9B Mr. M. M. Pattisoii Muir on Ckemical Classification, 



But it is not always easy to tell when we are dealing with 

 unsaturated compounds. We have many reasons for believing 

 that the compounds of the members of the nitrogen group 

 which contain three atoms of monad radicle united with one 

 atom of the given element are really unsaturated compounds — 

 that the members of this group are really pentavalent. The 

 formation of saturated or unsaturated compounds will depend 

 largely on the total force exercised by the combining atoms ; 

 and this will be again conditioned by the circumstances under 

 which combination takes place ; so that the mere fact that 

 a compound shows no readiness to take up additional atoms is 

 not a certain proof that the compound is really saturated. It 

 is only from a study of the general characters of the combi- 

 ning atoms, and of the analogies of the compound, that (as I 

 shall endeavour to point out in the sequel) probable conclu- 

 sions can be deduced as to the state of saturation of a compound. 

 Conclusions as to valency drawn from the consideration of 

 unsaturated compounds will naturally be misleading. 



17. There are certain compounds, to all appearance definite 

 chemical bodies, which are incapable of becoming gases with- 

 out undergoing decomposition. Sal ammoniac, for instance, 

 when heated, splits up into ammonia and hydrochloric acid ; 

 but when the mixed gases are allowed to cool, sal ammoniac 

 is re-formed. Compounds which are thus dissociated by heat 

 have been called by Kekule and others molecular compounds, 

 in contradistinction to those atomic compounds which are 

 stable in the state of gases. If the structure of those com- 

 pounds which are thus decomposed by heat is essentially dif- 

 ferent in kind from that of those which are stable at high tem- 

 peratures, it will of course be impossible at present to make 

 use of the former in attempts to solve the general question of 

 valency. Apart altogether from any opinion as to the exist- 

 ence of '^ molecular " compounds, the mere fact that certain 

 bodies are broken up when heated, and therefore cannot be 

 obtained as gases, at once shuts out such substances from the 

 cateo-ory of those compounds which are to be taken into account 

 when attempting to deduce the valency of an element. But 

 inasmuch as many chemists still appear inclined to press solid 

 and liquid compounds into their service when considering 

 valency, and as it appears to me that the errors which are 

 thus certain to be made (in the present state of our know- 

 led tre) can be rendered very apparent by considering the 

 question of '' molecular compounds " in general, I shall endea- 

 vour briefly to describe the two main views which have been 

 advanced concerning these bodies, and to recapitulate shortly 

 a few of the points in favour of each. 



