TRANSACTIONS OF SECTION B. 815 



sodium chlorate, and benzil, produce rotation only when in the crystallised state ; 

 the dissolved (or fused) substances are inactive. Others, like oil of turpentine, caui- 

 plior, and siicar, are optically active when in the liquid state or in s&lution. In the 

 former case the molecules of the substance have no twisted structure, but they unite 

 to form crystals having such a structure. As Pasteur expressed it, we may build up 

 a spiral staircase — an asymmetric figure — from symmetric bricks; when the stair- 

 case is again resolved into its component brii-ks, the asymmetry disappears. (I 

 will explain presently the precise significance of the terms symmetry and 

 asymmetry as used in this connection.) In the case of compounds which are 

 optically active in the liquid state, the twisted structure must be predicated of the 

 molecules themselves ; that is, there must be a twisted arrangement of the atoms 

 which form these molecules. 



The earliest known experimental facts regarding the rotation of the plane of 

 polarisation by various substances, solid and liquid, were discovered by Arago and 

 by Biot. 



After this preliminary statement as to what is understood by optical activity, 

 we may consider Pasteur's special contributions to the solution of the problems 

 involved 



Pasteur tells us, in the well-known ' Lectures on the Molecular Asymmetry of 

 Natural Organic Products,' which he delivered in 1860, before the Chemical 

 Society of Paris, that his earliest independent scientific work dealt with the 

 subject of crystallography, to whicb he had turned his attention from a conviction 

 that it would prove useful to him in the study of chemistry. In order to perfect 

 himself in crystallographical methods, he resolved to repeat all the measurements 

 contained in a memoir by De la Provostaye on the crystalline forms of tartaric 

 acid, racemic acid, and their salts. These two sets of compounds have the same 

 composition, except that they frequently difl'er in the number of molecules of 

 water of crystallisation which they contain ; but whereas tartaric acid and the 

 tartrates are dextro-rotatory, racemic acid and tberacemates are optically inactive. 

 It was probably this circumstance that decided Pasteur in his choice of a subject, for 

 it appears that, even as a student, he had been attracted by the problem of optical 

 activity. In the course of the repetition, however, he detected a fact which had 

 escaped the notice of his predecessor in the work, accurate observer as the latter 

 was — nan\e]y, the presence, in the tartrates, of right-handed hemihedral faces, 

 Avhich are absent in the racemates. Hemihedral faces are such as occur in only 

 balf their possible number ; and in the case of non-superposable hemihedry, to 

 which class that of the tartrates belongs, there are always two opposite hemihedral 

 forms possible : a right-handed or dextro-form, and a left-handed or Isevo-form. 

 Which is right, and which is left, is a matter of convention ; but they are opposite; 

 forms, and difi'er from one another exactly as the right hand of the human body 

 differs from the left : that is, they resemble one another in every respect, except 

 that they are non-superposable — the one cannot be made to coincide in space with 

 the other, just as a right hand will not fit into a left-hand glove. The one form is 

 identical with the mirror image of the other: thus the mirror image of a right 

 hand is a left hand. Such opposite hemihedral crystalline forms are termed 

 enantiomorphs ; they have the same faces and the same angles, but differ in the 

 fact that all positions in the one are reversed in the other for one dimension of 

 space, and left unchanged for the other two dimensions; this being the geometrical 

 transformation which an object appears to undergo when reflected in a plane 

 mirror. Enantiomorphism is possible only in the case of asymmetric solid figures : 

 these alone give non-superposable mirror images. Any object which gives a 

 mirror image identical with the object itself — a superposable mirror image — must 

 have at least one plane of symmetry. 



The hemihedry of the tartrates discovered by Pasteur is in every case in the 

 same sense — that termed right-handed — provided that the crystals are oriented 

 according to two of the axes which have nearly the same ratio in all the tartrates. 

 Pasteur was inclined to connect the molecular dextro-rotatory power of the 

 tartrates with this right-handed hemihedry ; since in the racemates both the 

 hemihedry and the rotatory power were absent. A similar connection, which. 



