HOBSON: ACTIVITY OF HAWAIIAN REEF FISHES 



the different species, as relating to the activity- 

 described in the present report. This informa- 

 tion is important because the size of a given fish 

 is reflected in its behavior during twilight, as 

 discussed below. 



A number of species prominent on Hawaiian 

 reefs are not mentioned in this report. My cov- 

 erage is confined to species in which I recognized 

 behavior relating specifically to the twilight 

 transition period on the three reefs where twi- 

 light behavior was studied. Some other species 

 not prominent on these reefs are abundant else- 

 where and were studied during other phases of 

 the overall project. These other segments of 

 the work in Kona yielded no data inconsistent 

 with what is reported or discussed below. 



STUDY LOCATIONS 



The reef habitat on the Kona coast is restrict- 

 ed to a narrow shelf close to shore. From the 

 base of a rough, basalt shore cliff, the sea floor 

 in most locations slopes gently downward for 

 distances of between 50 anda600 m from shore. 

 At this point, where the water is generally about 

 20 to 25 m deep, the bottom drops abruptly to 

 great depths. 



Most of the data for this report were collected 

 from three inshore locations, all at depths be- 

 tween 5 and 10 m. Each location has a distinctly 

 different substrate. Station 1 is in Kealakekua 

 Bay among massive heads of the coral Porites 

 pukoensis rising 2 to 3 m above the sea floor. 

 Station 2 is in Honaunau Bay among an exten- 

 sive field of the coral Porites compressus grow- 

 ing in fingerlike branches 10 to 15 mm wide. 

 Station 3 is also in Honaunau Bay, but among 

 massive basalt boulders 2 to 3 m across that are 

 largely overgrown with encrusting algae and 

 corals. These boulders are variably interspersed 

 with the forms of both P. pukoensis and P. com- 

 pressus described above. 



The remarkably constant weather and water 

 conditions that characterize the Kona coast held 

 true throughout the 15 months of observations. 

 Consequently, variables inherent in changing en- 

 vironmental conditions were minimized, making 

 it easier to recognize activity patterns associ- 

 ated with the twilight transition periods. 



TRANSITION FROM DAY TO NIGHT 



It is difficult to determine exactly when the 

 daytime situation begins to move toward that 

 prevailing after dark. In large part, this diffi- 

 culty reflects variations in activity of many di- 

 urnal fishes because of differing water trans- 

 parencies, and even more when variable cloud 

 cover causes light levels to fluctuate. These var- 

 iations are difficult to distinguish from those 

 associated with the lesser light levels of the ad- 

 vancing afternoon. Additionally, as the day 

 progresses, activity of diurnal fishes changes 

 subtly in many other ways that may or may not 

 relate to a transition to the nocturnal mode. 

 The situation prevailing throughout the day 

 (Figure 1), including variations with changing 

 light levels and other factors, is described else- 

 where (Hobson, in preparation). This report 

 is concerned with the more striking transfor- 

 mations that occur during the transition between 

 daylight and darkness, beginning about 15 min 

 before sunset and lasting until about 45 min 

 after sunset — a span of about 1 hr. Three major 

 periods are readily recognized: 1) the cover- 

 seeking of the diurnal fishes; 2) an interim pe- 

 riod; and 3) the mass emergence of nocturnal 

 species. 



COVER-SEEKING OF THE DIURNAL FISHES 



Initial Overt Phase of the Transition 



An early cue to the developing transition lies 

 with the plankton-feeding fishes that swim dur- 

 ing the day in stationary aggregations up in 

 the water column. Prominent among these are 

 the damselfishes Ahudefduf abdominalis (142: 

 105-162 mm), Chromisleucurus (57: 37-70 mm), 

 C. ovalis (124: 121-138 mm), C. vanderbilti (38: 

 17-46 mm), and Dascyllus albisella (79: 42-95 

 mm) . The distance of each fish from the sub- 

 strate is related to its size, because the larger 

 individuals swim at higher levels in the water 

 column. These fishes descend progressively 

 closer to the sea floor as light diminishes. Mem- 

 bers of the smallest species, Chromis vanderbilti, 

 are the first to take cover; they seldom move 

 more than a meter above the reef on even the 



717 



