August 1 8, 1887] 



NATURE 



371 



satisfactorily explained by the constitutional formula, and 

 are, so to speak, embodied in the formula. 



A constitutional formula is thus founded on reactions 

 and predicts reactions. In this lies its chief valua. A 

 constitutional formula which is not founded on reactions 

 possesses a very slight value indeed. The constitutional 

 formula of a complex mineral silicate, for example, is not 

 an expression of the reactions of that silicate, inasmuch 

 as the silicate has not hitherto been induced to yield any 

 variety of reactions worth mentioning: it is merely the 

 simplest, and perhaps the most symmetrical, way of 

 arranging the component atoms consistently with their 

 valency, and in accordance with certain analogies in the 

 constitution of salts of oxygen-acids. Not even the mole- 

 cular weight of the silicate is known, and this knowledge 

 is the first step towards the construction of a constitutional 

 formula which shall have any great value. But the 

 beginner, who has not always the genesis of the various 

 formulas before his eyes, is apt to put all constitutional 

 for.nulae into one category, and to view all with equal 

 trust or distrust according as his temparament happens 

 to be sanguine or sceptical. 



The chief opponents of constitutional formulii; are to 

 be found among inorganic chemists. Constitutional 

 formulae are essentially a creation of organic chemistry. 

 We have seen that they mainly originated in the necessity 

 of explaining the phenomena of isomerism. Now iso- 

 merism, which is the rule in organic chemistry, is entirely 

 the exception in inorganic chemistry. The constitutional 

 formulae of inorganic chemistry are thus an artificial 

 growth : they are the result of an attempt to transplant 

 ito inorganic chemistry methods and analogies derived 

 tro n organic chemistry, and it cannot be affirmed that 

 these borrowed growths have altogether flourished in the 

 new soil. Where the organic formuUe have guided the 

 chemist through the labyrinths of the various classes of 

 compounds, predicted reactions, laid down the number of 

 possible isomerides, and shown the way to the synthesis 

 of natural compounds so high in the scale of complexity 

 as alizarin and indigo, the same methods applied to in- 

 organic chemistry have led to no tangible result higher 

 than that of checking a few doubtful formulae by means 

 of the laws of valency, The reasons of the failure have 

 already been indicated. But the partial failure of consti- 

 tutional formuUe in inorganic chemistry is hardly an argu- 

 ment against their use in organic chemistry, where they 

 Ivivc achieved the most signal success. 



Up to this point we have regarded the constitutional 

 formula simply as a symbolic device, by means of which 

 reactions and cases of iso nerism may be expressed and 

 predicted. The question now arises : Is it anything 

 beyond this ? A constitutional formula is primarily a 

 ertain definite arrangement of atomic symbols. Is there 

 mything like this atomic arrangement in the molecule 

 'self, or even anything corresponding with it? 



It is in the highest degree improbable that there is any- 

 thing li/:c it in the molecule itself, but quite possibly there 

 is something corresponiin^ with it. That the constitu- 

 tional formula cannot be like the molecule in the sense of 

 being a picture of it is manifest from a variety of con- 

 siderations. To mention one out of several : a constitu- 

 tional formula represents the atoms as points connected 

 wit.i one another in a certain definite way by lines of 

 attraction, without reference to any actual positions in 

 spice which these atoms may be supposed to occupy ; for 

 the sake of convenience they are represented as lying in 

 the plane of the surface on which the formula is drawn. 

 Now the kinetic theory of gases informs us that the 

 atoms within the mnlecule are not to be conceived of as 

 occupying their positions in a state of rest : each executes 

 some form of vibration or rotation. This view is quite 

 compatible with the existence of definite relations of 

 attraction between given atoms within the molecule. To 

 borrow an illustration from astronomy, we might in the 



constitutional formula of acetic acid, for instance, regard 

 the two carbon atoms as the two suns of a double star, 

 and the atoms directly attached to the carbon suns as 

 planets — one with a satellite. The parts may execute 

 their respective motions without disturbing the stability 

 of the whole, any more than the stability of the solar 

 system is disturbed by the motions of its parts. Now, it 

 is evident that a constitutional formula which represents 

 the atoms as motionless in a plane cannot be a true image 

 of the molecule — cannot be li/ce it. 



That the constitutional formula, however, in some way 

 corresp mds with the molecule, is shown, not only by the 

 chemical evidence which we have already discussed, but, 

 what is more important, by a number of physical con- 

 siderations. That the physical properties of a substance 

 are dependent on the arrangement of the atoms within 

 the molecule is evident from the fact that in isomeric 

 compounds the melting-point, boiling-point, specific 

 gravity, and other physical properties generally vary 

 for each isomeride. A comparison of the physical pro- 

 perties of similarly-constituted compounds shows that in 

 many cases very definite relations can be traced between 

 constitution and physical properties. 



Very important information has been gained in this way 

 by studying the behaviour of organic compounds towards 

 light. Thus a number of these compounds when in the 

 liquid state, or in solution, cause the. plane of a ray of 

 polarized light, if passed through them, to turn through 

 a certain angle. It was observed by Le Bel that 

 all such optically active substances contained in their 

 constitutional formulae at least one asymmetric carbon 

 atom — that is, a carbon atom united to four dissimilar 

 atoms or groups ; and an ingenious hypothesis has been put 

 forward by Le Bel, and in greater detail by Van't Hoff, to 

 account for this concatenation. The researches of Glad- 

 stone and Dale, Landolt, Bruhl,and others on the molecular 

 refraction of organic liquids have demonstrated an intimate 

 connexion between the refractive power of a liquid on the 

 one hand and its constitution on the other, so that obser- 

 vations on the refractive power may be employed in 

 ascertaining the constitution of such compounds ; and 

 Perkin has recently shown that the " magnetic rotatory 

 power" of organic liquids — the power which such liq lids 

 possess, when placed in a strong magnetic field, of turning 

 the plane of the polarized ray — may be utilized in the 

 same manner. Again, the selective absorption which 

 organic liquids exercise on light of different wave-lengths 

 is closely connected with the constitution of these liquids ; 

 and the presence of certain organic radicals in the formula 

 of a compound is manifested by certain definite absorption- 

 bands which make their appearance in the photographed 

 spectra of the infra-red (Abney and Festing) and of the 

 ultra-violet (Hartley). 



Other remarkable relations between constitution and 

 physical properties are manifested in what is termed the 

 milecular volume of organic liquids at their boiling-points 

 — a subject first investigated by Kopp, and later by 

 Thorpe, Ramsay, Lossen, and others. By the molecular 

 volumes of compounds are understood the relative volumes 

 which quantities of these compounds taken in the pro- 

 portion of their molecular weight occupy. Kopp found 

 that the molecular volume of a liquid organic compound 

 at its boiling-point is the sum of the atomic volumes of its 

 elements ; and that, whereas the atomic volumes of carbon 

 and hydrogen arc constant, the atomic volume of oxygen 

 varies with its mode of combination, having two distinct 

 values : one value for an oxygen atom attached with both 

 its affinities to the same atom of another element, and a 

 second value for an oxygen atom attached to two different 

 atoms.' Sulphur exhibits a similar definite variation in 

 atomic volume in accordance with the mode of distribu- 

 tion of its affinities. 



' Kopp dist'.njfuished " intra-adical " and " extraradical " oxygen. The 

 above is a re-statement of his views in terms of modem formu'ae. 



