Larval development of Cyrtograpsus affinis ( Dana ) ( Decapoda , Brachyura , Varunidae ) from Rio de la Plata estuary , reared in the laboratory *

The crab family Varunidae includes marine, brackish and freshwater species (Anger 1995). The genus Cyrtograpsus comprises three species that inhabit the temperate waters of the southwestern Atlantic: C. angulatus Dana, 1851, C. altimanus Rathbun, 1914 and C. affinis (Dana, 1851). The latter was included in Hemigrapsus by Rathbun (1918) but it was re-transferred to Cyrtograpsus by Boschi (1964). Adults of C. angulatus and C. altimanus have wide geographical distributions and form dense and conspicuous littoral populations (Spivak, 1997). They coexist in rocky intertidal habitats (Scelzo and Lichtschein de Bastida, 1979) and show different abilities for invading estuarine habitats (Spivak, in press). In addition to morphological differences, these Cyrtograpsus species markedly differ in size. C. angulatus adults are three times larger than C. altimanus: the maximum observed male body sizes are 58.8, and 19.0 mm carapace width respectively (Spivak, in press). C. affinis, a sublittoral species, is the smallest Cyrtograpsus and seems to be morphologically similar to juvenile C. altimanus (Boschi, 1964; Rathbun, 1918). Most of the scarce findings of C. affinis were made near the mouth of Rio de la Plata estuary: 94% of the 66 specimens examined by Rathbun (1918: 266) were collected “off Rio de la Plata”; Boschi (1964:39) reexamined part of this SCI. MAR., 64 (1): 29-47 SCIENTIA MARINA 2000

material and studied 4 new specimens that also correspond to that area.
A sample of males and females Cyrtograpsus sp. were collected in a subtidal muddy estuarine habitat in the Rio de la Plata, near the coasts of Montevideo, Uruguay, during a fishing cruise of the "BIP Dr. Eduardo Holmberg" (Instituto Nacional de Investigación y Desarrollo Pesquero, Argentina).They were similar to C. altimanus but smaller (maximum observed male carapace width was 11.2 mm) and morphometrically different (males have larger chelipeds); they were assigned to C. affinis.(Spivak, unpublished).
The larval development of C. altimanus and C. angulatus was described from laboratory reared individuals (Scelzo and Lichtschein de Bastida 1979;Rieger and Vieira, 1997).The aim of this paper is to describe the larval development of C. affinis reared in the laboratory.

