Distribution and drift of the crab Carcinus maenas (L ...

Journal of Plankton Research Vol.18 no.ll pp.1981-2000, 1996

Distribution and drift of the crab Carcinus maenas (L.) (Decapoda, Portunidae) larvae over the continental shelf off northern Portugal in April 1991 Henrique Queiroga Departamento de Biologia, Universidade deAveiro, Campo de Santiago, 3810 Aveiro, Portugal

Introduction

The green crab, Carcinus maenas (L.), stands as the most characteristic decapod that inhabits European estuaries and rocky shores. Its wide geographical distribution, which spans from the Scandinavian Peninsula to the Gibraltar strait (Almaca, 1962), and local high abundance of juveniles and adults (Klein-Breteler, 1976; Beukema et al., 1978; Sobral, 1985), place C.maenas as a species of prime ecological importance. In spite of this, we know little about the coastal distribution and ecology of its larvae, apart from general statements about the seasonal or spatial distribution (Lebour, 1928; Vives, 1979; Roff et al, 1986; Lindley, 1987) and the more recent works by Queiroga et al. (1994) and Queiroga (1995) describing the mechanisms of larval transport in estuaries. This picture is in contrast with the research effort that the ecology of juvenile and adult crabs has received through the years (e.g. Edwards, 1958; Naylor, 1962; Wolf and Sandee, 1971; Walne and Dean, 1972; Klein-Breteler, 1975a,b, 1976; Jensen and Jensen, 1985; PihJ, 1985). Large populations of the species exist in the Portuguese coastal area. The most important is probably the one from the Ria de Aveiro, as it supports a commercial fishery that is unique in European waters. Estimates of commercial landings range from 500 to 1000 tons per year, although the number of discarded animals under the legal size, which are normally not returned to the Ria ah've, may be three or four times higher than the catch (Sobral, 1985; Gomes, 1991). Published reports © Oxford University Press

1981

Downloaded from plankt.oxfordjournals.org by guest on September 27, 2011

Abstract The coastal distribution and abundance of Carcinus maenas larval stages were studied from plankton samples collected in a grid of 79 sampling stations, organized into six cross-shore transects extending from the coast to -1701cm offshore, between the Mondegoand Averivers,on the northwest coast of Portugal. The samples were collected in April 1991 with a modified B6 multinet plankton sampler, which was towed from a maximum depth of 200 m and provided a vertical resolution of up to five standard strata at each station. Current and wind data were available for a period that extended beyond the period covered by the observations. All the species' larval stages were found in the samples, but their distribution was confined to the inner and middle shelf stations. Vertically, 88% of the larvae occurred in the top 30 m and another 11% in the 30-60 m stratum. The zoeal stages I and II were concentrated (-90%) in the surface layer, but a gradual ontogenic displacement to deeper waters was observed from then on, the megalopa being equally distributed between the 0-30 and the 30-60 m strata. Horizontally, there was a clear association of the first zoea with the estuarine inlets, while the older zoeal stages were dispersed progressively offshore. Evidence was found that the megalopa experienced an onshore transport that did not affect the previous stages. This transport is consistent with the observation of an onshore flow component at 40 m. It was not possible, however, to examine the hypothesis that this flow conveys the last larval stage to the coast, but not the previous ones. The dependence of the along-shore flow component on wind stress lends support to the hypothesis that the larvae are advected from the north as the upwelling season progresses.

H-Qodroga

on the occurrence of crab larvae along the Portuguese coast are virtually nonexistent. The only study on the subject seems to be the one by Paula (1987), who described the inner shelf distribution of larval crabs in the Sines area, southwest Portugal. This study reported the occurrence of C.maenas larvae throughout most of the year, with no resolution to stage level. The present paper analyses the spatial distribution and drift of the larvae of C.maenas from a collection of samples and hydrological observations made off the Portuguese northwest coast during April of 1991. This sampling campaign was conducted as part of the Eastern Boundary Currents programme of the World Ocean Circulation Experiment II (Hagen et al., 1994; H.Ch.John, Biologische Anstalt Helgoland, unpublished data). A summary of the reproductive biology and ecology of the species is presented first, followed by a description of the main features of the oceanography of the northern Portuguese coast. This information forms the background against which observed patterns of distribution are interpreted and inferences about larval transport are made.

