Oct. 2S, 1880I 



NA'j' ORE 



609 



and the molecule will be graphically represented as 



But we also know of a compound the 



^\ /' 



H^C-C^H 



molecule of which contains two carbon, but only tour 

 hydrogen atoms, this is represented as h/^""*^\H' ^^^ 

 a third hydrocarbon, CoH,, is represented as H— C=C— H. 

 In the first molecule the carbon atoms are commonly said 

 to be " singly-linked," in the second " doubly-linked," 

 and in the third "trebly-linked." We do not as yet 

 attach any definite physical conception to these phrases ; 

 a compound said to contain " singly-linked " carbon atoms 

 is, as a fact, incapable of combining with hydrogen or 

 other monovalent element, whilst a compound said to 

 contain " doubly-linked "' carbon atoms can combine with 

 two monovalent atoms for each pair of doubly-linked 

 carbon atoms it is represented as containing ; and a 

 compound said to contain " trebly-linked " carbon atoms is 

 capable of combining with four monovalent atoms for 

 each pair of trebly-linked carbon atoms in the graphic 

 formula thereof. 



These are instances of isomerism said to be due to 

 differences in the linking of the atoms of the isomeric 

 molecules. But according to the generally accepted 

 theory isomerism may arise among hydrocarbons in 

 which all the carbon atoms are singly-linked ; such 

 isomerism is due to different relative arrangements of 

 parts of the molecule. We may suppose all the carbon 

 atoms arranged in a chain, or we may suppose ramifica- 

 tions of these atoms; thus the hydrocarbons represented as 

 H H H 



H H H H H-C-C-C-H 



I'll , III 



H— C— C— C— C— H and ^t I h would be 



I I 1 I 

 H H H H 



H I H 

 H— C— H 



H 



isomeric. 



Thomsen deals only with isomerism due to differences in 

 the Hnking of atoms. If from a certain number of dis- 

 sociated carbon and hydrogen atoms a compound be pro- 

 duced containing only " singly-linked " carbon atoms, 

 that compound is not capable of taking up any more 

 hydrogen ; but if a compound be produced containing 

 " doubly-linked '' carbon atoms, that compound is capable 

 of taking up more hydrogen. But in the act of combining 

 with more hydrogen, heat will be evolved ; hence the heat 

 of formation of the first compound is greater than that of 

 the second. The heat of formation of an isomeric com- 

 pound containing "trebly- linked" carbon atoms would 

 be less than that of either of the preceding. 



Thomsen, from the results of his own and other 

 experiments, has calculated the heat of formation, from 

 amorphous carbon, of a pair of singly-linked, a pair of 

 doubly-linked, a pair of trebly-linked, and a pair of quad- 

 ruply-linked carbon atoms. From these values he has 

 calculated the heats of formation of isomers containing 

 singly, doubly, or trebly-linked carbon atoms. The cal- 

 culations involve certain assumptions, but the applica- 

 tions of his results to actual hydrocarbons show very close 

 agreement between the calculated and the actually deter- 

 mined " heats of formation." 



Thomsen furnishes us with a thermal value for the 

 formation of each of the three possible linkings of the 

 group C, in the molecule of compounds. The value of 

 this result to the chemist is great ; a determination of the 

 heat of combustion of a hydrocarbon may now yield him 

 much information as to the stru;ture of the molecule of 

 that hydrocarbon. 



Thomsen's results also strengthen the commonly- 



accepted theory of isomerism, and they point towards a 

 dynamical explanation of this theory and to the possibility 

 of attaching a definite physical idea to the phrases 

 "singly" or "doubly-linked" atoms. 



As Thomsen has succeeded in tracing a quantitative 

 connection between the heats of formation of certain 

 molecules containing carbon and the linking of the 

 carbon atoms in these molecules, so Briihl has shown that 

 the linking of carbon atoms exerts a definite, measure- 

 able influence on the molecular refractions of compounds 

 of this element. 



Landolt showed many years ago, that in many com- 

 pounds, the atoms of each elementar)- substance, possessed 

 a definite specific refractive capacity independently of the 

 way in which the atoms might be grouped. 



lilolecular refraction is defined as V' , )■ M> where 



fi, = refractive index, d= density of substance, and M = 

 molecular weight. 



The difference between the molecular refraction of a 

 compound containing carbon, hydrogen, and oxygen, and 

 that of a compound containing the same number of car- 

 bon and hydrogen atoms, but free from oxygen, gave 

 the atomic refraction of oxygen. Numbers were thus 

 found expressing the atomic refraction of carbon, hydro- 

 gen, oxygen, and a few other elements. Gladstone and 

 bale showed, however, that the obsen-ed molecular 

 refractions of many carbon compounds, especially the 

 compounds existing in essential oils, were greater than the 

 refractions calculated from Landolt's numbers : it seemed 

 that the grouping of atoms did exert, in certain cases, an 

 influence on the refractive power of molecules. 



Briahl finds that certain groups of isomeric carbon 

 compounds possess but one molecular refraction ; in 

 these groups the refractive power of the molecules is 

 independent of the grouping of the atoms ; in other 

 isomeric groups, however, the molecular refraction varies. 

 The members of the latter groups of isomers are always 

 represented in structural formula; as containing " doubly- 

 hnked" carbon atoms. Now if the molecular refraction 

 be conditioned by the linking, but not by the grouping, 

 of the atoms in the molecule, it follows that th»» atomic 

 refraction of each monovalent element must be a constant 

 number, inasmuch as there is but one way of linking a 

 monovalent atom to other atoms. Such isomers as ethy- 

 lene chloride, C1H„C— CH.,C1, and ethylidene chloride, 

 CloHC—CH,, should possess the same molecular refrac- 

 tion. But the atomic refraction of any polyvalent atom, 

 c.o;. oxygen, must vary according as the atom is linked by 

 one, two, or more " bonds " to other atoms : such isomers 



CH2 CH3 



II I 



as allylic alcohol CH , acetone C = 0, and pro- 



1 I 



H— O— CHo CH3 



/CH2 

 O I 



pylene oxide \CH should possess each a distinct mole- 



I 



CH3 

 cular refraction. Briihl's actual results confirm these 

 deductions. There is then a definite value for the atomic 

 refraction of the carbon, or oxygen, atom according as that 

 atom is "singly-linked" or "doubly-linked" to other 

 atoms : in other words, the ' molecular refraction of a 



N / . . 



compound containing the group — C— C— is different 



from that of the isomer containing the group ^C = Cx^. 

 and the molecular refraction of a compound containing 

 the group C = is different from that of the isomer 



