Laboratory observations on the feeding behavior and feeding rate of the nemertean Procephalothrix simulus.
Knowledge about the ecology of nemerteans is still limited, despite numerous studies reporting on their feeding behavior or interactions with their prey (Gibson, 1998). In 1985, McDermott and Roe published an excellent review of what was known about the diets, the feeding behavior, and the feeding ecology of nemerteans up to that time. Since then, additional studies on the feeding biology of nemerteans have been published, most of which focused on the impact of nemertean predators on the populations of their prey organisms (Bourque et al., 2001, 2002; Van Son and Thiel, 2006; for other references, see the overview by Thiel and Kruse, 2001), or on the impact of nemertean egg predators on the egg mortality of their decapod hosts (Wickham, 1986; Kuris and Wickham, 1987; Shields and Kuris, 1988a, b; Wickham and Kuris, 1990; Shields et al., 1990a, b; Kuris et al., 1991; Torchin et al., 1996).
The fact that nemertean predators exert a significant impact on the populations of their preferred prey has been documented for several nemertean species, primarily hoplonemerteans and heteronemerteans (McDermott, 1984, 1988; Nordhausen, 1988; Rowell and Woo, 1990; Thiel and Reise, 1993; Kruse and Buhs, 2000; Bourque et al., 2001, 2002; Van Son and Thiel, 2006). Although very little is known about influences on the feeding behavior of nemerteans, a few studies have revealed that their predation rates might be affected by various factors. For example, McDermott (1984) found that temperatures of 20 [degrees]C or above were unfavorable for feeding and survival of Nipponnemertes pulcher (Johnston, 1837). The predation rates of Carcinonemertes epialti Coe, 1902, at 4 and 10 [degrees]C were significantly lower than those at higher temperatures (15 and 20 [degrees]C) (Shields and Kuris, 1988b). The feeding rate of Lineus viridis (Muller, 1774) was affected by tide and photoperiod (Thiel, 1998), and the average predation rates of Prosorhochmus nelsoni (Sanchez, 1973) in treatments with at least five individuals of Hyale maroubrae Stebbing, 1899, were usually higher than those with only one or two prey items (Thiel et al., 2001). Finally, the foraging activity of Cerebratulus lacteus (Leidy, 1851) was increased when the density of C. lacteus was high (Bourque et al., 2002). Most of these reports are anecdotal studies hinting at the importance of these environmental factors on predation rate. Few rigorous tests of the influence of extrinsic factors on the prey consumption of these important benthic predators have been conducted.
Three principal feeding patterns have been described for nemerteans, namely suctorial feeding, macrophagous feeding, and suspension feeding (McDermott and Roe, 1985). All observations on the feeding biology of cephalothricid nemerteans revealed that they were macrophageous feeders (Jennings and Gibson, 1969; McDermott and Roe, 1985), which was defined by the consumption of entire prey organisms rather than by taking fluids and organs from the inside (McDermott and Roe, 1985). Procephalothrix simulus Iwata, 1952, is one of the most common nemerteans along the coasts of Qingdao and vicinity, China, and may be very abundant in certain intertidal habitats. Knowledge about the feeding biology of this nemertean is very limited. We only know that it can prey upon the oligochaete Tubifex sp. and can forage by chemoreception (Wang and Sun, 2006). In the present study, experiments were carried out to examine the feeding behavior of P. simulus and the effects of prey density, temperature, salinity, and photoperiod on its predation rates.
Materials and Methods
Specimens of Procephalothrix simulus were collected from coarse sand or under stones at Taiping Jiao, Qingdao, China. In the laboratory, the nemertean consumed all the foods with an animal origin that were offered, including unidentified marine nematodes, Tubifex sp., Saccocirrus gabrillae Marcus, 1946, Capitella capitata (Fabricius, 1780), meat of the shrimp Litopenaeus vannamei (Boone, 1931), and meat of crabs from the genus Grapsus. Two annelids, Tubifex sp. and S. gabrillae, were chosen as prey items in the present study because both could be obtained easily and had similar shape and size. The freshwater oligochaete Tubifex sp., which is a preferred prey of P. simulus, was obtained from a local pet market. The marine polychaete S. gabrillae was collected from the same habitats as P. simulus, where both worm species were abundant in spring and autumn.
Feeding behavior experiments
In these experiments, nemertean worms were maintained in the laboratory in seawater (practical salinity [S] = 31) at 16-17 [degrees]C and fed with Tubifex sp. for at least 7 days. To increase the appetite of the worms, feeding was stopped 2 weeks prior to experiments.
Predation on Tubifex sp. and Saccocirrus gabrillae by nemerteans not size-screened. To observe the feeding behavior of P. simulus, two experiments were conducted using S. gabrillae and Tubifex sp. (Tubifex was immobilized after being transferred to seawater) as prey items. To understand the overall patterns of feeding behavior of this nemertean, randomly selected individuals were used; that is, both nemerteans and prey items were selected regardless of their size. The wet weight of nemertean worms was 9-39 mg (mean [+ or -] SD = 19.8 [+ or -] 8.1). The body length of the prey S. gabrillae was 11.2-22.9 mm (mean [+ or -] SD = 17.7 [+ or -] 2.8), and that of Tubifex sp. was 9.3-26.6 mm (mean [+ or -] SD = 16.8 [+ or -] 4.1). The average body weight of these two prey items was about 2.0 mg and 2.2 mg, respectively.
At the beginning of each replicate, a single nemertean was placed together with about 20 prey items in a petri dish (diameter: 12 cm; height: 2.2 cm). The behavior of the nemerteans was carefully observed. If a prey organism was attacked, subsequent events and their duration were recorded. If a nemertean did not attack the prey after 15 direct body contacts with prey items, the nemertean was replaced with another one. A total of 100 attacks were recorded in the Tubifex experiment, and 200 attacks were recorded in the S. gabrillae experiment. A nemertean worm was used for observing only a single attack.
Comparison of the feeding behavior of nemerteans of different size. Three treatment groups, nemerteans with body masses of 8-12 mg, 18-22 mg, and 28-32 mg, were used to examine the effect of predator size on feeding behavior. S. gabrillae, body length 16-18 mm (average body weight about 1.6 mg), was chosen as the prey. For each treatment, 90 attacks were observed. The experimental procedure was the same as described in the previous section.
