REPRODUCTIVE ASPECTS OF MARINE CLADOCERANS Penilia avirostris AND Pseudevadne tergestina ( CRUSTACEA , BRANCHIOPODA ) IN THE OUTER PART OF GUANABARA BAY , BRAZIL

The objective of this study was to report on some aspects of marine cladoceran reproduction in Guanabara Bay. Samples were collected during 1985 using a conical net with 200 μm mesh. Two species were identified: Penilia avirostris and Pseudevadne tergestina. Both species disappeared from the plankton in winter. Two months before this phenomenon, parthenogenetic females displayed reduction in brood size, while gamogenetic individuals appeared among the populations. This sequence of events is generally typical of marine cladoceran populations in temperate waters.


INTRODUCTION
Most marine cladocerans are restricted to coastal waters, where they make up a significant part of the zooplanktonic community at given periods.This confers on these animals a major trophodynamic role, as they can be an important food item for carnivorous zooplankton, as well as pelagic fish and their larvae (e.g., Cheng & Chao, 1982).Despite their high densities, marine cladocerans may disappear from the plankton during certain seasons of the year, which in temperate regions is generally in winter (e.g., Onbé, 1974Onbé, , 1978aOnbé, , 1985;;Ramirez & Perez Seijas, 1985).
Cladocerans present two forms of reproduction: parthenogenesis and gamogenesis.Throughout most of the year only females occur, reproducing themselves by parthenogenesis and giving birth to new females.
The explosive increase in their densities is the result of the high reproductive potential in parthenogenetic generation.At given periods, when environmental conditions become unfavorable, these females produceboth males and females.Following copulation and fertilization, the female generally produces a single large egg, called a "resting egg", with a resistant external membrane.The animal developing from this egg is always a female, thus restarting the parthenogenetic cycle (Barth, 1972).
According to Nogueira et al. (1988), marine cladocerans (Penilia and Pseudevadne genera) in Guanabara Bay suffer wide-ranging seasonal density variations.Based on these initial observations, the present study aims at describing the transition from parthenogenetic to gamogenetic reproduction of the species found in this region.

METHOD
Guanabara Bay is located in Rio de Janeiro, southeast Brazil (22 o 41'-22 o 56'S and 43 o 02'-43 o 18'W).This bay has been classified as a coastal estuary with shallow and partially mixed waters (Amador, 1982).The climate is tropical humid with a rainy season occurring in summer (December-March) (IBGE, 1977).
From January to December 1985, in the morning (7:00-11:00)at all 5 stations (Fig. 1) sample collections were carried out, as well as temperature and salinity data recorded.Zooplankton samples were obtained during horizontal surface hauls using a 0.5 m diameter and 2 m long conical net (200 µm mesh) provided with a flowmeter.Samples were preserved in 4% formaldehyde according to Griffiths et al. (1976).In the laboratory cladocerans were counted and, when samples were large, subsamples were taken with a Folsom plankton splitter (McEwen et al., 1954).At least 100 individuals were counted (Frontier, 1981).Identification of species was carried out under a stereomicroscope following descriptions given by Della Croce (1974) and Ramirez (1981).Immediately following qualitative and quantitative analysis at least thirty animals were randomly taken from each sample (when in sufficient number) and placed in glycerine droplets for body clearing.After 24 hours, the individuals were analyzed under an optical microscope for measurement of body length, sex, brood size (number of embryos per female), and embryo development.For Pseudevadne, length was taken as the distance between the anterior margin of the head and the tip of the body shell.However, Bainbridge (1958) found this distance to vary according to the developmental stages of embryos in this genus, therebye precluding its use as a standard.For Penilia, length was considered as being the distance between the anterior border of the head and the posterior margin of the body-shell dorsal articulation line.Measurements were made with a micrometer eyepiece to the nearest 10 µm.
Sex was determined according to the descriptions given by Della Croce (1974) and Onbé (1978b).Individuals observed were classified as parthenogenetic (females bearing parthenogenetic embryos) and gamogenetic (females bearing males and resting eggs).
The females' brood pouches were carefully dissected with a fine needle in order to determine the stage of parthenogenetic embryo development.Classification followed the descriptions given by Della Croce & Bettanin (1965) and Onbé (1974Onbé ( , 1977Onbé ( , 1978a)).Premature and mature embryos were separately considered in this study.For Penilia, embryos in stages XI-XII were considered mature, whereas those in all other stages were considered premature (Della Croce & Bettanin, 1965).For Pseudevadne, embryos in stages I-II were classified as premature, whereas those in stages III-IV were considered mature (Onbé, 1974(Onbé, , 1977(Onbé, , 1978a)).

