586 



Fishery Bulletin 101(3) 



ing March-April. In total, the trimodal oocyte 

 frequency represents a reservoir of oocytes prior 

 to FOM, one batch of oocytes beginning FOM, 

 and one batch completing FOM. Hydrated oo- 

 cytes were not observed in balao prior to 1 100 h. 

 However, in several ballyhoo collected around 

 dawn (i.e. at approximately 0600-0700 EST), 

 the nucleus in oocytes of the advanced batch 

 was still visible along the chorion but the cy- 

 toplasm was lightening in color. This suggested 

 that initiation of hydration at daybreak briefly 

 preceded nucleus breakdown. During the fol- 

 lowing 12-hour period, oocytes in this maturing 

 clutch advanced from late nucleus migration to 

 nucleus breakdown, and increased in diameter 

 from 1.5-2.0 mm in the morning to 2.0-3.0 mm 

 in the afternoon. Modal egg size for each of three 

 running-ripe ballyhoo (i.e. females with hydrat- 

 ed, ovulated eggs in the ovarian lumen) was 

 2.35, 2.60, and 2.80 mm diameter The complete 

 size range of these ovulated eggs was 2.2-3.4 

 mm diameter These females were collected at 

 or just before sunset (time: 1810-1855). Most 

 efforts to sample across the full 24-hour cycle 

 failed, apparently because halfbeak do not bite 

 hooks after sundown. A sample of 12 ballyhoo 

 was collected one night, however, by randomly 

 throwing a cast net on dense schools of fish. 

 These fish, collected between 2200 and 2359 

 hours during March 1997, all appeared to have 

 recently spawned. Histological preparations 

 demonstrated that they had fresh postovulatory 

 follicles, and they contained a distinct clutch of 

 oocytes in early nucleus migration. Whole oo- 

 cytes from these fish collected at night were not 

 archived in formalin; therefore they were not 

 measured for comparisons to whole oocytes col- 

 lected at other times during the diel cycle. These 

 patterns of diel reproductive periodicity also ap- 

 peared to apply to balao, but the available data 

 were not conclusive. 



Spawning habitat 





o 



a 



-Ballyhoo, n=1,791 



Female GSI 



Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct 

 1997 Month 1998 



Figure 2 



Mean (±95'^ confidence limits) gonadosomatic indices by month for bal- 

 lyhoo (Hemiramphus brasiUensis) and balao (H. balao) females. Values 

 are calculated for fish larger than size at 50% maturity, reported by 

 McBride andThurman (2003) as >198 mm fork length (PL) for ballyhoo 

 and >160 mm FL for balao. n = number of females. 



Perinucleolar Cortical Vitellogenic Nucleus Hydration 

 154/7 Alveolar 24/8 Migration -|20/7 



159/24 360/57 



Most advanced ooctye stage 



Figure 3 



Mean (±95% confidence limits) gonadosomatic indices of female bally- 

 hoo (Hemiramphus brasiUensis) and balao (H. balao) in various stages 

 of oocyte development. A two-way ANOVA demonstrated a significant 

 effect of both developmental stage and species (P<0.0001 ) on gonadoso- 

 matic index. Numbers indicate the number of fish, by stage, for each 

 species. Hydration = nucleus breakdown. 



In our study it was shown that hydrated oocytes 

 can be inferred from a threshold criterion of GSI >6.0 

 (Fig. 3), and ripe females (i.e. with a batch of hydrated 

 oocytes) will spawn within hours. Ripe ballyhoo females 

 were distributed throughout the fishing grounds in both 

 the Atlantic Ocean and Florida Bay (Fig. IB). In the Atlan- 

 tic, ripe ballyhoo females were caught in water depths from 

 1 to 20 m (mode: 6-10 m, 36.3'7( of the sets containing ripe 

 ballyhoo in the Atlantic Ocean). In Florida Bay, ripe bal- 

 lyhoo females were caught in areas that were exposed at 

 low tide and out to 6-m deep (mode: 2-4 m; 57.9'7f of the 

 positive sets in Florida Bay). Ripe ballyhoo females were 

 mainly associated with hard bottom or vegetated habitats 

 in both areas. In the Atlantic Ocean, ripe ballyhoo females 

 were collected above platform reefs in 51.79i of the sets, 

 above seagrass beds in 37.9% of the sets, near patch reefs 



in 5.2% of the sets, and over bare substrate in 5.2% of the 

 sets. In Florida Bay, these fish were also associated with 

 hard bottom substrates, specifically with vegetated bank 

 habitat, in 44.77r of the sets and with seagrass beds in 

 55.3% of the sets. 



Ripe balao females were distributed throughout the 

 Atlantic fishing grounds but not in Florida Bay (Fig. IC). 

 In the Atlantic, they tended to occur in deeper water than 

 did ripe ballyhoo females (range: 2-20 m; mode: 10-20 m, 

 51.37f of the sets containing ripe balao). The habitat as- 

 sociations of ripe balao females were similar to those of 

 ripe ballyhoo females in the Atlantic Ocean, but typically 

 reflected areas offshore rather than inshore of the reef 

 In the Atlantic Ocean, ripe balao females were collected 

 above platform reefs in 58.0% of sets, above seagrass beds 