In the laboratory, P. simulus was maintained at room temperature (20-25 [degrees]C) and at a practical salinity of S = 31 for at least 7 days, by supplying S. gabrillae as diet. Seven days before each experiment, the nemerteans were transferred into petri dishes (diameter: 12 cm; height: 2.2 cm) each containing 100 ml of seawater (one individual per dish), and given a 4-day acclimation to the experimental conditions. During this period, the temperatures for maintaining the worms to be used in temperature trials were adjusted to treatment levels at a rate of 4 [degrees]C per day; and the practical salinities for the salinity experiment were adjusted to treatment levels at a rate of S = 5 per day. Feeding was stopped 72 h prior to the experiment. The mean [+ or -] SD for the wet weight of the nemerteans was 33.5 [+ or -] 4.9 (range 27-47) mg. The mean [+ or -] SD for the body length of S. gabrillae was 13.1 [+ or -] 2.2 (9.0-20.2) mm (average body weight about 1.7 mg).
Four experiments were conducted to determine the effects of (i) prey density, (ii) temperature, (iii) salinity, and (iv) photoperiod on the feeding rate of P. simulus. Each experiment had 5-7 treatments (see below), and each treatment had 20 replicates (with the exception of the "prey density" experiment, which had 10 replicates per treatment). All trials were performed in the above-mentioned 12-cm petri dishes, each containing 100 ml of seawater. At the beginning of an experiment, 1 nemertean and 10 prey items were assigned to each dish (except in the prey-density experiment). The duration of each experiment was 7 days. The number of S. gabrillae preyed upon by the nemertean in each dish was recorded at about 0800 every day, and prey items were replaced to starting conditions.
All experiments were conducted at a temperature of 25 [+ or -] 0.5 [degrees]C, a practial salinity of S = 31, and a photoperiod of 12 h light/12 h dark (12L/12D), unless a factor was the treatment factor in the experiment. The seawater used was filtered through a 3-[micro]m cartridge filter, and a complete water exchange was performed every day.
(i) Effect of prey density. The experiment was conducted at six prey densities (1, 5, 10, 20, 40, and 60 individuals of S. gabrillae per petri dish).
To examine whether the oxygen decrease and metabolic waste products could impact the feeding rate when prey density was very high, a supplementary trial was conducted. After the S. gabrillae prey were held in petri dishes (100 ml of seawater and 60 individuals per petri dish) for 24 h, they were removed from the dishes. The seawater left in the petri dishes was then used as treatment medium, and fresh seawater was used as control. The test was conducted for 24 h at a prey density of 10 individuals per petri dish. The feeding rate of nemerteans in the treatment (1.55 [+ or -] 0.12 prey items nemertea[n.sup.-1] [day.sup.-1]) and in the control (1.50 [+ or -] 0.21 prey items nemertea[n.sup.-1] [day.sup.-1]) showed no significant difference (t = 0.208, df = 18, P = 0.838), suggesting that water quality alteration from high prey densities could not result in significant impact on the feeding rate of P. simulus.
(ii) Effect of temperature. The temperature experiment was conducted at seven temperature levels (5, 10, 15, 20, 25, 30, and 32 [degrees]C), and the upper temperature was determined according to the temperature tolerance of P. simulus (Zhao and Sun, 2006). In addition, the lower temperature limit for the nemertean to consume prey items was determined by supplying S. gabrillae under various lower temperatures.
(iii) Effect of salinity. This experiment was conducted at five levels of practical salinity: S = 10, 20, 30, 40, and 45. Hyper-salinity media were prepared by adding NaCl to natural seawater (S = 31), and diluted media were prepared by adding distilled water to seawater.
(iv) Effect of photoperiod. Five photoperiod treatments, 24L/0D (continuous light), 18L/6D (0200-0800 dark), 12L/12D (2000-0800 dark), 6L/18D (0800-1400 light), and 0L/24D (continuous dark), were established. A programmed time controller was used to maintain the periods of light and dark. In the 0L/24D treatment, the nemerteans were transferred to another petri dish in darkness while the petri dishes were checked for the number of prey items consumed.
In each experiment, potential differences among treatments were determined by one-way ANOVAs (all data passed the test for homogeneity of variance) followed by Tukey's multiple range tests, or by Student's t-test when testing the difference between two means. Differences were considered significant at a probability level of 0.05. All analyses were performed using SPSS 11.0 statistical software.
Patterns of feeding behavior
In general, Procephalothrix simulus consumed entire prey organisms rather than taking fluids and tissues from inside the prey, thus showing a typical macrophagous feeding strategy. Prey organisms were apparently located by chemoreception. Nemertean worms that had been gliding slowly in their containers became restless and repeatedly dilated the mouth when Tubifex was added. The initiation of foraging activity was even more pronounced when the extracts of Tubifex were added to the dishes. When the head of a nemertean contacted a prey item, the nemertean normally everted its proboscis in a spiral coil around the prey and pulled it toward the ventral mouth. Then the nemertean retracted the proboscis, losing contact with the prey. Usually, after the prey was immobilized or partially immobilized, the body anterior to the mouth was raised and the prey item was ingested as a whole. Occasionally a nemertean was observed to forage while on its back or side. Sometimes the nemertean did not lose contact with the prey and ingested it directly (especially for Tubifex). Non-active food items (Tubifex) were occasionally ingested without the aid of the proboscis, but these cases were mostly observed when prey items were directly contacted by the nemertean's mouth.
