Population biology of Callinectes ornatus Ordway , 1863 ( Decapoda , Portunidae ) from Ubatuba ( SP ) , Brazil *

The Portunidae family presents more than 300 described species. According to Williams (1974), the genus Callinectes Stimpson, 1860 is made up of 14 species confined almost exclusively to shallow, often brackish coastal waters. Eleven of these species are distributed throughout the Atlantic Ocean. The swimming crabs of the genus Callinectes are intensively exploited in estuaries and bays of the entire American continent, especially along the east coast (Perry and Malone, 1985). Although representatives of the entire genus Callinectes are known as blue crabs, this name is most commonly applied to C. sapidus Rathbun, 1896 which is the most economically important species occurring along the Atlantic coast. Other species are: C. bocourti, C. danae, C. exasperatus, C. larvatus, C. ornatus and C. similis. Literature about blue crabs is extensive, probably because they are highly commercially exploited in North America. Concerning the biological aspects of the species of the genus Callinectes, some papers can be pointed out: Kretz and Bücherl (1940) studied their anatomy and physiology; van Engel (1958,1990) SCI. MAR., 63 (2): 157-163 SCIENTIA MARINA 1999


INTRODUCTION
The Portunidae family presents more than 300 described species.According to Williams (1974), the genus Callinectes Stimpson, 1860 is made up of 14 species confined almost exclusively to shallow, often brackish coastal waters.Eleven of these species are distributed throughout the Atlantic Ocean.The swimming crabs of the genus Callinectes are intensively exploited in estuaries and bays of the entire American continent, especially along the east coast (Perry and Malone, 1985).
Although representatives of the entire genus Callinectes are known as blue crabs, this name is most commonly applied to C. sapidus Rathbun, 1896 which is the most economically important species occurring along the Atlantic coast.Other species are: C. bocourti, C. danae, C. exasperatus, C. larvatus, C. ornatus and C. similis.Literature about blue crabs is extensive, probably because they are highly commercially exploited in North America.
Concerning the biological aspects of the species of the genus Callinectes, some papers can be pointed out: Kretz and Bücherl (1940) studied their anatomy and physiology; van Engel (1958Engel ( ,1990) ) investigated the fishery biology and reproduction of C. sapidus; Haefner and Schuster (1964) studied the growth of C. sapidus; Devries et al. (1983) analyzed the reproduction periodicity of C. arcuatus; Román-Contreras (1986), Hines et al. (1987), Buchanan and Stoner (1988), Lipcius and van Engel (1990) studied the population dynamics and distributional aspects of several species.
In a particular reference to Callinectes ornatus Ordway, 1963, Williams (1966) and Gore (1977) dealt with its diagnostic features and its distinction from C. similis.Haefner (1990aHaefner ( , 1990b) ) investigated the morphometry and natural diet of C. ornatus.Negreiros-Fransozo andFransozo (1993 and1995) analyzed comparatively some aspects of the biology and the ecological distribution of C. ornatus and C. danae in a South American bay.
The biology of tropical decapods is still poorly reported and new information is necessary for a more complete and comparative comprehension of their biological strategies.
The swimming crab C. ornatus is distributed in the western Atlantic from North Carolina down to Florida, Bermuda, Gulf of Mexico, Central America, Antilles, Guyannas and Brazil (from Amapá to Rio Grande do Sul).It can be found near river mouths and bays but is most common in moderate salty waters, inhabiting sand, mud or shell bottoms (Williams, 1974).The depth range of this species is from the intertidal zone to 75 m of depth (Melo, 1985).
The species C. ornatus constitutes an artisanal commercialized species in the Ubatuba region.Recently an increase in crustacean rearing activities of this species has been observed and in order to proceed, a more detailed knowledge of their biology, particularly with reference to growth, reproduction and dietary aspects is necessary.
This paper describes the results of some aspects of the biology, including population structure, sexratio and reproductive season, of a tropical shallow water swimming crab, C. ornatus in Ubatuba, a region of the Brazilian coast.

