Biological characteristics of a newly established green crab (Carcinus maenas) population in the southern Gulf of St. Lawrence, Canada.

ABSTRACT Various biological characteristics of the exotic green crab (Carcinus maenas) were studied in 2000 and 2001 to describe its population structure and reproductive strategy off the coast of Prince Edward Island (PEI), Canada. This particular population is found at the northern limit of the green crab distribution off the east American coast. Results showed that the number of female crabs was higher during springtime compared with the rest of the year. Male crabs were found in higher numbers from August to December. Ovigerous females were observed from early July to mid-September, their numbers peaking in early July. This was confirmed by a gonadosomatic analysis, which showed that the relative gonadal wet weight dropped from late July to mid-September 2001. Fecundity was estimated at 140,000-200,000 embryos per mass. Mating occurred after larval release, which occurred from late August to December with a peak in September. Mature males were molting from June to December, but mainly in July. This was observed a few weeks before mating. Size at maturity was estimated at a carapace width of 21.30-34.49 mm for males. Size at maturity for females was estimated at 28.66-41.43 mm. Overall, our results showed that most reproductive characteristics of the green crab's life cycle tend to occur about a month later in PEI in contrast to what is usually observed in Maine (USA).

KEY WORDS: green crab, Carcinus maenas, northern distribution, population, reproduction, ovigerous females, gonadosomatic analysis, fecundity, mating, size at maturity

INTRODUCTION

The green crab (Carcinus maenas) was originally distributed along the eastern Atlantic coast, from Norway to Mauritania including southern Iceland (Broekhuysen 1936, Crothers 1968, Grosholz & Ruiz 1996). Since the early 1800s, the green crab was observed off the coasts of Australia (Zeidler 1978), Tasmania (Gardner et al. 1994), and South Africa (LeRoux et al. 1990) as well as off the western coasts of the USA (Cohen et al. 1995) and Canada (Jamieson et al. 1998). The green crab's worldwide distribution demonstrates the ability of this species to exploit a wide range of habitats, being tolerant to a large spectrum of environmental conditions (Wheatly 1981, Abello et al. 1997).

One of the earliest green crab introductions occurred off the eastern American coast. Individuals were first documented in New York and New Jersey in 1817. Their expansion then progressed northward reaching the state of Maine in the early 1900s (Audet et al. 2003). The species colonized, the following 50 y, estuarine habitats along the Maine coastline up to the Bay of Fundy in Canada (Scattergood 1952, Glude 1955, MacPhail et al. 1955). The northern limit of the distribution of the green crab has expanded to Prince Edward Island (PEI) in 1997 within the southern Gulf of St. Lawrence (SGSL) (Audet et al. 2003).

In northwestern Europe, the green crab's entire reproductive cycle usually lasts about a year. Gonadogenesis may, however, occur twice a year (spring and winter) in the case of large females (Broekhuysen 1936, Demeusy 1963). Though a high proportion of females extrude eggs during winter, ovigerous females are found all year long between the Scandinavian Peninsula and the Gibraltor Strait (Wheatly 1981, Van der Meeren 1992). Individuals generally mate once a year during the midsummer to early-fall period (d'Udekem d'Acoz 1993). Differences in the life cycle may, however, be observed between North Sea (d'Udekem d'Acoz 1993) and Portuguese populations (Almaca 1982).

Variations in reproductive strategies may also be observed among newly invaded coastlines. The reproductive strategy of green crabs observed off the Pacific coast of the USA, for instance, appears similar to the European pattern, females producing eggs in winter and spring (Cart & Dumbauld 2000, Hunt & Behrens Yamada 2001). In Maine, green crabs seem to reproduce once a year, bearing eggs only in spring (Berrill 1982).

The PEI coastline presents similar surface water temperature than the Norwegian and Iceland coasts during the winter period. In the summer, however, the temperature pattern of the surface waters varies from one side of the Atlantic to the other (Thurman 1997). The general effects of the Gulf Stream on the northeastern American coast and the topographic profile of the SGSL sea floor create suitable habitats where the green crab may exploit spatial and food resources.

The present study outlines the population structure of a newly established green crab population and describes various aspects of its reproductive biology. Biological events such as mating, molting, egg bearing periods and ovarian development were studied over time from a PEI lagoon. Size at sexual maturity and fecundity were also determined. This study provides important basic information to better understand the effect of latitude and associated climate on the reproductive strategy of the green crab in the northern limit of its North American distribution. The environmental characteristics of most PEI's estuaries and lagoons may be ideal for rapid proliferation of the green crab.

MATERIAL AND METHODS

Study Site

Green crabs were collected from Basin Head on the northeastern coast of PEI, Canada, in 2000 and 2001. Basin Head is a marine lagoon of 0.59 [km.sup.2] dominated by the eelgrass Zostera marina (Fig. l). The southern section of the basin is characterized by a 500 m long sandy bottom channel. This channel represents the unique opening toward the open waters (Northumberland Strait). The lagoon's mean depth is less than 1.0 m; deeper troughs (up to 3.0 m) are observed in the main basin and in the channel. The south shore is surrounded by salt marshes, whereas agriculture fields are found on the northern shore. The northern section bares mud and sandy-mud substrata rich in benthic life. A detailed description of the water system is available in Sharp et al. (2003).

Sampling Protocols

The first sampling season was conducted from July to November 2000. Three sampling sites were used: sites 1-3 (Fig. 1). The second sampling season was carried out from May to December 2001, with the addition of sites 4-8. Sites 2, 3, and 5 were situated in the main basin where the greatest depths occurred (2-3 m). Muddy sediments and extensive beds of Ulva lactuca characterized these sites. Sites 1, 4, and 6 were placed within the first km of the northeastern arm of the basin. Sites 7 and 8 were placed in the southern sandy channel.