Feeding patterns of nemerteans not size-screened. When Tubifex was used as prey, the feeding behavior of P. simulus could be categorized into five patterns: (1) Feeding without use of proboscis: when a nemertean contacted a specimen of Tubifex with its mouth region, the nemertean did not evert its proboscis but directly ingested the prey by dilating its mouth. (2) Everting proboscis during ingestion: this pattern was similar to the former one except that the proboscis was everted during prey ingestion; it seemed that proboscis eversion normally occurred when a prey was difficult to ingest--for example, when the relative size of the prey was large or when ingestion started from a nonterminal portion of the prey; in that case the prey would be bent into the shape of a V or a J, as reported for Paranemertes peregrina Coe, 1901 (Roe, 1970). (3) Everting proboscis but retracted before ingestion: the proboscis was everted when the head of a nemertean contacted a prey but retracted before the ingestion started. (4) Prey entwined by proboscis and ingested: when the anterior edge of a nemertean's head contacted a prey, it everted its proboscis to entwine the prey, and ingestion started while the prey was entwined. (5) Feeding with the proboscis everted for two or three times: this pattern was similar to the last one, but the proboscis was re-everted once or twice during the ingesting process. Results shown in Table 1 indicated that all attacks in this experiment were successful (the prey being attacked was always ingested by the nemertean), and 25% of predation events did not involve the assistance of the proboscis. The duration of an entire feeding sequence ranged from 31.8 to 239.8 s (mean [+ or -] SD = 89.5 [+ or -] 42.3); and the more complex the feeding behavior was, the longer the feeding process took (Table 1).
When the polychaete S. gabrillae was used as prey, direct ingestion (without use of proboscis) was not observed. The attacks of P. simulus on S. gabrillae, which invariably were accompanied by the eversion of the proboscis, did not always result in prey ingestion. The performance of P. simulus following an attack could be categorized in the following five patterns: (1) AI (Ingestion while attacking with proboscis): when a nemertean contacted a prey, the former everted its proboscis, entwined the prey, and then ingested it. The duration of the whole feeding process was about 250.1 [+ or -] 228.6 s (29.6-1071.4 s). (2) API (Attacking with proboscis--disengagement--prey paralyzed--ingestion): when a nemertean contacted the prey, it everted its proboscis and the prey was entwined by the proboscis for a moment. During this period the prey was paralyzed. Ingestion occurred after the proboscis was retracted. The time for a nemertean to ingest a speciment of S. gabrillae was 284.8 [+ or -] 189.2 s (46.5-633.5 s). (3) ANPI (Attacking with proboscis--disengagement--ingestion [prey not paralyzed]): this was similar to API except that the prey was not paralyzed when ingestion started. The time for a nemertean to ingest a specimen of S. gabrillae was 350.1 [+ or -] 307.3 s (60.3-971.8 s). (4) AP (Attacking with proboscis--disengagement--prey paralyzed but not ingested): prey was attacked and paralyzed by the proboscis of a nemertean, but the nemertean did not try to ingest the prey. The time for a nemertean to paralyze a apecimen of S. gabrillae was 321.9 [+ or -] 172.5 s (55.6-864.6 s). (5) ANP (Attacking with proboscis--disengagement--prey not paralyzed and not ingested): prey was attacked by the everted proboscis of a nemertean, but the prey was not paralyzed, and the nemertean did not try to ingest the prey.
In the S. gabrillae experiment, only 46% of all attacks resulted in successful ingestions, and attacks always relied on the assistance of the proboscis (Table 2). The feeding sequence lasted longer than that in the Tubifex experiment. The mean feeding duration of the simplest successful predation pattern (AI) in the S. gabrillae experiment was 250.1 [+ or -] 228.6 s (Table 2), almost three times as long as the average duration for feeding on Tubifex (89.5 [+ or -] 42.3 s).
Difference in the feeding patterns among nemerteans of different sizes. The observation mentioned in the previous section that prey items with relatively larger sizes were more difficult for P. simulus to ingest was further proven by this experiment. The rate of successful attacks increased when the body weight of nemerteans increased, while the duration of the feeding events (the entire feeding time, entwining time, paralyzing time, or ingesting time) decreased with increasing predator sizes (Table 3). Table 3 also showed that 30.0% of the attacks by 28-32-mg nemerteans resulted in uninterrupted ingestion (AI), while this was only 21.1% and 10.0% for 18-22-mg and 8-12-mg worms, respectively. In summary, small nemerteans required more time and effort than larger nemerteans when preying upon S. gabrillae.
Effect of prey density. The average predation rate of P. simulus on S. gabrillae varied from 0.71 [+ or -] 0.26 to 1.36 [+ or -] 0.72 prey items nemertea[n.sup.-1] [day.sup.-1] in treatments of five prey densities. The results shown in Figure 1 indicated that the average predation rate in the treatment with five prey items was slightly higher than that in the other treatments (Fig. 1), but the difference was not statistically significant (P = 0.161).
Effect of temperature. Laboratory tests showed that P. simulus did not prey on S. gabrillae at 2 [degrees]C, but did consume the prey at 3 [degrees]C. The feeding rate of P. simulus increased significantly when the temperature was elevated from 5 [degrees]C to 30 [degrees]C, but the extremely high temperature (32 [degrees]C) inhibited predation in this nemertean (Fig. 2). The peak value observed at 30 [degrees]C was 2.58 [+ or -] 0.61 prey items nemertea[n.sup.-1] [day.sup.-1].
Effect of salinity. Figure 3 indicated that feeding rates of P. simulus at practical salinities of S = 30 and 40 were significantly higher than those at other salinities, and the predation rate at S = 10 was significantly lower than at other salinity levels. The highest predation rate obtained at S = 40 was 2.27 [+ or -] 0.66 prey items nemertea[n.sup.-1] [day.sup.-1], almost four times the value at S = 10 (0.64 [+ or -] 0.30 prey items nemertea[n.sup.-1] [day.sup.-1]).
[FIGURE 1 OMITTED]
Effect of photoperiod. The feeding rates of P. simulus under different photoperiods are shown in Figure 4. The results demonstrated that photoperiod had a significant effect on the feeding rate of the nemertean (P = 0.001). The feeding rates of nemerteans under the photoperiods of 12L/12D, 6L/18D, and 0L/24D were significantly higher than those under 24L/0D ([P.sub.24L, 12L] = 0.008, [P.sub.24L, 6L] = 0.002, [P.sub.24L, 0L] = 0.034), suggesting that predation may be inhibited by light. The peak value observed at 6L/18D was 2.51 [+ or -] 0.61 prey items nemertea[n.sup.-1] [day.sup.-1], while the lowest value observed in the 24L/0D treatment was 1.73 [+ or -] 0.59 prey items nemertea[n.sup.-1] [day.sup.-1].
