<I>Eupolymnia Nebulosa</I> (Polychaeta, Terebellidae)

Home , Eupolymnia nebulosa, Terebellidae

BULLETIN OF MARINE SCIENCE, 48(2): 420-431, 1991

LARVAL RELEASE FROM THE EGG MASS AND SETTLEMENT OF EUPOLYMNIA NEBULOSA (POL YCHAETA, TEREBELLIDAE)

Michel R. Bhaud

ABSTRACT The study of distribution and abundance of soft-substrate organisms can be examined by the recruitment mechanisms of the species involved. A study was carried out on Eupolymnia nebu/osa (Polychaeta, Terebellidae), starting from the release of larvae from the egg mass through the initial tube formation. The liberation of larvae, e){tending over 10 days, was counter-balanced by a marked synchrony at the time of settlement which occurred over a 2-day period for nearly 70% of the larvae. Despite the morphological differences present at the time of liberation, settlement took place at the same morphological stage. The duration of the planktonic life varied in relation to the time ofrelease from the egg mass. The larvae were provided with sufficient energy reserves to reach the first benthic stage. At this point, the most essential requirement is the construction of the first tube. Later, two basic requirements arise: food, and material for the prolongation of the tube. Both are satisfied by input from sedimentation. The synchronized settlement period caused a negative inter-individual effect, which tended to distance the individuals from each other, resulting in an even distribution pattern.

The dynamics of benthic communities are subject to fluctuations, which are due for the most part to the different life history strategies of the species involved. Therefore, study of the most abundant species, whose presence in a community is likely to be the most influential, would appear to be essential. In the Bay of Banyuls, Eupolymnia nebulosa is one such species; the characteristics of its life cycle have been described previously (Bhaud, 1988a; 1988b; Bhaud et aI., 1987; Bhaud and Gremare, 1988). These can be summarized as follows: the growth of oocytes within the coelom occurs between September (10-30 ~m) and March (150-170 ~m). During the spawning period, which extends from the beginning of March until the end of May, the females deposit gelatinous masses containing fertilized eggs close to their tubes. This spawning period is synchronized. In observations over 5 years (1983-1987) the most marked peak occurred during the last quarter of the moon in April. Juvenile growth takes place during the seasonal increase in water temperature. An experimental study has shown that there is a positive correlation between an increase in temperature and growth rate, when sufficient food is available (Bhaud, 1988a). In spite of this information, release oflarvae from the egg mass into the plankton, and the transition from the water column to the sediment, remain largely unknown. Therefore, a study of these phases between spawning and recruitment has been undertaken to address the following questions. What conditions control the liberation of the larvae? What is the duration of planktonic life? What type of relationship is established between the first benthic stages and the sediment, and does the latter act simultaneously or separately as a physical support and as a source of nourishment for the larvae?

MATERIALS AND METHODS

Collection of the Egg Masses. - In the study area, E. nebu/osa has a large vertical distribution ranging from the infralittoral upper levels (Laubier and Paris, 1962) to more than 300 m at the mouth of a submarine canyon (Reyss, 1964). Adults inhabit a tube and can be found at the interface of rocks and

