Hesp et al.: Timing and frequency of spawning and fecundity of Rhabdosargus sarba 



653 



further in size and all of their lipid and yolk material 

 had coalesced (Fig. 4E). Under the dissecting micro- 

 scope, these hydrated oocytes were of similar appear- 

 ance to the corresponding oocytes at 21:30 h (Fig. 4F). 

 Although mature fish with ovaries containing the above 

 stages in oocyte hydration were frequently found in 

 nearshore shallow waters, the numbers of such fish in 

 these waters declined markedly after about 00:30 h 

 and none of the few fish caught there after this time 

 contained recently formed POFs. However, fish with 

 ovaries containing newly formed POFs were caught in 

 offshore deeper waters. 



Histological examination demonstrated that, when 

 hydrated oocytes were present in the ovarian duct, 

 the ovary contained recently formed POFs, which are 

 formed by the thecal and granulosa layers of the oocytes 

 that surround the zona radiata externa (Fig. 5A). Newly 

 formed POFs (0-6 h old) possess a conspicuous lumen 

 and their granulosa cells contain prominent darkly 

 stained nuclei (Fig. 5B). These newly formed postovula- 

 tory follicles were first observed in the ovaries of females 

 caught at ca. 01:30 h and were present in the ovaries of 

 several fish caught in the ensuing four hours. In con- 

 trast, no newly formed POFs were found in the ovaries 

 of R . sarba at dusk, i.e., ca. 18:30 h. At this time, the 

 POFs comprised one of two morphological forms. The 

 first and least degenerate form was less well organized 

 than newly formed POFs and its nuclei were becoming 

 pycnotic (Fig. 5C); the second form was smaller and 

 highly degenerate and its nuclei had become far less 

 visible or undetectable (Fig. 5D). The least degenerate 

 of the two forms of POFs in ovaries of fish caught at ca. 

 22:00 and 01:00 h (Fig. 5, D and E) represents stages in 

 degeneracy that are intermediate between those of the 

 two different forms described above for the ovaries of 

 fish caught at 18:30 h. These POFs were thus compact 

 and, although some of their nuclei were still detectable, 

 they were markedly pycnotic. 



Chromatin nucleolar oocytes 

 Perinucleolar oocytes 

 D Cortical alveolar oocytes 

 Yolk granule oocytes 



30 



20 



10 



 Chromatin nucleolar oocytes 



E3 Perinucleolar oocytes 



fJJ Cortical alveolar oocytes 



B Yolk granule oocytes 



100 200 300 400 500 



Oocyte diameter (urn) 



Figure 2 



Percent frequency distributions for the oocyte diameters 

 of different oocyte stages in histological sections of stage- 

 VI ovaries of two female Rhabdosargus sarba. 



Influence of salinity and tides on spawning 



Both a and /5 atretic oocytes were frequently observed 

 in the ovaries of R. sarba. The chorion (zona radiata) 

 of the early a atretic vitellogenic oocyte was distorted, 

 fragmented, and had moved inwards (Fig 6A). By the /3 

 atretic stage, the yolk and lipid had been resorbed and 

 a large proportion of the oocyte volume was occupied by 

 vacuoles (Fig. 6B). 



Sixty-two percent and 72% of the stage-V and stage- 

 VI ovaries sectioned in July and August, respectively, 

 were at mid or late atretic state 1, i.e., 11-50% of their 

 yolk granule oocytes were a atretic (Fig. 1). However, 

 the prevalence of these mid-late state-1 ovaries declined 

 precipitously to 28% in September, as salinities rose 

 markedly, and remained at a similar level until the end 

 of spawning in late November. 



Histological sections showed that, in July and August, 

 only 39 r /c of the 57 pairs of ovarian lobes of R. sarba 

 that were macroscopically assigned as stage V and 

 stage VI contained migratory nucleus oocytes, hydrated 



oocytes, or POFs, i.e., were at stage VI. However, in the 

 following two months, 76% of the 88 pairs of ovarian 

 lobes of R. sarba, that were macroscopically assigned 

 as stage V or stage VI, were shown by histology to be 

 at stage VI. 



During September, when spawning activity was great- 

 est, the prevalence of spawning (PS) was positively 

 correlated (P<0.05) with maximum daily tidal height 

 (T). PS = 91.72T + 20.73 (/- 2 = 0.46, number of sampling 

 occasions=10) (Fig. 7A). 



Data for the same days as those used to provide the 

 points shown in Fig. 7A demonstrated that the preva- 

 lence of spawning (PS) is inversely correlated with the 

 difference in hours between the time when spawning is 

 believed to cease (ca. 06:00 h, see later) and the time 

 of high tide. PS = -8.26(T) + 78.22 (r 2 =0.49, number of 

 sampling occasions=10) (Fig. 7B). Thus, the prevalence 

 of these "spawning" females was greatest on those days 

 when the time that the tide was about to change from 

 flood to ebb coincided with the time when R. sarba is 

 considered to cease spawning. 



