Whereas high levels of flavonoid extractives are responsible for the characteristic A,,, at about 280nm in the ultra-violet spectra of red wines, there is wide variability in the spectral features of natural white wines, in which the phenolic content is often minimal. For the latter, A,,, may occur anywhere within the range 265-285 nm, with relatively strong absorbance at 320 nm arising from the presence of hydroxycinnamate esters. Massive fining of many white wines and juices with polyvinylpyrrolidone (10% w/v) has shown residual non-phenolic U.V. absorbing constituents to be sufficiently uniform to enable direct spectral measurement of total hydroxycinnamates in white juices, white wines and roses as (E320-1 .4) absorbance units. For white grape juices and white wines, total hydroxycinnamates are directly quantitated as mg litre-' 'caffeic acid equivalents' and total flavonoid extractives may also be estimated. Total phenolic components of red wines are accounted as (Ez80-4) absorbance units. The spectral interpretations enable comparative estimation of flavonoid extractives in all natural wine types.

There is a very strong interaction between sulphur dioxide (SOa) and o-dihydroxy phenols in response to the Folin-Ciocalteu colorimetric assay, currently the established procedure for analysis of total phenols in wines and plant extracts. Because of this previously unsuspected interference, the method is shown to be generally quite invalid in application to white wines; measures of total phenols may be magnified several-fold by the SO2 artefact. By use of model solutions, it has been shown that the extent of interference depends primarily on the molar ratio of SO2:o-dihydroxy phenols. The mechanism of the artefact can be interpreted with reference to the chemistry of the photographic process, in which the oxidation-reduction reactions involved in image development are similarly mediated by the presence of bisulphite in alkaline solution.

Polyamides are shown to be effective adsorbents for removal of the bitter principle, limonin, from pasteurised Washington Navel orange juice. The extent of limonin removal varies with different juices, but sufficient can be removed by one or two treatments to bring the concentration below the organoleptically detectable level. The commercial use of polyamides to prepare non‐bitter juices from Navel oranges is discussed.

Methods are described for the general analysis of total ultraviolet (u.v.) absorbing constituents of white wines. An elution profile, which is characteristic of the particular wine, is obtained by continuous monitoring of effluent from a dextran gel column at 280 nm. Surprisingly, about half of the total U.V. absorbance in the Riesling wine examined was found to be due to non-phenolic wine constituents, which are separately eluted in the gel column analysis. These are nucleotidic materials, to which little attention has previously been given in oenology.Data relating to the protein-tannin fraction, and to nucleotidic and phenolic constituents, are directly obtained by recording the wine elution profiles at 265, 280 and 320 nm. The interpretation of the elution curves and the significance and general utility of this new analytical method are discussed.

No abstract