MATERIALS AND METHODS
The specimens were collected by otter-trawl (with 10 mm cod mesh), every other month from January 1991 to May 1993 in the Ubatuba bays (from 23°20' to 23 o 35' S and from 44 o 50' to 45 o 14' W), São Paulo, Brazil (Fig. 1).The samples were taken in a depth range from 4 to 18 m along the coast.The sampled area represented 15,000 m 2 every other month.
Identification of species was based on Williams (1974).C. ornatus has morphological peculiarities that can be used in identification: i.e., it often presents small submesial frontal teeth, almost completely rudimentary and the first gonopods do not overlap the 5th thoracic sternite.The adult male is easily separated from other congeners.
In the laboratory, the following data were recorded: sex, carapace width (except the lateral teeth: with a caliper with an accuracy of 0.1 mm) and for females, the ovigerous condition.
The adult swimming crabs were distinguished from the juveniles by the examination of their abdomens.Juveniles have an abdomen sealed to the thoracic sternites while adults have a flexible one.
The median values of the densities of swimming crabs obtained for each demographic category (AM = adult male; AF = adult female; JM = juvenile male; JF = juvenile female) were analyzed by means of the Kruskal-Wallis one way analysis of variance on ranks (p< 0.05).To compare the densities of C. ornatus in the Ubatuba region, data from a previous work by Pires-Vanin (1989) was used for depth ranges from 0 to 100 m and data of the present project in 1991 ranging from 4 to 18 m.
In order to evaluate the sex-ratio and to estimate the reproductive period based on ovigerous female rates, a multinomial proportion comparison test was used and complemented by the Tukey test (Goodman, 1964(Goodman, , 1965)).
The temperature data were obtained from the Oceanographic station of the University of São Paulo located in Ubatuba.Each value used corresponds to a monthly mean water temperature.
A linear correlation of Pearson was used to study the relation between ovigerous-ratio and mean water temperature, during each sampling period.The significance level adopted was 5 %.

Population structure
A total of 3,829 specimens of C. ornatus was obtained, of which 2,394 were males (62.5%) and 1,435 were females (37.5 %).The density median value obtained for this swimming crab during the study period was 15.3 specimens per 1,000 square meters.
The differences in the median values for density between each demographic category of C. ornatus in the sampled area were greater than would be expected by chance and there is a statistically significant difference.The density of adult males differed from that of adult females, juvenile males and females (Fig. 2).
As can be seen in Figure 3, the density of C. ornatus in the Ubatuba region differed in depth range (P<0.01), the highest density being observed in the region from 4 to 18 m of depth.Unfortunately, there is no available data for the comparison of demographic categories in such a wide depth range.
The size frequency distributions for each collecting month are shown in Figure 4 The sex-ratio for all collected individuals was 1.7:1.0.Male occurrence was significantly (P<0.05)higher than female in eleven months, but was not significant in July/91, September/91, January/92 and March/92 (Table 1).Sex-ratio seasonality was not verified.
The male proportion of C. ornatus increased towards larger sizes.Sex-ratio did not differ from 1:1 in the small sizes, but becoming progressively and significantly biased towards males in the largest sizes, finally reaching 100% of males in the largest sizes (from 53.0 mm of CW).

Reproductive period
The occurrence of adult non-ovigerous females was higher than ovigerous females, except in November/92.In general, ovigerous females were found throughout the year.However, none was sampled in November/91, July/92, September/92, March/93 and May/93.The greatest proportions of ovigerous females for C. ornatus occurred in January/91/92/93, May/92 and November/92 (Fig. 5).
POPULATION BIOLOGY OF C. ORNATUS FROM BRAZIL 161 However, only in November/92 did the incidence exceed 50%.The correlation analysis carried out for water temperature with the proportions of ovigerous females did not reveal any relation (p>0.05) (Fig. 6).

