TRANSACTIONS OF SECTION B. 545 



The authors have isolated and fully characterised by the modern methods of 

 cultivation an organism, a small bacillus, which sets up fermentation not only in 

 solutions of plucose cane-suj^ar, milk-sugar, and starch, but also in solutions of 

 maunite, glycerine, and calcium glycerate. The fermentations of maunite and 

 glycerine have alone been so far studied. 



In each case the products are essentially the same, viz., principally ethyl, 

 alcohol, and acetic acid, together with smaller quantities of formic acid and a trace 

 of succinic acid. The alcohol was separated by distillation and the quantity de- 

 termined, whilst the several acids were estimated by conversion into their barium- 

 salts in the case of the acetic and formic acids, whilst the succinic acid was ex- 

 tracted and weighed in the free state. 



In the case of the mannite fermentations it wjis found that the amount of 

 alcohol and acetic acid formed, stood in the proportion of two molecules of alcohol 

 to one molecule of acetic acid, whilst in the glycerine fermentations there were 

 three molecules of alcohol to one of ncetic acid. 



Of particular interest is the fact that the organism has no fermentative action 

 on dulcite, the isomer of mannite, which thus furnishes a very striking instance of 

 the selective power of micro-organisms between the most closely allied isomeric 

 bodies. The authors were also unable to cause the organism to ferment solutions 

 of either erythrite, ethylene glycol, calcium lactate, tartrate, citrate, or glycollate. 



In view of the characteristic products — ethyl, alcohol, and acetic acid— of the 

 fermentations, the authors propose for the organism the name of Bacillus 

 ethacetkus. (See also ' Proc. Roy. Soc' xlvi. 345.) 



4. The Constitution of the Aromatic Nucleus.^ 

 Bij S. A. Sworn, B.A., Assoc. E.G. ScI. 



The stability and other properties of the aromatic nucleus are to be explained 

 by assuming that the six atoms are situated at the most symmetrical positions in 

 space — viz., at the angular points of a regular octahedron. Such positions are 

 most consistent with the compactness of the molecular structure. It must not, 

 however, bo supposed that the atoms form a rigid structure. By their po.<itions in 

 space we mean merely the mean positions about which they oscillate. That they 

 are in vibration is evident from Hartley's work on the absorption by aromatic 

 compounds of the ultra-violet rays. In order to gain an exact knowledge of the 

 nature of this vibration, it is certainly necessary to know the points about which 

 vibration occurs — that is, the mean positions of the atoms. 



Two octahedral formul£e have been proposed — viz. those of Meyer and Thonisen. 

 The former is a modification of Ladeiiburg's prism symbol. The latter is a deve- 

 lopment of the diagonal symbol of Claus. They are in a manner complementary 

 to one anotlier, as when superposed they form a complete octahedron. They 

 explain equitllj- svell the relationships of the benzene substitution derivatives. No 

 other octahedral symbol will account for these relationships. 



Each is characterised by para-linkage, and it may be shown that experimental 

 ■work necessitates the assumption that there is direct para-linkage in the aromatic 

 nucleus. A symbol, .«uch as that proposed by Armstrong, is, if not meaningless, 

 quite opposed to experimental facts. (See Part I. pp. 5-11.) 



Thom.fens .'fymbol must be preferred to that of Meyer, inasmuch as the latter 

 is inadequate for the representation of many experimental facts. (See Part II. 

 pp. 12-15.) A further development of Thomsen's symbol affords a full explanation 

 of the meta- and para-laws of substitution. (See Part 111. pp. 15-18.) 



Lastly, the acceptance of any symbol which adequately explains the properties 

 of benzene necessitates a modified view of the Van 't Hoff theory. (See pp. 2, 3, 

 and IG.) This theory must be regarded, not as affording any solution of the true 

 positions in space of the atoms, but merely as indicating in a very rough way the 

 general direction in which particular atoms lie. 



Isomerism, as explained by the Van 't Iloff theory, is merely an answer to the 



' Published in extenso in the Phil. Mag. 1889, Nov. and Dec. 

 1889 N N 