Modified eel traps were used as the main fishing gear in 2000 and 2001. These traps were 1 m long, cylindrical (diameter: 30 cm), and made with a plastic coated wire mesh (1 [cm.sup.2]). Only one end of the trap had a funnel opening (10 cm). Four traps per site were deployed on the same line biweekly during a period of 24 h. The eel traps were 5.5 m apart and baited with frozen mackerel or herring. Traps situated in very shallow areas were exposed to air at low tide. The catch per unit effort was adjusted according to the period of time traps were emerged. Fyke nets were used during the 2001 season to capture a broader range of green crabs. These nets (mesh size: 15 x 15 mm) were equipped with a 4.57 m leader from trap to shore. This fishing gear allowed the capture of ovigerous females, because these individuals are less attracted by baited traps. Two fyke nets were deployed on the north shore of the main basin and left in the water for a 24-h period (Fig. 1). A juvenile collector was used to capture small crabs. These collectors were similar to the one described by Serfling & Ford (1975). They were made with 0.38 [m.sup.2] plastic coated wire mesh (25 x 25 mm), filled with oyster shells (Crassostrea virginica). A second mesh net (3 x 3 mm) covered the bottom and the sides of the box to prevent juveniles from escaping at low tide. Two juvenile collectors were placed at every sampling site and left in the water for a two-week period. Fry nets were finally used in the intertidal area during the 2001 season to capture juveniles under beds of sea lettuce. Fry nets were used at low tide during a two-hour period on the north shore of the basin. All individuals were transported live to the laboratory where they were kept at 4[degrees]C. Measurements were taken within a week to prevent desiccation.

[FIGURE 1 OMITTED]

Water temperature and salinity were measured with a portable temperature-salinity meter (YSI model 30) at each sampling station. A submersible data logger (VEMCO) was used to monitor water temperature and salinity at each sampling site biweekly from July 2001 to May 2002.

Laboratory Procedures

Carapace width (CW) and the crusher's cheliped height (CH) were measured on live individuals with a modified vernier caliper (Watson & Wells 1970). Specimens were grouped in 1 mm size classes for length-frequency distributions. The abdominal width (AW) was also measured in females for maturity analysis. Wet body mass (BM) was measured and specimens with missing appendices were recorded as incomplete. Incomplete crabs were not used to describe the length/weight relationships. Allometric growth of the CH (males) and the AW (females) were used to estimate the morphometric size at 50% maturity in relation to CW (Tessier 1960, Somerton 1980). A significant change in the size of the male crusher cheliped is expected after its puberty molt (Lee & Seed 1992, Mariappan et al. 2000). Males with a regenerating crushing cheliped were not used for this analysis. Variations in the allometric growth of the female could be detected by a significant shift from the AW growth curve.

Molting periods were evaluated according to the crab's abdominal coloration. Both male and female green crabs bear an orange-red coloration during a prolonged intermolt period. A pale green to yellow coloration indicates a very recent molt (McGaw et al. 1992, Wolf 1998). A gonadosomatic index [GSI = (wet gonad mass/CW) * 100] was determined for females captured in baited traps and fyke nets. The ovary condition was also established from coloration (Bondgaard et al. 2000). Six developmental groups from immature to fully mature embryo were used: (1) white; (2) beige; (3) yellow, (4) pale-orange; (5) orange; (6) dark-orange.

A fecundity analysis was carried out with 33 ovigerous females caught in fyke nets. The developmental stage of the eggs was noted to estimate the length of the development period: orange egg mass with eggs (embryos) mostly composed of vitellogen to dark brown egg mass composed of eggs with embryonic eye pigmentation. The egg masses, still attached to pleopods, were placed in a 10% formaldehyde and seawater solution for short-term preservation. The eggs were separated from the pleopods and then dried at 60[degrees]C. The eggs were freed from pleopods and other debris and weighted. A subsample of 100 eggs was isolated from each egg mass and weighted. After measuring the weight of subsamples, we could extrapolate this information to their entire egg mass and estimate the fecundity of females bearing uneyed and eyed eggs.

[FIGURE 2 OMITTED]

Data Analysis

Monthly size variations were analyzed using ANOVAs. Nonparametric Kruskal-Wallis tests were used when normality and homoscedasticity requirements were not met. SNK tests were carried out for all post hoc comparisons. Regression coefficients in size-weight relationships were compared by testing slopes for equality of variance (ANOVA). Line elevation equalities were compared with an analysis of covariance (ANCOVA). Simple linear regressions were used to estimate the fecundity of females carrying uneyed and eyed embryos in relation to their CW. ANCOVAs were used to test for differences between uneyed and eyed embryos' regression lines. The type of allometric growth for various relationships was established by comparing regression slopes with the isometric value (1 or 3) using Student t-test.