420 BHAUD: RELEASE AND SETTLING OF E. NEBULOSA 421 coarse sediment. During the reproductive season, egg masses were always submerged and were located with a glass bottom viewing tube and removed by hand. Additional samples were taken by divers between 5-10 m depth. Samples were usually collected from a 70-m stretch of a small semi-enclosed bay close to the laboratory. Daily observations were made in order to determine the spawning dates. Range of Recruitment Times. - The following procedures were developed to determine the range of recruitment times of individuals emerging from a single egg mass at the beginning of the reproductive season. They were based on the progressive withdrawal or addition of sediment, contained in petri dishes, upon which the larvae settled. The beginning of the recruitment period was established using a tank filled with 50 liters of sea water and containing a single egg mass of known spawning date. A series of 12 petri dishes containing substrate were deposited on the bottom of the tank. The dishes were removed at a rate of one a day and isolated in another tank supplied with filtered water (0.2 !tm). The appearance of tubes in the withdrawn dishes signalled the beginning of settlement. The end of the recruitment period was established in a similar way: the petri dishes containing sediment were kept in a tank supplied with filtered running sea water, and were transferred at a rate of one a day to a tank containing a single egg mass of known spawning date. The end of the settlement period was evident by the absence of tubes in the introduced dishes after this point. The same procedures were carried out in May 1986, at the end of the reproductive season, with the following alteration: the transfer of the petri dishes from the tank with the egg mass to the incubation tank was made via an intermediary tank for a period of 2 days to allow complete settlement to occur, while avoiding possible contamination from the dishes already placed in the tank. Sediment Used. -Silt was obtained following sedimentation from laboratory seawater supplies and sieved on mesh of 60 !tm. A particle analyzer was used to give the cumulative size distribution. This process indicated that 76% of the particles were less than 10 !tm. A rough estimate of the quantity of organic material, 1.13 mg·g-' dry weight, was obtained by the loss of weight on ignition at 400·C of a dried sediment without carbonates. This sediment constitutes the standard sediment referred to in the subsequent text. Laboratory Rearing of the Juveniles. - The juvenile stages were kept in a tank 60 x 40 x 16 em, filled with 50 liters of sea water. Water was circulated through the tank at a rate of 0.5 liters·min-' and was drained from the surface by a siphon. Suspended particles in the sea water were allowed to settle out at two points of decantation. The egg masses were placed on a platform 7 cm from the bottom. Petri dishes with standard sediment were placed on the bottom of the tank. Food was added after juveniles became established. This food supplement was a suspension of 1.5 mg dry weight of tetra mine which had been pulverized in a Potter tube containing 12 ml of filtered sea water. The calorific value of the suspension was 4.10 cal, mg-' as determined with a calorimetric bomb of the type designed by Phil- lipson (1964). After addition of food, the water supply to the tank was stopped for 4 h to allow the particles to settle to the bottom to ensure that food was not a limiting factor. Distribution Pattern for the First Tubes. -A single egg mass was placed on a support close to the the surface of the tank, the bottom of which was covered with a fine layer of standard sediment. The quantity used was 3 mg (dry weight)'cm-2, giving a sediment thickness of about 0.1 mm. This enabled the larvae to reach the hard bottom of the tank which was a prerequisite for successful settlement (Bhaud, 1990). When the first tubes appeared, two photographs were taken. A grid dividing the photographs into units of known area allowed the numbers of tubes in each square to be counted and the parameters of the distribution calculated. In addition, the spatial distribution of first settling juveniles was characterized by measurement of the distance between the opening of an individual tube and those of its neighbors. Larval Requirements at the Time of Settlement.-Between 50 and 60 planktonic larvae, which had just emerged from the mucus mass, were introduced to a container holding 80 ml of sea water. The water was changed daily. A 2-ml suspension of silt which served as both food and material for tube- construction was added after the introduction of the larvae. The optimum quantity of food for this number of individuals was not known and so temperature was used to control the consumption of food by the larvae. The experiment was carried out at four different temperatures (6, 9, 12, and l5OC, each with two replicates), over a period of 30 days. Juvenile Requirements. - Larvae fed by means of cilia located around the mouth and the ventral side of the first tentacle. Ifadditional food is not added to the aquarium, the sediment in front of the tube is utilized for food and tube construction. An evaluation of juvenile requirements for construction material and food was carried out by supplementing the cleared zones close to the juvenile tubes with additional sediment using a pipette held 5 cm above the substrate. Four different characteristics are given to justify the use of this species for the experimental analysis of settlement: (1) after the release of all the larvae from the egg mass, an experiment of short duration (2 to 3 days) is sufficient to obtain settled stages; (2) settlement is easily detected by the appearance 422 BULLETIN OF MARINE SCIENCE, VOL. 48, NO.2, 199]

