A survey has been made of the biomass of the macrobenthos at twenty stations in the English Channel off Plymouth. The object was to provide a basis for following long-term fluctuations in the fauna.

A brief survey of physical conditions in the area is given, and a gradeanalysis of the soil at each station has been made.

Core-samples show that the sediment is shallow in many places and rock has been taken at 36 cm. or less below the surface at a number of stations.

Samples totalling ½m.2 were taken at each station with a modified 'scoopsampler', covering 1/10m.2, which is briefly described.

Sources of error in sampling are considered in detail. Some species may evade the sampler and others live too deep in the sediment to be taken. A comparison against a Petersen grab and a new ' suction-corer' show that the scoop-sampler does take a reasonably quantitative sample. The number of species taken in successive hauls, when plotted against the log of the area sampled, approaches a straight-line relationship similar to that obtained by Williams (1950).

A statistical analysis is made of a series of samples taken at one station, and the variance between the two samples in each haul and between successive hauls calculated. At this pàrticular station the fauna in successive samples is shown to be fairly random.

A sieve of 2-2 mm. mesh was employed. Compared with a finer sieve losses in terms of numbers may be quite large, but the total weight taken is little affected.

The host-response of the polynoid Polynoë scolopendrina Savigny, commensal with the terebellid Polymnia nebulosa (Montagu), is relatively specific within the family Terebellidae.

Evidence is presented that an unstable or closely bound attractant is present on the outside of the host and perhaps in tube material, but apparently absent in the mucus secreted by the host. Efforts to demonstrate its presence in ground-up whole host resulted in failure.

The polynoid Lepidasthenia argus Hodgson demonstrates a similar specificity of response to its host Amphitrite edwardsi Quatrefages.

The polynoid Harmothoë spinifera Ehlers, commensal with Amphitrite gracilis Grube, responds to the alternate host Polycirrus caliendrum Claparede as strongly as to its own host, and more strongly than to other species of Amphitrite. Likewise, Polynoë scolopendrina, commensal with Polymnia, responds with greater intensity to the alternate host, the eunicid Lysidice ninetta Audouin and Milne-Edwards, than to any non-host terebellid. This behaviour may be i terpreted as evidence that the specificity of at least some single commensal species for hosts which are not closely related to each other may depend upon the production of similar attractants by these hosts.

The responses of three commensal populations of Harmothoe lunulata (Delle Chiaje), commensal respectively with the ophiuroid Acrocnida brachiata (Montagu), the holothurian Leptosynapta inhaerens (O. F. Miiller), and the terebellid Amphitrite johnstoni Malmgren were tested against seven alternate hosts. Each gave maximum response to its own host. However, the population from Acrocnida responded as strongly to Leptosynapta as to its own host, while giving no responses at all to any of the other five.

Removal of the eyestalks in Leander serratus does not result in an earlier moult, a shorter intermoult period, or a higher moult rate. There is no evidence of an eyestalk moult-inhibiting hormone in this species. The evidence points to the existence of an eyestalk moult-accelerating hormone. The possible correlation of the presence of the moult-inhibiting hormone and seasonal moulting is discussed.

The behaviour of lugworms in sand was studied by recording kymographically the water movements through their burrows and observing the timing of defaecation. The worms ranged in weight from 2 6 to 186 g. Three types of apparatus were used, differing in the amount of space allowed to the worms.

The worms often traced a characteristic pattern described elsewhere, consisting of regularly spaced diphasic excursions of the irrigation graph with a defaecation on the apex of each first phase. The interval between excursions was about 15–20 min. for worms weighing about 3 g. and increased with size to about 40 min.

The worms sometimes traced similar periodic bursts of irrigation activity without defaecation.

Many of the tracings were confused in pattern. The worms often defaecated—sometimes sporadically, sometimes at regular intervals—in the absence of the characteristic diphasic excursions. Restriction of a worm to a small space appears to favour the production of confused tracings.

The relation between the irrigation patterns and the timing of defaecation is discussed.

