Govoni et al.: Early life history of Xiph/as gladius in the western North Atlantic 



783 



Figure 4 



Photomicrographs of a 30.0-mm-PSL swordfish larva: (A) fundic stomach; (B) gastric glands in stomach 

 mucosa (Gg=gastric glands; M=mucosa; L=stomach lumen with larval fish remains; E=eye of eaten larval 

 fish). 



Discussion 



The first check on sagittae apparently corresponds with 

 the completion of yolk and oil globule absorption and the 

 beginning of feeding. Sanzo (1910) and Yasuda et al. (1978) 

 reported hatching 3 days after fertilization when larvae 

 were ~4 mm live total length (LTD, or 3.8 mm live stan- 

 dard length (scaled from their drawings), and complete yolk 

 and oil globule absorption 8 d after fertilization (or 5 DAH) 

 when larvae were -5 mm LTL, or 4.3 mm live standard 

 length. Larvae from the present material had completed 

 yolk and oil absorption between 3.8 and 3.9 mm PSL and 

 had the first check 3 increments after the core increment. 

 Temperature and feeding influence the growth rate of fish 

 larvae and their otoliths, but larvae are typically collected 

 in water 25 ±1''C (Arata, 1954; Taning, 1955; Tibbo and 

 Lauzier, 1969; Markle, 1975; Govoni et al., 2000), as were 

 the larvae collected for age determination. This tempera- 

 ture is common to the Gulf Stream and its progenitor cur- 

 rents (Schmitz et al., 1993; Hitchcock et al., 1994), and is 

 similar to the temperature used to rear larvae (Sanzo, 1910; 

 Yasuda et al., 1978). The difference in length at complete 

 yolk and oil absorption between Sanzo (1910) and Yasuda 

 et al. (1978) and the present collections probably owes to 

 shrinkage of larvae with death and preservation (e.g. Thei- 

 lacker, 1980). 



The second check follows concomitant changes in diet 

 and morphological features that take place between 8 and 

 13 mm PSL or from 7 to 11 DFH. An acceleration in so- 

 matic growth follows the second check within a day or so. 

 Young swordfish larvae eat copepods; older larvae other 

 larval fishes. The most striking morphological change of 

 larval swordfish is in the jaws. Swordfish larvae <13 mm 



SL have beak-like jaws that are typical of the larval scom- 

 broid fishes (Collette et al., 1984), particularly those of 

 the wahoo (Acanthocybium solandrl) and scaleless tuna 

 (Gymnosarda unicolor); older larvae develop bill-like jaws 

 with elongate rostral cartilages anterior of the premaxillar- 

 ies and equally elongate mandibles (McGowan, 1988). The 

 constitution of the alimentary canal changes as well. The 

 development of a functional stomach with gastric glands 

 in larval swordfish, which typically arises during the meta- 

 morphosis of fishes (Govoni et al., 1986a), is evident in the 

 larvae of other scombroid fishes where it is accompanied 

 similarly by a change in diet from zooplankton to fish (Kaji 

 et al., 1999; Shoji et al., 1999). A switch from zooplankton to 

 fish is common among istiophorid larvae, but it is neither 

 as exclusive nor abrupt (Voss, 1953; Lipskaya and Gorbu- 

 nova, 1977) as it is with swordfish. Accelerated growth af- 

 ter such a dietary shift is also a common trait of scombroid 

 larvae (Shoji etal, 1999). 



Swordfish larvae grow rapidly, faster than other larval 

 fishes with reportedly rapid growth. Growth rates reported 

 in the present study are for larvae that have survived pre- 

 dation and possibly variable feeding success; these rates 

 do not necessarily represent average larval growth of the 

 overall population. Growth rates of larvae >11 mm PSL 

 (13 DAH), 5.6 mm/d, are nonetheless faster than the lar- 

 val growth rates of other fast-growing larvae that survive 

 in the sea, e.g. sablefish (Anoplopoma fimbria) (Boehlert 

 and Yoklavich, 1985), and the oceanic-pelagic common dol- 

 phinfish (Coryphaena hippurus) reared in the laboratory 

 at high food densities without predation (Hassler and Hog- 

 arth, 1977; Kraul, 1991). The growth rate of larval sword- 

 fish <13 mm LJFL, 3 mm/d, is slower than the maximum 

 (16 mm LJFL/d) and sustained (10 mm LJFL/d) growth 