DISCUSSION
Unimodality in size frequency distributions usually provides evidence of continuous recruitment, without interruptions, in the size classes and constant mortality rates.
From the sex-ratio obtained (1.7:1), some inferences can be made.There are some possible explanations for such a pattern, which support the hypothesis that this pattern might be due to reversal or sexual differences in longevity, mortality and growth rate.As suggested by DuPreez and McLachlan (1984) regarding the portunid Ovalipes punctatus, the deviation between the sex proportions may be due to differential adult mortality rates.However, in this study of C. ornatus none of these parameters are known.Nonetheless, one female can mate several males during the same reproductive period and the adult females can move offshore where the temperature and/or salinity conditions are good for spawning (Negreiros-Fransozo and Fransozo, 1995).
According to Giesel (1972), deviation from the expected sex-ratio can regulate the population size by acting on the reproductive potential.Díaz and Conde (1989) proposed that this condition can be seen in species with different growth rates and/or different life expectancy for each sex.The C. ornatus obtained data might fit this model.
Although the samples were taken as deep as 18 meters and the species occurs down to 75 meters, the size range of the swimming crabs described in the present paper is comparable to others in literature (Table 2).Hartnoll and Bryant (1990) mention that in sampling living decapods there are, in many cases, substantial problems in obtaining unbiased samples, since the methods of collecting and sorting normally select against smaller specimens, and different size-groups may be spatially segregated.Habitat partitioning by different sizes of swimming crabs have been studied by Hines et al. (1987) and was also reported by Negreiros-Fransozo and Nakagaki (1998).
The possibility of habitat partitioning and salinity tolerance of C. ornatus must be considered.As assumed by Negreiros-Fransozo and Fransozo (1995), all sexual and maturation phase groups of C. ornatus are present in the Fortaleza bay, Ubatuba (SP), but there is an absence of males and juvenile males of C. danae in the same area.
In the case of C. ornatus, probably the juvenile habitat is different from the adult one, as it can be often found in the shallower sites, but this aspect has not yet been investigated in detail in the tropical and subtropical areas.
As shown by the comparison test used to analyze the swimming crab density at different depth strata, density decreases with increasing depth such that bellow 50 m there are very few specimens.This information confirms the current data are consistent for an appraisal on C. ornatus population biology.
According to Wenner et al. (1974), the percentage of ovigerous females can vary with the season and the reproductive cycle due to changes in temperature, food supply and/or other factors.However, in tropical or subtropical regions this is not verified.Compared to temperate regions, the oscillation of temperature arising out of the latitudinal gradient, is not so intense.This makes possible the continuity of the physiological processes of growth and reproduction throughout the year.
A diffrent spawning strategy by C. ornatus and C. danae was observed by Pita et al. (1985) in Santos region (São Paulo littoral) and by Negreiros-Fransozo and Fransozo (1995) during a previous period in a nearby area.The larval discharge by these species in different periods of the year could 162 M.L. NEGREIROS-FRANSOZO et al.Pita et al. (1985) 28.6/83.925.9/100.9Brazil Florianópolis (SC) Branco et al. (1990) 20/90 25/85 Brazil Bermudas, USA Haefner (1990a) 12.1/84.410.5/64.2 Ubatuba (SP) present study 9.3/84.615/74.8Brazil be a strategy to reduce competition and, consequently, adjust their reproductive cycles to a period of better developmental conditions.Nevertheless, a more detailed study must be accomplished in order to get a precise conclusion on the interaction of the reproductive biology in Callinectes.
FIG. 2 -Callinectes ornatus: median density of each demographic category obtained in the Ubatuba region (between 4 and 18 m of depth).(Demographic categories: AM = adult males; AF = adult females; JM = juvenile males and JF = juvenile females).

TABLE 1 .
-Comparison of the sex-ratio of C. ornatus sampled from Jan/91 to May/93 in the Ubatuba (SP) region of Brazil, analyzed by Multinomial proportions test.(The proportion without asteriks in the line means no difference between male and female, P>0.05).

TABLE 2 .
-Comparison of the carapace width (mm) of C. ornatus from different regions.