RESULTS
Mean monthly values for surface water salinity varied from 24.7 to 35.6 (Fig. 2).The lower values recorded from January to March (summer) reflected the seasonally high precipitation levels typical of this region.Temperatures were characteristic of the two well-defined seasons: mean values varied from 21.2 o C in July (winter) to 28.5 o C in February (summer) (Fig. 2).

DISCUSSION
The seasonal distribution, size, and fecundity variations of the marine cladocerans Penilia avirostris and Pseudevadne tergestina were studied in Guanabara Bay.The sequence of events which took place in plankton of the bay is generally typical of marine cladoceran populations: concomitant with the reduction of female size in P. avirostris and P. tergestina, males appeared in the populations, females displayed less fecundity at the same time that they started to produce resting eggs, and the organisms disappeared from the plankton.This fact reinforces the reproductive standard described by several authors, who have observed that the first gamogenetic individuals of marine cladoceran species appear usually at the population maxima, and sexual reproduction becomes most intense when the population is about to disappear from the planktonic communities (e.g., Gieskes, 1971;Specchi et al., 1974;Onbé, 1974Onbé, , 1977Onbé, , 1978a;;Rocha, 1982).The seasonal pattern just described may be explained by Berg's theory (1931): after long study of the transition from parthenogenetic to gamogenetic reproduction in Daphnia populations, Berg formulated the hypothesis that the change in reproductive mode reflects unfavorable external conditions and that the main determining factors are temperature, food supply levels, populational density, chemical agents, and excretion products.According to this hypothesis, the variation in these factors (either isolated or combined) creates a state of "depression" in parthenogenetic females resulting in a reduction in average body length and the production of parthenogenetic embryos.The parthenogenetic eggs, which had previously produced only parthenogenetic females, now also produce males.Sexual reproduction results in the formation of resting eggs, which weather adverse conditions in the sediment.For copepods, resting eggs are generally viewed as the result of a mechanism to ensure survival during periods of environmental adversity, with photoperiod, temperature, and fish predation acting as independent cues signalling deteriorating of environmental conditions (Marcus, 1996).
On February 12, extreme temperature and salinity were recorded for Guanabara Bay.Samples taken 21 days later (March 5) evidenced depressed P. avirostris populations.Possibly temperature and salinity were directly or indirectly responsible for depressed conditions in the populations of this species but, on this point, the results of this study can offer no certainty.However, so far an apparent consensus seems to prevail among researchers that temperature variations are conditioning factors in reproductive changes found in marine cladocerans in temperate regions.This is because these organisms are very evenly distributed from spring to autumn.But with the start of winter, depressive features are in evidence and, shortly thereafter, the species disappear from the plankton.Undoubtedly, determining the factors depressing for marine cladocerans is of extreme relevance, as these organisms display significant reproductive changes in in response to a number of environmental conditions and, thus, these changes may become effective ecological indicators, as has been proposed by Gieskes (1971).
In this study, the mean number of embryos was similar for P. avirostris and P. tergestina females bearing premature and mature embryos.A few authors (see Platt & Yamamura, 1986) have suggested that certain species of marine cladocerans may partially reabsorb their brood when embryos are still in initial developmental stages.For example, Bainbridge (1958) observed several disintegrated embryos and indistinct tissue spots in the brood pouch of some E. nordmanni females.Ramirez & Perez Seijas (1985) found that, for P. intermedius and E. nordmanni, embryos in initial developmental stages were always more numerous when compared to mature embryos in the female's brood pouch.The results of the present study eliminate the hypothesis that P. avirostris and P. tergestina reabsorb part of their broods in the beginning stage of development.
Acknowledgements -This research is part of a PRONEX project, and was supported by the Coordenadoria de Aperfeiçoamento de Pessoal de Ensino Superior (CAPES), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and the Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ).

Fig. 3 -
Fig. 3 -Variations in brood size of Penilia avirostris and Pseudevadne tergestina.The data were obtained at 5 stations from January to December 1985 in Guanabara Bay.The missing line represents absent data.Point = mean; box = standard error; whisker = standard deviation.