The size at maturity ([t.sub.i]) was described using a continuous piecewise linear model [g(x|[theta])] to identify the size at the intercept point(s) (Roa et al. 1999). Parameters ([[alpha].sub.0], [[alpha].sub.1], [[beta].sub.1], [[delta].sub.1], [t.sub.1], and [t.sub.2]) were determined with a maximum log-likelihood fitting method using the following algorithm:

g(x|[theta]) = [[alpha].sub.1]x + [[alpha].sub.0] + [epsilon],(x [less than or equal to] [t.sub.1]); [[beta].sub.1]x + [b.sub.0] + [epsilon], ([t.sub.1] < x [less than or equal to] [t.sub.2]); [[delta].sub.1]x + [d.sub.0] + [epsilon], ([t.sub.2] < x).

where [[alpha].sub.1], [[beta].sub.1], [[delta].sub.1] = growth rates; [t.sub.i] = CW at which a significant inflection of the growth rate occurs; [[alpha].sub.0] = initial intercept; [b.sub.0] = ([[alpha].sub.1] - [[beta].sub.1] [t.sub.1] + [[alpha].sub.0] and [d.sub.0] = ([[beta].sub.1] - [[delta].sub.1]) [t.sub.2] + [b.sub.0] = ([[alpha].sub.0], [[alpha].sub.1] and [t.sub.2]); and [epsilon] ~ N (0, [[sigma].sup.2]). The confidence interval was established around the size (t) at which 50% of the population is expected to be mature by nonparametric bootstrap (n = 1,000) (Rice 1995). A likelihood ratio test was used to justify the piecewise linear model--in comparison with a linear model-with a 95% level of significance.

RESULTS

Environmental Factors

The maximum water temperature was observed in August 2001 (Fig. 2). It then slowly decreased and stabilized around -2[degrees]C during the winter months. Water temperature started to increase by the end of March 2002. Salinities varied between 22.2[per thousand] and 28.9[per thousand] from May to December 2001. Traps were less efficient in mid-December when water temperature was approximately 4[degrees]C.

[FIGURE 3 OMITTED]

Population Structure

The number of green crabs collected in modified eel traps was 610 in 2000 (429 males and 181 females). The CW ranged from 32.60-76.44 mm for females and 37.90-89.76 mm for males. A total of 2,216 crabs were collected in 2001 (1,374 males and 842 females): 1,385 with modified eel traps (906 males and 479 females), 606 with fyke nets (310 males and 296 females) and 225 juvenile crabs with megalopae collectors and fry nets (158 males and 67 females). The CW ranged from 16.12-73.74 mm for females and 9.96-82.88 mm for males (modified eel traps and fyke nets).

The mean CW for females captured in modified eel traps (Fig. 3A) failed to show any significant variation in time in 2000 (F = 2.44, P = 0.10). The mean CW for males, however, was significantly higher in early August ([chi square] = 32.55; P < 0.001). In 2001, the mean CW of females significantly increased from May to December (F = 26.88, P < 0.001). The SNK test confirmed that the mean CW of females was smaller from May to June and larger from November to December. The mean CW of males gradually increased from May to late July and then remained stable from August to December. Overall, males were significantly larger late in the season compared with those captured early in the season ([chi square] = 9.56, nonparametric posthoc, P = 0.048). A similar trend was observed in 2001 for females and males captured with fyke nets (Fig. 3B). Ovigerous females observed in July 2001 had a mean CW significantly larger than nonovigerous females captured at the same time (P = 0.004). Juveniles captured in fry nets were abundant under sea lettuce beds from mid-July to September.

[FIGURE 4 OMITTED]

The highest number of females captured in modified eel traps was observed in July 2001 (Fig. 4A). Other peaks were observed in early June, early September, and early November. The highest number of males was also observed in July with other peaks in mid-August and mid-November. The number of females was significantly higher than males during springtime ([chi square] = 20.33, P < 0.001). Males outnumbered females from August to December. Numbers collected from fyke nets showed similar results, females outnumbering males early in the season (Fig. 4B). No abundance peak was, however, observed in November in contrast to results from modified eel traps. Ovigerous females were only captured in June and July. The number of males and nonovigerous females was low during that period.

Small-scale movements were observed in Basin Head in 2001 (Fig. 5). Catches from the central basin (sites 2, 3, and 5) were dominated by males from July to December. A significant gender shift was observed in the Northern water arm (sites 4 and 6). This was particularly true for site 4 where males dominated the catches from July to August, whereas females were more abundant from September to November. The number of both males and females increased in the channel (sites 7 and 8) in November.

Size-Weight Relationship

The BM to CW relationship was described for female and male crabs using the power regression BM = [aCW.sup.b]. The model explained a high percentage of the observed variability (P < 0.001) for both regressions (Fig. 6). The regression coefficient (b) for males was higher than the one observed for females (ANCOVA: F = 106.37; n = 2906; P < 0.0005). The growth curve observed for males (Fig. 6A) showed a larger increase in BM compared with the one observed for females (Fig. 6B). This was true for CW [greater than or equal to] 39 mm. Growth was more important for females <39 mm. The BM to CW relationships was negative (allometric versus isometric growth comparison) for both males (t = 15.4, df = 1,836, P < 0.001) and females (t = 15.7, df = 1,066, P < 0.001), as well as for the entire population ([BM.sub.t] = 0.0003 [CW.sub.t.sup.2.9200]; t = 20.0; df = 2904; P < 0.001).

[FIGURE 5 OMITTED]

Molting Periods

The relative abundance of green- and red-colored crabs over time is presented in Figure 7. Overall, a greater proportion of males had a green coloration from May to August ([chi square] = 134.10; P < 0.001). The red coloration was dominant from September to November ([chi square] = 31.74; P < 0.001). The proportion of green- and red-colored males was similar in December ([chi square] = 2.16; P > 0.05). The green coloration was observed among all sizes from May to December. Small males (<60 mm CW)) remained in a green form after the molting period. Green- and red-colored females were found in similar proportions in May ([chi square] = 1.25; P > 0.05). Orange-red-colored females then dominated from June to August, regardless of the size ([chi square] = 67.45; P < 0.001). The number of green-colored females was more important from September to December ([chi square] = 30.04; P < 0.001). Recently inseminated females (green-colored individuals bearing a swollen spermatheca) peaked in September (31.7%). Only a few inseminated females were observed in November (2.2%) and December (3%).

