553 



DISPERSION. 



DISPERSION. 



54 



passing gradually and insensibly from deep red to an attenuated violet, 

 in the following order, as described by Newton, and since very generally 

 concurred in, red, orange, yellow, green, blue, indigo, violet. This 

 experiment, which first opens the analysis of light, is easily made by 

 letting a beam of sunlight pass through a small circular hole in a 

 shutter, in a darkened room, on a glass prism such as above described, 

 the refracted and dispersed beam being received on the opposite wall, 

 ceiling, or floor, according to the position of the prism. 



When the image of the sun, or a star, caudle, &c.. is thus formed by 

 admission through a small hole, and the refraction of the prism, the 

 coloured apace G </, which has the same angular breadth in a direction 

 parallel to the axis of the prism as it would have had if the prism 

 were removed and the light received directly on the screen at the same 

 distance from the place of the prism (the screen being in both cases 

 supposed to be held perpendicularly to the incident light), but which is 

 considerably elongated in the perpendicular direction, is called the 

 spectrum ; and that angle of the prism BAG the sides containing which 

 B A, A c, have been traversed by the ray D E F G, is called the refracting 

 angle of the prism. 



Suppose, now, that a small orifice, o, is made in the screen at some 

 point of the spectrum, so that rays of any particular colour, green for 

 example, may be transmitted through it; and let the transmitted 

 portion be again submitted to refraction through another prism, this 

 beam being supposed very small, to ensure its purity or near uniformity 

 of colour. It will not, after refraction, be again decomposed, or undergo 

 any alteration of colour ; but if the first prism be turned round its axis, 

 which will cause different colours in succession to fall on o, while the 

 direction of the ray incident on the second prism remains unchanged, 

 being that of a line joining o with the first prism, it will be found that 

 as we pass from the red to the violet the ray will be more and more 

 bent round by the refraction of the second prism. This shows that 

 light incident on the first prism, when once decomposed into homo- 

 geneous elements by refraction, is then, at least by refraction, not 

 further decomposable, but each element retains its own colour and its 

 own refrangibUity, or disposition to be bent by refraction. 



Thin experiment will not perfectly succeed with sun-light as above 

 described, because however small be the hole in the window the incident 

 beam will not be a mere ray, on account of the finite angular diameter 

 of the sun, but a cone of very sensible angle, so that the first spectrum 

 will be impure, from the finite size and consequent overlapping of the 

 coloured circles corresponding to the elementary kinds of light. It 

 might be rendered pure, but at great expense of light, by limiting the 

 beam by a screen with- a small hole placed at some distance from the 

 iir.it hole. But it is far better, according to the method first described 

 by Newton, to combine the prism (which in this case must be placed at 

 a considerable distance from the hole in the window) with a convex 

 lens, and receive the spectrum on a screen placed at the focus of the 

 lens conjugate to the hole in the window. The lens alone collects into 

 a point the divergent pencil of rays of any one kind ; the prism alone 

 bends it round as a whole, but differently for the different kinds, with- 

 out (at least in the position of minimum deviation) affecting the diver- 

 gency ; and the two combined will give a pure and brilliant spectrum, 

 but very narrow. To obtain an equally pure and brilliant spectrum, but 

 of considerable breadth, we have only to replace the hole by a narrow 

 aperture parallel to the axis of the prism. If our object be merely to 

 see a pure spectrum, without placing objects in it, we may replace the 

 lens and the screen by the eye and the retina, that is, merely view, 

 through a prism applied to the eye, a slit transmitting light. In this 

 way very pleasing and instructive experiments may easily be made on 

 the action of absorbing media (such as coloured glasses, &c.) on light. 

 Instead of the naked eye, a telescope may be employed ; and in this 

 way accurate measures may be taken, supposing the telescope to be 

 properly mounted, and furnished with cross wires. 



We have seen that compound light, the sun's for example, may be 

 decomposed into its homogeneous constituent rays by refraction through 

 a transparent prism. Conversely it may be recompounded into light 

 similar to the original, merely by making the rays, thus separated, by 

 another refraction to occupy the same place. This may be effected by 

 placing a prism of exactly similar material and form to that already 

 used, with its refracting angle turned in a direction opposite to that of 

 the former, so that the near faces of both prisms may be parallel ; for 

 the rays entering the second prism are in the same condition as if we 

 supposed their direction inverted, that they may repass through the 

 first ; and therefore they emerge in a similar compound ray with the 

 original, which may also be easily confirmed by experiment. 