Table I. Temporal data on the release of larvae (2 and 3), the disintegration of the mucus mass (4 and 6), and the appearance of the first tubes (5 and 7) as a function of the spawning date of the egg mass (1)

4 6 7 I 2 3 Mucus mass 5 Differ- Differ- Egg mass First larvae Last larvae disintegration First tubes, ence eoce (spawing date) appearance appearance data appearance 4 - I 5 - I

10 March 14 March: +4 24 March: +10 28 March I April 18 22 18 March 22 March: +4 31 March: +9 2 April 7 April 15 20 31 March 4 April: +4 13 April: +9 15 April 17 April 15 17 31 March 4 April: +4 13 April: +9 13 April 17 April 13 17 31 March 4 April: +4 13 April: +9 17 April 17 April 17 17 21 April 25 April: +4 2 May: +7 5 May 6 May 14 15 23 April 27 April: +4 4 May: +7 7 May 8 May 14 15 15 May 18 May: +3 23 May: +5 24 May 27 May 9 12 18 May 21 May: +3 26 May: +5 27 May 28 May 9 10 11 June 14 June: +3 19 June: +5 19 June 19 June 8 8 of tubes; (3) the number of larvae ranges between 3,000 and 20,000 per egg mass which allows experiments to be carried out in large volumes of water; and (4) the synchrony of the larval settlement simplifies the interpretation of the results.

