354 POLARIS ATIOX. 



under all circumstances, exhibit the colours of a diametral 

 section, if the surrounding medium is of approximately equal 

 density. 



The middle zone, however, acts in the longitudinal view, as 

 before remarked, like two superposed plates of crystal. Observation 

 can therefore, at most, decide whether the axes of elasticity 

 of the longitudinal and latitudinal directions correspond, or 

 intersect obliquely ; and, in the latter case, whether the middle 

 line of the acute angle, which the major axes together 

 form, lies longitudinally or latitudinally, and which of the two 

 ellipses is the upper one. After what has been noted above 

 (pp. 348 349) on this point, no further explanation is needed to 

 show that these data are in general insufficient for constructing 

 the ellipsoid of elasticity. It is nevertheless important to 

 test, one after another, the different cases which observation 

 may afford, and to determine for each individual case the 

 results obtained. 



ment of this kind gives of transverse sections of cylinders, is produced by cones 

 of light whose aperture corresponds with that of the diaphragm that is, of the 

 polarising Nicol. The total effect produced by such a cone, as regards double 

 refraction, is approximately the same for the eye as if the differently inclined 

 rays had traversed the object in the direction of the axis of the Microscope. 

 A plate of crystal, whose optic axis is perpendicular, thus in reality exhibits 

 no trace of double refraction. The action is quite different of the special 

 polarising apparatus of Noerrenberg, Dove, &c., which are sometimes 

 designated as Microscopes, though incorrectly so. The real image which is 

 here formed, and viewed through the eye-lens, is not the image of the 

 object under examination ; the rays which intersect in the plane of the image 

 represent on the contrary, when produced backwards, a pencil of parallel rays, 

 which traverses the object at an inclination to the axis of the Microscope the 

 greater in proportion to the distance of the point of intersection from this 

 axis. In that point of the image, therefore, incident rays of definite inclination 

 produce interference, and upon this circumstance is based the formation of 

 rings as they are observed in plates of calc-spar cut perpendicularly to the 

 axis. 



The " cross of the first order" occurs only in the polarising apparatus, but 

 not in the polarising Microscope (very low amplifications except ed). Valentin 

 commits the mistake in this as in other points (<'.//., combination with a selenite 

 plate, determination of the axial direction, &c.) of basing his theoretical 

 explanations upon the action of a special polarising apparatus, and then 

 applying the results he has obtained without further consideration to obser- 

 vations with the polarising Microscope. It is manifest that such a method, 

 combined with the arbitrary assumptions mentioned at the outset, must lead 

 to error. 



