Page:The New International Encyclopædia 1st ed. v. 18.djvu/493

* SPECTKO-PHOTOMETKY. 425 SPECTROSCOPE. curve in the accompanying diagram. Fig. 4. A comparison of the intensitydistribution in the spectra of the electric incandescent and arc liglits with that of the sun gives the following results, calling unity the relative intensity of each with respect to the sun for sodium light (D line) : WATE-LBNGTH Inc. Him Arc sun 800 UU 11.86 4.88 2.68 1.25 1.00 0.38 0.17 0.10 0.05 1.67 Line A. 1.37 •■ B 1.28 •• c 0.97 •• D 1.00 ■■ E •' F •. 0.77 0.56 •• G 0..S3 •• H 1.21 The reflecting power of a polislied plane sur- face or the absorbing power of a thin film or plate of an absorbing medium can be determined for various colors by means of a .spectrophotometer, by placing the mirror or the thin film in front of the adjustal)le slit T', and comparing the inten- sity of the light after reflection or transmission with the intensity of the light coming directly to the other collimator of the photometer. For ex- ample, a polished silver surface reflects 95 per cent, of the red light falling on it normally, while for the blue only 60 per cent, is reflected. A mirror of solid cyanine reflects 15 percent, of the orange, 2 per cent, of the green, and 6 per cent, of the violet. Bibliography. The methods of observation and of calibration, and the sources of error are fully treated in the following articles: Lunimer und Brodhun, Zeitschrift fi'rr Instrumentenkunde, 1889, p. 42, and 1892, p. 137 (Berlin) : Murphy, Astro-Physical Journal. 1897, p. 1 (Chicago) ; Brace. Philosophical Maga::in€, 1899, p. 420 (Lon- don), and Astro-Physical Jonrnnl, 1900, p. 6 (Chicago) ; Capps, Astro-Physical Journal, 1900, p. 25 (Chicago) ; Tuckermann, Astro-Physical Journal. 1902, p. 145 (Chicago). SPECTROSCOPE (from Lat. spectrum, ap- pearance, image, apparition -j- Gk. (TKOtreiFy tkopein, to view). An instrument designed to investigate the nature of the radiations emitted by various sources of 'light,' it being understood that this term includes all waves in the ether, although only those within certain narrow limits of wave-length aft'ect the sense of sight. It is shown in the article on Spectroscopt that radia- tions are being sent off in the ether from all natural bodies, and that these are in the form of waves of different wave-length. When these waves pass through a spectroscope they are dis- persed in such a way that tlie waves of certain wave-length are brought to focus at a point dif- ferent from that to which waves of a different wave-length are brought. In this way the radia- tions from any source are analyzed and spread out in what is called a 'spectrum.' The essential features of a spectroscope are then: first, a slit, or extremely small source of radiation; second, some means of producing dispersion: third, a lens or other means of focusing the radiations at the eyepiece of a telescope, upon a screen, or upon some suitable recording instrument. Tf the radia- tions are of such a nature as to affect a photo- graphic plate, that is. if they are in the ultra- violet or in the visible portion of the spectrum, methods of photography may be used in connec- tion with the spectroscope. (See Spectrocrapii.) If the radiations are in the infra-red, that is, if the wave-lengths are so long that they do not af- fect the sense of sight, instruments must be used which are sensitive to such radiations; for in- stance, a thermometer, bolometer, radio-micro- meter, radiometer, or otlicr Iicat-rogistering de- vice. It has been found that if a plate of glass covered with some phosphorescent substance, such as Balmain's paint, is exposed to light and then carried into a darlvcned room, it will con- tinue to be luminous, for some time; but, if ex- posed in a spectroscope to infra-red radiations, the pliosphorescenee at those points reached by the radiations is destroyed. This furnishes a method, therefore, for the study of these long waves. As ordinarily constructed, a spectroscope has the same general appearance as a spectrometer (q.v.). There is, however, in addition, in case prisms are used to produce dispersion, some aux- iliary apparatus for the purpose of enabling the observer to record numerically the positions oc- cupied by the waves which he is observing, in comparison with other waves. One method which is in common use is to attach to the instrument a tube containing at one end a transparent scale, and at its other end a lens, the tube itself being so placed that when the scale is illuminated by a lamp the waves proceeding from it fall upon the last face of the last prism and are reflected in such a manner as to be brought to focus in the same plane as are the radiations under investiga- tion. By this means there is produced across the spectrum a series of lines regularly spaced and numbered, and the position of any radiation can be recorded. Various means are used to produce dispersion, but the two methods most generally adopted are ( 1 ) to interpose a prism or train of prisms between the collimator and telescope, or (2) to allow the light from the collimator to fall upon a diffraction grating (q.v.). The dispersing action of a prism and of a grating has been explained before (see Diffraction and Diffraction Gratings) ; but there are several important differences be- tween the spectra obtained by these two instru- ments. Prismatic spectra are said to be 'irration- al' because there is no simple relation between the material and shape of the prism and the dis- persion produced by it: and. further, because prisms of the same shape .and size produce quite different spectra, in the sense that the relative deviations of the same waves differ widely when different prisms are used. See Dispersion. The spectra produced by plane gratings, on the other hand, obey a definite law, there being an extremely simple relation between the constants of the grating and the deviation of a given train of waves, which is independent of the material of the grating, and which enables one by simple means to measure the wave-length of the radia- tions being studied. If a concave grating is used in place of a plane one, as was first done by Rowland in 1882, it is not necessary to have lenses in the spectroscope, the essential parts of the instrument then being simply a slit, a grat- ing, and some receiving apparatus, such as a pho- tographic plate. There is a further advantage in the use of a concave grating (which may also be obtained with a plane grating if suitably ad- ■