Jordan and Bruce: Larval development of three roughy species 



85 



Figure 5 



Development of dermal spination on scale 

 of Opticus sp.: (A) 8.0mm, (B) 10.2mm. 



ing aggregations have been located 

 (Beardsell 1984). 



The 26mm Hoplostethus atlanticus 

 specimen, caught in a demersal trawl at 

 400-950 m off St. Patricks Head, east- 

 ern Tasmania (CSIRO H1141), repre- 

 sents the smallest H. atlanticus reported 

 to date (Fig. 6). Kotlyar ( 1984 1 previ- 

 ously recorded a 36 mm H. atlanticus 

 taken by bottom trawl in the Atlantic 

 Ocean at a depth of 965-990 m. The 



26 mm juvenile has characters that are common to the other 

 trachichthyids identified, such as a deep body, large mouth, cranial 

 ridges and opercular spination, heavily-pigmented gut and pelvic 

 fins, and distinct anal- and dorsal-ray bases. It is highly likely that 

 such characters are retained in the larvae of H. atlanticus, and 

 Hoplostethus larvae in general. 



Several scenarios may explain why Hoplostethus atlanticus lar- 

 vae have not yet been located. 



1 Bimonthly sampling frequency is too coarse to capture larvae 

 during their pelagic stage. Although this cannot be discounted due 

 to the lack of information on larval duration, the likelihood of com- 

 pletely missing all larvae seems low. 



2 Larvae occur further off the shelf or slope than sampled (>18km 

 from the shelf break). 



3 Larvae may occur in greater depths than those sampled in 

 'standard' ichthyoplankton surveys. This may be the most reason- 

 able scenario, and has previously been suggested by Kotlyar (1984) 

 for Hoplostethus less than 15-19 mm (based on the capture of three 

 H. melanopterus juveniles 15.0-18.2 mm in Isaacs Kid trawls at 

 1000-1500 m in the Sulu Sea). Larvae may also occur close to the 

 bottom on the continental slope, as supported by the capture of our 

 26 mm specimen and that by Kotlyar ( 1984). 



Hoplostethus atlanticus adults occur in depths of 500-1200 m (May 

 & Maxwell 1986 ). Spawning has been confirmed from both the east 

 coast of Tasmania (Lyle et al. 1989) and New South Wales (Williams 

 1989). Hoplostethus atlanticus eggs (2. 12-2.45 mm in diameter 

 with a conspicuous orange oil droplet) have been collected in plank- 

 ton tows above 400m in Tasmanian (Lyle et al. 1989) and New 

 Zealand waters (Beardsell 1984), and are presumably bouyant. The 

 lack of larvae in surface waters, however, suggests that egg density, 

 the presence of a thermocline, or a combination of both factors 

 may confine eggs and larvae to deep water. Although deep-water 

 plankton sampling is logistically more difficult than shallow- 

 water sampling, such sampling should be carried out if we are to 

 fully understand the early life histories of certain deep-water 

 species. 



Figure 6 



Hoplostethus atlanticus juvenile, 26 mm. 



Acknowledgments 



We are grateful to R. Thresher, 

 J. Gunn, J. Leis, J. Paxton, 

 P. Last, and two anonymous re- 

 viewers who made many useful 

 suggestions for improving 

 the manuscript. We thank A. 

 Miskiewicz and A. Steffe for 

 supplying additional specimens. 

 This work was supported in 

 part by FIRDC research grant 

 87/129. 