The rays issuing from the second face of the refracting prism, may 

 also be collected by means of a large convex lens, placed at a consider- 

 able distance from the prism. If the rays after passing through the 

 lens be received on a screen of white paper, which is first held close to 

 the lens and then moved away, the spectrum will contract in width, 

 until at the focus conjugate to the prism the colours disappear, the 

 original compound white light being reproduced, after which the 

 irs reappear in the reverse order. 



Th,e prismatic analysis of light, together with the phenomena relative 

 to the transmission and absorption of light, enabled Newton to conclude 

 that the colours of natural bodies are not inherent qualities of those 

 bodies, but depend on their powers of reflecting, transmitting, or ab- 

 sorbing the rays of some colours more than others from the compound 



light incident on them ; for all bodies placed in homogeneous light of 

 any colour appeal- themselves to be of that colour, though they are 

 most luminous (making allowance for the different intensity of diffe- 

 rent parts of the spectrum) when placed in that coloured light which 

 they reflect most copiously. Hence also arise the different colours 

 of coloured liquids or glasses. [ABSORPTION.] 



Many of the prismatic colours may be imitated by mixing colours 

 taken, as they lie in the spectrum, of greater and less refraugibility, as 

 orange from red and yellow, &c., but such compound colours are not 

 identical with the homogeneous light of the same colour, being imme- 

 diately decomposed when viewed through a prism. 



If the original prism B A o be turned gradually round its axis, pre- 

 senting always to the incident light the same refracting angle A, the 

 spectrum Gy may be made to descend towards K, but after arriving at 

 a certain point where the deviation, that is the inclination of D E pro- 

 duced to F G, is a minimum, it then re-ascends, and it is usual to make 

 the chromatic experiments in this definite position of minimum de- 

 viation. This occurs when the position of the prism is such that the 

 angles of incidence and emergence, or their complements DEB, a F c, 

 are equal ; for when the moving point G has reached its lowest place, 

 it is for a moment in the condition of a fixed point like the point D, 

 through which we may suppose the incident beam admitted ; hence 

 rays proceeding from D, notwithstanding a small variation of incidence 

 arising from the rotation of the prism, reach G, as if it were a fixed 

 point ; and since in dioptrics it is of no consequence to the path in what 

 direction we suppose the rays to move, it follows that rays proceeding 

 from G, notwithstanding a small alteration of the angle c F G, would 

 arrive at the fixed point or orifice D ; and consequently the data for 

 the determination of the angles D E B, a F c, in the position of minimum 

 deviation, are precisely the same, and therefore these angles must theu 

 be equal. 



This being premised, the following easy calculation will give the 

 necessary angle of incidence to produce a minimum deviation. 



Since the angles of incidence and emergence are equal, the angles 

 formed by the interior ray E F with both sides of the prism are equal, 

 or the triangle A E F is an isosceles ; let 2 o be the refracting angle of 

 the prism, then drawing A M perpendicular to E F, we have L E A M = o, 

 which being the complement of A E M, is necessarily the angle of 

 refraction; if, therefore, p. be the index of refraction for rays of any 

 given colour, the angle of incidence P, corresponding to a minimum 

 deviation, is given by the equation, 



Sin F n sin a .... (1) 



Also the deviation of the ray at either refraction is equal to the 

 excess of P, the angle of incidence, over the angle of refraction, which 

 in this case is equal to o. If then D be the whole deviation, or the 

 angle between the directions of D E and t o, and i = 2a, the whole 

 angle of the prism, we get from (1) 



D + i i 



Sin = ft Sin g . . . . (2) 



The angles t and D, and consequently the value of n, admit of very 

 exact determination by using a telescope properly mounted, provided 

 we can find in the spectrum an object sufficiently precise to fix on. 

 The transition from one colour into another is far too gradual to allow 

 us to fix on the limit which separates them ; but fortunately we have 

 been put in possession of perfectly definite standard objects by the 

 beautiful discovery made by Wollaston and Fraunhofer of the existence 

 of dark spaces, narrow bauds transverse to the length of the spectrum, 

 and now generally designated Fraunhofer's lines. 



These bands are best observable by forming the spectrum of a lumi- 

 nous line instead of a point, by means of a prism of great purity, and 

 viewing it through a telescope of good magnifying power, though some 

 of them may, when carefully pointed-out, be recognised by the unassisted 

 eye, and after one recognition are in future easily found ; and for naked 

 eye observations a prism of moderate purity will suffice. They are 

 spaces totally deficient of light, of very unequal width, and exceedingly 

 numerous ; it is also to be remarked that these bands, always the same 

 in number and relative position for the same light, are different, or 

 altogether wanting, when the source of light is varied. Thus sun- 

 light, moon-light, planet-light, sky-light, derived from a common 

 source, have the same lines, but several of the fixed stars have dark 

 Hues of their own, while artificial liyhts rarely if ever exhibit dark 