MATERIAL AND METHODS
Males and females of Cyrtograpsus affinis were collected in the Rio de la Plata estuary (35°07' S, 56°02' W) during a fishing cruise of the "BIP Dr. Eduardo Holmberg" (Instituto Nacional de Investigación y Desarrollo Pesquero, Argentina) on November 19, 1997.The sample was taken with a bottom trawl, 1cm mesh size.The bottom (10 m deep) was muddy and had abundant empty bivalve shells.Water salinity and temperature at the bottom were 21.2 PSU and 18.1°C, respectively.One ovigerous crab, transported to the laboratory at Departamento de Biología, Universidad de Mar del Plata, was maintained in aquaria containing natural sea water until the eggs hatched (December 12, 1997).The larvae were transferred to beakers of 500 ml capacity for mass culture.Natural sea water was used at a temperature of 20°C and salinity of 35 PSU.Larvae were subjected to continual artificial light regime: 8/16 h (L/D).First zoeae were reared and fed with the algae Chaetoceros calcitrans.Zoea II were fed with a mixed diet of algae C. calcitrans and Artemia sp.nauplii.From zoea III to V, Artemia sp.nauplii was offered ad libitum.
Drawings and measurements were made using a Wild MZ6 and Olympus BH compound microscope, both equipped with a camera lucida.All measurements were made by an ocular micrometer.Drawings were based on 5 larvae, and measurements on 10 larvae per stage.In zoea larvae, rostro-dorsal length (RDL) was measured from the tip of the rostral spine to the tip of the dorsal spine; carapace length (CL) from the base of the rostrum to the posterior margin; carapace width (CW) as the distance between the tips of the lateral spines.In the zoea I stage, dorsal and rostral spine length (DS and RS, respectively) were measured from the base to the tip of each one; carapace height (CH) was estimated as RDL -(DS + RS).In the megalopa stage, carapace length (CL) was measured from the base of the rostrum to the posterior margin; anterior carapace width (CW1) as the distance between the tips of the anterolateral protuberances and posterior carapace width (CW2) as the maximum width.The long natatory setae on the distal exopod segments of the first and second maxillipeds are truncated in the Figures 7 and 8. Also, long aesthetascs of the antennules are truncated in the Figure 3.
Zoea I larvae of Cyrtograpsus angulatus and C. altimanus were obtained from ovigerous females collected from Mar Chiquita coastal lagoon (37°45' S; 57°26' W) and the rocky seashore of Santa Clara del Mar (37°50' S; 57°30' W), respectively.These larvae were also dissected, measured, and their morphology compared with the original descriptions of the larval development of these species by Rieger and Vieira (1997) and Scelzo and Lichstein de Bastida (1979).
Quantitative relationships between morphometrical variables were described with least-square regressions (after G tests of homogeneity for normal distribution).Allometric growth of zoeae was studied using the potential model (y = ax b ).The null hypothesis of isometry (b =1) was tested in Cyrtograpsus affinis by means of a t test (Zar, 1984: 271) after logarithmic transformation of the data.
Samples of larvae (zoea I to megalopa) and adult females of C. affinis, and zoea I of C. altimanus and C. angulatus were deposited at the United States National Museum of Natural History, Washington, under the catalog numbers USNM 260933, 260934, and 260935 respectively.
Description and figures are arranged according to the standard proposed by Clark et al. (1998).

RESULTS
The complete zoeal development of Cyrtograpsus affinis took place through five zoeal and one megalopa stages.Mean sizes, and first day of appearance of each zoeal stage are represented in the Table 1.

Description
Zoea I Carapace (Fig. 1A).Globose, smooth and without tubercles.Dorsal and rostral spines long and straight.Lateral spines well developed.A pair of posterodorsal setae.Anterodorsal region, posterior and ventral margin without setae.Eyes sessile.
Antenna (Fig. 4A).Well developed protopod that did not reach at the middle of rostral spine and bear-ing two rows of spines.The exopod is elongated, bears 2 lateral spines and is acute in its distal half.
Telson (Fig. 11C).Posterior margin with 4 pairs of serrulate setae.One row with 9-10 teeth in the inner distal part of each furcal arm.Otherwise unchanged.
Telson (Fig. 11D).One row with 11-12 teeth in the inner distal part of each furcal arm.Otherwise unchanged.

First crab
General morphology (Fig. 2B).Carapace slightly longer than broad with 3 denticulated teeth on the anterolateral margin.Frontal margin rounded, strongly denticulated and with a medial dip.Posterior margin of ocular orbit denticulated.Cheliped equal in size.Cheliped and pereiopods setation as figured.

Morphometry and growth
The proportions between some morphological traits changed through zoeal growth in Cyrtograpsus affinis.A potential relationship exists between RDL and CL and between CW and CL (Fig. 13a).The slope of these lines is 1.07 for RDL and 0.75 for CW indicating slightly positive (P<0.01) and negative (P<<0.001) allometric growth, respectively.On the other hand, RDL, CL and CW increase exponentially throughout the zoeal development (Fig. 14a).Within stage growth (zoea n+1/zoea n), total growth (zoea 5/zoea 1), and mean growth (average size increase within stage) are shown in Table 2 and 1979) and from drawings of Rieger and Vieira (1997, p. 607), respectively (n = 5).RDL: rostro-dorsal length; CL: carapace length; CW: carapace width.(1979) and from drawings of Rieger and Vieira (1997, p. 607), respectively.There is a potential relationship between RDL and CL; its slope is 1.03 in both species (Fig. 13b).RDL and CL increase exponentially during development (Fig. 14b).Within stage growth, total growth, and mean growth calculated from RDL and CL values, are shown in Table 2.
Size differences among zoeae I of Cyrtograpsus species are detected in all examined morphological traits (Fig. 15).RDL, CW, DS and RS are larger, and CH smaller, in C. angulatus larvae than in C. altimanus and C. affinis larvae.On the other hand, clear differences among the three species are found in RDL, DS and RS, C. altimanus larvae being the smallest.Since larvae are from only one hatch, data are not independent and cannot be used for statistical analysis.

