I Govoni et a\ Distribution of larval Xiphias gladius off thie soutfieastern United States 



67 



Figure 2 



The position of the Gulf Stream (diagonal line pattern) illustrated from 

 a 3-d average of sea-surface temperatures compiled on 20 June 1984 

 with data from infrared sensors on a Global Stationary Environmen- 

 tal Satellite operated by the National Weather Service (closed circles 

 denote approximate locations of multiple neuston collections where 

 swordfish larvae were present; open circles, locations where swordfish 

 larvae were absent; fractions are the number of positive collections 

 over the total number of collections). 



Length of larvae 



The overall distribution of the length of swordfish lar- 

 vae in the Caribbean Sea, Gulf of Mexico, and off the 

 southeastern Atlantic coast of the United States pro- 

 vided a view of the occurrence of the smallest larvae. 

 Because most published have records reported either 

 standard length (SL) or total length (TL), both mea- 

 sures are reported in our study. We employed no con- 

 version, although the length of the caudal finfold or 

 fin ranges from 3% to 12% of TL for larvae from 6 to 

 192 mm (calculated from Arata [1954]). We employed 

 no correction for the shrinkage of larvae due to death 

 or preservation, because preservation fluid was not 

 routinely reported in the literature and some speci- 

 mens were measured alive. Although the length from 

 the anterior eye orbit to the tip of the notochord is per- 



haps a better measure of length in swordfish larvae, 

 because the rostrum is often broken (Potthoff and Kel- 

 ley, 1982), this measure was not uniformly available 

 in published records. Standard or total lengths (ante- 

 rior tip of the upper jaw to the posterior tip of the noto- 

 chord or hypural plate (SL) or caudal fin (TL)) of sword- 

 fish larvae and locations of collection were taken ft-om 

 Arata ( 1954), Arnold ( 1955), Tibbo and Lauzier (1969), 

 Markle (1974), Post et al. (1997) and MARMAP. Stan- 

 dard length was measured on larvae from SEAMAP, 

 CF8408, CH8807, and CH9703 (ft-om neuston as well 

 as plankton collections taken within the Gulf Stream 

 fi-ontal zone with a MOCNESS (Wiebe, et al., 1976) 

 system with 505 m mesh nets), and on a larva fi"om 

 a neuston collection taken in a Sargassum raft at 

 33°38.7'N/076°02.7'W in 1991. Specimens were consid- 

 ered larvae if they were <190 mm SL or TL, because 