Six species of luminescent polynoids of the Plymouth fauna have been studied, namely Lagisca extenuata, Gattyana cirrosa, Harmothoë liinulata, Polynoë scolopendrina, Acholoë astericola and Malmgrenia castanea. Their scales are luminescent, and the light is produced by granular eosinophilic photocytes, which form a unicellular layer on the lower surface of the scale. The nervous supply of the elytrum is described, and the luminescent response is shown to be under nervous control. Luminescent responses from all six species have been recorded by the use of a photomultiplier cell and oscilloscope. The normal response has been found to consist of a series of rhythmic flashes, from 9 to 1 per sec, lasting up to 1 min. Some characteristics of the luminescent responses are given, and the part they may play in the normal life of the animal is discussed.

The sizes, length and breadth, of the cyprids of Balanus balanoides, B. crenatus and Verruca stroemia have been measured on material collected. during the spring outburst and in the later part of the year.

The size-frequency curves for Balanus balanoides and B. crenatus show at least two distinct modesduring the spring, equivalent to populations of two distinct sizes. It is suggested that these correspond to populations developed in different environments. Measurements of the nauplii taken from adults at different points on the shore indicate that the size of the nauplii is dependent to some extent upon the conditions of development, the conditions also affecting growth and total nauplii output.

Swedish samples, in which the parents had developed under more uniform conditions, showed only a single population as regards size.

In B. crenatus and Verruca stroemia, which produce further broods during the year, the size decreases in these later broods. The rate of development of the eggs and embryos, probably dependent upon temperature, is suggested as the important factor.

One hundred each of stage VI male and female and stage V Calanus finmarchicus were picked at random, and, by means of a projection microscope, measurements made on a number of segments in both metasome and urosome.

The stage VI female population was mixed containing 95% of a smallsized population and 5% of a larger size-group. The males constituted a single population.

At this time of the season the stage VI male and female metasomes are the same length, but there are differences in the sizes of the individual segments, the most striking of which is that of the head

The stage V animals in many of the measurements could be quite clearly separated into two size-groups of about 50% each and it is suggested that these are potential male and female populations. This separation at 50/50 has been noticed in numerous other collections of stage V, but it shouldbe emphasized that the differences are population differences in regard to those measurements considered.

Evidence is presented which shows that the state of the antennules after preservation in 5% neutral formalin is unlikely in stage V animals to be a reliable guide to their sex in the next stage.

Glossus humanus (L.) (Isocardia cor Lam.), a large marine eulamellibranch, occurs in deep water in soft muddy substrates. Restriction of the species to a soft substrate is probably correlated with the inflated form of the shell, the weak musculature of the foot, and the poor elastic properties of the ligament.

Growth of the mantle/shell is the resultant of radial, transverse and tangential components, the normal axis being a turbinate spiral. The anterior region of the ligament is progressively split during growth owing to the displacement posteriorly of the secreting surfaces. The relationships of the valves, ligament and mantle isthmus are discussed.

The deeply plicated ctenidia are specialized for dealing with large volumes of water containing little suspended material.

The anatomy and functioning of the stomach is described in detail. There is a complex ciliated mechanism selecting particles on the basis of size. Such a mechanism, it is suggested, must be associated with intracellular digestion. The flap-like major typhlosole prevents material carried by the intestinal groove from returning to the circulation within the stomach, while the retention of an isolated opening for the left pouch is probably correlated with the need for firm fixation of the gastric shield.

Spawning occurs at the end of September.

The length, breadth, and in part the height, of individual Balanus balanoides and B. crenatus have been measured over several seasons, and the mean specific growth-rates calculated. The B. balanoides had settled at various known intertidal levels, and both species under conditions of permanent submersion.

Under all conditions, except on the highest intertidal panels, length increases rapidly subsequent to settlement and during the early summer. B. crenatus can reach its virtual maximum size (20–25 nun. rostro-carinal length) in a single growing season, under the conditions studied. Late-settled B. crenatus continue to grow further into the autumn, and begin to grow again early in the spring of the second season at the end of which the virtual maximum size is reached. The seasonal growth of both intertidal and submerged B. balanoides is similar, but the growth-rate is a function of submersion, which is, of course, dependent upon the tidal level. However, maximum growth is not reached in the first season and there is consequently relatively more growth in the second season than in B. crenatus.

