refractive and dispersive Powers of the Atmosphere, $c. 209 



It may be noticed that the alternating colours observed in 

 the fluctuation, and recorded in the 5th column of the table, 

 are cceteris paribus most observable when the stars are near 

 to the horizon ; and I have observed that the maximum of the 

 intense red colour which distinguishes the alternate change 

 happens at the elevation of about 10 degrees above it. 



The phaenomena noticed by Kepler in the new star dis- 

 covered by him in 1604 in Serpentarius, and now lost, if rightly 

 recorded, afford a striking instance of the converse of what 

 we usually observe. For in that large and memorable star, 

 the colours were continually changing when the star was high, 

 but when it got near to the horizon it was uniformly •white. It 

 is much to be wished that observations had been made on 

 the changeable stars and on those that are now quite lost. 

 The Stella mira may afford interesting observations. Comets 

 ought also to be very accurately observed. The great comet 

 of j 680, which is expected again in 1833, will probably be the 

 subject of much curious research. 



The ordinary distribution of stars into the 1st, 2d and 3d 

 degrees of magnitude, and so on, though it may serve com- 

 mon purposes, is nevertheless very imperfect; the stars re- 

 corded as being of the first magnitude differing as much among 

 themselves as some of them do when compared with those of 

 the reputed 2d magnitude. In the above table of large stars, 

 I have endeavoured to arrange them in the order of their ap- 

 parent size and intensity of light, the best way of ascertaining 

 which is by observing the relative degrees of light in which 

 each can first be seen, either at evening or morning, measured 

 by a photometer. But it is evident that this plan will include 

 also their brilliancy, which is a thing quite different from their 

 apparent size. Some small stars have a more piercing light 

 than others for their size, as the Pleiades, for example. In 

 order therefore to ascertain and compare the relative apparent 

 size alone, we must trust a good deal to our own judgement. 



It is my intention to pursue the above subjects, and to make 

 as accurate comparisons as I am able between the light of 

 individual stars, dispersed by means of prismatic glasses 

 adapted to telescopes, and the light of various combustible 

 substances in a state of ignition viewed through prisms. 



There have appeared to me to result some very extraor- 

 dinary phenomena, from the adaptation of a prismatic glass to 

 different parts of the telescopes: but as I cannot account for 

 them on any known principles, I only state the fact as a hint, 

 in order that others may repeat the experiments. Finally, we 

 must ever bear in mind that the retina, and the cerebral parts 

 in connection with it, are a necessary part of the optical ap- 



VoL 63. No. 311. March 1824. D d paratus 



