lift Dr Young's Theory of the colours observed 



The same figure shows the case of obUque reflexion, the 

 rays proceeding from F and reaching the eye at C by reflex- 

 ion. If from the mean of the common perpendiculars AD> 

 BE, we take away equal parts in the paths of the rays FAC, 

 FBC, the tint will depend on the difference between the se- 

 micircles AE, and DB, which may be easily replaced by the 

 sines of the angles in the expression 



AE — DB = wzX. 

 The difference which exists between the colours and systems of 

 grooves and those produced by narrow apertures, though 

 the same principles of interference apply for the two cases, will 

 become more sensible if we observe in Fig. 1, that if AB is a 

 narrow aperture, for which BG = >^ the eye will not receive 

 from this aperture a single ray, while in the system of grooves, 

 it is from this part which the eye receives most intensely this 

 species of light. Hence we must not refer the colour of the 

 interval AB of the grooves to the transparent part AK con- 

 sidered as a narrow aperture, since the tint which would arrive 

 at C ought to vary with the width AK of the aperture, which 

 it does not do. 



If any additional proof was required that light has no ten- 

 dency to propagate itself according to any determinate direction, 

 it might be easily found in the great obliquity of the rays 

 transmitted by grooves in relation to the plane of the system. 

 If we receive on MN, Fig. 2, a solar ray FA, the eye being 

 placed at C, we may make the rays FA so oblique to MN, 

 and the direction AC so inclined on the other side to the same 

 plane, that the rays FA almost parallel to MN will be obliged 

 to take a direction nearly retrograde to arrive at C, which al- 

 most completely verifies the hypothesis of Huyghens. 



AiiT. XVII. — Theory of the colours observed i7i the ewperi- 

 ments of Fraunhofer. By Thomas Young, M. D. F. R. S. 



As this Journal is the only English work in which a very full 

 account of the discoveries of Fraunhofer have been published, 

 we hasten to lay before our readers the following paper by Dr 

 Thomas Young, which at the present moment derives a fresh 



