I OF ARTS AND SCIENCES. 303 



least, a distinction was not always observed between the vari- 

 ous classes of phenomena which relate to optical meteorology. 

 1. There are those which depend on reflection alone. 2. 

 There are those which depend on refraction alone. 3. There 

 are those which result from the combined effect of reflection 

 and refraction. 4. There are those which depend on diff'rac- 

 tion, or the interference of light. Rainbows belong to the 

 third class. Halos, properly so called, belong generally to the 

 second ; but in their more complicated forms, when accom- 

 panied with streams of light and mock suns or moons, they 

 belong to the third class. Coronas, properly so called, belong 

 to the fourth class. Halos and coronce are frequently con- 

 founded together, although they have each its own very 

 decided characteristics. Halos are produced by refractions, 

 either with or without reflection. The ordinary value of the 

 diameter of these circles, which is either 47° or 94°, has sug- 

 gested the theory that they are produced by refraction in 

 prisms of snow or ice, in which the refracting angle is 60°. 

 The arrangement of the colors is prismatic, the red occupying 

 the place of the smallest circle. Experiments with a polari- 

 scope show that the light of the halo is polarized by refraction. 

 The crystals which produce the refraction are supposed to 

 exist in the region of the cirrus cloud, so that the halo is often 

 taken as evidence of the first approach of that cloud, even 

 when the cloud itself cannot be distinctly seen. Mr. Lovering 

 exhibited various specimens of plates of fibrous crystals, cut 

 perpendicular to the fibre, which produce upon light changes 

 very analogous to those wrought on a larger scale by the at- 

 mosphere in the production of halos, and which serve, there- 

 fore, to give an idea of the structure of those clouds which 

 develop these extraordinary optical phenomena of the air. 

 Volume XVIII. of the Journal of the Royal Polytechnic School, 

 published at Paris in 1847, was also exhibited. M. A. Bravais 

 has occupied the whole of this volume in an analytical discus- 

 sion of halos, and has enriched his work with copies of the 

 most distinguished appearances of this kind on record, se- 

 lected from all the journals of Europe and America. 



