Nov. 6, 1879] 



NATURE 



25 



These photometers could only be used for comparing objects 

 very near together, as double stars, or satellites. For greater 

 intervals, another device was tried. Two achromatic prisms of 

 small angle were placed in front of the telescope, so as to 

 cover the central portion of its object-glass. Two images of 

 any object would thus be formed, separated by an interval de- 

 pendent on the angle of the prisms aud on their relative positions. 

 By turning one or both of the prisms, the directions of the two 

 images may be altered at will, and their distance varied between 

 the sum and the difference of the angular deviation of the prisms. 



:^"- 



EFiGS 



After bringing the images near together, they could be com- 

 pared by one of the photometers described above. This method 

 was tried with two circular prisms, having a diameter of 4"4 cms., 

 and producing a deviation of about l°'3. A telescope was used 

 having an aperture of 10 cms. The light of any objects nearer 

 than 2°'6 could be measured with this instrument. The constant, 

 or proportion of the light transmitted by the prisms, w as easily 

 determined by comparing, by the photometer, the two images 

 of the same object. This instrument, like those previously 

 described, has the great advantage that both objects are seen 

 undeMhe same magnifying power, and therefore closely resemble 



each other, even when the condition of the air is not good. This 

 plan cannot be used for large intervals, since, if the angles of 

 the prisms are large, the images will be coloured by the 

 secondary spectrum, and it would also be difficult to find the 

 objects. With a large telescope, the prisms could not be 

 reached easily by the observer, and the large diameter required 

 would be an objection to their use. 



The idea suggested itself to Frof. Pickering, that such photo- 

 meters might be used to compare the colours of the components 

 of double stars, by measuring //d' relative light of different 'portions 

 of their spectra. A combined spectroscope and photometer, 



shown in Fig 3, was devised. A is a Nicol, placed in front of the 

 eye-piece B. The graduated circle C C is attached directly to the 

 tube carrying the Nicol ; and the indices D D are fastened to the 

 tube G, which slides into the telescope. H is a direct-vision 

 prism, by which the images of the stars are converted into linear 

 spectra. F is a diaphragm placed at the focus, and having a 

 slit in it "02 cm<. broad, parallel to the edges of the prism. It 

 is, therefore, perpendicular to the spectra, and permits a short 

 portion of each to pass through. These appear as tv o stars, of 

 a colour which may be varied with the position of the objects 

 observed, as regards the axis of the telescope. Their relative 

 light was measured by forming two images of each, by a plate of 

 Iceland spar, E, which was used instead of a double-image 

 prism, since the rays were not parallel. The light was then 

 measured by turning the Nicol. 



All of the photometers described above are open to the objec- 

 tion that the loss of light is very great. Under the most favour- 

 able circumstances only '20 to '40 of the light is used ; so that, 

 with the large telescope of aperture 38 cms., faint objects 

 appear no brighter than with a telescope having an aperture of 18 

 to 24 cms., with a common eye-piece. To remedy this objection 

 which was greatly felt during the observations of the satellites 

 of Mars, a class of photometers of wholly different form was 

 tried. 



In these the image of some bright object, assumed as a 

 standard, is reflected into the field of the telescope, and its light 

 reduced by a known amount, until it is no brighter than the 

 object to be measured. An unobstructed view of the latter is 

 obtained meanwhile, with an eye-piece of the usual form. The 

 first of these instruments is represented in Fig. 4. 



The image of the faint object formed by the telescope is 

 viewed by the eye-piece A. The light of the bright star taken 

 as a standard, passes outside the telescope, and falls upon the 

 prism B, by which it is reflected through the objective D of a 

 small auxiliary telescope, and falling on the prism F, is brought 

 into the field of view. The faint object is thus seen in one half 

 of the field with the full aperture of the telescopt ; while the 

 bright standard appears in the other half of the field, its image 

 being formed by the small telescope, c and E are two Nicols, 

 of which E may be rotated, and the light pas-ing through it 

 reduced at will. G G is a graduated circle, attached to 

 the tube carrying D and E, and measuring the reduction of the 

 light by an index H, which is fixed. The whole photometer 

 may be turned around the axis of the large tele cope, the 

 tube carrying the prism enables the latter to rotate around 

 the axis of the auxiliary telescope, and, finally the prism 

 may be tipped around an axis parallel to its edges. 

 Either two of these motions enable the observer to bring 

 any object into the field of view of the small telescope. 

 Practically, the second and third motions were used for 

 the purpose. The first of these movements was reserved 

 almost exclusively for the purpose of placing the prism so that 

 it would conceal the bright star or planet with which the faint 

 object was to be compared, when their distance apart was small. 

 Otherwise, its light as seen in the large telescope, would be so 

 intense as to interfere with the proper estimate of the light of 

 the faint object. I is a lamp, by which the half of the field 

 covered by the prism F may be illuminated, so as to render it as 

 bright a- the other half of the field. A piece of blue glass, K, 

 served to vary the colour of the light. 



Great difficulty was experienced in obtaining good images of 

 bright stars with the small telescope, on account of the Nicols 

 used. Since the rays passing through E are convergent, aber- 

 ration i, caused by the obliquity of its faces, even if they are 

 plane and parallel. Hence the Nicols were removed, and a new 

 form tried. The lenses of a double-image micrometer being 

 taken out, two V-shaped pieces of brass were attached to the 

 slides carrying the divided lens. A square hole, or "cat's eye," 

 was thus formed, whose dimensions could be altered at « ill, by 

 turning the micrometer screw. This arrangement is shown in 

 M N, Fig. 6. Placing it near the objective D, Fig. 4, the light was 

 varied by changing the aperture of the small telescope. All these 

 instruments, however, were heavy, difficult to adju-t, and not 

 easily removed and replaced. These defects were remedied by 

 still another form, represented in perspective in Fig. 5, and in 

 section, on a scale of one-fifth, in Fig. 6. The same letters 

 are used as in Fig. 4, for the corresponding parts. The 

 faint object is viewed with the eyepiece A, while the 

 light of the bright object, passing outside of the telescope, is 

 reflected by the prism B into the object-glass D, \\ hose aperture 



