PRESENT PROBLEMS OF TNORGANKJ CHKMTSTKY. 213 



niolefiilcs of inorganic compounds, which arc iisiuilly solid, forms 

 another bar to invcsti<2:ation. It is clear that sulphuric acid, to choose, 

 a conunon instance, possesses a very complicated molecule, and the 

 fused nitrates of sodium and potassium are not correctly represented 

 by the sim[)le fonnuhc NaNOg and KNO.,. Any theory of the struc- 

 ture of their derivatives must take such facts into consideration; but 

 we appear to be getting nearer the elucidation of the molecular 

 weights of solids. Again, the complexity of solutions of th" most 

 conunon salts is maintained by many investigators. For example, a 

 solution of cobalt chh)ride, while it undoubtedly contains, among 

 other constituents, simple molecules of CoCl,, also consists of ions of 

 a complex character, such as (CoCl,)". And what holds for cobalt 

 chloride also undoubtedly holds for many similai- compounds. 



In determining the constitution of the compounds of carl)on, stereo- 

 chemistry has played a great part. The ordinary structural formulte 

 are now universally acknoAvledged to be only pictorial, if, indeed, that 

 word is legitimate. Perhaps it would be better to say that they are 

 distorted attempts at pictures, the drawing of which is entirely free 

 from all rules of perspective. But these formulfe may in almost 

 every case be made nearly true pictures of the configuration of the 

 molecules. The benzine formula, to choose an instance which is by 

 no means the simplest, has been shown by Collie to be imitated by a 

 model which represents in an unstrained manner the behavior of 

 that body on treatment with reagents. But in the domain of inor- 

 ganic chemistry little progress has been made. Some ingenious ideas 

 of the geologist Sollas on this problem have hardly received the atten- 

 tion which they deserve. Perhaps they may have been regarded as 

 too speculative. On the other hand, Le Bel's and Pope's proof of the 

 stereo-isomerism of certain compounds of nitrogen. Pope's demon- 

 stration of the tetrahedral structure of the alkyl derivatives of tin, 

 and Smiles's syntheses of stereo-isomeric sulphur compounds give us 

 the hope that further investigation Avill lead to the classification of 

 many other elements from this point of view. Indeed, the field is 

 almost virgin soil, but it is well worth while cultivating. There is 

 no doubt that the investigation of other organo-metallic compounds 

 will result in the discovery of stereo-isomerides ; yet the methods of 

 investigation capable of separating such constituents have in most 

 c-ases still to l)e discovered. 



The number of chemical isomerides among inorganic compounds 

 is a restricted one. AVerner has done much to elucidate this sub- 

 ject in the case of complex ammonia derivatives of metals and their 

 salts; but there appears to be little doubt that if looked foi-. the 

 same or similar phenomena would be discoverable in compounds 

 with much simpler formulse. The two forms of So.., sulphuric an- 

 hydride, are an instance in point. No doubt fornuition under dif- 