Several authors have extensively described the reproductive biology of C.maenas. Crothers (1968) gives a comprehensive review of its biology, based on the bibliography available at the time. What follows is a synopsis that includes recent reports Ovigerous females move to the lower parts of the estuaries where hatching occurs during nighMime ebbing tides (Queiroga et al, 1994). Newly hatched larvae migrate vertically in synchrony with the tidal cycle, attaining their highest position in the water column during ebb (Queiroga, 1995). This migration, under behavioural control (Zeng and Naylor, 19%), ensures that all the larvae are exported to the sea shortly after spawning. Larval development includes four zoeal stages and one megalopa (Rice and Ingle, 1975). The megalopa metamorphoses to first crab instar 4-9 weeks after hatching, depending on the temperature at which development took place (Dawirs, 1985; Mohamedeen and Hartnoll, 1989; Nagaraj, 1993). The megalopa is the stage that reinvades the estuary. This stage is more abundant in estuarine waters during night-time flooding tides when it occupies a higher position than during ebb. This pattern of occurrence results in a net upstream transport (Queiroga et aL, 1994; Queiroga, 1995). The chronology of the reproductive events depends on latitude. In the northern end of the species range (Maine, eastern USA coast), the occurrence of ovigerous females is restricted to the warmer months (Berril, 1982). As one proceeds south, ovigerous females tend to distribute throughout the year, as in Great Britain and the Wadden Sea (Broekhuysen, 1936; Naylor, 1962); in the Portuguese coast abundance maxima occur in winter (Almaca, 1982; Marques and Costa, 1983; Queiroga, 1995). The larvae can be found in Europe's coastal waters during most of the year (Lebour, 1928; Paula, 1987), but the abundance maxima of the zoeae occur from April to July (Rees, 1952; Lindley, 1987). Settlement of megalopae, either on rocky shores or in estuaries, starts in June and extends into the summer (Broekhuysen, 1982


Carcinus maenas biology

Distribution and drift of Cardnus mamas

1936; D6meusy, 1958; Klein-Breteler, 1976). In the Portuguese estuaries, maxima of abundance of first zoea were found between February and April, but secondary peaks were also identified in June and July (Goncalves, 1991; Paula, 1993; Queiroga, 1995). In the Ria de Aveiro, located on the northwest coast of Portugal, Queiroga (1993) described recruitment maxima during spring. Oceanography of northern Portugal

1983


The geographical position of Portugal, on the northern border of the subtropical anticyclone belt and on the eastern margin of a large oceanic basin, determines most of its climate and coastal ocean oceanography (Fiuza, 1982; Fiuza et al, 1982; Frouin et al., 1990). The seasonal migration of the subtropical front and the Azores high, whose centre moves from 27°N in winter, to 33°N in the summer, regulates the wind regime. Therefore, weak westerly winds predominate during winter, while stronger northerly winds dominate the summer atmospheric circulation. These winds, the 'Portuguese trade winds', force southward currents in nearsurface layers and are upwelling favourable. Within intermediate layers, there is, below 43°N, a geostrophic eastward flow, resulting from the meridional pressure gradient with high southern and low northern values off the Portuguese coast. This onshore flow of the large-scale motion field is diverted to the north when it encounters the slope, then running along the slope and the outer shelf (Frouin et al., 1990; Haynes and Barton, 1990). The southward-directed component of the 'Portuguese trade winds' balances the pressure gradient. Therefore, as the southward wind forcing lessens during the winter, this current rises to the surface. The intra- and interannual variability of the upwelling indices and wind stresses indicates that upwelling maxima occur consistently during the months of July-September in the Portuguese coast, although intense episodes can also occur during spring (Wooster et al, 1976; Fiuza et al, 1982; Sousa and Fiuza, 1989). However, even during the summer, upwelling is an intermittent phenomenon that depends, on a time scale of hours, on local wind events. In any case, the intensification of the north winds induces southward currents with an Ekman offshore transport, with the divergence zone over the shelf, while a decrease generates coastal convergence of the surface layer and a reversal of the meridional transport in deeper layers (Fiuza, 1982,1984; Jorge da Silva, 1992). Most of the spatial and temporal variability of the currents over the Portuguese shelf, thus, depends both on spatial and temporal structures of the wind stress (Haynes and Barton, 1990;Hagen et al, 1993,1994). As a consequence of the large frequency range in this variability and the lack of systematic studies, details of the shelf circulation are not well studied. The most detailed report seems to be the one by Jorge da Silva (1992), who described the circulation on the northwest Portuguese shelf during spring and summer of 1987, which encompassed several episodes of increase and decrease of upwelling favourable winds. In broad terms, the increase in the northerly wind accelerates the equatorward jet that extends vertically down to the bottom, accompanied by offshore advection of near-surface waters. Compensation flows are established along the slope, to the north, and along the bottom (John and R6,19%) or at some intermediate depth above the