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
Procephalothrix simulus is a carnivorous nemertean that has no strict selectivity for particular food items (see Materials and Methods). It can capture living polychaetes and nematodes, but it seems to prefer inert food items (immobilized Tubifex). Prey items are apparently detected by chemoreception (its foraging activity can be stimulated by prey organisms and prey extracts). These results are in accord with previous observations on nemerteans of the family Cephalothricidae (Jennings and Gibson, 1969; Wang and Sun, 2006). Cannibalism, which has been reported in some nemerteans (Gibson, 1972), was also noted in this species. It was mostly found in freshly collected specimens (starved worms?), and in all cases the body fragments of damaged individuals were swallowed by intact worms.
[FIGURE 4 OMITTED]
Nemerteans that use the macrophagous feeding strategy have been known either to catch their active living prey with their proboscis, killing or at least partially immobilizing it with the proboscis secretions before ingestion, or to feed directly on inactive or decaying food material without prior use of the proboscis (Gibson, 1994). Both feeding strategies were observed in the present study, in which P. simulus did not always evert the proboscis when the prey was inactive (Tubifex) (Table 1), but invariably everted the proboscis when attacking active prey (Saccocirrus gabrillae) (Table 2). Similarly, in two Cephalothrix species, Cephalothrix rufifrons (Johnston, 1837) and Cephalothrix linearis (Rathke, 1799), the proboscis was never everted when taking inactive food (Jennings and Gibson, 1969). Observations on other nemerteans--for example Lineus ruber (Muller, 1774) and Cerebratulus lacteus--also showed that they did not evert the proboscis when consuming inert food (Jennings and Gibson, 1969; Bourque et al., 2002). When a predation event was aided by the proboscis, P. simulus sometimes lost proboscis contact with the prey before ingestion, or re-everted the proboscis during the feeding sequence. This was similar to Paranemertes peregrina and Lineus viridis, which might lose contact with a prey item after the prey was paralyzed (Roe, 1970; Nordhausen, 1988), but differed from L. ruber, which did not lose proboscis contact with the prey while searching for a place to start ingestion (Beklemishev, 1955; Jennings, 1960).
The reported duration of feeding in nemerteans varies greatly, and data on this aspect were obtained mostly from observations on monostiliferous hoplonemerteans. For example, the entire feeding process of the polychaete-feeding nemertean Paranemertes peregrina could take place in 2-3 min (Roe, 1970); the mean feeding duration of the amphipod-feeding nemertean Oerstedia dorsalis (Abildgaard, 1806) was about 7 min, with the entire feeding sequence (from the attack to the termination of feeding) ranging from 11 to 34 min (McDermott and Snyder, 1988); in other amphipod-feeding nemerteans, the feeding duration of Zygonemertes virescens (Verrill, 1879) was 3-10 min (McDermott, 1976), that of Tetrastemma melanocephalum (Johnston, 1837) was [greater than or equal to] 4 min (Bartsch, 1973), and that of Amphiporus lactifloreus (Johnston, 1828) and Nipponnemertes pulcher might take more than half an hour (Jennings and Gibson, 1969; McDermott, 1984). Feeding duration might be determined not only by predator species, but also by some extrinsic factors. Besides the fact that P. simulus showed different patterns of feeding behavior when preying upon different prey items (see preceding paragraph), the nature of prey also has a significant effect on feeding duration. When preying upon the non-active oligochaete, P. simulus had an average feeding duration of 89.5 [+ or -] 42.3 s; when the prey was the active living polychaete, P. simulus took much longer to finish an ingestion (the duration of the simplest feeding pattern [AI] was 250.1 [+ or -] 228.6 s, Table 2).
The relative size of predator and prey is another factor that affects the feeding behavior and feeding duration of P. simulus. The rate of successful attacks by P. simulus increased and the duration of the feeding sequence decreased with increasing nemertean body mass (Table 3). For other macrophagous nemerteans, Jennings and Gibson (1969) reported that the ingestion process of L. ruber and Ramphogordius sanguineus (Rathke, 1799) might take minutes to hours depending on the size of the prey items. Similarly, observations on three species of hoplonemerteans showed that the time for immobilizing prey items was inversely related to the size of the nemertean and directly related to the size of the prey (McDermott, 1976). Studies on other animals also showed positive correlations between the sizes of the predator and its preferred prey (e.g., Sanchez-Salazar et al., 1987; Brown, 1997; Menn and Armonies, 1999; Mayfield et al., 2001).
The feeding behavior of P. simulus might also be determined by where in its body the prey was attacked. If a prey was ingested starting in its median body region, it would be bent into a V- or J-shape, and the nemertean tended to evert its proboscis several times during ingestion. On the basis of similar observations in Paranemertes peregrina, Roe (1970) suggested that the limiting factor in ingestion is prey diameter rather than prey length. McDermott and Roe (1985) concluded that "the available size of prey is generally limited by the nemertean diameter or dilatability of the mouth; therefore prey is usually of wormlike proportions." Another interesting example documented by Bourque et al. (2002) showed that when Cerebratulus lacteus attacked the bivalve Mya arenaria Linnaeus, 1758, the retreat of the nemertean into sediment was related to the type of proboscis eversion. In these observations, C. lacteus never retreated into the sediment unless its proboscis penetrated the clam M. arenaria, but 80% of the nemerteans retreated into the sediment (waiting for the clam to be weakened by toxins) when the proboscis was everted into the mantle cavity of the clam.
Thiel and Kruse (2001) supposed that "numbers of available prey items probably strongly affect feeding rates of nemerteans as has been shown for polychaete or flatworm predators." Although this is possible, most studies done up to now do not show a strong relationship between prey numbers and feeding rate. For example, significant effects of prey density on the feeding rate of nemerteans were not found in either the experiment by Thiel et al. (2001) or the present study. One possible reason is that nemerteans normally consume only a few prey items each day and the number varies greatly among individuals, resulting in large standard deviations when average feeding rates are calculated. However, Thiel et al. (2001) noticed that Prosorhochmus nelsoni usually had higher average predation rates in treatments with at least five individuals of Hyale maroubrae than in those with only one or two prey items. A similar tendency was also observed in the present prey-density experiment (the average predation rate in treatment with one specimen of S. gabrillae per dish was lower than that in the other treatments). Thiel et al. (2001) suggested that nemerteans in the low-density treatment repeatedly consumed all prey items in a petri dish, after which there were no prey left to be consumed until new ones were added during the next control.
