of the Production of Colours. 295 
observed, the blue and violet light is more dispersed by re- 
fraction, than in proportion to the difference of the appropriate 
dimensions deduced from the phenomena of thin plates. Hence 
it happens, that when a line of the light proceeding to form an 
image of the rings of colours of thin plates, is intercepted by a 
prism, and an actual picture is formed, resembling the scale de- 
lineated by Newton from theory, for estimating the colours of 
particles of given dimensions, the oblique spectrums, formed by 
the different colours of each series, are not straight, but curved, 
the lateral refraction of the prism separating the violet end 
more widely than the red. The thickness corresponding to the 
extreme red, the line of yellow, bright green, bright blue, and 
extreme violet, I found to be inversely as the numbers 27, 30, 
35, 40, and 45, respectively. In consequence of Dr. Wollaston’s 
correction of the description of the prismatic spectrum, com- 
pared with these observations, it becomes necessary to modify 
the supposition that I advanced in the last Bakerian lecture, 
respecting the proportions of the sympathetic fibres of the 
retina; substituting red, green, and violet, for red, yellow, and 
blue, and the numbers 7, 6, and 5, for 8, 7, and 6 . 
The same prismatic analysis of the colours of thin plates, 
appears to furnish a satisfactory explanation of the subdivision 
of the light of the lower part of a candle : for, in fact, the light 
transmitted through every part of a thin plate, is divided in a 
similar manner into distinct portions, increasing in number with 
the thickness of the plate, until they become too minute to be 
visible. At the thickness corresponding to the ninth or tenth 
portion of red light, the number of portions of different colours 
is five; and their proportions, as exhibited by refraction, are 
nearly the same as in the light of a candle, the violet being the 