H.Quetroga

shelf (Jorge da Silva, 1992), to the shore. The relaxation or reversal of the wind causes a reduction of the speed, or even a reversal, of the coastal jet, and induces coastal convergence at all depths. The tridimensional pattern of the water column can be rather complex, with offshore displacement of the coastal jet and reversal of the along-shore and onshore flows with depth, depending on the intensity and spread of the wind events. Method

E

A

Rio Ave

I

9-10

N

\

o

o

X

•

•

•

•/

41*30'

c

Is,

y

B

\ Laixoes \ ^ Porto

1

8-9 o

•

•

•

a

O

D

V

D

Rio Douro

\ \

7-8 Co

a

•

•

«

•

a

a

u

•

P Ria de Aveiro

4-5

i

W\ D

_ 41"

j

•

a

f

o

•

•

X

•

Aveiro

n If I

!

401301

3-4

a

E D

a

•

•

•

2-3 F

o

"H. a

c

Q

n

7

o

/ Figueira da Foz

\

•

•

X I

11" W

•

°

\

* / Rio Mondego 1

I

1

40"

10"

Fig. L Survey area (inset) and stations sampled off northern Portugal. Letters A-F refer to transect lines. Stations, within each transect, were numbered from 1 to 12-14, in ascending order from the shore. Arrows give the direction (dates above arrows) of sampling. Open squares represent daytime, filled squares night-time and crossed squares twilight CTD casts/plankton tows. The bold cross shows the location of the current meter's mooring. The arrow in the inset map shows the location of the weather station.

1984


Andres et al. (1992) and John and Re" (1996) described in detail the layout of the standard stations and sampling methods used during the Eastern Boundary Currents programme. For the reader's sake, they will be briefly recounted here together with specific information relative to the spring cruise. A grid of 79 sampling stations (Figure 1), organized into six cross-shore transects, was sampled from 2 to 10 April 1991, on board the German RV 'Heincke'. Transects were named A-F, from north to south, and 12-14 stations were allocated to each. Stations within each transect were numbered from 1 to 12-14, starting from the shore, regardless of sampling direction. Plankton samples were collected with a B6 Multiple-Opening-Closing-Net (MCN) with 300 \x.m mesh, equipped with

Distribution and drift of Carcinus maenas

where v is the wind velocity, u and v are, respectively, the eastward and northward wind components, pa is the air density (1.12 kg nr 3 ) and CD is the drag coefficient (0.0012). The data were then averaged to produce daily wind stress values. Besides the plankton samples described above, another set of samples, collected by the staff of the Instiruto Portugues de InvestigacSo Maritima (LPIMAR), was analysed. These samples were collected at 22 stations organized into four crossshelf transects located between Figueira da Foz and Porto, in July and October of 1991 and February of 1992. Oblique hauls with a FAO net were made at each station between the surface and 50 m. 1985


flowmeters, and towed in an oblique path from a maximum depth of 200 m to the surface. Individual samples from the same haul were obtained from four 30 m wide strata, down to a depth of 120 m, and from 120 to 200 m. In shallow stations, the number of sampled strata was adjusted according to bottom depth, but always using the described standard steps, except for the deepest range, which was started 2-5 m above the bottom depending on sea state. The mean size of all analysed samples (see below) was 2.78 m2 (s = 0.941, n = 224). CTD casts were obtained at each station with a ME-OTS 1500 mini CTD, attached to the plankton net. Carcinus maenas larvae were identified to stage according to Rice and Ingle (1975). Owing to the limited time available to process the samples, aliquots were obtained with a Folsom plankton sample splitter. At least a quarter of each sample was analysed and, if any C.maenas larvae were found, the whole sample was counted. Furthermore, as it became apparent that the distribution of the larvae was restricted to the inner and middle-shelf stations (see Figures 4-8), only every second station from the 200 m depth line offshore was studied. Other brachyuran zoeae and megalopae were also counted, with no resolution to lower taxa or stage in the zoeal series. Abundance values were expressed as catch per unit effort (CPUE), in numbers of each larval stage per square metre of sea surface, either per individual stratum in the case of vertical distributions, or integrated over the total sampled vertical range when the interest was on horizontal distributions. Vertical and horizontal contour plots of the abundance of the larval stages were drawn using a least squares regression method of interpolation. For the purpose of the vertical contours, the larval stages were considered to be concentrated in the middle point of the strata at each station. Besides the plankton and CTD data described, current velocity and direction were measured with three current meters moored near Station C3 (see Figure 1) at depths of 40,71 and 98 m, from March 19 to April 12 (Hagen et ai, 1994). A wind time series furnished by the Institute de Meteorologia, obtained at the weather station of Cabo Carvoeiro, was used to analyse, in a graphical form, the influence of wind on flow of the surface layer and the potential for along-shore transport during the development time of the larvae at sea. Eastwards and northwards wind stress components were computed from measurements made at 00, 06,09,12 and 18 h, using the equation:

H.Qneiroga

Results Hydrological situation

12 8 4

1

CM

b Uj

a. CM

o

\5

Eastward wind stress

. . A V L A A . ,/\

0 -4 -8 8 4 0 -4 -8 -12 -16 -20 -24

Vv

1

11 21 1

V"

11

21

1

11

21

1

11 21

1

11

21

1

11

DAYS

Fig. 2. Daily averaged eastward and northward wind stress components calculated from measurements made at the weather station of Cabo Carvoeiro.

1986


A comparison of the wind data collected at Cabo Carvoeiro (Figure 2) with the winds measured on board at the sampling stations (Figure 4 of Hagen et aL, 1993) shows that the intensities at the weather station are somewhat lower than at sea. John and R6 (19%) also report phase differences of up to 2 days along the northwest Iberian coast. However, the data of the weather station did not show directional bias and largely agreed with local conditions. Upwelling-favourable wind conditions were found at Transect C on April 7. The stations of this transect were sampled at the beginning of the event, and Transects B and A during the following days. The intensity measured at Cabo Carvoeiro during this particular wind event was -0.85 times lower than the local wind. Vertical contour distributions of salinity and temperature at Transects C-A show upwelling-like structures at these sections, over the shelf and the continental slope, with intrusions of colder upwelled water of higher salinity. Salinity and temperature at the shelf stations ranged from 34.0 to 35.9%o and from 12.5 to 14.5°C, respectively. At Transects D, E and F, the upper 200 m were slightly stratified, the vertical temperature gradient not exceeding 1.5°G Salinity and temperature variation at the shelf stations of these transects was narrower than that found at the three northern transects: 35.3-35.8%o and 13.0-14.0°C, respectively. The progressive vector diagrams constructed from the current measurements (Figure 3) show a correlation between resulting current direction at 40 m and wind stress (Figure 2), within the period March 19-April 13. Particularly evident are the

Distribution and drift of Carchuu mamas

10 km

41°30'

10 miles

N

19/3

40°30'

40° 10° W Fig. 3. Progressive vector diagrams constructed from the current velocity and direction data obtained at the mooring station, between March 19 and April 13,1991. T\vo consecutive crosses limit time intervals of 48 h (adapted with permission from Hagen el at., 1994).

onset of the northerly winds on March 22 (that reached TN —19 X 10~2 Pa, corresponding to average wind velocities >11 m s~x at the weather station) and the northeasterly wind event found at Transect C on April 4 (TN — 4 X 10~2 Pa, TE ~ 4 X 10"2 Pa; average wind velocities >6 m r 1 at the weather station). Both resulted in a shift of the flow at 40 m, from NE to S in the first case and to SSE in the second, with a much larger advection during the first event. A net onshore transport was observed between March 19 and April 7 at 40 m. The east-west component inverted direction from April 7 to 12, probably responding to the northerly upwelling-favourable winds found at Transect C The current meters at 71 and 98 m showed a reduced or zero onshore drift. The time lags between the onset of the wind events and the presumed surface effects are within the local inertia period of 18.2 h (Jorge da Silva, 1992). 1987


