Marphysa “sp.”, commonly known by the name of “goose”, is one of the most appreciated bait species among those collected and commercialized in Portugal because of its high resistance in the hook and its reputed ability to attract valuable fish species such as the seabreams Sparus aurata and Diplodus sargus. Though it is difficult to value this resource because the sales are not declared (Castro 1993), harvesting is one of the most important socio-economic resources for local fishermen, and sales of €500000 per year are estimated (unpublished data) (Garcês and Pereira 2010).

The long-term exploitation of this genus requires a good knowledge of the biological and ecological characteristics of natural populations in order to develop rearing techniques for intensive production with the aim of increasing supply, reducing harvesting and fostering sustainable use.

With this purpose in mind, a small-scale experiment was conducted in order to address the way in which intraspecific density can influence somatic growth in juveniles of Marphysa “sp.” when food and sediment are not limiting factors. These effects were studied over a four-month period in laboratory facilities.

Further development of polychaete culture is considered of economic importance beyond that associated with the bait supply industry which acted as the initial catalyst for this research.

During the study, the physico-chemical water parameters remained approximately constant. There were no significant differences between replicates (p≥0.05) for the parameters under study. However, over the 4 months gradual and slight variations in water temperature (19-25°C), salinity (30-33), pH (7.6-8.1) ammonium (0.4-0.0) and nitrites (0.2-0.03) were nonetheless observed. The slight increase in ammonia and nitrite values during the experiment may be correlated with the confrontations and the associated mortality, but they did not reach high values that could have influenced this mortality. The control of these parameters reveals a perfect functioning of the biological filter. There was a higher dispersion value in dissolved oxygen concentrations according to the density, which means an increased consumption by worms. However, there were no significant differences (p≥0.05) between the replicates or in each density over time (Fig. 3).

During this study, juveniles of Marphysa “sp.” were observed to capture and ingest food particles on the sediment surface using their jaws, which are located ventrally as the worm moves over the substratum. The feed intake was observed to vary with density. At low density, the food particles were almost immediately consumed by the worms, which sometimes emerged completely from the gallery, getting fully exposed in the sediment surface. At medium and high densities, feeding was mainly at night and the worms never left the gallery, collecting food in a circular area and fleeing immediately into the gallery at the slightest sign of danger. At high density, the food was not entirely consumed during the first two months. However, it was not possible to determine the frequency with which searching activities resulted in particle consumption. Worms were observed seizing particles with their forceps jaws on many occasions, on both the substratum and over their tube or over the tube of a neighbour without confrontation.

No significant differences were found in growth between replicates of each treatment during the four months of the study, although differences tended to be greater between replicates in the high-density treatment. A two-factor ANOVA showed significant density and time effects (p<0.001) and a significant interaction between density and time (p>0.001) for all the growth variables in every month. Means of total length, weight and number of segments were inversely related to density. In all treatments, juveniles grew steadily during the four months of the study (Fig. 4), but a comparison of the monthly average increase shows that the three parameters under study evolved differently for each density treatment.

At low and medium density levels, the estimated monthly increase in the number of segments was higher in the first month and decreased progressively over time. Under high density, however, there was a progressive increase in the number of segments from the second month. Only increase in weight was, in general, continuous at all densities.

After four months, at the lowest density the polychaetes formed 61.5±4.2 segments, gained 0.1±0.01g in dry weight and grew 4±0.2 cm, at average density they formed 51.8±4 segments (0.08±0.008 g, 3±0.2 cm) and at high-density they formed 43±6 segments (0.05±0.01g, 2.4±0.3 cm) (see Fig. 4). It is noteworthy that at low density a significant decrease in the three parameters was observed in the last month (p>0.0001).

In this study, this high aggressiveness and territorial behaviour of Marphysa “sp.” is well evidenced by the higher mortality (35%) and greater incidence of broken and regenerating posterior segments (19.3%) observed under high density in the first month and by the presence of worms with lesions under low density (30%) (Fig. 5).

Although the incidence of broken posterior segments was overall slightly higher (16%) at medium density, mortality was much lower (15%). The sharp decline both in mortality and in the incidence of broken posterior segments that occurred in the following months reveals that the worms tended towards an equilibrium density. However, confrontation never ceased, as shown by 5.2% and 1.3% of individuals with lesions in high and medium densities respectively in the fourth month of the study (Fig. 5). Average estimated mortality after four months was 44.9%, 18.9%, and 0% at high, medium and low-density levels, respectively.

Despite this very high mortality, the average evolution of the existing population (Fig. 6) shows that significant differences in the number of individuals (p>0.001) continued to exist between the high and medium densities throughout the study, though it did not call into questiohn the existence of three densities.

The existence of confrontations in the low-density tanks (with 16.7% and 13.3% of broken and regenerating individuals, respectively) confirms the high aggressiveness and territoriality evidenced by this species and suggests that, up to a certain point, the density can be higher than the optimum without necessarily resulting in death.