RESULTS The Release of Larvae from the Egg Mass into the Plankton. - The time necessary for the disintegration of the mucus and the appearance of the first benthic tubes varied with the spawning date (Table 1). For an egg mass spawned on 10 March 1984, the first larvae emerged on 14 March; 10 days later (28 March) the mucus mass was almost empty and had disintegrated. The first tubes were visible 22 days after spawning (1 April). However, these events occurred with a shorter time period later in the breeding season. For an egg mass spawned on 18 May, the first larvae were liberated on 21 May, the gelatinous mass was empty by 26 May, and the first tubes were visible on 28 May, only 10 days after laying of the egg mass. The variation in the developmental parameters with respect to the egg mass depends on the position in the reproductive season. This variation in the release oflarvae seems to be related to a temperature effect: release oflarvae accelerated with increasing temperature from 11°C in early March to 16°C by the end of May. This release is either a direct consequence of increased activity and faster development of the larvae or an indirect consequence of the disintegration of the mucus. The collection of discrete gelatinous egg masses after all the larvae were released, especially at the beginning of the reproductive season, suggests there is no correlation between the disintegration of the mucus and the release of larvae, The release as a consequence of larval activity is a more likely explanation. The disintegration of mucus occurs over a longer period than that taken for the release of larvae: three empty mucus bags have been kept for 17 days in the laboratory (at 20°C) and have been colonized by bacteria without observable disintegration, Therefore, it appears that bacteria and temperature playa limited role in the disintegration of mucus and cannot be used to explain the release oflarvae directly. Field observations suggest the more efficient role of agitation by waves in the break-up of mucus. The observation of an egg mass under a compound microscope shows that the timing of escape by the larvae depends on their position within the egg mass. Those located peripherally are released first. The spawning mass is a sphere of approximately 4 cm in diameter, with larvae evenly distributed throughout, except at the center. This central portion consists of an axis of 1 cm in diameter through BHAUD: RELEASE AND SETTLING OF E. NEBULOSA 423 which pass the tentacles of the female. Eggs are transported from the nephridial openings of the female in mucus strings along the tentacles, and are deposited in consecutive layers around the central core. Thus larvae are located at variable distances from the surface of the mucus mass and release is earlier for larvae at the periphery. However, there is no difference in the developmental rates oflarvae regardless of their location within the egg mass. The morphology of the first and last larvae to be liberated varies greatly. The former measure 430 /.Lm in length and have three pronounced constrictions posterior to the metatroch. The first setiger is indicated by two pairs of lateral setae: one pair are club-shaped setae and the other are capillary, but these are not always visible at the time of liberation. Capillary setae may appear later, and the first larvae to be released bear two provisional setae on each of the two first setigers. The last larval stages to be freed measure 510 /.Lm and have four setigers which each bear one pair of provisional club-shaped setae and one pair of capillary setae. The fifth segment has one pair of full-grown uncini, and on the sixth a pair of newly-formed uncini may be visible. Comparison oflarvae of the same age within the mucus mass with those already released reveals no obvious morphological differences. However, there is a size difference. Within the mucus mass the constrictions ofthe hyposphere are visible, and the larvae are elongate: the average length is 231 /.Lm (N = 42 and SD = 15.30). In the water of the aquarium, the larvae have a more pyriform appearance, the constrictions less defined and the body shorter: the average length is 198 /.Lm (N = 46 and SD = 10.04). This difference can be explained by observation of the larval behavior. In the mucus the larvae have a sinuous movement and their bodies are elongate. The provisional setae of the first two segments are clearly visible. In contrast, free swimming larvae in the water rely on the ciliary move- ments of the prototroch for their movement, the body is compact and the first provisional setae are difficult to see. Microscopic observation shows that the provisional setae are still enclosed in the setiger sac, however, when a larva is placed between a coverslip and slide and compressed slightly, the tegument rup- tures and first two, then four provisional setae can be seen. The Variability oj Recruitment. - The spawning date of the egg masses used in the experiments at the beginning of the reproductive season was 29 March 1986. The first tubes were expected 16 days later; therefore the introduction or withdrawal of the 12 substrate dishes commenced 6 days before and continued until 5 days after this date. Settlement began on 13 April for series 1 and 2 and the end ofthe settlement period occurred on 16 April (series 1) and 17 April (series 2). Consequently, larvae of exactly the same age all settled within a period of 4- 5 days. The spawning date of the egg masses used at the end of the reproductive season was 3 May 1986. The first tubes were expected 11 days later and the transfer· of substrates was carried out between 9 May and 20 May. For series 1, the benthic settlement period began on 13 May and was completed by 17 May. For series 2, the beginning and the end of the settlement period occurred on 14 and 18 May, respectively. In both series all tubes were constructed over a period of 5 days. Larvae released from one egg mass all settle within 4-5 days of each other; this feature remains constant throughout the entire breeding season. Comparison of several spawning egg masses indicates little variation of settling period between egg masses. However, within this period, settlement of larvae released from an egg mass is not homogeneous. Results of an experiment to ascertain the number of settling larvae each day (Fig. 1) show that the mean number of settled larvae 424 BULLETIN OF MARINE SCIENCE, VOL. 48, NO.2, 1991

~ 300t ...... • ,. 0.0 1\ .S , \ ';:i \ .w I

(from 4 spawning masses) were 6, 69, 48, and 120 respectively (percentages of 2.5, 28.4, 19.7, and 49.4) during a 4-day period. Therefore, approximately half of the total number oflarvae settle within 1 day and almost all of the larvae within 3 days. After day 8, settlement of larvae ceases. The fluctuation in the number of larvae recorded after this time can be attributed to differences in settlement between test containers. In conclusion, the long period of time over which the liberation of larvae from the egg mass occurs (10 days at the beginning of the reproductive season) is compensated for by a marked synchrony at the time of settlement of the larvae. Spatial Distribution oj the First Benthic Individuals. -A distribution pattern is easily obtained in the laboratory by placing an egg mass in a tank (50 liters), at BHAUD: RELEASE AND SETTLING OF E. NEBULOSA 425

N: 35 N: 35 N :48 m: 38 m : 21.6 m: 5.6 (j2: 10 0-2: 6.9 0-2: 3.2 U (cm2): 12 U (cm2): 9 U (cm2): 2 d (t.cm-2) : 3.2 d (t.cm-2) : 2.4 d (t.cm-2) : 4.3

