On Crystallography* 357 



parallel to the axis, is at the same time parallel to a plane 

 which should pass by the great diagonals of the bases of 

 the priniiiive form. T have but a verv few observations ou 

 this subject as yet ; but it is probable that all the substances 

 which have the double refraction fall within one or other 

 bf the fore^goiiig cast-s, which s\vt of themselves the limits 

 'of all the positions which refrangent surfaces may have 

 relatively to the prin^itive ibrm. But as the position pa- 

 rallel to the axis is variable in its turn among several limits, 

 which correspond with the diagonals and the Fides of the 

 bases of the primitive form, it will be requisite to know 

 which of these last limits is that which agrees with every 

 substance. 



F shall explain, when I come to the article emerald, how 

 a mistake led me to these results ; and 1 even confess that 

 T am still in uncertainty as to the refraction of some sub« 

 stances, not having had time sufficiently to multiply my 

 inquiries, in order to ascertain if a crys.tal of this de- 

 scription which presented only a single image of the ob- 

 jects, would not exhibit two alter having been cut in a cer^ 

 tain manner. 



I shall detail, in speaking of every substance, what I have 

 observed with respect to its refraction, and I propose to 

 make some fresh experiments on this delicate point con- 

 nected with minerals, which I have been only able to glance 

 at as yet. We shall find, under the article carbonated 

 lime, the detail of the particular results which [attained re- 

 lative to the double refraction of this acidiferous substance, 

 which suits njore easily than the others this kind of in- 

 quiries. 



Another observation, which will not be altogether useless 

 when we are occupied with generalizing the theory of the 

 phsenomenon in question, consists in this — namely, that 

 all the substances in which the integrant molecule is re- 

 markable for i'.s symmetry have the simple refraction. Of 

 Ihis kind are those which have for their primitive furm the 

 cube, the regular octahedron and the rhomboidal dodeca- 

 hedron. 



As yet we have only subjected to the experiments which 

 concern this object, bodies taken from among those which 

 we coinmonlv designate by the name of stones. I have 

 Extended tht^e experiments to several of those which are 

 Called salis^ as well as to inflammable substances, and to 

 metallic substances oxidized and united to other principles, 

 and I have foe. :i 'hat there is no class of minerals w^hicli 

 lioes not present bodies endowed with double refraction. 



Z 3 There 



