38 SCIENCE PROGRESS 



(Annalen, 191 8, 416, 203) with a view to establishing some 

 generalisations ; it is found that, in order that a substance may- 

 be oxidised by silver oxide in ammoniacal solution, it must con- 

 tain a = CHOH group combined with two — CH 2 OH, = CHOH 

 or — COOH groups, or with any two of these groups. Several 

 other conclusions which have been arrived at may be found by- 

 reference to the original paper. 



Reference was made in these columns (Science Progress, 

 191 8, xii. 392) to the synthesis of Zingerone the pungent prin- 

 ciple of ginger by Nomura. The same author has recently 

 (J. Chem. Soc, Abstr. 191 8 (I) 438) undertaken an investiga- 

 tion into the taste of ketones allied to this compound and finds 

 that certain unsaturated ketones have hardly any taste, but on 

 reduction yield more soluble compounds which are distinctly 

 pungent. The degree of pungency is considerably influenced 

 by the position of the hydroxyl group in relation to the ketonic 

 side-chain and is specially marked when there is a methoxy 

 group in the meta position. 



An interesting paper on colour variation in Anthocyanins 

 is published by Shibata, Shibata and Kasiwagi (/. Amer. Chem. 

 Soc. 1 91 9, 41, 208), but for particulars the reader is referred 

 to the original, as reasons of space do not permit of its abstrac- 

 tion here. 



MINERALOGY. By Alexander Scott, M.A., D.Sc. 



G. A. Rankin and H. E. Merwin {Amer. Jour. Sci. 45, 301, 

 191 8) have investigated the ternary system MgO — A1 2 3 — Si0 2 , 

 so far as the equilibrium between the various solid phases and 

 the melt are concerned. The substances stable in contact 

 with the melt comprise the three simple oxides, the four com- 

 pounds previously found in the ternary systems, and one 

 ternary compound 2MgO, 2A1 2 3 , 5Si0 2 . This last, like 

 clinoenstatite (MgSi0 3 ), dissociates below its melting-point 

 and is capable of taking up a considerable amount of silica in 

 solid solution. It exists in two modifications, the low-tem- 

 perature form being obtained by the crystallisation of a glass 

 below 950 C. and undergoing an apparently irreversible trans- 

 formation to the other form at somewhat higher temperatures. 

 The properties of both forms agree closely with those of the 

 natural mineral cordierite, but the latter differs from the former 