5 -

n N 31 46 16 28 2 10 Number of tubes Figure 2. The substrate, which represents the reception area, is divided into units of known size, and the number of tubes in each one is counted. The number of surface units (vertical axis) containing a given number of tubes (horizontal axis) is displayed for three different situations.

a distance of7 cm above the standard sediment. A uniform distribution of recruits is observed which suggests that negative inter-individual relationships exist. In the three studied situations the ratio (12·m-1 is always lower than 1 (Fig. 2). The mean distance between two openings is 839 J.Lm (N = 87 and SD = 27). This distance is almost exactly twice that which can be explored by each individual remaining attached to its tube by means of the posterior uncinal plates. This is obvious when the mode of feeding and the exploitation of the environment immediately bordering the tubes are observed. With the gradual growth of the tube, both openings are exploited which results in a passage cleared of sediment. The individual projects itself outside its tube while holding on with the uncinal plates of the posterior part of the abdomen. Food collection is carried out using the ventral ciliated area fringing the mouth and the bud of the first tentacle. In spite of the above likely explanation, an alternative hypothesis is that the observed distribution pattern is due to the circulation of water within the aquarium. The same experimental design was carried out in still water to test this hypothesis. The resulting distribution pattern in the still water experiment was entirely dependent on the larvae. In the tank (60 cm x 40 cm) the distribution is very irregular with the majority of the tubes situated close to the egg mass support. However, at a distance of 10-25 cm from the support, the distribution pattern is modified and the tubes become more scattered and longer (Fig. 3). When the density of individuals is increased, the deposited food is used up rapidly and the non-renewal of the latter results in the exploitation of neighboring areas; however, each individual is restricted by the zone occupied by its neighbors and consequently the availability of sediments is reduced. The Significance oj Sediment Collection by Juveniles. - Sediments are required by juveniles for tube construction and food. It is not known which of the two needs must be satisfied at the time of settlement. The following observations can be made. I) Tube construction occurs when additional food is present. Recent observations (Bhaud, 1990) also show that when the substrate is suitable almost all 426 BULLETIN OF MARINE SCIENCE, VOL. 48, NO.2, 1991

r", I B I r·..• I I I , I r-1 I I 5 I r-'I I I I I I I I I I I 1. r-' I I I I L I -~ 4 7 10 12 17 23 Tube length (mm) Figure 3. The influence of recruit density on the tube lengths in the absence of sediment renewal. The comparison of the lengths of tubes at two distances: 8 cm (A) and 23 cm (B) from the center of the egg mass, shows a very significant difference in size. the larvae arriving close to the interface settle and construct a tube. Settlement and tube construction occur almost simultaneously so that the second can be considered an expression of the first. 2) In stages with seven segments each bearing one pair of capillary setae, the stomach wall still contains lipids after settlement, which implies that an energy reserve exists at the time of settlement. 3) The nature of adult life indicates that tube construction is essential. Therefore, during the

-l

Figure 4. Illustrations of the interaction between food requirements, construction material requirements and animal density. Sedimentary material is deposited close to the tubes as the surrounding area is gradually cleared. A: No renewal of sediment; tube construction in an area where the density of individuals is 4igh. B: No renewal of sediment; tube construction in an area where the density of individuals is low. In both these cases the areas surrounding the tubes that have been cleared of sediment are visible. C: Renewal of sediment. (Photographs by 1. Lecomte.) BHAUD: RELEASE AND SETILING OF E. NEBUWSA 427 428 BULLETIN OF MARINE SCIENCE, VOL. 48, NO.2, 1991 short early benthic phase, the energetic needs are secondary with respect to tube construction. In the case of an energy shortage, it appears that greater mobility, and therefore the abandonment of the tube, would be an advantage. However, a tube is essential to the animal before it can collect sediment in order to feed. Conditions that prevent tube construction, even when food is present (for example a sediment that is too fluid), results in death (Bhaud, pers. obs.). The effects of food limitation can be demonstrated by observation of the behavior of a juvenile aged 20 days. When there is reduced sedimentation, the animal explores the environment immediately surrounding the tube and, subsequently, increases the length of the tube. When sedimentation is greater, the need for nutritional material is satisfied without exploration and there is an accumulation of sedimentary material at one opening of the tube. These observations may be integrated into an experimental design with the possibility of distinguishing between the two different requirements: tube building and food. The zone near the tube is supplemented with sediment as soon as it has been cleared by the animal. If the length of the tube is critical, the tube will continue to lengthen; if, however, the need is nutritional, the animal slows the growth of its tube, and an accumulation of sediment results at one of the tube openings. The latter case corresponds to the results obtained, indicating that the construction of a new tube, or an extension of the original one, reflects the nutritional needs of the juvenile (Fig. 4).

