is saturated with the vapour of the metal now set free, and is tinted with a colour characteristic of the metal. With a little Sodium chloride (common salt), for example, we obtain the homogeneous yellow light of Sodium; salts of Lithium and Strontium colour the flame of a carmine red tint; salts of Potassium clear violet; salts of Barium green; and salts of Calcium yellowish red. Analysts had no doubt long employed these characters to demonstrate the presence of the above metals, but the colour of the flame continued to be an uncertain means of recognition until prismatic decomposition was applied as a means of investigation. It was almost impossible, for example, with the naked eye to distinguish between the red flame of Lithium and that produced by Strontium, but if the two are looked at through the spectroscope they exhibit perfectly distinct spectra, which are exhibited on the Spectrum plate (see Frontispiece, Nos. 6 and 8). If, again, a specimen of Sodium salt with which only a trace of Lithium is mingled be examined, the presence of the latter cannot be recognised with the naked eye, because its feeble red stain is completely overpowered and concealed by the brilliant yellow of the Sodium. The spectroscope, however, shows distinctly the red line

of Lithium close to the yellow Sodium line, each in its place, thus disclosing the chemical composition of the substance in question.

This qualitative method of chemical analysis is termed spectrum analysis, and although the spectra of some coloured flames had been known for some time, and their applicability as chemical tests recognised, Bunsen and Kirchhoff were the first who laid down the scientific grounds on which alone a method of investigation could be raised, and who must therefore be regarded as the true discoverers of spectrum analysis. Bunsen and Kirchhoff showed first that the positions which the bright lines of the spectrum occupy are independent of the temperature of the flame; in fact, that the same red colour is obtained and the same two lines, a red and a reddish yellow, are seen in the spectroscope whether the Lithium chloride be volatilised in the flame of a Bunsen's burner or in the much hotter flame of the oxyhydrogen blowpipe. It is to be noted that the brilliancy of the several lines increases with increasing temperature, and thus it may happen that by means of intense heat lines come into view which at lower temperatures are too feeble to be perceived. If, for example, Lithium be volatilised in the electric flame, a blue line is visible in its spectrum, which occupies exactly the same position as the blue line of Strontium. In the flame of the Bunsen's burner it exhibits only the two abovenamed lines. Moreover, the two observers just mentioned demonstrated that different combinations of the same metals give invariably the same spectrum, whence the conclusion is irresistible that the lines seen in any instance may be regarded as positive evidence of the actual presence of the metals in question.