M'Keevor's foyage to Hudson's Say. 71 



that of which we have been speaking, depends upon these 

 principles. Dr. Hamilton, of this city, was, it seems, the first 

 who attempted to discover any positive evidence of the elec- 

 trical nature of the aurora boreal is. The only proof, however, 

 which he advances is an experiment of Hawkesbier, by which 

 the electrical fluid is shown to assume appearances resembling 

 the aurora borealis, w|ien it passes through a vacuum. He 

 observed, that when the air was most perfectly exhausted the 

 streams f electrical matter were then quite white ; but when 

 a small quantity of air was let in, the light assumed more of a 

 purple colour. The flashing of the light, therefore, from the 

 dense reg'ion of the atmosphere into such as are more rare, and 

 the transition through mediums of different densities, he con- 

 siders as the cause of the aurora, and of the different colours 

 it assumes. Dr. Halley, and, more lately, Mr. Dalton, have 

 advanced many ingenious arguments in favour of the opinion 

 that this phenomenon depends on the quantity of magnetic 

 fluid existing in the atmosphere, the polarity of magnets hav- 

 ing been observed to be disturbed during its appearance. It 

 has been proved, however, particularly by Perecotte, that this 

 disturbance does not always take place on such occasions; 

 and, as the same circumstance is observed to happen when the 

 atmosphere is in a positive state of electricity, the theory itself 

 becomes highly questionable. Whatever may be the imme- 

 diate cause of this phenomenon, it is evidently connected with 

 the condensation of vapour from the air, as, during their 

 appearance, there always is observed a copious deposition of 

 dew, or hoar-frost. Hence, perhaps, this meteor* is so common 

 in those latitudes Avhere the vericular vapour hangs near to the 

 earth's surface, and when its evaporation and precipitation are 

 slowly taking place. 



The halos which occasionally surround the sun and moon de- 

 serve to be next considered. This appearance occurs only when 

 there is a slight fog in the atmosphere. They rarely accompany 

 the sun, owing to these vapours being so readily dissipated by 

 the calorific rays of that planet. This phenomenon appears to 

 be occasioned by the rays of light striking against a cloud, or 

 body of vapour, which, although considerably uniform and 

 dense, is still so rare as to allow them to be scattered at the 

 point of incidence, and which are thus reflected and refracted, 



* See, on this subject, Robertson's History of the Atmosphere. Dr. Halley 's 

 Philosophical Transactions, No. 347, p. 406. Kirwan's Transactions of the 

 Royal Irish Academy, 1778, p. 80. Franklin's Experiments and Observa- 

 tions, 1769, p. 49. Philosophical Transactions, Vol. xlviii. Part 1. p. 358. 

 Priestly *s History of Electricity. 