Relation between Temperature and Food Requirements. - The construction of the first tubes by 50 to 60 larvae was followed in a small volume of sea water. The number of living animals (LA), the intensity of tube construction (TC) and the number of animals without tubes (AT) were monitored for each dish and for each observation date (Table 2). All animals (LA) survived the 30-day period at all temperatures except 6°C. The intensity of tube construction (TC) was related directly to the water temperature. The number of tubes constructed was 75 at 15°C, 70 at 12°C, 32.5 at 9°C and zero at 6°C after 23 days. The number oftubes constructed gradually increased during the settlement period at 12 and 15°C. The number of tubes exceeded the number of introduced larvae which indicates that the young recruits leave their tubes or that the tubes fragment. The expression "number of tubes" is used to describe the total number of fragments because while tubes are formed in a continuous manner, they break up during manipu- lation. The number of tubes is therefore a measure of the construction activity. A series of size measurements, at the end of the experiment (30 days), indicates a mean length of 9.3 mm per tube (N = 20 and SD = 0.52). Examination of the third parameter (AT) during the settlement period showed that there is a progressive decrease in the number of animals without tubes. At 6°C and 9°C, this decrease was slow and never complete, during the experimental period. Initially the large number of animals without tubes corresponded to juveniles which were too young to settle. Subsequently, there was a phase of tube construction and the number of animals without tube was reduced. As settlement progressed a balance was established between the number of animals and available food. Later, the number of animals without tubes increased again. These observations suggest that as food becomes limited, the animals begin to search for new supplies, initially by extension ofthe tube and later by leaving the tube. A comparison may be made between the dual requirements for tube construction and feeding. When temperature conditions do not limit metabolic activity, i.e., the experiments at 12°C and 15°C, the number oflarvae without tubes falls quickly towards zero, indicating the natural tendency of the animal to con- BHAUD: RELEASE AND SETTLING OF E. NEBULOSA 429

Table 2. For a given quantity of food, fixed at the beginning of the experiment, three elements are monitored: the number of living animals (LA); the intensity of tube construction (TC) and the number of animals without tubes (AT). For 12 and 15°C, the number of animals without tubes has a minimum value between 15 and 23 days after the start of the experiment; this period corresponds to an equilibrium between the animals' size and their sediment consumption. Block at the right hand side at the base of the table indicates juveniles of sufficient large size to be placed in circulating water without food transport

Experimental temperatures OC

Day· 9 12 15 9 LA 50 50 58 44 64 46 TC 0 0 0 0 32 6 AT 50 50 58 44 32 40 15 LA 50 50 58 44 64 46 TC 0 0 0 0 59 56 AT 50 50 58 44 15 0 23 LA 50 50 58 44 64 46 TC 0 0 27 38 80 59 AT 50 50 49 6 2 I 30 LA 35 35 58 44 64 46 TC 0 0 32 38 79 73 AT 35 35 44 6 49 18

• Days after experiment began. struct a tube. When there is an energy deficit (for instance when growth is rapid and feeding is increased which is particularly evident in the experiments at 12°C and 15°C after a 23-day period), construction continues indicating that the search for food is achieved by extension of the tube. With a continued decrease in the amount of available food, the number of animals without tubes increases which is in agreement with the general observation that worms leave their tubes and die.

