24 On the apparent Changes 



transferred to intermediate points of AFBG, and the luminary 

 appears, with a smaller vertical, and extended horizontal 

 diameter, of the form of two half ellipses combined on the same 

 major axis, the lower considerably more eccentric than the 

 upper. All rays not passing as above, will be dispersed and 

 lost, or stopped and extinguished, and in an undulatory state 

 of the strata of vapours, the observed undulatory changes of 

 outline and limb, particularly of the lower limb will be pro- 

 duced. All these rays, vertical as well as horizontal, in passing 

 on to the eye of the spectator, will not only be thus inflected 

 and bent, but will be variously distributed, and dispersed into 

 various colours, by the first and successive orders of particles 

 by which they pass, and not only divided, but by degrees en- 

 tirely separated from the rest of the direct light, in the order 

 of colours from blue to red. First, all the blues will be dis- 

 persed, and separated from the rest, and scattered over, and 

 variously reflected by the whole atmosphere, giving it, when 

 seen free from clouds, the usual coerulean blue, in the manner 

 described in a paper on the Colours of Waters, in the 9th 

 Number, Vol. V., p. 81, of the Royal Institution Journal, and 

 never before clearly accounted for. After this separation, the 

 colour of the luminary becomes yellow, until by the increased 

 action of the denser strata of lower vapours, into which it 

 descends, the yellow is entirely separated and dispersed after 

 gilding with its colours the lower surfaces of the horizontal 

 morning and evening clouds, leaving the sun of a bright red 

 sustainable by the eye, and of a lustre continuing to decay, as 

 long as the orb continues to be seen. 



The phaenomena, thus dependent upon the vapours of the 

 atmosphere for existence, will vary also with these vapours, 

 their quantities contained in air, and their states of perfect or 

 partial solution therein, of more or less absolute separation up 

 to that of rapid precipitation in the form of drops of water. 

 The hygrometer fitted to determine and to measure these 

 changes, of numbers of particles, and of condition of air de- 

 pending thereon, would obviously be the instrument to be 

 used in observations, rather than the barometer and thermo- 



