66 



this applying especially to the time when the display commences. 



On the other hand, the observations seem to afford most 

 conclusive evidence that the variable element water is, and 

 that dust is not, the cause for these magnificent displays. 

 Professor Tyndall, in his experiments into the formation of 

 clouds by the action of light when scrupulous freedom from dust 

 ivas ensured, found sometimes as many as four reversals of the 

 plane of polarization, which may at times be also seen in the 

 sky when cirrus clouds are forming. He showed that the 

 position of the neutral point depended upon the degree o£ 

 attenuation of the fumes or vapour employed. He showed 

 that the effect of introducing various fumes among the common 

 air motes did not change their character of polarization, simply- 

 making it more bright, whilst he showed that various vapours 

 did alter the plane of apparent polarization. I liere quote 

 portion of his remarks wherein he says : — " But when a puff of 

 aqueous cloud or of the fumes of hydrochloric acid, hydriodic 

 acid, or nitric acid is thrown into the beam there is a complete 

 reversal of selenite tints. Each of these clouds twists the 

 plane of polarization 90°." Then again, his experiments on 

 the Alps show that the alpen glow exhibits the phenomena of 

 apparent cross-polarization, whilst the tints he describes must 

 have equalled in beauty if not in extent those of the crepuscular 

 glimmer. The glows in the locality mentioned could not have 

 been due to dust, for his experiments on bacterial life go far 

 to show that no dust was carried into those localities, and 

 irrespective of this the tints are observed at the times when 

 vapour would most likely be condensed. 



The fact observed by Arago that the light of the halos is 

 partially polarised in planes tangeutial to their circumference 

 seems to show that they consist of refracted light, whilst the 

 light of the rainbow, being completely polarised in radial 

 planes, shows that water in various forms is competent to fulfil 

 all conditions required in the crepuscular glimmer. 



There are certain relations to the barometer and to tempera- 

 ture which I should be happy to communicate at some future 

 time, the points not thoroughly dealt with for want of time 

 being as follows : — The streamers ; their angles, positions, and 

 colours. Dew point; its relations. Thebarometer; its relations. 

 The red glow on the dark side of the sky, and the order of 

 colours then coming into view. The ordinary twilight existing 

 often as long as the red tint, stars being visible through both. 

 The clouds being seen as dark objects on high background any 

 evening whether there be red tints or not. The streamers 

 never occuring when clouds are absent. Their existence when 

 there has only been the yellow sunset tints, the rays having 

 similar tints. And various other points all favouring the 

 vapour theory. 