DISCUSSION AND CONCLUSIONS The major point of interest in this study of part of the life cycle of Eupolymnia nebulosa is the synchrony of settlement. All of the larvae released from a single egg mass undergo settlement during a 5-day period with 70% of the larvae settling within a 2-day period. The youngest tubicolous worms have five segments bearing club-shaped and capillary setae and one pair of full-grown uncini (except segment 1). The only potential variation, which occurred in 5 out of23 specimens, was on segments six and seven which may lack or possess a partially to fully-grown pair of uncini. The synchrony of settlement reduced the size range of the juvenile stages. This is not the only characteristic of the life cycle which favors the pro- duction of young benthic individuals with a small size range at settlement. The correlation of the egg mass with the ascending part of the curve of seasonal temperature (Bhaud, 1988a), and the synchronization offemale spawning (Bhaud and Gremare, 1989) also contribute to the small size range. During the completion of the growth phase the growth rate is positively correlated with increasing temperature. This type of correlation, previously reported for oocyte growth (Olive, 1980), results in a reduction in the size range. The fact that several characteristics in the life cycle produce the same result suggests that reduced juvenile size range has an adaptive value. It appears that the different lengths of time spent in the egg mass have no effect on the speed of development. The existence of a mucus egg mass in the life cycle of E. nebulosa could restrict the passage of oxygen to the larvae situated centrally 430 BULLETIN OF MARINE SCIENCE, VOL. 48, NO.2, 1991 and released at the end of the liberation period. This could cause a difference in developmental rates for eggs from the same egg mass and liberated at different times. However, no difference was noted in the morphology of the larvae regardless of the time of liberation. This suggests that there is no variation in the amount of oxygen available to larvae regardless of their position in the mucus which is contrary to the findings of Chaffee and Strathmann (1984) and Strathmann and Chaffee (1984). In spite of the morphological differences observed at the time of liberation, settlement occurs at a fixed morphological stage, and it is the duration of the planktonic life that varies in relation to the time of release from the egg mass. To readdress the problems posed in the introduction, it is possible to make the following observations. Release of larvae from egg masses is correlated with increasing water temperature and the resultant higher metabolic activity of the larvae. Larvae experience a shorter period within the egg mass at the end of the reproductive season when the water temperature is 5°C higher. The planktonic larvae are lecithotrophic and remain in the water column for 5-10 days. The settlement period is short; larvae from one egg mass all settle within a 5-day period. Settling larvae initially require a sediment which they can manipulate to construct the first tubes. Secondly, they require food and may lengthen their tubes to reach a food source. Larval survivorship may reach 100% under laboratory conditions. These results lead to two types of predictions concerning, on the one hand, the density dependent constraints affecting the juveniles and, on the other hand, the importance of deposited material in the future location of adults. The high density of individuals at the recruitment stage and the heavy mortality of the first recruits are two frequently described characteristics in studies of pop- ulation dynamics in the natural environment, even though the cause of this mortality is not conclusively known (Bachelet, 1987). The low mortality observed in the laboratory between larval release and settlement suggests that a high density of juveniles may be present at the time of settling in the field. The observations on larval distribution made in the laboratory indicate that density dependent factors are active even though the small size range of juveniles at settlement prevents initial, size biased, interspecific competition for food. It is not possible to extrapolate these inferences directly to the field situation because of the greater surface area available to the larvae during settlement. Larval distribution in the field may, however, be affected by a highly probable negative interaction between adults and juveniles which has been observed during laboratory experiments (Bhaud, pers. obs.). The area controlled by the adults covers a circular area around the tube with a diameter of at least 60 em. Taking into account the conditions necessary for larval settlement, in particular the presence of a fine sediment fraction for food and tube building, it is possible to predict the location of adults. In the field, the zone inhabited by juveniles until maturity will be characterized by a low energy current regime which facilitates high sedimentation. Day (1967) noted that Terebellidae live in calm areas, such as bays, lagoons and crevices where there are favorable hydrodynamic conditions for the deposition of organic particles.