Most infaunal animals irrigate their burrows with oxygen-rich overlying water. They are active 20% to 50% of the time, alternating between active and quiescent periods in a rhythmic fashion, according to species (Garcês and Costa 2009, Galaktionov et al. 2004, Kristensen 2001). Marphysa “sp.” dwells in L-shaped burrows and display vigorous bio-irrigation activity by pumping oxygenated water via peristaltic motions. According to Webber 1978, this genus has a high affinity to oxygen and high sensitivity to changes in pH, suggesting a permanent life in galleries subject to low oxygen concentrations.

According to (Lemieux et al. 1997), the ability to explore a site largely depends on the ability to defend the gallery with a minimum of injuries. These authors have found a significant correlation between aerobic metabolism and the duration of the fight in Nereis virens. They argue that short-term confrontations are advantageous because they increase survival while still limiting the mobilization of anaerobic metabolism. On the other hand, high aerobic activity may be related to high metabolic rates, so the benefit of fights may be somewhat offset by a high metabolic costs for the antagonists, decreasing the energy available for growth and reproduction.

Effects of increased density were also observed to lead to modifications in burrow shape. The absence of connections between the burrows as the density increased suggests that territoriality is certainly the main factor involved. A preference for the centre of the aquarium was observed during this study, i.e. colonization was made from the centre to the periphery of the aquaria at all densities. Scaps et al. (1998), in a study of Perinereis cultrifera, found that the area around the burrow opening is not defended and therefore does not constitute a territory, and they postulate that P. cultrifera probably competes for burrow space but may share food. By contrast, Levin (1982) and Taghon (1992) showed that spacing between neighbours in Spionid species can be affected by the particle supply mechanism and the source of food particles. In Paralvinella sulfincola, a polychaete worm living near deep-sea hydrothermal vents, Grelon et al. (2006) showed that the size of the territory and the control exerted over intruders depends on worm-density and individual body weight, reinforcing the importance of size and aggressiveness in the defence of the foraging area. The method of feeding may also partly explain the observed behaviour. The use of sand and the type of diet, which was deposited directly onto the substrate surface, forced the worms to use their jaws to capture the food particles, therefore exposing their anterior region.

The results obtained suggest that confrontation may have significantly (p<0.001) reduced growth, particularly at medium and high densities and in the first month (Figs 4, 5). The rising growth, which occurred at high density (Fig. 4) in the last two months, can be explained by the sharply declining number of individuals and therefore a decrease in confrontation with the concomitant establishment of undisputed territories. The same pattern was observed for the increased frequency of individuals injured, regenerated and killed. The negative density-dependent effects on growth of Marphysa “sp.” juveniles are similar to those reported for several other polychaete species (Grelon et al. 2006, Safarik 2006, Nesto et al. 2012). The negative density-dependent effects on growth of juvenile and late juvenile M. sanguinea obtained by Parandavar et al. (2015) are similar to those obtained in our study. According to these authors, Marphysa sanguinea juveniles possess the ability to grow better at low density, and survival is negatively influenced by high rearing density. In the present study, survival was negatively influenced by high rearing density, as has also been already reported for N. arenaceodentata (Dillon et al. 1993), D. aciculata (Safarik et al. 2006) and Hediste diversicolor (Nesto et al. 2012). The decreasing survival percentages are likely to be related to the increasing intraspecific competition. Rearing density-related patterns in growth and mortality provide evidence for negative density-dependent interactions with neighbouring worms. According to Safarik et al. (2006), in Diopatra aciculate the negative effects of density on growth are due to intraspecific interactions, including competition for resources other than food. Territoriality can be fundamental for the survival of this species because the individual territory serves primarily for defence and refuge from predators, but also appears to be the primary foraging ground.

The existence of confrontations in the low-density tanks (with 16.7% and 13.3% of broken and regenerating individuals, respectively) confirms the high aggressiveness and territoriality evidenced by this species and suggests that, up to a certain point, the density can be higher than the optimum without necessarily resulting in death. However, in this study the worms were limited to a rigid physical space, and as they grow, they increase not only the body volume but also the territorial surface, obviously increasing the likelihood for encounters. Competition for space has also been shown for Neanthes arenaceodentata, which exhibit aggressive behaviour at high density, as demonstrated by an increase in the frequency of polychaetes regenerating posterior segments (Bridges et al. 1996).

Studies undertaken by (Esnault et al. 1990, Scaps et al. 1993) showed a similar reduction in growth, biomass and survival at high densities, even in the presence of abundant food. Stress associated with frequent handling of Hediste diversicolor and changes in water temperature and salinity have also been found to induce negative effects on growth (Scaps et al. 1993, Garcês and Pereira 2010). The decreasing survival percentages are likely to be related to the increasing intraspecific competition. Rearing density-related patterns in growth and mortality provide evidence for negative density-dependent interactions with neighbouring worms.