1889.] on an attem'pt to ap])ly to Chemistry, &c. 511 



coal tar stand prominently forward, surpass the synthetical powers of 

 Nature itself. Yet this teaching, as applied to the structure of carbon 

 compounds, is not on the face of it directly applicable to the investi- 

 gation of other elements, because in examining the first it is possible 

 to assume that the atoms of carbon have alw^ays a definite and equal 

 number of affinities, while in the combinations of other elements this 

 is evidently inadmissible. Thus, for example, an atom of carbon 

 yields only one compound with four atoms of hydrogen and one with 

 four atoms of chlorine in the molecule, while the atoms of chlorine 

 and hydrogen unite only in the proportions of one to one. Simplicity 

 is here evident and forms a point of departure from which it is easy 

 to move forward with firm and secure tread. Other elements are of 

 a diflerent nature. Phosphorus unites wdth three and with five atoms 

 of chlorine, and consequently the simplicity and sharpness of the 

 application of structural conceptions are lost. Sulphur unites only 

 with two atoms of hydrogen, but with oxygen it enters into higher 

 orders of combination. The j)eriodic relationship which exists among 

 all the proj)erties of the elements, such, for example, as their ability 

 to enter into various combinations, and their atomic weiights, • 

 indicate that this variation in atomicity is subject to one perfectly 

 exact and general law, and it is only carbon and its near analogues 

 which constitute cases of permanently preserved atomicity. It is 

 impossible to recognise as constant and fundamental properties of 

 atoms, powers which, in substance, have, proved 'to be variable. But 

 by abandoning the idea of permanence, and of the constant saturation 

 of afiinities — that is to say, by acknowledging the possibility of free 

 affinities — many retain a comprehension of the atomicity of the 

 elements " under given conditions " ; and on this frail foundation they 

 build up structures composed of chemical molecules, evidently only 

 because the conception of manifold affinities gives, at once, a simple 

 statical method of estimating the composition of the most comj)licated 

 molecules. 



I shall enter neither into details, nor into the various consequences 

 following from these views, nor into the disputes which have sprung 

 up respecting them (and relating especially to the number of isomers 

 possible on the assumption of free affinities), because the foundation 

 or origin of theories of this nature sufi'ers from the radical defect of 

 being in opposition to dynamics. The molecule, as even Laurent 

 expressed himself, is represented as an architectural structure, the 

 style of which is determined by the fundamental arrangement of 

 a few atoms, while the decorative details, which are capable of 

 being varied by the same forces, are formed l3y the elements entering 

 into the combination. It is on this account that the term " structural " 

 is so appropriate to the contemporary views of the above order, and 

 that the "constructors" seek to justify the tetrahedric, plane, or 

 prismatic disposition of the atoms of carbon in benzole. It is evident 

 that the consideration relates to the statical position of atoms and 

 molecules and not to their kinetic relations. The atoms of the 



