Zeldis et al.: Development of Hoplostethus atlanticus eggs in the water column 



375 



of seawater (500 mL-min -1 ) filtered to 5 microns. 

 Water was cooled to 6°C by a refrigeration system, 

 then heated to desired culture temperatures by 

 aquarium heaters controlled with digital thermistors. 

 The 6°C inlet and heater for each bin were within a 

 central PVC "upwelling stem" driven by an airstone 

 and provided stable culture temperatures during the 

 study (±0.2°C). The jars and bins were disinfected 

 with weak ammonia about every three days. 



Culture temperatures used were 6, 8, 10, and 12°C, 

 which spanned the temperature range from the depth 

 of spawning to the sea surface in orange roughy 

 spawning areas. The eggs used for the 6°C culture 

 were fertilized from stripped fish (on voyage 

 TAN9306), and the eggs used for the 8, 10, and 12°C 

 cultures were captured from the plankton (on voy- 

 age TAN9206). Growing captured planktonic eggs 

 was generally more successful than starting with 

 stripped eggs and sperm. Only a few hundred strip- 

 fertilized eggs were successfully grown, as opposed 

 to thousands of captured eggs. Strip-fertilizing was 

 most successful when free-flowing, ovulating females 

 and freely expressible males were taken from the 

 wings of the trawl and when cooled seawater and 

 dilute sperm infusions were used with the eggs. The 

 eggs that were selected from the plankton samples 

 for culturing were those that floated in the plankton 

 catches. 



Eggs were staged by using the criteria and figures 

 for killifish Fundulus heteroclitus from New (1966), 

 in which most relevant features of orange roughy 

 development were identifiable, up to and including 



New's stage 25. Subsequent stages were defined by 

 the degree the tail was free of the yolk (as a propor- 

 tion of total tail length), and stage 30 was defined as 

 the hatching stage. Stage numbers used in this pa- 

 per are assumed to represent the midpoints of the 

 time periods of the stages. Eggs were sampled from 

 the jars and examined under a binocular microscope 

 every 3 or 4 hours at 6°C, and less frequently (see 

 Fig. 2) at 8, 10, and 12°C, until stage 9 (morula for- 

 mation). The 6°C eggs were sampled more frequently 

 because they were the only culture temperature used 

 on voyage TAN9306 and more time was available to 

 sample and observe eggs. After morula formation, 

 eggs from all temperatures were sampled once or 

 twice a day until stage 19 (blastopore closure) and 

 then once every 1 or 2 days until hatching was im- 

 minent. After examination, the sampled eggs were 

 either replaced in the culture vessels (if they were 

 rare) or more commonly preserved in 49c buffered 

 formaldehyde in seawater. 



Since the eggs used in the 8°C, 10°C, and 12°C se- 

 ries were caught from the plankton, their absolute 

 ages at the start of their series (at stage 5) were not 

 known. To estimate the absolute ages of the observed 

 stages, the cumulative time up to each stage since 

 stage 5 was regressed against the observed stage for 

 each series for each temperature. This meant all re- 

 gression lines effectively had a common x-axis inter- 

 cept at stage 5, with age = 0. The absolute age at 

 stage 5 for each temperature was estimated by con- 

 tinuing the lines to on the x-axis and taking the 

 absolute values of the resulting negative y-axis in- 