The present observations demonstrated that P. simulus consumed prey items throughout a temperature range of 3 [degrees]C to 32 [degrees]C. Feeding rates increased when the temperature was elevated from 5 [degrees]C to 30 [degrees]C, but they significantly decreased when the temperature was elevated to 32 [degrees]C. The year-round temperature of seawater in Qingdao is about 3 [degrees]C to 27 [degrees]C (Office of Qingdao Local Records, 1997). In the littoral zone, however, as noted by Zhao and Sun (2006), temperatures could be much lower in winter and much higher in summer. This suggests that the feeding of P. simulus may vary greatly among different seasons and even completely cease on extremely cold or hot days. For another nemertean, McDermott (1984) reported that temperatures of 20 [degrees]C or more were unfavorable for feeding and survival of Nipponnemertes pulcher. The tendency for the feeding rate to be stimulated by the elevation of temperature at lower temperature ranges and inhibited by very high temperature has been demonstrated in many aquatic animals (e.g., Dou et al., 2000; Moens and Vincx, 2000; Dubber et al., 2004; Sheng et al., 2006), but the inflection point may vary among animals. The temperature-induced elevation of feeding rate in lower temperature ranges should be mostly a manifestation of the "rate effect" or "[Q.sub.10] effect," which is a basic effect of temperature on the physiological process of animals (Somero, 1997). The significant inhibition of food intake at sublethal temperatures (e.g., 32 [degrees]C is near the upper survival limit of P. simulus; see Zhao and Sun, 2006) is likely to reflect the disruption of physiological mechanisms by thermal stress. For example, sublethal temperatures can induce structurally minor but biologically significant changes to protein conformation or denaturation (Somero, 1997).
With respect to the effect of salinity on food intake, no data have been published for nemerteans. Zhao and Sun (2006) demonstrated that P. simulus possessed strong salinity tolerance. The present salinity experiment, though not able to determine the salinity limits for feeding, show that P. simulus consumes prey in a wide range of practical salinity (at least from S = 10 to 45). Its feeding rate is relatively high at S = 30 and 40, but decreased significantly in media diluted to S = 20 and 10 and in medium with salinity as high as 45 (Fig. 3). The salinity of the seawater of Qingdao is quite stable, normally varying from S = 31.5 to 32.6 (Office of Qingdao Local Records, 1997). Thus the feeding of P. simulus should be scarcely limited by the natural salinity alternation of the local seawater, although it may be temporarily impacted by heavy rainfall, which can greatly decrease the salinity level in the intertidal zone. The phenomenon that salinity changes could affect food intake has also been documented in many marine animals from other phyla (e.g., Kinne, 1971; Vernberg and Vernberg, 1972; Cheung, 1997; Pechenik et al., 2000; Normant and Lamprecht, 2006). However, none of these studies gave explanations for the proximate cause that salinity stress affected food intake. Although some nemerteans are known to possess the capacity for volume/weight regulation, documented data showed that they could not restore volume/weight to the normal level, and such capacity was limited by the exposure salinity levels (e.g., Ferraris and Schmidt-Nielsen, 1982; Ferraris and Norenburg, 1988; Zhao and Sun, 2006). Since maintenance of a proper cellular volume (and therefore of a proper spatial organization of macromolecules) is a basic requisite of life (Gilles and Delpire, 1997), cellular function may be damaged when nemerteans are exposed to a salinity stress, thereby affecting their metabolic processes.
The feeding rate of P. simulus reared under photoperiods of 12L/12D, 6L/18D, and 0L/24D were significantly higher than that under 24L/0D (Fig. 4). The lower food intake under continuous light is likely related to the fact that light can inhibit the activity of the nemertean. Our field observations showed that many individuals of P. simulus came out of their hiding places and crawled on the sediment surface during low tide at night, but almost all the nemerteans hid in sand or under stones during low tide in daylight. High foraging activity at night has also been reported for several other nemertean species (Roe, 1970, 1976; Thiel et al., 1995; Thiel, 1998). The mechanism by which light controls the behavior of nemerteans has not yet been determined. Thiel (1998) suggested that the nemertean preference for nocturnal activity may be due in part to the risk of desiccation during daytime low tides or to reduced competition with fast visual epibenthic predators. Viherluoto and Viitasalo (2001) presumed that in the mysid shrimp Mysis mixta (Lilljeborg, 1852), light induced an endogenous reaction resulting in cessation of movement and decrease of feeding, which may also be the case for skotophilous nemerteans.
In summary, P. simulus is a carnivorous predator or scavenger that consumes a wide variety of foods. It employs the typical macrophagous feeding strategy with particular variations (use of proboscis, immobilization of prey) that might be determined by the nature of prey items, the relative size of nemerteans and prey organisms, and the position where a prey is attacked. Feeding of P. simulus takes place in wide ranges of temperature and salinity and under different light conditions, but temperature, salinity, and photoperiod have significant impact on the feeding rate of this nemertean. The diversity in food spectrum, variation in feeding behavior patterns, and eurytopicity of feeding to environmental factors suggest that this nemertean is a species successfully adapting to the variable environmental conditions of the intertidal habitat. Thiel and Kruse (2001) commented that (because nemerteans' predation is unique in relying on their rapidly everted proboscis and highly potent toxins; and their chemosensory system is strongly developed, permitting them to remain on the trail of a prey item once having "smelled the rat") nemerteans might play a crucial role in marine habitats in which they occur in high abundance. The present results, which consist mainly of laboratory observations on feeding behavior and single-factor effects on feeding rate, have limited applicability in evaluating the role of P. simulus in natural systems; however, as mentioned in previous sections, P. simulus might be very abundant and aggregated in certain microhabitats in some seasons, suggesting that it may exert predation pressure on the populations of potential prey organisms such as polychaetes and nematodes.
The study is supported by the National Natural Science Foundation of China (30270235). We are grateful to Professor Dejian Yang for identifying the polychaetes, and to Dr. Ying Liang for kindly reading the manuscript. Our thanks also go to Dr. Martin Thiel and two anonymous reviewers for their important critiques on the earlier version of the manuscript.
