PheHimitia of the Images through Iceland CryJtaK T99 



ef different (hapes and fizes were employed. The faoic things happened by candle-light, 

 and alfo by refraiSted homogeneal light. In fliort, I repeated moft of my experiments oa 

 flexion with Iceland cryllal, and found that they were nt>t changed at all in their relults. 



Obfervaiioh 4. Having great reafon to doubt the accuracy of an experiment tried by Mf, 

 Martin, and in which, by a prifm of Iceland cryilal, he thought fix fpedlra were produced, I 

 was not much furprifed to find that a prifm made by polifliing the two contiguous fides of a 

 parallelopiped of Iceland cryftal produced only two equal and parallel images, in whatever 

 pofuion the prifm was held. But though, from the imperfedl account which Martin gives of 

 this appearance, it was impofllble to difcover his error from his own words, yet chance led 

 tne to find out what moft probably bad mifled him ; for, looking at a candle through the op- 

 pofite fides of a fpecimen of Iceland cryftal, I faw four coloured images (befides two white 

 ones) of the candle. Thefe were parallel to one another, and in the fame line as reprefented in 

 iig. 7, plate VII. where E reprefents the two regular images, G and F two others coloured' 

 very irregularly, and changing colours as the cryftal was moved horizomally, fometimes ap- 

 peared each two fold, and its two parts of the fame or different colours. A and B were regular-^- 

 ly coloured, and evidently formed by refradlion, and refieded back from the fides. On turn- 

 ing the cryftal round fo that its pofition might be at right angles to its former pofition, . 

 the images moved round, and were in a line perpendicular to A B, as C D. All this hap- 

 pened in like manner in the fun's rays ; and On viewing the fpecimen I found it was fplit 

 and broken in the infide, fo as to be lamellated in direclions parallel or nearly fo to the fides : : 

 ©n thefe plates there were colours in the day-time by the light of the clouds j and it is- 

 cvident that it was thefe fradtures which caufed the irregular images G and F, for other 

 fpecimens {hewed no fuch appearance. 1 would therefore conclude, that Iceland cryftal 

 feparates the rays of light into two equal and fimilar beams by refraiflion, and no more*. 



As to the caufe of the reparation, 1 would hope that fome information may be obtained 

 from the experiments I have related : for from them it appears, that this fingular property 

 extends no farther than to the adtion of the particles of Iceland cryftal on the particles of 

 light in their paffage through the body ; and from Obf. 4, it is farther evident, that it is not 

 owing to the different properties which Sir Ifaac Newton conjedures the different fides of 

 rays to have; for, if this were the caufe, when the rays pafs between two pieces of cryftal 

 an uncommon flexion would take place. Laftly, another fa£t (mif-ftated by Bartolinf and 

 Rome de Lifle:];) fhews that the unufual refra£5:ion takes place within the body, while thc. 

 «ther, like all refraftions, begins at fome finall diftance before the rays enter. . 



The writers juft now quoted affert, that if the cryftal be turned round fo as to aiTumc 

 different pofitions, there is one in which the line appears fingle. The fa£t is very different, , 



* Mentioning this account of Martin's miftake to Profeflbr Robifon of this univerfity, I was pleafed to find 

 a full confirmation of it. It was that excellent philofopher who fliewed tlie appearance to Martin ; but he no: 

 vnderftanding it, took the liberty of publilhing the obfervation as his own, after firft mangling it in fuch a way- 

 it to give him indeed fome pretext for the appropriation. The Profeflbr merely mentioned his having com- 

 municated k to Mr. Martin: ho\v the latter ufed it we have fhewn in the text- the theory of the appearance is 

 fomewhat more complex than appears by my obfervations. I was therefore pleafed to find that the Profeffiir was 

 in poflcffion of the true account of it, which is however foreign t9 the ptefcnt purp ofe. B^ 



j- Experimenta Cryftalli abridged in Phil. Tranf. V0I..V1 



J CriiiallcgraphJe, voLi.. 



