Starch is one of the most important natural products. It finds many industrial uses and applications, and, in different forms, is used throughout the world as a staple item of the human diet. Consequently, the structure and properties of starch have deservedly formed the subject of extensive investigations both by chemists and bio-chemists. Because of the inherent difficulties in the many problems which arose, and the continual need for development of new techniques, progress prior to 1925 was slow and uncertain. However, the subsequent elucidation of the precise structural configurations of the simpler sugars e.g. glucose, maltose and cellobiose, gave a firm foundation for future work, and steady progress ensued. One of the most important advances was the realisation of the apparent inhomogeneity of starch; this was followed in 1941 by the first quantitative fractionation into simpler components. As a result, the highly polymeric nature and main structural features of the molecule are now well estab¬ lished. In general, starches can be separated into two distinct entities, both of which are high polymers of α-D-glucopyranose units joined by α-1:4-glycosidic linkages: (a) Amylose - a mixture of very long, essentially inbranched, chains of these units. (b) Arnylopectin - a mixture of much shorter, highly branched chains of these units, the inter-chain, or branching, linkages being essentially α-l;6. Yet many problems still remain. Some are concerned with the clarification of controversial issues on which the present experimental evidence, as reported by various workers, is either at variance or considered to be inconclusive; the majority, however, are concerned with the fundamentally important details of fine structure and the molecular size and shape of the starch components. It is now accepted that there is little prospect of progress on these particular problems unless full use is made by the carbohydrate chemist of physical and enzymic methods to supplement purely chemical approaches. When the researches reported in this thesis were begun, it was apparent that, before the established physical techniques used in polymer chemistry for the determination of molecular size and shape could be profitably applied, work on the following topics was an essential preliminary in order that the materials to be studied could be adequately and accurately characterized:- (1) The development of a differential potentiometric iodine titration technique for the determination of the % amylose in starches and their fractionated components, and for the possible characterization of other a-1:4-glucosans. It was desirable that the apparatus should be considerably more sensitive and accurate than any of those previously reported, yet also be capable of simple operation as a routine analytical method. (2) A study of the oxidation of starches by the periodate ion in view of the low quoted percentage accuracy of the technique, and the then existing lack of agreement in the literature on the following points; (a) conditions for the quantitative estimation of the formic acid released, (b) the length of time required for complete oxidation, and the possibility of over-oxidation, (c) the presence of intact glucose residues after oxidation, suggesting that inter-chain linkages other than 1:6 existed in amylopectin. (3) A study of the effect of the presence of contaminating protein on results obtained by use of the potentiometric iodine titration and periodate oxidation techniques. (4) The preparation, and purification, of starches from several botanical sources which had not previously been studied, so that the use at any stage of industrial samples of unknown origin, treatment, and history could be avoided. By preparation of the starches under the mildest possible conditions, so as to minimise degradation, these same samples could later be used for physico-chemical studies of molecular size and shape. The opportunity was taken, after the extraction of each starch, to apply a graded series of extractions to the residual plant material; the fractions isolated were analysed for protein, ash, and uronic acid content, and the polysaccharide content was investigated by chromatographic analyses of the sugars liberated on hydrolysis.