[FIGURE 6 OMITTED]

Reproductive Cycle and Fecundity

The ovarian development was strongly related to oocyte coloration. The ovary showed a deep-orange coloration prior to the spawning period when oocytes were fully developed. The ovary was transparent-white after the embryo production and extrusion, before the development of new oocytes. Females in pre and early-vitellogenesis occurred from mid July to late September but peaked from mid to late August in 2000 (Fig. 8). Highly developed ovaries were then observed from late October to late November. Highly developed ovaries were still observed in females sampled in May 2001. Nearly all females sampled between early June and early July were bearing highly developed ovaries. Females in pre and early vitellogenesis were then observed from mid-July to early November.

The GSI was low in mid-August 2000 (Fig. 9A). It then increased gradually until the end of the season. The number of crabs captured during the period of egg extrusion was, however, low and might not be representative (n [less than or equal to] 14). Ovigerous females were not captured because fyke nets were not used in 2000. In 2001, the GSI increased from May to mid-June and dropped in early July (Fig. 9A). It then increased again until August when it dropped for a second time. The number of large females was important during that period (Fig. 9B). A gradual increase was finally observed the rest of the year.

The percentage of ovigerous females captured with fyke nets in 2001 dropped from 100% on July 4 to 20% on July l7 (Fig. 10). Ripped ovaries were observed in all nonovigerous females on July 12 and 19% of females on July 17 had empty egg capsules attached to their pleopods. The latter indicates that embryos were probably hatched around that period. The other females were characterized by white- to pale-orange ovaries. Over 21% of females had empty egg capsules on July 25. The presence of egg capsules was observed until mid-September. Females with egg capsules attached to their pleopods had ovaries with an advanced stage (stage 5-6: orange). These results suggest that spawning probably occurred in the first half of July and that it was short lasting.

[FIGURE 7 OMITTED]

The color variation of the egg mass tends to show that embryogenesis lasted 8-13 days for most females (Fig. 11). The number of embryos reached 195,833 [+ or -] 83,673 (mean [+ or -] SD) in females carrying eggs at an early development stage. The number of eggs with developed eye pigments was 140,374 [+ or -] 60,717. The CW ranged from 41.54-68.30 mm for females with a complete brood. The number of eggs to AW relationship is presented for both early ([r.sup.2] = 0.916, P < 0.0001) and advanced ([r.sup.2] = 0.751, P < 0.0001) embryo masses in Figure 12. Both relationships showed that the number of embryos increased with increasing AW. The regression coefficients differed significantly (ANCOVA: F = 7.95, n = 33; P < 0.0005) suggesting that large females may lose more embryos during the incubation period than smaller ones. The average loss of embryos from an early to advanced developmental stages was about 16.5%.

[FIGURE 8 OMITTED]

Size at Maturity

A three components continuous piecewise linear model was used to study the log-log relationship between AW and CW for females (Fig. 13A). Two intersection points were observed on the curve, each representing a size at which growth was modified:

g(CW|[theta]) = 1.09CW - 1.36 + [epsilon], (CW [less than or equal to] 3.36), n = 146; 1.42CW - 2.46 + [epsilon], (3.36 < CW [less than or equal to] 3.72), n = 59; 1.12CW - 1.28 + [epsilon], (CW > 3.72), n = 858.

[FIGURE 9 OMITTED]

Three growth periods were described by the model for the females. Juvenile females were characterized by a slight positive allometric growth. The first inflection point was observed at 28.66 mm CW (CI 95%: 24.87-30.37 mm). At that size, the growth rate of the abdomen abruptly increased in relation to the body size. A second inflection was observed at 41.43 mm CW (CI 95%: 36.90-43.79 mm). This inflection may define a size at maturity for females. This phase was only seen in 5.9% of females and occurred within a narrow size range (28.66-41.43 mm). The growth coefficient during this size interval was very strong (1.422 [+ or -] 0.065 mm) and clearly corresponded to a period of intense energy input in the enlargement of AW.

[FIGURE 10 OMITTED]

[FIGURE 11 OMITTED]

A similar log-log model described the maturity function in males (Fig. 13B):

g(CW|[theta]) = 1.27CW - 2.40 + [epsilon], (CW [less than or equal to] 3.06), n = 172; 0.99CW - 1.52 + [epsilon], (3.06 < CW [less than or equal to] 3.54), n = 69; 1.40CW - 3.01 + [epsilon], (CW > 3.54), n = 1387.

The male allometric growth curve revealed a significant inflection at of 21.30 mm CW (CI 95%: 18.15-23.06 mm). A second inflection was observed at 34.49 mm CW (CI 95%: 31.58-47.08 mm). This "transitional" phase is described by an isometric growth rate (0.99) where the CH growth rate, in relation to CW, becomes allometric (1.40).

