62 



<iolours seen by him and a friend in Kent in 184^1, thns : — 

 " Bands of colour not in circular form around tlie sun, follow- 

 ing the sinuosities of the clouds. The colours commencing 

 from the white area forming the interior and proceeding 

 outwards to its edge were — 1st, white ; 2nd, very pale pink ; 

 v3rd, blue green, a very strong colour ; 4th, at the edge purplish 

 pink considerably intense, beyond which pure blue sky. The 

 «ame tints were on other clouds having no reference to the 

 more or less proximity to the sun, but depending on the thick- 

 ness of the cloud or the length of the path within it traversed 

 b}^ the visible rays. In fifteen minutes the bands grew broader 

 and tints stronger, and a tendency to form a corona round 

 the sun." " It seems imjDossible," he says, "to regard these 

 colours other than the resultants of the superposition of a series 

 of interference fringes following a regular progression of 

 breadth due to an increasing size in the drops from the 

 exterior to the interior of the clouds." 



Forbes has shown that steam in the act of expanding, and 

 while still transparent, is highly absorptive of the violet, blue, 

 and a portion of the green rays of the spectrum. Supposing 

 the vapour only in the form it had on the outside of these 

 clouds, the red would reach us when the other tints had become 

 faint, and yet cloud forms not exist definitely. 



We have also the well known fact of hygrometric rela- 

 tions of air to its temperature, and also the different tints 

 the air has before clouds form and as soon as they do form 

 — the depth of air being seen through the numerous gaps. 

 Thus the pure ultramarine is most marked after cb)uds form. 

 And before they appear at certain temperatures the sky has a 

 greenish blue appearance. If the upper portion of the atmos- 

 phere were at a slightly higher temperature, and if the radiation 

 of heat by the earth had increased we might expect transparent 

 vapour at a higher altitude than usual, or at altitudes in 

 greater quantity than previousl}''. There is no necessity for 

 assuming the quantity to be great, as if so it would cause 

 precipitation to suit the ordinary temperature. The constitu- 

 tion of this vapour would certainly change, and we might 

 expect it to form into its approximate condition to watery 

 vapour (like air charged up to the highest dew point) and to 

 reverse its phase about sunrise with the reverse order of change 

 in temperature. 



There are further observations tending strongly to prove 

 that the cause lies in some form of water, as I shall imme- 

 diately show ; but I would just refer to several difficulties, if 

 we assume that this water came from space. Thus, if it con- 

 sisted of a small- sized cloud of vapour or ice the probabilities 

 against it striking the earth would' be enormous. If it were a 