Larval morphology
Based on DNA and larval evidences re-definition of the former grapsid subfamily Varuninae H. Milne Edwards, 1853 seemed necessary (Cuesta and Schubart, 1997;Schubart and Cuesta, 1998;Schubart et al., in press)  sus H. Milne Edwards, 1837 and Chasmagnathus de Haan, 1835) are now included in the family Varunidae (Schubart et al., in press).Cyrtograpsus is the second polyspecific genus of Varunidae for which larval development has been described for all species.
Larval development of Cyrtograpsus affinis corresponds in its general morphology to those of C. altimanus and C. angulatus.The few differences found in zoeal morphology are listed in Table 3.The megalopae stages of C. altimanus, C. angulatus and C. affinis have practically the same setation in antenna, mouthparts, pleopods and uropods.The main observed differences among the megalopae are in the morphology of antennule and the mandible: the setation of basal segment of antennular peduncle (3 setae in C. angulatus, 2 in C. altimanus and 5 in C. affinis), the number of aesthetascs on the three terminal segment of antennular flagellum (5,7,6 in C. angulatus, 7,6,4 in C. altimanus and 7,7,4 in C. affinis), and the number of setae on mandibular palp (0,8 in C. angulatus, 0,6 in C. altimanus and 0,5 in C. affinis).
In the present study, newly hatched zoea I of C. altimanus and C. angulatus have been examined.Opposed to previous descriptions of Scelzo and Lichtschein de Bastida (1979) and Rieger and Vieira (1997), we have not found intrageneric differences in the setation of antennule and coxal endite of the maxilla.Also, we have not observed the presence of exopodal setae on maxillule of the zoea I of C. altimanus (see Table 3).Therefore, the rest of differences in setation of other stages, shown in Table 3, must be considered with care.Obvious differences between Cyrtograpsus zoeae can be observed only in morphometry as explained below.
LARVAL DEVELOPMENT OF CYRTOGRAPSUS AFFINIS 45 Based on larval morphology, the genus Cyrtograpsus is closely related to Brachynotus de Haan, 1853.These genera can be separated from the rest of grapsoid genera on the basis of the development of the zoeal telson: Cyrtograpsus and Brachynotus acquire only one pair of serrulate setae in the posterior margin of the telson, as Gaetice depressus (de Haan, 1833) and Helice leachii Hess, 1865.However, Cyrtograpsus and Brachynotus can be easily differentiated from the latter species due to the presence of lateral carapace spines.Also Metaplax distincta H. Milne Edwards, 1852 presents 4 pairs of serrulate setae in the posterior margin of the telson in the last zoeal stage, but differs from Cyrtograpsus and Brachynotus zoeae in the setation of the second maxilliped endopod: 1 seta on the proximal segment in Metaplax distincta versus unarmed in Cyrtograpsus and Brachynotus.The zoeae of the remaining species of Varunidae, with known larval development, acquire more than one pair of serrulate setae through development.
There are no diagnostic morphological differences between Cyrtograpsus and Brachynotus larvae, but both genera have a very distinct distribution and it would be unlikely to find larvae of these genera in the same plankton samples.Brachynotus is restricted to the Mediterranean Sea and eastern Atlantic coast in vicinity of Mediterranean Sea while Cyrtograpsus is present only in South-Western Atlantic and South-Eastern Pacific.Hartnoll (1982: 163) stated that in Crustacea "there is an almost total lack of comprehensive biometric data on larval stages, in contrast to the extensive morphological descriptions".In spite of the time passed since Hartnoll's review, this lack of information seems to continue.The proportions between some morphometrical traits change through zoeal growth in Cyrtograpsus affinis: the distance between the tips of the lateral spines (CW) grows slower, and the distance from the tip of the rostral spine to the tip of the dorsal spine (RDL) a little faster, than the distance from the base of the rostrum to the posterior margin (CL).The slope of the potential relationship between RDL and CL is similar in the other Cyrtograpsus species (Fig. 13).Size, measured as RDL, CL and CW, increase exponentially throughout the zoeal development of Cyrtograpsus spp.(Fig. 14).An exponential size increase in the larval stages of some decapod crustacean species ("Brook's law", Kurata, 1962) was shown by Rice (1968), but other species reveal a linear rather than an exponential pattern (Anger, 1998).The growth within the zoeal stages in Decapoda was examined by Gore (1985), who calculated a series of growth factors using morphometrical data compiled from literature descriptions (not from illustrations, even if available).The total (lat zoea/zoea 1) and mean growth factors of the Grapsidae (18 species) averaged 2.59 and 1.28, respectively.These values are in agreement with those calculated in Cyrtograpsus affinis, C. altimanus and C. angulatus although it is important to note that growth factors differed according to the trait measured (Table 2).