DISCUSSION

Crab Activity, Population Structure, and Recruitment Period

Results from the modified eel traps showed that female green crabs were active and mobile during springtime and in the fall compared with the summer months. Females need to gain weight and regenerate their ovaries early in the season prior to the spawning period. The increase in the mean CW observed from May to late June tends to confirm this. The postovigerous period observed late in the season is also a critical time for the females. This period allow them to store energy reserves before the winter period. Again, the increase in the mean CW from July to early November tends to confirm that females were active and foraging during that period. Females were less active and less mobile during summertime when they were bearing eggs and freshly molted, a period during which they do not eat and limit their movements to avoid predators. Females were then less likely to be captured in baited traps. Our results showed that the number of males remained high all season and that they actively forage from May to early August. These seasonal observations for females and males tend to be confirmed by various green crab studies in northern Europe (e.g., Naylor 1962, Rasmussen 1973, Munch-Petersen et al. 1982). The numbers gathered from fyke nets gave similar results than the ones from the modified eel traps, but the variability was more important. Fyke nets have the tendency to capture what may pass near the nets regardless of the size compare with traps that are more selective.

[FIGURE 12 OMITTED]

Our results tend to show that female and male green crabs moved toward the sandy channel late in the season. Many authors proposed that green crabs migrate to deeper and more temperature stable waters in the fall (Broekhuysen 1936, Crothers 1968, Styrishave & Anderson 2000). According to Crothers (1968), females migrate earlier than males and ensure an optimal embryogenesis. There is, however, no evidence that green crabs migrated outside from Basin Head for the winter. Considering the cold water temperature and the presence of an ice cover during the winter, it is possible that individuals reduce their metabolism and hibernate in Basin Head. The reproduction cycle of the green crab in Maine (Berrill 1982) and PEI (this study) is restricted to a few months in the summer and fall. Consequently, an offshore migration of females might not be observed in the northeastern American region because there is apparently no winter brooding. Our results showed that crabs tend to end their wintering period when the water temperature reaches at least 10[degrees]C. Cohen et al. (1995) suggested that green crabs are not feeding at temperatures lower than 7[degrees]C. Crab movement may then be minimal. Our observations showed that the green crab may feed at temperature less than 6[degrees]C. This was, however, only observed during late fall and not during springtime.

Results obtained from the megalopae collectors and fry nets tend to suggest that the green crab recruitment period in Basin Head might occur during November. Our observations also tend to suggest that recruitment may be occurring in sea lettuce beds.

Molting Period and Reproduction

The transition from red- to green-colored green crabs, corresponding to the female's molting and mating periods, occurred suddenly and lasted a short period of time. Our results revealed that mating and molting occurred at the end of July for most females but can last until late September for a few of them. Female molting occurred just after larval release and male molting, leaving a small period of time (4-8 wk) for the males to harden their carapace before mating.

Green crab populations have been shown to synchronize their life cycle to various seasonal patterns. From the North Sea to Portuguese waters, female green crabs are able to produce eggs year round and females may undergo one or two ovarian cycles per year (Broekhuysen 1936, Demeusy 1958, Demeusy 1963, Crothers 1967, Almaca 1982, Queiroga 1996). The present study confirms the ability of the green crab to adapt to a wide range of environmental conditions. The optimal temperature window for successful reproductive activities for crustaceans in the SGSL is, however, limited. Our results showed that the green crab exhibits a one-year reproductive cycle as observed in many regions (Table 1). Certain reproductive events are, however, restricted to a few months. This is in contrast to similar European regions under the same latitude. In Maine (USA), 82% of ovigerous females are found from May to June (Berrill 1982). The same life cycle events in the SGSL occurred primarily in July. Our results suggest that spawning probably occurred in the first half of July and that it was short lasting. The high proportion of ovigerous females found in the SGSL suggests that there is only one brooding period.

Except for mating and female molting, which occur at the same time (from mid-July to mid-October), a two- to eight-week time gap separates most of the green crab's reproductive events between populations of Basin Head and Maine (Table 1). This suggests that SGSL temperatures in spring and early summer delays the reproductive activities of green crabs until sufficient food supply can support growth in males and ovary development in females. High water temperatures in Basin Head observed during the summer (20[degrees]C to 26[degrees]C) may accelerate the egg incubation time (Veillet 1945) and the crab's metabolism allowing them to produce larvae and to mate at the same period of the season in both regions. The apparent shorter reproduction period observed in green crabs of Basin Head reflects a great capacity to adapt to environmental conditions.

[FIGURE 13 OMITTED]

Our results also showed that females started to develop ovaries in late May. The ovaries rapidly matured in mid-June. Females were then involved in further reproduction processes from early July to late September. Ovigerous females were first observed in early July. Maturation of embryos occurred rapidly, which allowed the larvae to appear early in the season. Postlarval stages could have undergone multiple molts before the arrival of the cold season. Our results showed that the occurrence of ovigerous females is peaking in July for about two weeks and that maturation of embryos also took place over a two-week period.

Size at Maturity

It is at the puberty molt that female green crabs will mate for the first time having reached morphometric and gonadal maturity. However, as observed for snow crabs (Chionoecetes opilio), green crabs may reach gonadal maturity before being physically able to mate (Comeau & Conan 1992). According to our results, females have two different sizes at which a critical molt occurs (28.66 and 41.44 mm). The interval between both critical sizes is short and includes only 5.9% of the population. During this interval, the AW is clearly growing faster than the CW, thus representing sizes at which puberty could be reached for some females. The female maturity occurred at a maximum mean size of 41.44 mm. The AW growth rate then decreased at the adult stage because of energy requirements for reproductive activities. The smallest females with eggs or with egg capsule residues, measured 38.74 mm and 38.18 mm (CW), respectively. A preliminary histological study of green crab ovaries confirmed that females from Basin Head were mature at a CW > 40 mm (Khadra Benhalima, pers. comm.). The observation of two different inflections of the female growth curve, as observed for other brachyuran species (Negreiros-Fransozo et al. 2003), could be considered as a transition period before the complete maturation at 41.44 mm. In this case, the morphometric maturity seems to correspond to the gonadal maturity. The skewed size-distributions of green crabs observed in Basin Head may, however, biased to a certain point this interpretation.

