Ebenezer Ford was born on 22 September 1890 at Hove, Sussex, and died on 14 October 1974 at the age of 84. His father, George Horace Ford, was a man of insatiable intellectual curiosity and was a great stimulus to his sons. He took them on walks to ‘explore’ the countryside of Sussex, including the sea shore, which much interested ‘Ebb’, as he was known. Both sides of the family came from yeoman non-conformist stock, which instilled into him a keen sense of duty and desire to serve the public. Ebb was much involved in country affairs and had many opportunities of taking part in Wealden farm life. He was thus essentially an open-air boy with a growing interest in nature.

This paper estimates the turbulent transport of heat, fresh water and nutrients through the seasonal thermocline. In order to obtain a representative estimate of the annual variation it is necessary to average data obtained at a single position over a number of years. The number of years of sampling required depends on the particular parameter being studied. This can be seen by comparing the annual variation with the standard deviation, σ, for a specific month. This is analogous to the signal to noise ratio. For example, the annual range of temperature (or signal) is about 8°C for the water of the western English Channel. The variability of temperature in the same area (or noise) is about o.6°C. This figure is the standard deviation for the temperature obtained in a specific month at a fixed depth when data from a large number of different years are available. This variability can be attributed to a number of causes, to list a few; horizontal patchiness, local weather, advection, long term climatic changes. Hence for temperature the signal to noise ratio is about 13 and the annual pattern of events will clearly be established on the basis of monthly values obtained at a position during a single year.

Studies on the distribution of epifaunal species on their host have concentrated on those of plants, particularly laminarian algae (Fowell, 1939; Ebling, Kitching, Purchon& Bassindale, 1948; Sloane, Ebling, Kitching & Lilly, 1957; Scarratt, 1961, Ph.D. Thesis, University College of Wales, Aberystwyth; Norton, 1971), Sargassum algae (Kato et al., 1961; Nishihira, 1971; Ryland, 1974) and eel grasses (Nishihira, 1968; Nagle, 1968). Recently, however, the distributions of the animals living on ascidians Microcosmus sp. (Monniot, 1965) and the bryozoan Flustra foliacea (L.) (Stebbing, 1971) have been investigated.

The marked pattern of vertical zonation in the nearshore distributions of benthic fauna in sea-lochs seems to be controlled by the degree of fluctuation and extent of depression in salinity (Gage, 1974). The different vertical patterns in the abundances of species can be related to their known tolerances to reduced and fluctuating salinity. An overall faunal discontinuity seems to occur at a depth which roughly corresponds to the maximum depth to which markedly brackish conditions are experienced.

Eulalia viridis is the most common littoral phyllodocid at Cullercoats Bay, where it occupies crevices and small holes in the well-sutured sandstone and limestone rocks. It is particularly common in rocks having a heavy encrustation of barnacles. The reproductive biology of this species in particular, and of phyllodocids in general, is poorly known. Meyer (1938) has described the development of larvae of Eulalia viridis both from artificial fertilizations and from egg cocoons said to belong to Eulalia viridis collected in January at Plymouth, U.K. The observations made during the present investigation differ in a number of important respects from those of Meyer and his conclusions cannot be considered appropriate to all populations of E. viridis.

The phenomenon of epitoky in polychaetes, in which maturing individuals of benthic species undergo metamorphosis and then swim to the surface of the sea for spawning, has been extensively documented (see Fage & Legendre, 1927; Clark, 1961 for references). The death of individuals after spawning has been widely recorded and has been assumed to apply to all swarming species of polychaetes. However, Herpin (1925) recorded the survival of Pionosyllis lamelligera and Eusyllis blomstrandi after they had swarmed and spawned. He noted the loss of the swimming chaetae present at the time of spawning and that the individuals survived for some time. Goodrich (1933) found that individuals of Odontosyllis enopla survived swarming and considered that they might live to breed again.

Some integumental organs of Amphipoda have been briefly described by Bate & Westwood (1863). Their diagrams are extremely small and, in some cases, inaccurate. Fleminger (1973) described the integumental organs of calanoid copepods of the genus Eucalanus. He shows that these are distributed over the body in a pattern peculiar to each species. Bate & Westwood do not mention patterns in the distribution of integu-mental organs in amphipods. They do, however, describe two peculiar structures in gammaridean species of the families Talitridae and Gammaridae; they do not mention these as occurring in any other families.

The haustoriid amphipods Bathyporeia pilosa and B. pelagica are common in intertidal sand deposits around the British Isles. Both species live in the surface sediments and B. pilosa is typically a high-shore species, while B. pelagica is restricted to beach levels below the high-water mark of neap tides. Gravid females are readily identified by the presence of blue-pigmented eggs held in the brood pouch where they undergo direct development. The extent of the breeding season varies with locality (Salvat, 1967; Fish & Preece, 1970) and on the west coast of Wales B. pilosa breeds throughout the year and B. pelagica from March to October (Fish & Preece, 1970). At this locality, however, the breeding cycles of the two species are similar in that they show reproductive peaks in spring and autumn and it has been suggested that the spring peak is caused by breeding of the overwintering generation, the progeny of which mature to reproduce in the autumn of the same year (Fish & Preece, 1970). In common with many amphipod species, brood size of Bathyporeia is related to body length. Salvat (1967) has shown that the brood size of B. pelagica on the west coast of France doubles from February to April, but apart from his study adequate data on brood size are lacking. This paper describes embryonic development, hatching and brood size of overwintering and summer generations of both species.