The mean specific growth-rates in both species decrease with increasing size and, with permanently submerged B. crenatus, there is no suggestion that during the first spring and early summer any change in the environmental factors was effective. On the intertidal panels, the mean specific growth-rates at a given size vary with the tidal level, and this effect is most marked in the smaller barnacles.

A thin membrane surrounding the faeces of many of the Caridea has proved to be chitinous. The membrane is not secreted by the fore-gut, but almost certainly by the anterior part of the intestine, and is therefore a peritrophic membrane sensu stricto.

On the evening of 20 February 1953, two plaice (Pleuronectes platessa L.) in the largest tank of the Aquarium were observed spawning. No information on the spawning behaviour of pleuronectids has been found in the literature; nevertheless, the observation may very well not be original, but is considered of sufficient interest to justify an account of it.

Six methods for estimating the numbers of nanoplankton organisms in sea water have been compared. Of these, centrifugation of a living sample, followed by counts of the numbers of organisms in the concentrate, appears to be the most satisfactory method from all points of view. The method is rapid, simple to operate, and gives a result which is strictly comparable with results from counts made under dark-ground illumination of unconcentrated samples, and from counts after concentrating with niters. By this method the organisms counted can usually be assigned to the class and order to which they belong, even if their systematic position cannot be determined more definitely.

The use of species-pure cultures is a necessary adjunct to the direct examination of sea-water samples when studying the distribution, abundance, and annual fluctuations of nanoplankton, as so little is known of the systematics of these organisms.

An examination was made of the stomach contents of plaice post-larvae taken from the Southern Bight spawning ground in the spring of 1950. They were found to be feeding mainly on a soft-bodied prey, which left characteristically shaped faecal pellets and an undigested matrix vesicle as the only evidence of its nature. Pellets from the post-larvae compared closely in shape with those of Oikopleura dioica, found in the plankton of the spawning area during the period of post-larval development.

In 1951, undigested Oikopleura were seen in the oesophageal regions of a few dissected specimens, in addition to typical indigestible remains in a large proportion of the total feeding post-larvae examined.

Records for 1946 suggest a similar exploitation of the Oikopleura population by plaice post-larvae, although the phenomenon was not repeated in 1947, when inclement weather may have interfered with the annual outburst of this prey.

The phenomenon of food selection by plaice post-larvae is discussed, with special reference to the part that vulnerability may play in limiting the number of species of prey available as food.

A feature of this and other investigations on post-larval feeding, was the occurrence of a high proportion of empty stomachs. An experiment designed to test a regurgitation hypothesis, revealed the existence of a marked diurnal pattern of feeding activity, apparently influenced by light conditions. The importance of light in feeding behaviour was confirmed by tank experiments. In this investigation a large proportion of the empty stomachs encountered can be explained by reference to diurnal fluctuations in feeding activity, supporting the view that plaice post-larvae are visual feeders.

The blood of Lophius piscatorius is poor in haemoglobin, the volume of red cells being only 17%. The plasma contains less than 40 g. of protein/1., of which only 6 7 % is albumin. This explains its low osmotic pressure. As is well known, it contains more crystalloids than mammalian blood, A(depression of freezing-point) being 0–84°, the same figure being found at Naples, Woods Hole and Plymouth. This rather high concentration is not due to organic constituents, that of total non-protein nitrogen being of the same magnitude as in mammals; it is mainly due to a high content of sodium chloride. Chloride is at a concentration of 15.3 m.equiv./1oo ml., sodium at 18.5, while in mammals they reach 9 and 15 respectively.

Total non-protein nitrogen concentrations in plasma are similar to concentrations in mammals; themain non-protein nitrogen constituents of plasma are neither urea, ammonia, uric acid or allantoin, but trimethylamine or trimethylamineoxide. Of our analysis 37% of non-protein nitrogen of plasma remain unidentified, so far as we can rely upon our chemical methods. The power of concentration of the kidney for non-protein nitrogen on the whole is not high; it varies up to fifteen times. But the degree of concentration by the kidney, small for most constituents, even for trimethylamine, seems to be very high for creatine which is the main representative of non-protein nitrogen. Of our urine analysis, 30% of non-protein nitrogen remain as yet unidentified.