The abrupt growth acceleration in the male CH is directly associated with sexual maturation, because the cheliped size is considered as a secondary sexual character (Mariappan et al. 2000). Thus, at a size of 34.49 mm, 50% of the male population from Basin Head would have undergone their sexual maturation. However, a preliminary histological analysis of the vasa deferentia demonstrated that males from Basin Head were starting to develop spermatophores at sizes >50 mm (Khadra Benhalima, pers. comm.). The large variability observed around the second inflection point may show that males are progressively maturing sexually until reaching 50 mm. The growth rate of the CW was proportional to the CH growth during the "transitional" phase, suggesting that the overall size of males is increasing. Females tend to choose larger males to mate, which might explain the tendency of males to attain a large CW as they increase their CH (Berrill & Arsenault 1982, Lee & Seed 1992, Abello et al. 1993).

Green crabs observed off the coast of Belgium are mature at a CW ranging from 23-45 mm (d'Udekem d'Acoz 1991). Berrill (1982) suggested that female maturity occurred between 34 and 45 mm of CW in green crabs from Maine. Size at maturity for females could be influenced by a strategically postponed puberty molt (d'Udekem d'Acoz 1993). In an area such as the North Sea, where the reproductive activities are prolonged, green crabs may mate from May to November. Young females tend to mate for the first time later in the season compared with older females. Old females have an increased reproductive success by attaining a larger size. Environmental and climatic factors particular to the SGSL may affect size at maturity, as well as food quality and availability (Aldrich 1983, Berkenbusch & Rowden 2000), which in turn may also have an effect on the sexual behavior of females.

Conclusion

The green crab has been present off the coast of PEI since 1997 (Gillis et al. 2000, Audet et al. 2003). Considering that the life span of green crabs from the coast of Maine is five to six years, this suggests that the green crab rapidly established itself in PE! within one or two generations. The rapid establishment of the green crab in a habitat such as Basin Head is typical of an opportunistic invasive species. The reproductive strategy of the green crab is subject to latitudinal and regional temperature variations. The reproductive pattern in PEI is similar to what is observed in New England, USA, (Berrill 1982) and in Denmark (Rasmussen 1973). Green crabs from Basin Head exhibit a relatively similar life cycle than the one observed from populations in Maine. Water temperatures on PEI, however, remain <10[degrees]C for at least eight months of the year. This suggests that the species may have adapted to a narrow breeding time window occurring during the warmer months. Egg bearing, for instance, which takes place strictly during the warmer months of the year, was occurring a few weeks later. Overall, our results show a delay of a few weeks compared with what is observed in Maine.

ACKNOWLEDGMENTS

The authors thank Nell MacNair, Mark McLeod, and Jeff MacEwen (Department of Fisheries, Aquaculture and Environment of PEI) for their field work contribution. The authors also thank Dr Khadra Benhalima and Dr Glen S. Jamieson for the provision of unpublished information. Additional thanks are addressed to Dr Jim Williams and two anonymous referees for reviewing this manuscript. This research was funded by the Department of Fisheries and Oceans and the Faculte des etudes superieures et de la recherche de l'Universite de Moncton.

LITERATURE CITED

Abello, P. A., C. G. Warman, D. G. Reid & E. Naylor. 1993. Differential mating success of right- and left-handed forms in the shore crab Carcinus maenas (L.) (Crustacca: Brachyura). In: J. C. Aldrich, editor. Quantified phenotypic responses in morphology and physiology. Proc. 27th Eur. Mar. Biol. Syrup. pp. 137-142.

Abello, P., A. Aagaard, C. G. Warman & M. H. Depledge. 1997. Spatial variability in the population structure of the shore crab Carcinus maenas (Crustacea: Brachyura) in a shallow-water tidal fjord. Mar. Ecol. Prog. Ser. 147:97-103.

Aldrich, J. C. 1983. Seasonal, geographical and size differences in oxygen consumption, digestive gland and gills in Carcinus maenas (L.), a study of condition for ecologists. P. Z. N. I.. Mar. Ecol 4: 147-162.

Almaca, C. 1982. Notes on the biology of Carcinus maenas (L.) from intertidal beaches of western Portugal. Quad. Lab. Tecnol. Pesca. Ancona 3:179-185.

Audet, D., D. Davis, G. Miron, M. Moriyasu, K. Benhalima & R. Campbell. 2003. Geographical expansion of a nonindigenous crab, Carcinus maenas (L.), along the Nova Scotian shore into the southeastern Gulf of St. Lawrence, Canada. J. Shellfish Res. 22:255-262.

Berkenbusch, K. & A. A. Rowden. 2000. Latitudinal variation in the reproductive biology of the ghost shrimp Callianassa filholi (Decapoda: Thalassinidea). Mar. Biol. 136:497-504.

Berrill, M. 1982. The life cycle of the green crab Carcinus maenas at the northern end of its range. J. Crust. Biol. 2:31-39.

Berrill, M. & M. Arsenault. 1982. Mating behavior of the green shore crab Carcinus maenas. Bull. Mar. Sci. 32:632-638.

Bondgaard, M., U. Narum & P. Bjerregaard. 2000. Cadmium accumulation in the female shore crab Carcinus maenas during the molt cycle and ovarian maturation. Mar. Biol. 137:995-1004.