This is one of a series of four papers dealing with vertical distribution of cephalopods in the North Eastern Atlantic at six stations near 20° W and at about 10° intervals from 60°N to 11° N (Clarke & Lu, 1974, 1975 a; Lu & Clarke, 1975). The present study is based upon a series of hauls made at discrete horizons between o and 2000 m with opening-closing nets during both daylight and darkness. The collections were made for the ecological programme of the National Institute of Oceanography, Wormley, Surrey, England (now part of the Institute of Oceanographic Sciences).

Salinity is a major limiting factor in littoral and estuarine environments and the salinity encountered in such habitats fluctuates with tide, season and weather. Despite this variable situation, most investigations of salinity tolerance and osmoregulation have involved laboratory experiments at steady salinities. The literature devoted to such investigations is extensive, and has been reviewed on several occasions, notably by Krogh (1939), Potts & Parry (1964) and Kinne (1971).

Assimilation efficiencies reported for bivalves vary, but different experimental conditions, in particular the concentration of the suspension at which the animals are fed, may affect the values found. Widdows & Bayne (1971) determined the assimilation efficiency for Mytilus edulis when fed Tetraselmis and they found that efficiencies decreased with increasing concentration of cells. Allen (1962) fed radioactively labelled Phaeodactylum to Mya arenaria and Venus striatula and suggested that the amount of P recovered in the faeces was proportional to the amount of algae ingested rather than to concentration of suspension directly. However, Winter (1969) concluded that for Arctica islandica and Modiolus modiolus there was no relation between assimilation efficiency and concentration of algae.

The echinoderm skeleton is interesting because its material is so different from that of other invertebrates. It is full of interconnecting cavities, and each piece of skeleton seems to be made of a single crystal of rather impure calcite (Nichols & Currey, 1968). An obvious question for a biologist is whether echinoderm skeleton is in some way better mechanically than other, more conventional, skeletal materials.

The classification of the Rissoidae has been, and continues to be, based largely upon shell characters. In a group of small animals of variable form this has resulted in a state of considerable taxonomic confusion. Ponder (1965, 1967, 1968), in his works on the New Zealand Rissoidae, attempted to clarify, what he termed, the existing ‘pigeon-hole classification’ in the light of more detailed anatomical investigation. He observed (Ponder, 1967) that ‘many wrong assumptions can be drawn from the study of the shell alone’. In an earlier work Johansson (1939) had also tried to relate the taxonomy of the Rissoidae to anatomical features.

Periodic structures indicating discontinuous growth of shell wall plates have been observed in thin sections of shell and on the outer surface of shell plates of many species of operculate (Bourget, 1975). When sectioned and etched the shell plates appear to be composed of successive thin layers of shell material, each layer consisting of a few rows of fine crystals. Where two layers are adjoined, a dark line, marking the junction of two incremental zones, is clearly visible. The thin layers of shell have been termed growth bands whereas the junction of adjoining layers will be referred to as the growth line.

A survey by Yonge (1962) of the byssal apparatus in the Bivalvia indicates that the formation of byssus is a larval characteristic which may or may not be retained in the adult. Some species such as the mussel, Mytilus edulis, retain the byssal attachment but have the ability to break the byssus and move over a limited area. Others, such as the scallop, Pecten maximus, lose the byssus entirely, become unattached and move by swimming. Yet others, such as the oyster, Ostrea edulis, relinquish all mobility at metamorphosis by cementing the shell to the substratum, and after metamorphosis lose all trace of the foot. Ostrea, Mytilus and Pecten therefore encompass a wide variety of life styles in the adult and are chosen as examples for detailed comparison of the byssus forming glands and other foot glands in the pediveligers.

The development of the planktonic veliger larva of Mytilus edulis L. culminates in a swimming crawling stage during which the foot is of considerable importance in the selection of a settlement site. This stage has been described for many other bivalves (see Bayne, 1965) and has been given the term ‘pediveliger’ by Carriker (1961). The pediveliger of Mytilus edulis is negatively phototactic and positively geotactic during velar swimming (Bayne, 1964b) and is usually confined to water layers close to the substratum. The foot may be protruded during swimming and if it comes into contact with the substratum it adheres and a period of crawling commences. During crawling the larva progresses on the ventral or posterior surface of the foot by means of ciliary and muscular action. Crawling may result in attachment by secretion of the first byssus thread or alternatively the foot is withdrawn and velar swimming is continued until the next exploratory crawling phase. Mytilus larvae attach most readily to filamentous substrates in the field (Blok & Geelen, 1958; Bayne, 1964a).

The larval cement of Ostrea edulis is formed from material secreted by the settling larva from two types of pedal gland (Cranfield, 1973b). Light microscope observations suggested that there were fibrous regions in this cement. The histochemistry and ultrastructure of the cement were therefore examined to establish the nature of this fibrous structure and to determine the part each of the two secretions might play in its formation.