Lophius kidneys perfused with the heparinized blood (venous) of the fish secrete urine in which total non-protein nitrogen is concentrated, magnesium highly concentrated, and chloride only slightly so or not at all. Oxygenation of the blood, or lowering the temperature of the perfusate from c. 20° to c. 5° C. does not appear to influence secretion. The blood flow through the kidneys increases with the perfusion pressure, the increase often becoming disproportionately large. The urine flow, on the other hand, above a certain critical level is largely independent of changes in perfusion pressure.

The commensal polynoid Acholoë astericola (Delle Chiaje) exhibits a powerful positive response on contact with its host, Astropecten irregularis (Pennant). This response is only relatively specific, Acholoë demonstrating it to a number of starfish that do not serve as host in Plymouth waters.

All living host tissue has some attraction for the commensal; and the attractant is apparently released in quantity in the stomach of the host.

Passage of time (apparently as long as tissue remains alive) and chilling do not markedly alter the attractiveness of isolated host tissue. Temperatures above 450 C. alter or abolish its attraction.

The extreme instability of the attractant has so far made it impossible to demonstrate its presence in extracts.

Evidence is presented that the attractant is rapidly oxidized.

Experiments with the scale-worms Gattyana cirrosa (Pallas) and Lepidasthenia argus Hodgson, commensal respectively with Amphitrite johnstoni Malmgren and A. edwardsi Quatrefages, indicate that a chemotaxis to the host may be of relatively minor importance in governing the behaviour of these commensals.

Experiments with isolated hearts of various Crustacea (Cancer pagurus, Maia squinado, Homarus vulgaris, Squilla mantis) were performed in order to test the postulated secretory function of the pericardial organs (nerve trunks of unusual structure lying in the pericardial cavity). Tests were made with extracts of these elements added to the fluid perfusing isolated hearts.

In Cancer, Homarus and Squilla the extracts caused a distinct increase in amplitude and frequency of the heart beat, and their effect proved to be very similar to that produced by adrenaline and noradrenaline tested on the same hearts. In Maia the extracts produced an increase in amplitude and decrease in frequency of the heart beat.

Extracts of the pericardial organs of Cancer pagurus gave the fluorescence reaction characteristic of adrenaline and related compounds. Blood taken from the pericardial cavity gave the same reaction, but that taken from the leg arteries did not.

It is assumed that the function of the pericardial organs in the Crustacea consists in liberating, through fine neuropile-like terminations of the nerve fibres, some hormone passing with the blood into the heart and producing on it a stimulating effect. There is certain evidence that the pericardial organs might release a second hormone having an inhibitory and perhaps also a stabilizing effect on the heart rhythm.

Laboratory experiments showed that freshly liberated Spirorbis larvae settle on surfaces which bear previously settled individuals, in marked preference to bare controls, but that larvae which have been kept swimming for 3 hr., or longer, exercise less choice when setting.

When larvae are freshly liberated there is a great difference between the rate of setting gregariously, and the rate of setting in isolation; but after they have been kept swimming for 3 hr. this difference becomes less, owing to the larvae setting more rapidly both gregariously and in isolation. After swimming for about 8 hr. rates of setting fall off, and many larvae fail to settle, apparently through weakness.

I am greatly indebted to Dr D. J. Crisp for advice regarding this problem, and to him and Mr G. M. Spooner for much help with statistical treatment.

A description is given of a new type of apparatus designed to sample plankton from the immediate vicinity of the sea bottom, and purely from this region. Details of the construction are given and the relative merits of the design are briefly discussed.

It could be used for quantitative hauls.

A small additional net serves to sample the smaller benthic fauna at the same time as bottom plankton is being sampled.

The 'anchor-dredge' is a lightly built one-sided instrument designed to catch deep-burrowing animals. The depth to which it digs is dependent on the pull of the warp and not on the weight of the dredge.

From a trawler, it is best to work the dredge from the forward gallows, and give it a 'tug' as the ship is moving astern.

Comparative tests and direct underwater observations have shown that the anchor-dredge samples the in-fauna more efficiently than do either large rectangular and oval dredges, or the conical dredge.

A description and plan is given of a larger model, designed for use from a 6o-ft. or 90-ft. ship.