Bartsch, I. 1973. Zur Nahrungsaufnahme von Tetrastemma melanocephalum (Nemertini). Helgol. Wiss. Meeresunters 25: 326-331.
Beklemishev, K. V. 1955. Predatory nemertines. Priroda 9: 108-109.
Bourque, D., G. Miron, and T. Landry. 2001. Predation on soft-shell clams (Mya arenaria) by the nemertean Cerebratulus lacteus in Atlantic Canada: implications for control measures. Hydrobiologia 456: 33-44.
Bourque, D., G. Miron, and T. Landry. 2002. Predator-prey relationship between the nemertean Cerebratulus lacteus and the soft-shell clam, Mya arenaria: surface-exploration activity and qualitative observations on feeding behaviour. Can. J. Zool. 80: 1204-1211.
Brown, K. M. 1997. Size-specific aspects of the foraging ecology of the southern oyster drill Stramonita haemastoma (Kool 1987). J. Exp. Mar. Biol. Ecol. 214: 249-262.
Cheung, S. G. 1997. Physiological and behavioural responses of the intertidal scavenging gastropod Nassarius festivus to salinity changes. Mar. Biol. 129: 301-307.
Dou, S., T. Seikai, and K. Tsukamoto. 2000. Feeding behaviour of Japanese flounder larvae under laboratory conditions. J. Fish Biol. 56: 654-666.
Dubber, G. G., G. M. Branch, and L. J. Atkinson. 2004. The effect of temperature and diet on the survival, growth and food uptake of aquarium-held postpueruli of the rock lobster Jasus lalandii. Aquaculture 240: 249-266.
Ferraris, J. D., and B. Schmidt-Nielsen. 1982. Volume regulation in an intertidal nemertine, Procephalothrix spiralis Coe. I. Short-term effects and independence on decerebration. J. Exp. Zool. 224: 307-319.
Ferraris, J. D., and J. L. Norenburg. 1988. Water and solute regulation in Procephalothrix spiralis Coe and Clitellio arenarius (Muller). Long-term acclimation to diluted seawaters and effect of putative neuroendocrine structures. Hydrobiologia 156: 161-173.
Gibson, R. 1972. Nemerteans. Hutchinson University Library, London.
Gibson, R. 1994. Nemerteans. Synopsis of the British Fauna (New Series), No. 24, 2nd ed. Field Studies Council, London.
Gibson, R. 1998. Epilogue--one hundred years of nemertean research: Burger (1895) to the present. Hydrobiologia 365: 301-310.
Gilles, R., and E. Delpire. 1997. Variations in salinity, osmolarity, and water availability: vertebrates and invertebrates. Pp. 1523-1589 in Handbook of Physiology, Section 13: Comparative Physiology, Vol. 2. W. H. Dantzler, ed. Oxford University Press, New York.
Jennings, J. B. 1960. Observations on the nutrition of the rhynchocoelan Lineus ruber (O. F. Muller). Biol. Bull. 119: 189-196.
Jennings, J. B., and R. Gibson. 1969. Observations on the nutrition of seven species of rhynchocoelan worms. Biol. Bull. 136: 405-433.
Kinne, O. 1971. A comprehensive, integrated treatise on life in oceans and coastal waters. Pp. 821-954 in Marine Ecology, Vol. 1, O. Kinne, ed. John Wiley, New York.
Kruse, I., and F. Buhs. 2000. Preying at the edge of the sea: the nemertine Tetrastemma melanocephalum and its amphipod prey on high intertidal sandflats. Hydrobiologia 426: 43-55.
Kuris, A. M., and D. E. Wickham. 1987. Effect of nemertean egg predators on crustaceans. Bull. Mar. Sci. 41: 151-164.
Kuris, A. M., S. F. Blau, A. J. Paul, J. D. Shields, and D. E. Wickham. 1991. Infestation by brood symbionts and their impact on egg mortality of the red king crab, Paralithodes camtschatica, in Alaska: geographic and temporal variation. Can. J. Fish. Aquat. Sci. 48: 559-568.
Mayfield, S., E. de Beer, and G. M. Branch. 2001. Prey preference and the consumption of sea urchins and juvenile abalone by captive rock lobsters (Jasus lalandii). Mar. Freshw. Res. 52: 773-780.
McDermott, J. J. 1976. Observations on the food and feeding behavior of estuarine nemertean worms belonging to the order Hoplonemertea. Biol. Bull. 150: 57-68.
McDermott, J. J. 1984. The feeding biology of Nipponnemertes pulcher (Johnston) (Hoplonemertea), with some ecological implications. Ophelia 23: 1-21.
McDermott, J. J. 1988. The role of hoplonemerteans in the ecology of seagrass communities. Hydrobiologia 156: 1-11.
McDermott, J. J., and P. Roe. 1985. Food, feeding behavior and feeding ecology of nemerteans. Am. Zool. 25: 113-125.
McDermott, J. J., and R. L. Snyder. 1988. Food and feeding behavior of the hoplonemertean Oerstedia dorsalis. Hydrobiologia 156: 47-51.
Menn, I., and W. Armonies. 1999. Predatory Promesostoma species (Platyhelminthes, Rhabdocoela) in the Wadden Sea. J. Sea Res. 41: 309-320.
Moens, T., and M. Vincx. 2000. Temperature, salinity and food thresholds in two brackish-water bacterivorous nematode species: assessing niches from food absorption and respiration experiments. J. Exp. Mar. Biol. Ecol. 243: 137-154.
Nordhausen, W. 1988. Impact of the nemertean Lineus viridis on its polychaete prey on an intertidal sandflat. Hydrobiologia 156: 39-46.
Normant, M., and I. Lamprecht. 2006. Does scope for growth change as a result of salinity stress in the amphipod Gammarus oceanicus? J. Exp. Mar. Biol. Ecol. 334: 158-163.
Office of Qingdao Local Records. 1997. Local Records of Qingdao: Oceanographic Records. Xinhua Publishing House, Beijing. (in Chinese).
Pechenik, J. A., R. Berard, and L. Kerr. 2000. Effects of reduced salinity on survival, growth, reproductive success, and energetics of the euryhaline polychaete Capitella sp. I. J. Exp. Mar. Biol. Ecol. 254: 19-35.
