304 On the double Images caujed by Atmofpherical RefraBion. 



It muft be remarked, that the vertical pofition of the furface contributed gieatly to encreafe 

 tic cffeft; for, fince the heated air rifes in the direftion of the furface, its afcent has in 

 this cafe no tendency to blend it with the adjacent denfer ftrata, and hence very different 

 degrees of denfity take place in the thicknefs of ^ of an inch j fo that, as the increments 

 of denfity are great, the refradlions will be proportionally fo ; but where the heated furface 

 is horizontal, the afcent of the rarefied air into the fuperincumbent denfpr ftrata renders 

 the variations far more gradual ; confequently, a heated furface of far greater extent muft 

 be requifite to produce equal refraftion. 



However, over extenfive plains, when the fun fliines, fome degree of inverfion is very 

 frequently to be feen ; but the inverted images are rarely well defined, unlefs over a 

 very even furface. One of the beft fituations for this purpofe is over a level open road, 

 with a gentle breeze blowing acrofs it. A current of air brings a cool ftratum more 

 clofely in contaft with the heated furface } and, fince refraftion depends on the increment 

 or difference of denfity in a given fmall pace, a very moderate breeze will thereby 

 render inverfion more perceptible j but a ftrong wind will reduce the temperature 

 of the furface, and may make the heated ftratum too thin for any objeft to be feen 

 through it from a diftance. 



In one inftance, when I faw a refraction of about 9 minutes, at the diftance of about 

 •f of a mile, a thermometer in the fand was loi"; at 4 inches above, 82"> ; and, at i foot 

 above, 76°. 



Over water, the evennefs of the furface is favourable to the produ£lion of fuch ap- 

 pearances ; but, fince the adtion of the fun is weak on a body fo tranfparcnt, a far greater 

 extent of furface is requifite to produce any perceptible inverfion. 



Being at Bognor one bright morning, when the fea was calm, I had an opportunity of 

 obferving the appearance of Selfea Bill, about fix miles diftant. The whole extent of 

 coaft, when viewed with a pocket telefcope magnifying about fixteen times, appeared 

 inverted from one end to the other ; and the lower part of a brick houfe upon the fliorc 

 was feen as diftin£l as the houfe itfelf. I judged the quantity of refradion, in this cafe, 

 to be about two minutes of a degree. 



This ftate of atmofphere appears to be not very uncommon ; for, at Shanklin Chine, 

 in the Ifle of Wight, a few days preceding, fimilar appearances were vifible in feveral 

 dircftions, but I neglefted to make any cftimate of the quantity of refraftion. 



In the inftance of the inverted vefTel feen by Mr. Huddart, (Phil. Tranf. for 1797, 

 Tab. I; Fig. 3.) at the diftance of eight miles the refraftion feems to have been about 3'. 



All the appearances defcribed by him, I am inclined to think, arofe from difference 

 of temperature alone. He offers a conjedure, that evaporation might occafion the lower 

 ftrata of the atmofphere to have a weaker refradive power; but from the following ex- 

 periments it feems to have a contrary eflFedt : 



Eiiper. 7. I took a plate of glafs, and, while I looked along the furface, I poured 

 upon it a ftnali quantity of ether. 



A line 



