616 



Fishery Bulletin 103(4) 



ratios at size indicating that males differentiated first. 

 This difference may be explained by the experimental 

 methods because the differentiation process was likely 

 similar between the cultured and wild fish. Chang et 

 al. 11995) showed that female development occurred 

 before male development based on levels of plasma sex 

 steroids. However, this finding was later corrected to 

 show that plasma sex steroid levels were the same for 

 males and females throughout sexual differentiation 

 (Chang et al., 1999). 



The formation of lobules with the proliferation of 

 germ tissue has been previously described as a male 

 developmental pattern (Stenger, 1959; Grier, 1981; 

 Grier and Taylor, 1998; Grier, 2000). The morphologi- 

 cal progression seen in the present study was similar 

 to that previously described in histological examina- 

 tions of differentiation in striped mullet in conjunction 

 with size (Stenger, 1959) and age (Chang et al., 1995). 

 However, ours is the first study to examine sexual dif- 

 ferentiation of both male and female striped mullet with 

 changes in size and age and to describe these changes 

 histologically. 



The undifferentiated gonad appeared to have male 

 morphological characteristics. The first morphological 

 signs of female differentiation were the movement of 

 deuterogonial germ-cell nests from the periphery of 

 the gonad. This pattern of development was similar to 

 the ontogeny of differentiation described previously for 

 striped mullet (Stenger, 1959). However, the presence 

 of the tissue stalk at the base of the suspensory tissue, 

 to which the ovary wall was attached, has not previ- 

 ously been described. The tissue stalk was present on 

 the majority (68%) of differentiating ovaries and only 

 a few (0.25%) of the differentiating testes. The pres- 

 ence of this stalk in differentiating testes indicated 

 that this characteristic alone should not be used to 

 determine genetic sex. However, the presence of the 

 tissue stalk, in addition to the rounded oogonial nests 

 throughout the gonad, strongly indicated that the speci- 

 men was female. There were no specimens observed to 

 be developing an ovary wall that also had developing 

 lobules or duct-work (male characters). Therefore, from 

 a morphological standpoint, the initial definitive identi- 

 fication of the differentiating ovary was the formation 

 of the ovary wall along with rounded germ cell nests 

 throughout the lobe. A primary duct at the center of 

 the developing ovary was present at this stage but any 

 secondary duct-work had begun to atrophy. It was also 

 observed that oogonial and oocyte proliferation could 

 occur throughout the lobe without a definitive ovary 

 wall, which would also be a strong indicator of the 

 female sex. 



Size and age at maturity 



Once sexual differentiation had occurred, the earliest 

 indication of spermatogenesis occurred at just under 

 250 mm (two specimens) and one year of age. However, 

 the majority of the developing specimens (89.9%) did 

 not show signs of spermatogenesis until they reached 



300 mm and age 2. The greater abundance of immature 

 males under 300 mm would also indicate that full matu- 

 rity was reached by this length. Almost all the males 

 over 325 mm were in some state of reproductive activity, 

 either developing or running, ripe, because most of these 

 larger males were captured only during the spawning 

 season. October, November, and December were the only 

 months when we saw these larger fish, except for some 

 atretic-stage specimens taken from freshwater during 

 the spring. The first signs of spermatogenesis for striped 

 mullet, both from eastern Florida (Stenger, 1959) and 

 South Carolina, were found in August. 



Greeley et al. (1987) did not age female striped mul- 

 let but used the growth schedule of Thomson (1966) to 

 conclude that striped mullet in eastern Florida reach 

 sexual maturity at 2.25 to 2.5 years, which is 1 to 2 

 years earlier than that previously reported (Jacot, 1920; 

 Broadhead, 1956; Anderson, 1958; Thomson, 1966). One 

 problem in earlier studies was the use of scale-based 

 age estimates (Jacot, 1920; Thomson, 1951, 1966; Ti- 

 moshek, 1973). The otoliths used in our study showed 

 more repeatability than would scales. Age schedules 

 based on scales were likely to contain problems with 

 the error terms and overestimation. Another factor may 

 have been the lack of a proper age-validaton protocol. 

 The lag in time between the actual birthdate and the 

 first increment formation was not incorporated into 

 the age model. A fish with a single annular ring that 

 appeared to be mature could actually have been up to 

 30 months old. Male striped mullet did not begin to 

 mature until they were one year old, and almost 100^ 

 had reached sexual maturity by age 3. Ripe and atretic 

 males were also found at age 1. 



Size at maturity for female striped mullet was re- 

 ported to be from 290 to 430 mm (Thomson, 1951, 1966; 

 Broadhead, 1956; Chubb et al., 1981). This wide range 

 in size at maturity depended on whether gonads were ex- 

 amined by gross morphological examination (Thomson, 

 1951, 1966; Broadhead, 1956) or histologically (Stenger, 

 1959; Chubb et al., 1981). Stenger (1959) found that oo- 

 cyte development occurred in specimens as small as 270 

 mm fork length (300 mm TL). Greeley et al. (1987) re- 

 ported the minimum size at maturity for female striped 

 mullet in eastern Florida was 230 mm SL (290 mm TL). 

 The minimum size at which a female was found to be 

 undergoing active vitellogenesis in the present study 

 was 291 mm. The first signs of female maturity were 

 evident in small numbers (15) in the 2-year-olds. The 

 first atretic females were also found at age 2. The age 

 at maturity for female striped mullet in our study was 

 similar to that found by Greeley et al. (1987) who used 

 length-based predicted ages. Therefore, it appears that 

 striped mullet in South Carolina have a similar matu- 

 rity schedule to those found in eastern Florida. 



Immature and inactive males and females were found 

 every month of the year. The presence of ripe males 

 from October through February and the presence of 

 developing females from August through March support 

 the idea of an extended spawning season from October 

 through April. 