ACKNOWLEDGMENTS

This work is a contribution to the French "National Program for the Determination of Recruitment" (PNDR) and was supported by a grant to the author from the Centre National de la Recherche Scientifique. Special thanks go to H. Woodward and A. Grehan for their valuable assistance in the preparation of the English manuscript. BHAUD: RELEASE AND SETrLlNG OF E. NEBULOSA 431

LITERATURE CITED

Bachelet, G. 1987. Processus de recrutement et role des stades juveniles d'invertebres dans Ie fonctionnement des systemes benthiques de substrat meuble en milieu intertidal estuarien. These de doctorat de l'Universite de Bordeaux I, Bordeaux, France. 478 pp. Bhaud, M. 1988a. Influence of temperature and food supply on development of Eupolymnia nebulosa (Montagu, 1818) (Polychaeta, Terebellidae). J. Exp. Mar. BioI. Ecol. 118: 103-113. ---. 1988b. Change in setal pattern during early development of Eupolymnia nebulosa (Poly- chaeta: Terebellidae) grown in simulated natural conditions. J. Mar. BioI. Ass. U.K. 68: 677-687. ---. 1990. Acquisition de la vie benthique par Eupolymnia nebulosa (Polychete Terebelidae): dispositifs experimentaux et premiers resultats. Vie Milieu 40(1): 36-43. --- and A. Gremare. 1988. Larval development of the terebellid polychaete Eupolymnia nebulosa (Montagu, 1818) in the Mediterranean Sea. Zoologica Scripta 17(4): 347-356. --- and ---. In press. Reproductive cycle of Eupolymnia nebulosa (Polychaeta, Terebellidae) in the western Mediterranean Sea. Proceedings of the 2nd International Polychaete Conference. J. B. Kirkegaard and M. E. Petersen, eds. E. J. Brill, Copenhagen. ---, ---, F. Lang and C. Retiere. 1987. Etude comparee des caracteres reproductifs du ter- ebellien Eupolymnia nebulosa (Montagu, 1818) en deux points de son aire geographique. Cr. Acad. Sc. Paris. t. 304, ser. 3: 119-121. Chaffee, C. and R. R. Strathmann. 1984. Constraints on egg masses. I. Retarded development within thick egg masses. J. Exp. Mar. BioI. Ecol. 84: 73-84. Day, J. H. 1967. A monograph on the Polychaeta of southern Africa. Part II. British Museum (Natural History) London. 468-878. Laubier, L. and J. Paris. 1962. Faune marine des Pyrenees-Orientales. Fascicule 4: Annelides Po- Iychetes. Supplement a Vie Milieu 13(1),80 pp. Olive, P. J. W. 1980. Environmental control of reproduction in Polychaeta: experimental studies of littoral species in northeast England. Pages 37-51 in W. H. Clark and T. S. Adams, eds. Advances in invertebrate reproduction, vol. II. Elsevier, North Holland Inc., New York. Phillipson, J. 1964. A miniature bomb calorimeter for small biological samples. Oikos 15: 130-139. Reyss, D. 1964. Bionomie benthique de deux Canyons sous-marins de la mer Cata1ane: Ie Rech du Cap et Ie Rech Lacaze-Duthiers. These doct. Univ. Paris. 251 pp. Strathmann, R. R. and C. Chaffee. 1984. Constraints on egg masses. II. Effect of spacing, size, and number of eggs on ventilation of masses of embryos in jelly, adherent groups, or thin-walled capsules. J. Exp. Mar. BioI. Ecol. 84: 85-93.

DATEACCEPTED: July 13, 1990.

ADDRESS: Universite P. et M. Curie, Laboratoire Arago, 66650 Banyuls-sur-mer, France.