FISHERY BULLETIN: VOL. 75, NO. 2 



"lagoon" adjacent to the Hawaii Institute of 

 Marine Biology (HIMB) on Coconut Island. The 

 northern edge of the reef adjacent to Lilipuna Pier 

 drops abruptly into a 3- to 10-m deep channel, 

 while the southern side is adjacent to the shore. 



The reef and channel area near Lilipuna Pier 

 are open to the effects of wind and waves within 

 Kaneohe Bay throughout the year. Occasionally, 

 the winds abate or shift and the bay's surface 

 becomes calm and glassy. The observations 

 reported here could only be made at such times 

 when the estimated wind velocity was less than 

 2.6 m/s (5 knots). At night near the end of the pier 

 a fixed low intensity incandescent light bulb casts 

 an arc of light out over a small area in the channel. 

 Observations were not made within the area 

 encompassing this arc of light. The waters in the 

 HIMB lagoon are usually calm or only slightly 

 rippled, being protected by a vegetation covered 

 coral rubble peninsula on its normally windward 

 side and thicker, higher, vegetation on its island 

 or leeward side. 



Kaneohe Bay is rimmed at approximately 1.6 

 km inland by mountains that rise to 762-960 m. 

 Throughout each day, dense clouds usually form 

 along these mountains, occluding the sun during 

 the late afternoon. This often results in twilight 

 conditions occurring earlier than would normally 

 be predicted for the bay's position of latitude and 

 longitude. 



METHODS 



The prey species of fish observed in this study 

 was P. insularum, approximately 20-60 mm SL 

 and approximately 0.03-2.45 g wet weight. Obser- 

 vations of the silverside's behavior were made 

 during calm periods in October (7 days) and 

 November (3 days) 1972 and September (5 days) 

 1973. All observations were made visually from a 

 height of 0-3 m above the surface of the water. The 

 morning observations commenced approximately 

 115 min prior to the time of sunrise. The evening 

 observation period terminated about 60 min after 

 the time of sunset. 



The only attribute monitored quantitatively 

 during the course of the observations was the 

 jumping escape behavior of the silversides in 

 response to attacking predatory fishes. Enumerat- 

 ing the jumps became a shorthand method of 

 quantifying the number of predatory attacks in 

 the calm areas studied because jumping was 



observed to be the primary means of escaping 

 predators once an attack occurred. Pranesus 

 insularum was the only prey species observed to 

 jump in the above areas during the periods of this 

 study. The success of predators at capturing prey 

 during the attacks was not determined. Hobson 

 (1968) used a similar method to quantify the 

 number of times leaping predatory cabrilla, 

 Mycteroperca rosacea, attacked flatiron herring, 

 Harengula thrissina, in the Gulf of California. 



During periods of darkness or reduced light, 

 when visual observations under existing light 

 were not possible, jumping by schools of prey could 

 be heard within the areas studied by careful 

 listening; this could only be done when there was 

 no wind and the surface of the water was calm. 

 The time at which schools broke up or reformed 

 during twilight was estimated by listening to 

 changes in the sound of jumps made by multiple 

 and single prey close by, or with a flashlight beam 

 which was quickly turned on and off in one spot, or 

 swept rapidly across the surface of the water from 

 above, and/or held underwater within 0.3 m of the 

 surface. Whether the silversides were schooling or 

 spread out could be readily determined when the 

 fish were illuminated by the beam of light. 



Light measurements were made above the 

 surface of the water with a photometer ( Weston 

 Ranger 9 universal exposure meter). 3 Readings 

 taken with this photometer were compared with 

 those made with a Gossen foot-candle meter and a 

 Spectra-Combi 5000 Model photometer (Photo 

 Research, Burbank, Calif.). The readings ob- 

 tained during twilight periods were comparable to 

 those given by Brown (1952). 



The observations and events reported here are 

 related to the time of sunrise, sunset, and the 

 periods of morning and evening civil and nautical 

 twilight. The two periods of twilight are defined by 

 the angular distance of the sun below the horizon, 

 0° to -6° for civil twilight, and -6° to -12° for 

 nautical twilight. Fish respond directly to the 

 amount and type of light present, which is 

 influenced by astronomical as well as local 

 environmental conditions. However, the use of 

 these terms and that of the corresponding angular 

 distance of the sun below the horizon is of 

 immense value when comparing the observations 

 of many investigators working in different loca- 



•'Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



416 



