The addition of non-adsorbing polymer to a suspension of colloidal hard-spheres causes phase separation via the depletion mechanism. At high enough concentration of polymer a variety of non-equilibrium aggregation behaviour is observed. Transient gelation is one such behaviour observed at the highest polymer concentrations. Transient colloid-polymer gels are metastable space-filling particle networks. They persist for some finite time before suddenly collapsing to form a dense sediment. Thus transient gels exhibit "delayed sedimentation" which is a phenomenon observed in many weakly aggregated suspensions. We have studied the collapse process occurring in the bulk of a gelled suspension using dark-field imaging and ultrasonic concentration profiling. At low polymer concentrations we observe delayed sedimentation behaviour. At the highest polymer concentration we observe a change in the settling behaviour. The suspension continuously sediments at a rate comparable to the initial slow settling rate of gels exhibiting delayed sedimentation. This is known as "creeping sedimentation ". We have also investigated the effect of varying suspension height and width on the delay time of a gel. We have found a critical height below which the gel exhibits a size-dependent delay time. Above this critical height the delay time is independent of height. The same behaviour is found when width is varied. We find good agreement between the results of this experimental study and a recent theory. Special thanks go to my supervisors Wilson Poon, Peter Pusey and Margaret Robins who between them have guided me through the jungle that is soft matter physics On the technical front I would like to thank Andy Schofield for making large quantities of PMMA colloid without which none of this work would have been possible. I would also like to thank "Uncle" Steve Illet for showing me the ropes all those years ago and teaching me how to live life in comfort. I'm also indebted to Falk Renth and Jerome Arrault for sharing with me their vast knowledge of all things optical (and doughnuts). Thanks also to David Hibberd at the Institute of Food Research for help and advice in performing the ultrasonics experiments and directing me to the best Thai restaurant in Norwich. For useful discussion and reading various parts of the thesis manuscript I'd like to thank Mark Haw and Mike Evans. I'd also like to acknowledge the rest of the "geisters" with whom I have had interesting scientific discourse: Mike Cates, Steve Meeker, and Abdellatif Mousald. I think that about wraps it up except to say a big thankyou to all the softies and the members of E=M.C.C. for keeping me sane(ish). Finaly, to my family for all their support during the last 25 years and to Mark for just being there.

The green amoeboid cells of Chlorarachnion reptans Geitler are completely naked and each contains a central nucleus, several bilobed chloroplasts each with a central projecting pyrenoid enveloped by a capping vesicle, several Golgi bodies, mitochondria with tubular cristae, extensive rough ER, and a distinct layer of peripheral vesicles. Complex extrusome‐like organelles occur rarely in both the amoeboid and flagellate stages. The only organelles entering the reticulopodia are mitochondria, but microtubules are also present. The chloroplasts contain chlorophylls a and b, but histochemical tests suggest that the carbohydrate storage product probably is not a starch. The chloroplast lamellae are composed of one to three thylakoids or form deep stacks. A girdle lamella and interlamellar partitions are absent. Each chloroplast is bounded by either four separate membranes, a pair of membranes with vesicular profiles between them, or three membranes; all three arrangements may occur in the same chloroplast. A periplastidal compartment occurs near the base of the pyrenoid where there are always four surrounding membranes. The compartment has a relatively dense matrix and contains ribosome‐like particles and small dense spheres; it extends over and into a deep invagination in the pyrenoid where its contents are enclosed in a double‐membraned envelope which is penetrated by wide pores. The zoospores are ovoid and each bears a single laterally inserted flagellum which appears to be wrapped helically around the cell body during swimming. The flagellum lies in a groove in the cell surface and bears fine lateral hairs. Neither a second flagellum or vestige of one, nor an eyespot, is present. A single microtubular root and a larger homogeneous root run from the flagellar base parallel to the emerging flagellum, between the nuclear envelope and the plasmalemma. In the simple flagellar transition region, fine filaments connect adjacent axonemal doublets. A detailed comparison of C. reptans with all other algal taxa results in the conclusion that it must be segregated in the new class Chlorarachniophyceae, the only class in the new division Chlorarachniophyta. The possibility that C. reptans evolved from a symbiosis between a colorless amoeboid cell and a chlorophyll b‐ containing eukaryote is considered, but the possible affinities of the symbiont remain enigmatic. The implications of the unique chloroplast structure of C. reptans for current hypotheses concerning the origin of chloroplasts are discussed.

The systematics of the Eustigmatophyceae are revised at the level of species, genus, family and order. All known species are included in the Eustigmatales, which is divided into four families: the Eustigmataceae Hibberd includes Eustigmatps Hibberd and Vischeria Pascher, each with three species; the Pseudocharaciopsidaceae includes only Pseudocharaciopsis Lee & Bold with two species; the Chlorobotryaceae includes only Chlorobotrys Bohlin with one species and the Monodopsidaceae includes Monodopsis Hibberd with one species and Nannochloropsis Hibberd with two species. Eustigmatophyta and Eustigmatophyceae are published as typified names for the division and the class, respectively, both based on Eustigmatos. Tribophyceae, based on Tribonema, is published as the typified name for the class previously called Xanthophyceae. Nannochloris coccoides Naumann is chosen as lectotype of the chlorophycean genus Nannochloris Naumann.

Information o n the internal structure of the Chrysophyceae is reviewed and some new data are presented. From ihese a model has been constructed which represents the combination of features comidered t o be the basic pattern of cell structure in this group. A brief review is given of modern research, which demonstrate.; that the choanoflagcllatcs should be rcmoved not only from the Chrysophyccae hut also from the plant kingdom.