Morphometry and growth
Measurements of morphological traits have been used to distinguish similar larvae of two Carcinus species (Rice and Ingle, 1975) and two Paralithodes species (Jensen et al., 1992); to differentiate among several co-occurring majid species (Pohle, 1991); and to characterize intraspecific variability in Cancer magister (Shirley et al. 1987) and Pachygrapsus transversus (Cuesta and Schubart, 1998).Morphometry may also be used for discriminating among Cyrtograpsus zoeae: at least, C. angulatus differs from the other two species in all examined morphological traits (Fig. 15).Most of the measurements included the carapacial spines lengths (RDL, RS, DS, CH, CW).However, the use of spines to distinguish zoeae from related species may be acceptable only within geographical regions, since their length vary with temperature and latitude (Shirley et al. 1987).Morphometrical information on the Cyrtograpsus species presented here came from populations collected in geographically proximate areas (from 32 to 38°S) with similar environmental temperatures.However, C. altimanus, and mainly C. angulatus, have extensive distributions and, consequently, more detailed studies on zoeal morphometry of these species will be needed.

AKNOWLEDGEMENTS
This paper was written as part of a Spanish-Argentine cooperative programme (Programa de Cooperación Científica con Iberoamérica) between the Instituto de Ciencias Marinas de Andalucía (CSIC) (Spain) and the Universidad de Mar del Plata (Argentina).It was funded by the Ministerio de Educación y Cultura and the Dirección General del Instituto de Cooperación Iberoamericana de la Agencia Española de Cooperación Internacional grant to Antonio Rodríguez and the Universidad de Mar del Plata grant 15/E082 to EDS.We are grateful to the authorities and colleagues from the Laboratorio de Biología Marina of the Universidad de Sevilla (Spain) for their hospitality, to the Instituto Nacional de Investigación y Desarrollo Pesquero, Argentina for the invitation to EDS to take part in the fishing cruise, and to Christoph Schubart from the University of Southwestern Louisiana for critical reading of an early version of the manuscript.

TABLE 1 .
-Mean (±SD) rostrodorsal length (RDL), carapace width (CW), carapace length (CL) in mm, and first day of appearence of the larval stages of Cyrtograpsus affinis reared in the laboratory.Two measurements of megalopae and first crab carapace width were made between antero-lateral spines and between posterior protuberances (spines) (CW1 and CW2 respectively).