The fish were collected at night with a light and a dip 

 net, and placed in 25 1 aquaria without aeration or 

 running seawater as soon as possible after collection. 

 The aquaria were located outdoors under an awning, 

 and therefore were exposed to the ambient diel light 

 cycle, but not to direct sunlight. The fish were 

 allowed to acclimatize for 12-24 h before treatment. 

 Usually there was mortality during this period, but 

 the proportion was not determined. All dead fish 

 were removed prior to treatment. 



The fish were exposed to 250 mg tetracycline/1 

 seawater for 12 h from sunset to sunrise, except in 

 experiment IVB when the immersion period was 

 from sunrise to sunset (Table 1). After an immersion 

 period, the aquarium was flushed with 90% water 

 changes until no visible color remained. The 

 tetracycline-seawater solution is yellow until ex- 

 posed to sunlight for more than ~3 h, when it turns 

 pink, due to oxidative photolysis. Following the treat- 

 ment, fish were maintained in clean seawater for 2-6 

 d by feeding either fresh wild plankton > 125 mm 

 diameter once a day (experiment I) or Artemia salina 

 nauplii 3-4 times/d (all subsequent experiments). 

 Artemia nauplii were more convenient for frequent 

 feedings than fresh wild plankton. Ninety percent of 

 the water in each aquarium was changed each morn- 

 ing by siphoning, to minimize handling the fish. Tank 

 water temperatures were measured over the diel 

 cycle during February 1982 (summer) at One Tree 

 Island. The temperature ranged from 25°C at 0630 h 

 to 30°C at 1800 h. Replacement water, added at 0700 

 h from the surface of the lagoon, measured 27°C. 



Larvae were killed at the end of each experiment by 

 placing them into 707c ethanol. Fish were subsequent- 

 ly measured to the nearest 0. 1 mm SL. Their otoliths 

 (both sagittae and lapilli) were removed and mounted 

 whole on glass slides without coverslips, using 

 Protexx. 



The following terms are used in this report for the 

 concentric rings seen in otoliths. A growth zone is a 

 wide ring which appears light or hyaline under 

 transmitted light. A discontinuous zone is the nar- 

 rower ring between two growth zones, often called the 

 opaque zone because it appears dark under trans- 

 mitted light. A growth increment, or simply an incre- 

 ment, is a growth zone plus a discontinuous zone. 



Otoliths were examined at 250-l,000X magnifica- 

 tion with a combination of incident ultraviolet light to 

 reveal the fluorescent tetracycline-marked rings, and 

 polarized transmitted light to count the rings. The 

 fluorescence microscope used ultraviolet light from a 

 50W mercury lamp. Excitation wavelength was 

 limited by a band pass filter (450-490 nm) and a long 

 pass suppression filter (515 nm). 



In most cases, one sagitta from each fish was read, 

 although occasionally the lapillus was used if its rings 

 were clearer. The area to be counted was selected by 

 scanning the margin of each otolith to find the place 

 where the greatest number of distinct rings could be 

 seen between the innermost fluorescent increment 

 and the edge. A datum was considered valid only if 

 identical counts were obtained in at least two out of 

 three blind readings. No other otoliths were con- 

 sidered in the analysis. Of 21 H. tropicalis otoliths 



Hypoathenna tropicalis 

 IA 



IB 

 MA 

 IIB 



IMA 



Total 



2( 1 ) 



4 



6( 2 1) 



6 



3 



21| 2 1) 



Spratelloides delicatulus 

 IIIA 6 



1MB 

 IVA 

 IVB 

 Total 



5( 2 2) 

 9 



9( 2 1) 

 29( 2 3) 



14.0 (13.6-14.4) 

 13.7 (12.8-14.7) 

 20.5 (16.2-27.2) 

 18.9(16.8-20.7) 



16.1 (15.4-17.2) 



17.5 (15.5-19 .1) 

 17 9(17.6-18.2) 



19 9(18.8-22.8) 



20 5 (17.9-22 9) 



2130, 8 July to 



0830. 9 July 1980 

 2130, 8 July to 



0830. 9 July 1980 

 1830. 31 Oct. to 



0630. 1 Nov. 1980 

 1830. 31 Oct. to 



0630, 1 Nov 1980 

 2000, 6 Nov to 



0700. 7 Nov 1981 



2000. 6 Nov to 

 0700. 7 Nov. 1981 



2000, 6 Nov. to 

 0700. 7 Nov. 1981 



1800. 31 Jan to 

 0630. 1 Feb. 1982 



0600 to 1 800 

 31 Jun 1982 



0830, 12 July 

 1730, 14 July 

 0730, 6 Nov. 

 0600, 7 Nov. 

 0545, 12 Nov. 



0545, 12 Nov. 

 1800, 9 Nov. 

 1 800. 6 Feb. 

 0715, 6 Feb 



2+1 



5 



4+1 



5+1 



4+1 



4+1 

 2 

 5 

 4+1 



1 



2 

 5 

 6 

 3 



17 



5 3 

 2 6 



7 12 



'Otoliths of two treated fish were destroyed by poor preservation. 



2 Number of fish discarded because of inconsistency between otolith readings. 



238 



