June 1st, 1SS7.] 



SCIENTIFIC NEV/S. 



87 



is represented in this diagram ; there is therefore no dis- 

 persion or splitting up into different coloured rays. 



We may stretch the analogy further by supposing that 

 some of the men are smaller, and take shorter and quicker 

 steps than others, and we may imagine the retardation to 

 be due to deep heather or stones. If these obstacles are 

 not extremely small compared with the size of the men, 

 it is clear that the smaller men will suffer most, and will be 

 most retarded, and, supposing them all to be marching 

 blindly, they will deviate most from the original direction. 



Bearing in mind that it is the wave front of the ether 

 vibration which we compare to the ranks of men, not the 

 vibrations to the individual men, this is analogous to what 

 occurs with light. 



Cauchy has shown that the separation of the different 

 coloured rays of light — or dispersion — may be accounted for 

 if we assume that the wave length of the vibration of light 

 is not enormously greater than the size of the molecules or 

 ultimate particles of matter. 



Taking a wave length at about one-5o,oooth of an inch, 

 it is probably not more than 10,000 times greater than a 

 molecule, which would be therefore one-5oo,ooo,oooth of 

 an inch, and this agrees fairly with estimates arrived at 

 from several totally independent sources. 



It is conceivable that the smallest men would be retarded 

 in greater proportion than that of their height compared with 

 men of medium stature ; and we find a corresponding 

 effect in the great deviation of the violet end of the spec- 

 trum when produced by a prism, compared with other 

 methods which will be described hereafter. 



By deviation is meant the refraction of a ray of any one 

 colour. By dispersion is meant the difference of the re- 

 frangibility, or capability of being refracted, of different 

 coloured rays. The larger or blunter the angle of the 

 prism up to a certain point, the greater is the deviation, and 

 correspondingly greater is the dispersion. But different 

 kinds of glass have different powers of refraction and of 

 dispersion. Heavy glass has a greater effect on the path of 

 a ray of light than light glass. Flint glass has greater dis- 

 persive power than crown glass. The dispersion of various 

 qualities of glass differs not only in degree, but in the 

 manner in which one colour is refracted compared with 

 another. 



There is another means of producing a spectrum which, 

 though at present not so common, has many important ad- 

 vantages ; the foremost of which are that the position of each 

 colour corresponds exactly to its wave length, and that a 

 scale of wave lengths can be easily applied to the spectrum. 

 There is no extension or spreading out of the violet, and the 

 colours are invariably in the same position with regard to 

 each other. 



This method is that of dijfractiou, and to this we owe the 

 colours of mother-of-pearl, and most of the hues of the 

 peacock's feather. 



Instead of passing a beam of light through a prism, it is 

 allowed to fall on a flat or curved piece of polished metal, 

 on which a number of equidistant parallel lines are 

 engraved. From 4,000 to 8,000 lines to the inch are 

 required to produce a good effect, but as many as 112,000 

 lines to the inch have been ruled for this purpose. Such a 

 plate is called a grating. 



In fig. 3. a section of such a grating is represented very 

 roughly at B D. A ray of light of a certain colour from A, 

 falling on the grating at B, is reflected in all directions in 

 the plane of the paper ; only two of these directions will be 

 considered at present, B E and B F. 



For the particular colour there is a certain wave length, as 

 has already been described in these columns. There will be 

 a certain number of wave lengths in the length E B. Let 



another ray of the same colour from C fall on the grat- 

 ing at D, and be reflected in all directions. Let D E and 

 D F be two of these. The wave lengths are roughly repre- 

 sented by the breaks in the lines, and with more distinct- 

 ness than by drawing the diagram with sinuous or wavy 

 lines. A short line and the following white space may 

 be taken to represent one wave length. It will be found 

 that at E, the waves coming from B and D coincide, but 

 at F they have, as it were, got out of step by half a wave 

 length. The result will be light at E, and darkness at F. 

 It is clear that the vibrations help each other at E, because 

 the lengths E B and E D contain an exact number of wave 

 lengths, differing in this case by one between K and D ; but 

 if the ray of light were of another colour the vibrations would 

 not coincide at E, but would be more or less out of step ; 

 there would be therefore little or no light. 



The position E therefore represents the full effect of light 

 of one colour, and in the same way, at other positions on the 

 line E F, the full effect of lights of other colours will be 

 found, and, in short, a spectrum will be displayed. In prac- 

 tice the spectrum is not thrown on a screen at E F, but is 

 observed through a telescope ; the same means are used for 

 observing a refraction spectrum. 



If the angle E B A is measured, and the distance B D 

 is known, the length K D can be calculated from the angle 

 G B H. By this method it is easy to measure the wave 

 length of a ray of light of any particular colour. 

 (To he continued.) 



A TRAVELLING PLATFORM. 



TO lessen the fatigue of visitors to the laternat'onal 

 Exhibition to be held in Paris in 1SS9, M. Eugene 

 Henard has suggested the use of an endless train or 

 moving platform, as shown in the accompanying illus- 

 tration. Fig. I. His idea is to have the platform on the 

 ground level, and to let it travel slowly, so thit any one can 



Fig. I. — Endless Moving Platform. 



step on or off" without the train being stopped. He also 

 proposes to let visitors use the train gratuitously, and to 

 cover the working expenses by having seats and elevated 

 cafe's, for the use of which a charge will be made. He be- 

 lieves that many persons will use these, as they can then 

 make a complete tour of the exhibition grounds without 

 fatigue. The proposed train is to consist of low railway 

 trucks carrying a simple timber platform, the rails being 

 laid in a trench, as shown in section in Fig. 2. The total 

 length of the proposed line is 2,080 metres. The horse- 

 power required is estimated at 640, and it is proposed to 