Broekhuysen, G. L. 1936. On development, growth and distribution of Carcinides maenas (L.). Archs. Neer. Zool 2:257-399.

Carr, E. M. & B. R. Dumbauld. 2000. Status of the European green crab invasion in Washington coastal estuaries: Can expansion be prevented? J. Shellfish Res. 191:629-630.

Cohen, A. N., J. T. Carlton & M. C. Fountain. 1995. Introduction, dispersal and potential impacts of the green crab Carcinus maenas in San Francisco Bay, California. Mar. Biol. 122:225-237.

Comeau, M. & G. Y. Conan. 1992. Morphometry and gonad maturity of male snow crab, Chionoecetes opilio. Can. J. Fish. Aquat. Sci. 55:262-279.

Crothers, J. H. 1967. The biology of the shore crab Carcinus maenas (L.). 1. The background--anatomy, growth and life history. Field Stud. 2:407-614.

Crothers, J. H. 1968. The biology of the shore crab Carcinus maenas (L.). 2. The life of the adult crab. Field Stud. 2:579-614.

Demeusy, N. 1958. Recherches sur la mue de puberte du Decapode Brachyoure, Carcinus maenas Linne. Arch. Zool. Exp. G E N 95: 253-491.

Demeusy, N. 1963. Influence des facteurs saisonniers sur la realisation de la puberte au sein d'une population de Carcinus maenas L. des cotes de la Manche. C. R. Acad. Sci. Paris 256:4762-4764.

d'Udekem d'Acoz, C. 1991. Reproduction in a population of shore crabs Carcinus maenas (Linnaeus, 1758) in the Southern North Sea. Belg. J. Zool. 121:Suppl. 118.

d'Udekem d'Acoz, C. 1993. Activites reproductrices saisonnieres des differentes classes de tailles d'une population de crabes verts Carcinus maenas (Linnaeus, 1758) dans le sud de lamer du Nord. Cah. Biol. Mar. 35:1-13.

Gardner, N. C., S. Kwa & A. Paturusi. 1994. First recording of the European shore crab, Carcinus maenas in Tasmania. The Tasmanian Naturalist 116:26-28.

Gillis, D. J., J. N. MacPherson & T. T. Rattray. 2000. The status of green crab (Carcinus maenas) in Prince Edward Island in 1999. Prince Edward Island Department of Fisheries and Tourism., Tech. Rep. No. 225. 309 pp.

Glude, J. B. 1955. The effects of temperature and predators on abundance of the soft-shell clam, Mya arenaria, in New England. Trans. Am. Fish. Soc. 84:13-26.

Grosholz, D. & G. M. Ruiz. 1996. Predicting the impact of introduced marine species: Lessons from the multiple invasions of the European green crab Carcinus maenas. Biol. Conserv. 78:59-66.

Hunt, C. & S. Behrens Yamada. 2001. Predation by the Red Rock Crab, Cancer productus on the European Green Crab, Carcinus maenas, Oregon State University Proceedings of the Second International Conference on Marine Bioinvasions, New Orleans, April 9-11, 2001. pp. 75.

Jamieson, G. S., E. D. Grosholz, D. A. Armstrong & R. W. Elner. 1998. Potential ecological implications from the introduction of the European green crab, Carcinus maenas (Linneaus), to British Columbia, Canada, and Washington, USA. J. Nat. Hist. 32:1587-1598.

Lee, S. Y. & R. Seed. 1992. Ecological implications of cheliped size in crabs: Some data from Carcinus maenas and Liocarcinus holsatus. Mar. Ecol. Prog. Ser. 84:151-160.

LeRoux, P. J., G. M. Branch & M. A. P. Joska. 1990. On the distribution, diet and possible impact of the invasive European shore crab Carcinus maenas (L.) along the South African coast. S. Afr. J. Mar. Sci 9:85-93.

MacPhail, J. S., E. I. Lord & L. M. Dickie. 1955. The green crab--a new clam enemy. Fish. Res. Board Can. Prog. Rep. Atl. Coast Sta. 63:3-12.

Mariappan, P., C. Balasundaram & B. Schmitz. 2000. Decapod crustacean chelipeds: an overview. J. Biosci. 25:301-313.

McGaw, I. J., M. J. Kaiser, E. Naylor & R. N. Hughes. 1992. Intraspecific morphological variation related to the moult-cycle in colour forms of the shore crab Carcinus maenas. J. Zool. 228:351-359.

Munch-Petersen, S., P. Sparre & E. Hoffmann. 1982. Abundance of the shore crab, Carcinus maenas (L.), estimated from mark-recapture experiments. Dana 2:97-121.

Naylor, E. 1962. Seasonal changes in a population of Carcinus maenas (L.) in the littoral zone. J. Anim. Ecol. 31:601-609.

Negreiros-Fransozo, M. L., K. D. Colpo & T. M. Costa. 2003. Allometric growth in the fiddler crab Uca thayeri (Brachyura, Ocypodidae) from a subtropical mangrove. Crustaceana 23:273-279.

Queiroga, H., J. D. Costlow & M. H. Moreira. 1994. Larval abundance patterns of Carcinus maenas (Decapoda, Brachyura) in Canal de Mira (Ria de Aveiro, Portugal). Mar. Ecol. Prog. Ser. 111:63-72.

Queiroga, H. 1996. Distribution and drift of the crab Carcinus maenas (L.) (Decapoda, Portunidae) larvae over the continental shelf off northern Portugal in April 1991. J. Plankt. Res 18:1981-2000.