Roe, P. 1970. The nutrition of Paranemertes peregrina (Rhynchocoela: Hoplonemertea). I. Studies on food and feeding behavior. Biol. Bull. 139: 80-91.
Roe, P. 1976. Life history and predator-prey interactions of the nemertean Paranemertes peregrina Coe. Biol. Bull. 150: 80-106.
Rowell, T. W., and P. Woo. 1990. Predation by the nemertean worm, Cerebratulus lacteus Verrill, on the soft-shell clam, Mya arenaria Linnaeus, 1758, and its apparent role in the destruction of a clam flat. J. Shellfish Res. 9: 291-297.
Sanchez-Salazar, M. E., C. L. Griffiths, and R. Seed. 1987. The effect of size and temperature on the predation of cockles Cerastoderma edule (L.) by the shore crab Carcinus maenas (L.). J. Exp. Mar. Biol. Ecol. 111: 181-193.
Sheng, J.-Q., Q. Lin, Q.-X. Chen, Y.-L. Gao, L. Shen, and J.-Y. Lu. 2006. Effects of food, temperature and light intensity on the feeding behavior of three-spot juvenile seahorses, Hippocampus trimaculatus Leach. Aquaculture 256: 596-607.
Shields, J. D., and A. M. Kuris. 1988a. Temporal variation in abundance of the egg predator Carcinonemertes epialti (Nemertea) and its effect on egg mortality of its host, the shore crab, Hemigrapsus oregonensis. Hydrobiologia 156: 31-38.
Shields, J. D., and A. M. Kuris. 1988b. An in vitro analysis of egg mortality in Cancer anthonyi: the role of symbionts and temperature. Biol. Bull. 174: 267-275.
Shields, J. D., R. K. Okazaki, and A. M. Kuris. 1990a. Brood mortality and egg predation by the nemertean, Carcinonemertes epialti, on the yellow rock crab, Cancer anthonyi, in southern California. Can. J. Fish. Aquat. Sci. 47: 1275-1281.
Shields, J. D., D. E. Wickham, S. F. Blau, and A. M. Kuris. 1990b. Some implications of egg mortality caused by symbiotic nemerteans for data acquisition and management strategies of red king crabs Paralithodes camtschatica. Pp. 383-395 in Proceedings of the International Symposium on King and Tanner Crabs, Lowell Wakefield Symposium Series, Nov., 1989. Anchorage, AK.
Somero, G. N. 1997. Temperature relationships: from molecules to biogeography. Pp. 1391-1444 in Handbook of Physiology, Section 13: Comparative Physiology, Vol. 2, W. H. Dantzler, ed. Oxford University Press, New York.
Thiel, M. 1998. Nemertines as predators on tidal flats--high noon at low tide. Hydrobiologia 365: 241-250.
Thiel, M., and I. Kruse. 2001. Status of the Nemertea as predators in marine ecosystems. Hydrobiologia 456: 21-32.
Thiel, M., and K. Reise. 1993. Interaction of nemertines and their prey on tidal flats. Neth. J. Sea Res. 31: 163-172.
Thiel, M., W. Nordhausen, and K. Reise. 1995. Nocturnal surface activity of endobenthic nemertines on tidal flats. Pp. 283-293 in Biology and Ecology of Shallow Coastal Waters, Proceedings of the 28th European Marine Biology Symposium, Iraklio, Crete, 1993, A. Eleftheriou, A. D. Ansell, and C. J. Smith, eds. Olsen & Olsen, Fredensborg, Denmark.
Thiel, M., N. Ullrich, and N. Vasquez. 2001. Predation rates of nemertean predators: the case of a rocky shore hoplonemertean feeding on amphipods. Hydrobiologia 456: 45-57.
Torchin, M. E., K. D. Lafferty, and A. M. Kuris. 1996. Infestation of an introduced host, the European green crab, Carcinus maenas, by a symbiotic nemertean egg predator, Carcinonemertes epialti. J. Parasitol. 82: 449-453.
Van Son, T. C., and M. Thiel. 2006. Multiple predator effects in an intertidal food web. J. Anim. Ecol. 75: 25-32.
Vernberg, W. B., and F. J. Vernberg. 1972. Environmental Physiology of Marine Animals. Springer-Verlag, Berlin. Chinese edition translated by T.-F. Song, published in 1982 by Agricultural Press, Beijing.
Viherluoto, M., and M. Viitasalo. 2001. Effect of light on the feeding rates of pelagic and littoral mysid shrimps: a trade-off between feeding success and predation avoidance. J. Exp. Mar. Biol. Ecol. 261: 237-244.
Wang, H., and S. Sun. 2006. Studies on the chemoreception of the nemertean, Procephalothrix simulus Iwata, 1952. J. Exp. Mar. Biol. Ecol. 336: 146-152.
Wickham, D. E. 1986. Epizootic infestations by nemertean brood parasites on commercially important crustaceans. Can. J. Fish. Aquat. Sci. 43: 2295-2302.
Wickham, D. E., and A. Kuris. 1990. Brood mortality estimation and role of symbionts in egg mortality of the tanner crab Chionoecetes bairdi in Alaskan waters. Pp. 397-402 in Proceedings of the International Symposium on King and Tanner Crabs, Lowell Wakefield Symposium Series, Nov. 1989. Anchorage, AK.
Zhao, Y., and S. Sun. 2006. Effects of salinity, temperature and pH on the survival of the nemertean Procephalothrix simulus Iwata, 1952. J. Exp. Mar. Biol. Ecol. 328: 168-176.
HAIYAN WANG, SHICHUN SUN*, AND QINGLONG LI
Mariculture Research Laboratory, Ocean University of China, 5 Yushan Road, Qingdao 266003, China
Received 24 February 2007; accepted 8 December 2007.