71 m

H.Qneiroga

Overall abundance and depth distribution

Zonation and drift The distribution of C.maenas larvae was restricted to the inner and middle shelf stations (see inset maps in Figures 4-8). Horizontally, the first zoeae were found close to the shore, associated with the inlets of estuaries. Older zoeae were distributed further offshore, with a variable number of maxima along a line 15-20 km from the coast. It is noteworthy that the contour lines for the megalopa again intercept the shore, a feature that was not found in the fourth zoea, and is indicative that the last larval stage experienced some onshore transport. Figure 10 shows the percentual distribution of C.maenas larval stages in the 0-30 m layer as a function of the sampling station, all sections combined. The cross-shelf distribution of the megalopae was bimodal, with -36% of the megalopae collected at the first station. This feature is consistent with the current meter data which show a net onshore component at 40 m. A similar pattern was not found in the 30-60 m stratum. The lack of evidence supporting an onshore transport of the megalopa below 30 m was expected because, being restricted to the inner-shelf stations, the larvae would eventually intercept the bottom during any coastward drift. The average water temperature recorded in the distribution area of C.maenas larvae, during this survey, was 13.5°C. The development time of the larval stages was extrapolated, for this temperature and 35%o salinity, from data obtained with individual- (Dawirs, 1985) and group-reared (Mohamedeen and Hartnoll, 1989) larvae (Table II). Since group-reared larvae grew consistently slower, at all temperatures, than individual-reared specimens, the extrapolated development times 1988


All C.maenas larval stages were found in the samples collected in April 1991. A total of 1183 larvae of the species, which comprised 4.0% of the total brachyuran larval catch, were counted. This percentage rose to 11.1% if only the stations where this species' larvae occurred are considered. In these stations, C.maenas was responsible for 11.2 and 7.7% of the total brachyuran zoeae and megalopae, respectively. A semiquantitative analysis of the IPIMAR samples (without standardization to individuals per square metre) showed that the counts of C.maenas larvae were up to two orders of magnitude larger in April 1991 than in July and October of the same year or in February 1992. Carcinus maenas larvae were not collected below 90 m (Figures 4-8, Table I). The data indicate a concentration of the larvae in the 0-30 m layer. Almost all the stations where Cmaenas larvae were found were sampled during the day (see Figure 1 and Figures 4—8); only two were sampled at twilight and one during the night. Figure 9 shows the average depth distribution of C.maenas larvae, all stages combined, during the day and at twilight. A Mest (Sokal and Rohlf, 1969) showed that the larvae were significantly deeper at twilight (P < 0.05), the variances of the vertical position being identical according to an F test (P > 0.90; Sokal and Rohlf, 1969). Table I indicates a tendency for older stages to move deeper, reflected in increasingly higher proportions of successive stages in the 30-60 m stratum.

riletrihntfnnftnHrfriftnff/irWiiiK M / I M / I

km 250

200

150

50

100

0

50 100 150 200

Section A

Section B 0.6

Section C

Section D

Section E

Section F

W 11°

10 c

50 100 150 200 50 100 150 200 50 100 150 200 m 8C

Fig. 4. Carcinus maenas first zoea. Vertical and horizontal (inset map) distribution. For the purpose of the vertical distribution, all the larvae are considered to be concentrated in the middle point of the strata (dots). Values are in individuals per square metre.

were used as two different estimates of the mean duration of each stage. The megalopae collected during this study would thus be 35-56 days old, using an average of the two growth rate estimates. This places the hatching moment of these larvae somewhere between February 10 and March 3. 1989


50 100 150 200 50 100 150 200

Total CPUE

120-200

90-120

60-90

30-60

0-30

Stratum (m)

(1.1) 0.0 (0.0) 0.0 (0.0) 69.1

1.9

(0.8) 0.0 (0.0) 0.0 (0.0)

242.9

0.8

(0.4) 0.0 (0.0) 0.0 (0.0)

2263

54.1 (78.3) 142 (20.6)

0.8

ZoeaUI

Zoeall

222.1 (91.5) 18.9 (7.8)

Zoeal

208.0 (91.9) 17.6 (7.8)

Cardnus maenas

23.2

(3.2) 0.0 (0.0) 0.0 (0.0)

0.8

13.5 (58.3) 8.9 (38.5)

ZoealV

10.1

4.9 (48.7) 5.1 (513) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0)

Megalopa

571.5

502.6 (87.9) 64.7 (11.3) 4.3 (0.8) 0.0 (0.0) 0.0 (0.0)

Total CPUE

12651

9281 (73.4) 2674 (21.1) 367 (2.9) 90 (0.7) 239 (1.9)

Brachyuran

1207

844 (70.0) 227 (18.8) 59 (4.9) 19 (1.6) 58 (4.8)

Brachyuran megalopae

Table L Vertical distribution and total CPUE of Cmaenas larval stages and brachyuran zoeae and megalopae. Values in parentheses refer to the percentage of each stage and taxon within each stratum


Distribution and drift of Carcinus maenas

km 250

200

150

100

50

0 50 100 Section A

Section B

Section D

Section E

Section F W 11°

10°

200 50 100 150 200 50 100 150 200 50 100 150 200 50 100 150 200 50 100 150 200 m

8°

Fig. 5. Carcinus maenas second zoea. As for Figure 4.