Rasmussen, E. 1973. Systematics and the ecology of the Isefjord marine fauna (Denmark). Ophelia 11:142-165.

Rice, J. A. 1995. Mathematical statistics and data analysis. 2nd ed. Belmont, California: Duxbury Press.

Roa, R., B. Ernst & F. Tapia. 1999. Estimation of size at sexual maturity: evaluation of analytical and resampling procedures. Fish. Bull. (Wash. DC.) 97:570-580.

Scattergood, L. W. 1952. The distribution of the green crab, Carcinides maenas (L.) in the Northwestern Atlantic. Maine Dept. Sea and Shore Fish. Circulation 8:2-10.

Serfling, S. A. & R. F. Ford. 1975. Ecological studies of the puerulus larval stage of the California spiny lobster, Panulirus interuptus. US National Marine Fisheries Services. Fish. Bull. (Wash. DC.) 73: 360-377.

Sharp, G., R. Semple, K. Connolly, R. Blok, D. Audet, D. Cairns & S. Courtenay. 2003. Ecological assessment of the Basin Head lagoon: a proposed marine protected area. Can. Manusc. Rep. Fish. Aquat. Sci. No. 2641.

Somerton, D. A. 1980. A computer technique for estimating the size of sexual maturity in crabs. Can. J. Fish. Aquat. Sci. 37:1488-1494.

Styrishave, B. & O. Anderson. 2000. Seasonal variations in hepatopancreas fatty acid profiles of two colour forms shore crabs, Carcinus maenas. Mar. Biol. 137:415-422.

Tessier, G. 1960. Relative growth. In: T. H. Waterman, editor. Volume I: Metabolism and growth. Academic Press, New York. pp. 532-560.

Thurman, H. V. 1997. Introductory oceanography. New Jersey: Prentice-Hall. 544 pp.

Van der Meeren, G. I. 1992. Location of spawning shore crabs, Carcinus maenas (L., 1758) (Decapoda, Brachyura). Crustaceana 63:92-94.

Veillet, A. 1945. Recherches sur le parasitisme des crabes et des Galathees par les Rhizocephales et les Epicarides. Ann. Inst. Oceanogr. Monaco 22:193-341.

Watson, J. & P. G. Wells. 1970. A gauge for measurements of crabs. J. Fish. Res. Board Can 27:1158-1161.

Wheatly, M. G. 1981. The provision of oxygen to developing eggs by female shore crabs (Carcinus maenas). J. Mar. Biol. Ass. UK 61:117-128.

Wolf, F. 1998. Red and Green forms in the common shore crab Carcinus maenas (L.) (Crustacea: Brachyura: Portunidae): theoretical predictions and empirical data. J. Nat. Hist. 32:1807-1812.

Zeidler, W. 1978. Note on the occurrence of the European shore crab C. maenas (Linn., 1758) in Australia. S. Aust. Nat 53:11-12.

DOMINIQUE AUDET, (1) GILLES MIRON (1) * AND MIKIO MORIYASU (2)

(1) Departement de biologie, Universite de Moncton, Moncton, Nouveau-Brunswick, E1A 3E9 Canada; (2) Centre des peches du Golfe, Peches et Oceans Canada, C. P. 5030, Moncton, Nouveau-Brunswick, E1C 9B6 Canada

* Corresponding author. E-mail: [e-mail]

TABLE 1.
Summary of various green crab (Carcinus maenas) reproductive events
occurring at different geographical locations.

                     Eastern Atlantic

Mating             Peak in August (1)
                   June to October (2)
                   August to September (3)

Egg bearing        Peak: March to April (1, 3)
                   November to December (1, 4)
                   Peak in Januarys
                   Oct to June and peak in January  (6)

Molting            May to June (1)
(male)             Peak in November (6)

Larval             Peak in July to August (1)
release            February to July (3)
                   February to April and June to July (7)

Maturity           36.00 to 42.00 mm (1)
(female)           23.00 to 45.00 mm (2)
                   28.00 to 30.00 mm (5)
                   21.5 mm (6)

Maturity           >44.00 mm (2)
(male)             27 mm (6)

                         Central
                        Maine (8)           Basin Head (PEI)

Mating             Peak in August         Peak from mid-August
                                          to September

Egg bearing        Peak from May          Peak in July
                   to June

Molting            June to October        Peak in July
(male)

Larval             Peak in September      Mid-July to the end
release                                   of September

Maturity           34.00 to 45.00 mm      36.90 to 43.79 mm
(female)

Maturity                    --            44.32 to 49.48
(male)

(1.) Broekhuysen (1936): The Netherlands.

(2.) d'Udekem d'Acoz (1993): Belgium.

(3.) Naylor (1962): South Wale, UK.

(4.) Wheatly (1981): UK.

(5.) Van der Meeren (1992): Norway.

(6.) Almaca (1982): Portugal.

(7.) Queiroga et al. (1994): Portugal.

(8.) Berrill (1982): Maine, Eastern USA (Atlantic).

Figure 11. Percentage (%) of females Carcinus maenas captured with
fyke nets in Basin Head (PEI) at different embryogenesis stages from
July 4 to July 17, 2001. The Embryogenesis stages were established from
coloration. Four groups were used: orange egg mass with embryos mostly
composed of vitellogen to dark brown egg mass composed of eggs with
embryos bearing eye pigments, brown (eyes); orange green;
dark orange; orange.

Egg development stage

             brown (eyes)   orange green   dark orange   orange

04-July-01                        1                         10
12-July-01        36                                         4
17-July-01        17                            1

Note: Table made from bar graph.