* To whom correspondence should be addressed. E-mail: firstname.lastname@example.org
Table 1 Statistics of the feeding behavior patterns of Procephalothrix simulus when the oligochaete Tubifex sp. was used as prey Patterns of feeding Number of attacks Feeding duration (seconds) behavior (% of total attacks) mean [+ or -] SD (range) Feeding without use 25 (25) 70.9 [+ or -] 46.0 of proboscis (31.8-239.8) Proboscis everted 8 (8) 86.0 [+ or -] 39.7 during ingestion (45.6-170.6) Proboscis everted 14 (14) 88.1 [+ or -] 38.8 but inverted (35.4-183.3) before ingestion Prey entwined by 46 (46) 95.3 [+ or -] 36.9 proboscis and (33.9-196.5) ingested Proboscis everted 5 (5) 111.9 [+ or -] 55.7 for two times (74.6-209.2) Proboscis everted 2 (2) 136.4 [+ or -] 78.3 for three times (81.1-191.8) Total 100(100) -- Table 2 Statistics of the feeding behavior patterns of Procephalothrix simulus when the polychaete Saccocirrus gabrillae was used as prey Patterns Duration (seconds) of feeding Number of attacks mean [+ or -]SD (range) behavior* (% of total attacks) (feeding events)[dagger] AI 63 (31.5) 250.1 [+ or -] 228.6 (29.6-1071.4) (Entire feeding time) API 14 (7.0) 284.8 [+ or -] 189.2 (46.5-633.5) (Ingesting time) ANPI 15 (7.5) 350.1 [+ or -] 307.3 (60.3-971.8) (Ingesting time) AP 28 (14.0) 321.9 [+ or -] 172.5 (55.6-864.6) (Paralyzing time) ANP 80 (40.0) -- Total 200 (100) -- * AI = Ingesting while attacking with proboscis; ANP = Attacking with proboscis-disengagement-prey not paralyzed and not ingested; ANPI = Attacking with proboscis-disengagement-ingestion (prey not paralyzed); AP = Attacking with proboscis-disengagement-prey being paralyzed but not ingested; API = Attacking with proboscis-disengagement-prey being paralyzed-ingestion. See text for further explanations of the abbreviations. [dagger] Entire feeding time: the time from the start of an attack to the termination of ingestion. Ingesting time: the time from the start to the termination of ingestion. Paralyzing time: the time from the start of an attack to the prey being paralyzed. Table 3 Statistics of the feeding behavior patterns of Procephalothrix simulus with different sizes (body mass 8-12 mg, 18-22 mg, and 28-32 mg), and the duration of the events arising after a prey (Saccocirrus gabrillae) was attacked by the nemertean Patterns of feeding Body mass (mg) of P. simulus[dagger] behavior* 8-12 18-22 AI Number of attacks (% of 9 (10.0) 19 (21.1) total attacks) Entire feeding time 272.6 [+ or -] 181.3 158.2 [+ or -] 114.3 (seconds) (120.5-697.0) (38.4-446.3) API Number of attacks (% of 0 (0.0) 4 (4.4) total attacks) Entwining time (seconds) -- 33.4 [+ or -] 15.8 (16.0-46.7) Paralyzing time (seconds) -- 234.6 [+ or -] 52.4 (186.4-297.7) Ingesting time (seconds) -- 40.5 [+ or -] 22.0 (28.1-73.4) ANPI Number of attacks (% of 31 (34.4) 45 (50.0) total attacks) Entwining time (seconds) 37.0 [+ or -] 67.5 27.4 [+ or -] 17.1 (3.6-362.0) (9.6-71.6) Ingesting time (seconds) 398.2 [+ or -] 286.1 270.0 [+ or -] 168.5 (75.0-1249.6) (26.8-721.5) AP Number of attacks (% of 3 (3.3) 7 (7.8) total attacks) Paralyzing time (seconds) 636.9 [+ or -] 468.2 27.2 [+ or -] 16.5 (238.6-1152.6) (5.3-65.1) ANP Number of attacks (% of 47 (52.2) 15 (16.7) total attacks) Entwining time (seconds) 30.9 [+ or -] 41.0 43.1 [+ or -] 45.0 (2.7-210.5) (10.0-181.0) Successful attacks in total 40 (44.4) 68 (75.6) (% of total attacks) Number of attacks in total 90 90 Patterns of feeding Body mass (mg) of P. simulus[dagger] behavior* 28-32 AI Number of attacks (% of 27 (30.0) total attacks) Entire feeding time 120.7 [+ or -] 90.4 (35.3-364.0) (seconds) API Number of attacks (% of 7 (7.8) total attacks) Entwining time (seconds) 29.0 [+ or -] 16.1 (15.5-52.7) Paralyzing time (seconds) 234.3 [+ or -] 103.2 (88.7-384.8) Ingesting time (seconds) 43.2 [+ or -] 18.2 (19.0-62.0) ANPI Number of attacks (% of 43 (47.8) total attacks) Entwining time (seconds) 23.9 [+ or -] 13.0 (6.7-77.8) Ingesting time (seconds) 178.6 [+ or -] 146.1 (25.4-537.6) AP Number of attacks (% of 0 (0.0) total attacks) Paralyzing time (seconds) -- ANP Number of attacks (% of 13 (14.4) total attacks) Entwining time (seconds) 27.2 [+ or -] 16.5 (5.3-65.1) Successful attacks in total 77 (85.6) (% of total attacks) Number of attacks in total 90 * AI = Ingesting while attacking with proboscis; ANP = Attacking with proboscis-disengagement-prey not paralyzed and not ingested; ANPI = Attacking with proboscis-disengagement-ingestion (prey not paralyzed); AP = Attacking with proboscis-disengagement-prey being paralyzed but not ingested; API = Attacking with proboscis-disengagement-prey being paralyzed-ingestion. Entire feeding time: the time from the start of an attack to the termination of ingestion. Entwining time: the duration that a nemertean entwined a prey using its proboscis (the time from the start of an attack to the disengagement of nemertean and prey). Ingesting time: the time from the start to the termination of ingestion. Paralyzing time: the time from the start of an attack to the prey being paralyzed. [dagger] Time data are expressed as mean [+ or -] SD (range).
|Printer friendly Cite/link Email Feedback|
|Author:||Wang, Haiyan; Sun, Shichun; Li, Qinglong|
|Publication:||The Biological Bulletin|
|Date:||Apr 1, 2008|
|Previous Article:||Origin and expansion of foliated microstructure in pteriomorph bivalves.|
|Next Article:||Phylogenetic analysis of caprellid and corophioid amphipods (Crustacea) based on the 18S rRNA gene, with special emphasis on the phylogenetic...|