The progressive vector diagram (Figure 3) shows a net southward drift at 40 m during the period March 19-April 13. A progressive vector diagram measures the temporal variability of the current in a fixed point in space. Usually, a particle of water enters a different velocity field once it passes the point of observation, and so the progressive vector diagram does not represent the true trajectory of the particle. However, it provides a measure of the potential for transport. During the southward wind event of March 22-25, for instance, a parcel of water at 40 m could 1991


Section C

150

H-Qoeiroga

km 250

200

150

50

100

0 50 100 150

Section A

200 50 100 150

Section B

200 50

Section C

Section D

Section E

Section F

W 11°

10°

9°

150 200 50 100 150 200 50 100

150 200 m

8°

Fig. 6. Carcinus maenas third zoca. As for Figure 4.

move -53 km in a southward direction in 3 days (17.7 km day"1). During the 25 days of study, the net southward component at 40 m was 34 km, or 1.4 km day-1. Figure 3 shows a dominance of southward wind stress during the months of February, March and April of 1991. Based on the congruent variation of wind stress and flow at 40 m during the periods covered by both sets of data, it appears that the megalopae collected during this study were advected from the north during their previous larval development. 1992


100 150 200 50 100

Distribution and drift of Cardnus mamas

km

250

200

150

50

100

Section A

Section B

Section D

Section E

Section F

W 11°

10°

9°

Fig. 7. Carcinus maenas fourth zoea. As for Figure 4.

Discussion

The present study was conducted during the season of greatest abundance of Cmaenas larvae in 1991 and the beginning of 1992. All larval stages were present in the coastal waters of northern Portugal. The clear association of the first zoea with the estuarine inlets indicates the estuarine origin of, at least, the great 1993


Section C

0 50 100 150 200 50 100 150 200 50 100 150 200 50 100 150 200 50 100 150 200 50 100 150 200 m

H-Qoelroga

km 250

200

150

100

50

0 50 100

Section A

150 200 50 100

Section B

Section D

Section E

Section F

W 11°

10°

50 100 150 200 50 100 150 200 50 100 150 200 m

9°

Fig. 8. Carcinus maenas megalopa. As for Figure 4.

majority of the larvae. The largest concentrations of the first larval stage occurred near the Ria de Aveiro. The vertical position of the larvae changed with the phase of day. Notwithstanding the fact that only one of the stations where the larvae were found was sampled during the night, an occurrence that prevented its inclusion in the statistical analysis, the larvae collected during twilight were distributed deeper than 1994


Section C

150 200 50 100 150 200

Distribution and drift of Carcinus manias

I 0. UJ Q

TWILIGHT

PHASE OF DAY Fig. 9. Carcinus maenas. Average depth of distribution of the larvae, all stages combined, during the day and at twilight. Symmetrical bars represent 95% confidence intervals.

those collected during the day. The variance of the vertical distribution, identical in the samples collected during the day and at twilight, suggests a vertical migration participated in by all the stages. However, since most of the stations (85%) were sampled during the light hours, it seems reasonable to analyse the vertical distribution of the stages in the whole set of samples. Virtually all larvae were collected above 60 m. More than 90% of the first and second zoeae were collected in the 0-30 m stratum. As one proceeds in the larval series, the percentage of each stage in the surface layer decreased gradually, the megalopae being equally distributed between the 0-30 and the 30-60 m strata. The present study used a sampling methodology that permits accurate integration of discrete depth samples over the entire vertical range of the strata. Thus, laMe D. Carcinus maaias. Estimated duration (days) of the larval stages at 13.5°C and 35V according to Dawirs (1985) and Nagaraj (1993) Stages

Dawirs

Nagaraj

Average

Zoeal Zoeall ZoeaHI ZoealV Megalopa Total to megalopa Total to first juvenile

7.2 6.4 7.1 9.0 17.9 29.7 47.6

10.2 8.6 9.1 12.6 24340.5 64.81

8.7 7.5 8.1 10.8 21.1 35.1 56.2

' Nagaraj did not measure the duration of the megalopa instar. This value assumes an average duration of the instars 36% higher, in Nagaraj's data, than in Dawirs' experiment.

1995


DAY

H.Queiroga

100

CO LU

O

CO LL

O LU

o