17 



the manner in which, according to this theory, an object viewed 

 through a medium of various densities, producing or rather repre- 

 sented by a curve of contrary flexure, and at the same time a con- 

 tiguous stratum of uniform density, will exhibit three different images; 

 the one through the uniform medium, in its proper place ; the other 

 through the convex part, somewhat higher, but inverted ; and the 

 third refracted by the concave part, still higher, erect, but somewhat 

 smaller. 



Grounded upon these principles, Dr. Wollaston proceeded next to 

 the set of experiments which are the subject of his second section. 

 The first and most explanatory of these experiments was made with 

 a square phial, about one-third filled with clear syrup, and the other 

 third with pure water, the two liquids forming by degrees, at the 

 plane of contact, a thin stratum of decreasing density from the syrup 

 upwards. Here the effect was obviously conformable to the theory, 

 an object viewed through these media being represented to the eye, 

 erect and in its proper place when seen through the syrup of uniform 

 density ; higher and inverted behind the adjacent variable medium ; 

 and still higher and erect behind the upper part of the variable 

 stratum. This effect of varied density was repeated by filling the 

 remainder of the phial with spirits of wine ; when at the plane of 

 contact between the spirit and the water, another variable stratum 

 was gradually produced, which exhibited the same phenomenon as 

 in the former instance. The next experiment proves that a difference 

 of temperature between adjacent strata of the same fluid will pro- 

 duce the same effect. And a third experiment, which may be con- 

 sidered as a corollary, showed that the air round a heated body (a 

 red-hot poker for instance,) will assume the same varied densities, 

 and exhibit precisely the same appearances. 



Under the third head the author observes, that though three images 

 have as yet been rarely seen in the atmosphere, yet this circumstance 

 does by no means invalidate the above theory ; since its appearing so 

 seldom may be well accounted for by the less rarefaction produced by 

 the heat of the sun, than by a red-hot iron, or the artificial means 

 above used. Over water, the evenness of the surface, he says, is fa- 

 vourable to the production of such appearances. 



Some observations are, lastly, added concerning Mr. Huddart's 

 opinion, that the peculiar state of the atmosphere which produced 

 the appearances he witnessed may have been occasioned merely by 

 the evaporation at the surface of the sea condensing the lower strata 

 of the atmosphere. Dr. Wollaston does not altogether accede to this 

 opinion ; but he does not absolutely deny that the cold produced by 

 this evaporation may in some instances occasion a density that may 

 enter as one of the data in the theory above laid down, though other 

 causes, such as the effects of the heat of the sun, currents of air, &c. 

 he thinks must co-operate. To the density, however, produced by 

 mere evaporation, he acknowledges may be ascribed the uncommon 

 elevation of the coast of France, lately observed at Hastings by 

 Mr. Latham ; and some of the appearances described by Professor 



