Radiant Heat through Crystals to the direction of Transmission. 19 



In this case, tlie differences, both quantitative and quahtative, 

 in both groups of rays are evident. 



The ease with which plates of mica of any required thickness 

 may be obtained, rendered this body pecuharly adapted to show 

 the manner in which the differences exhibited above were affected 

 by the thickness of the plates. 



In the first place, a plate was chosen of about one-third the 

 thickness of the plates made use of in the foregoing experiments. 

 The pi'oportion of Q to Q', that is, of the quantity transmitted 

 when the plane of polarization was parallel to that of the optic 

 axes, to the quantity transmitted when both planes were at right 

 angles to each other, was as 100 : 107. 



When, however, instead of sending the heat through one 

 plate of mica, it was sent through a number of plates whose 

 total thickness amounted to about six times that of the single 

 plate, under the same circumstances the proportion of Q to Q' 

 was as 100 : 119; a proof that the quantities of the two systems 

 of rays polarized at right angles to each other differ the more 

 from each other the greater the thickness of the crystal through 

 which they have to pass. 



To ascertain the influence of thickness upon the quality of 

 the transmitted rays, M. Knoblauch proceeded as follows : — 

 Having determined the rate of the quantities of heat transmitted 

 direct from the plate of mica to the thermo-electric pile, a dia- 

 thermanous bod};-, a green glass, for example, was introduced 

 between the mica and the pile, and the effects produced when 

 the planes were coincident and at right angles to each other 

 were compared. If the ratio observed, as before the introduc- 

 tion of the glass, showed itself here also, it would be a proof 

 that equal proportions of both groups of rays were transmitted 

 through the green glass. If the ratio were changed, it would 

 show that the two systems of polarized heat had traversed the 

 green glass in unequal quantities, and were therefore qualita- 

 tively different from each other after their passage through the 

 mica. 



It was found, in the case of a single plate of mica, that the 

 proportion of the two quantities mentioned, before entering the 

 glass, was as 100 : 107. After passing through the glass the 

 proi)ortion was precisely the same. After passing through 

 about three times the thickness of mica, the proportion, before 

 reaching the glass, was as 100 : 120; but after passing through 

 the green glass, as 100 : 170. Doubling the last thickness, 

 the proportion before entering the glass was as 100 : 120, but 

 after leaving the glass it was as 100 : 284. 



It thus appears that the difference of transmission of the two 

 differently polarized groups of rays, through the same diather- 



C2 



