FISHERY BULLETIN: VOL. 76, NO. 2 



(drifted) with the tides. The break up of schools 

 may have been a result of reduced predation 

 and/or lowered visual sensitivity thresholds 

 (Munz and MacFarland 1973). 



When a school was attacked, it usually split into 

 two or more segments and passed around behind 

 the predator to reform a single school again. When 

 a predator was successful in separating an indi- 

 vidual from a school, a chase occurred, the results 

 of which were seldom observed. Of the approxi- 

 mately 50 lizardfish stomach contents analyzed, 

 one contained a juvenile mullet. None of the 10 

 barracuda stomach contents analyzed contained 

 juvenile mullet. 



Potential invertebrate predators were abun- 

 dant in the various habitats where mullet occur- 

 red. However, only individuals of a single crab 

 species, Thalamita crenata, were observed stalk- 

 ing and extending their chelipeds toward passing 

 mullet. In one instance an individual crab did cap- 

 ture a juvenile mullet, but only after it had been 

 wounded by and escaped from a barracuda. 



DISCUSSION 



Mugil cephalus is a worldwide (lat. 42°N-42°S, 

 Thomson 1966) inhabitant of the estuarine inter- 

 tidal as well as freshwater and coastal marine 

 environments (Broadhead 1953, 1955; Hendricks 

 1961; Thomson 1963, 1966; Johnson and McClen- 

 don 1970). In Hawaii, selective pressures appear 

 to have favored prejuvenile and juvenile mullet 

 that are able to survive in the shallowest, warmest 

 estuarine intertidal waters, waters that are 

 characterized by temporal and spatial hetero- 

 geneity with respect to temperature, salinity, and 

 depth. Before discussing the adaptations evolved 

 by striped mullet making possible survival in es- 

 tuarine intertidal regions, a discussion of the en- 

 vironmental variables important to young mullet 

 in Hawaii might be in order. 



The monthly occurrence of mullet <50 mm SL 

 observed in 1972 and 1973 in Hawaii is presented 

 with data for 12 consecutive years (1962-73) of 

 recorded (skycover, rainfall, seawater tempera- 

 ture) and predicted (tidal) data in Figure 4. These 

 appear to be the most important environmental 

 factors that bear directly upon the lives of mullet 

 in the estuarine intertidal region. Indirectly, the 

 length of daylight (time from sunrise to sunset) 

 may also be important; it is shortest (about 10.9 h) 

 about 22 December each year, and longest (13.3 h) 

 about 21 June each year. 



Visual observations and collections of mullet 

 < 50 mm SL indicate that these mullet occur in the 

 Hawaiian intertidal estuarine regions during the 

 months when there are a maximum number of low 

 tides ^0.0 m (mean tide level at Honolulu is 0.2 m 

 (0.8 ft)). Perhaps of greater importance is the 

 occurrence of mullet when there is a minimum 

 number of high tides ^0.6 m (2.0 ft). In Maunalua 

 Bay, tide pools begin to form when the tide level is 

 approximately 0.06 m (0.2 ft). The number of tides 

 that would result in tide pool formation at noon 

 ( 1000-1400 h local time) begins to decrease during 

 the time of the year mullet are undergoing 

 metamorphosis in the intertidal estuarine region, 

 but is still maximal when prejuveniles first enter 

 the inshore areas. Thermal and salinity stresses 

 should be maximal during the noon time period. It 

 is not known whether this tidal-estuarine interti- 

 dal situation is unique to Hawaii or of more wide 

 spread occurrence. Also unknown is whether the 

 peak occurrence of young mullet during such tidal 

 relationships is fortuitous, or whether selection 

 pressures have resulted in a shift of the peak oc- 

 currence from either earlier or later in the 

 winter-spring season to its present "position" in 

 April. 



The extent to which stress occurs in the interti- 

 dal estuarine region may be ameliorated by low 

 ambient (oceanic) seawater temperatures and 

 maximum cloud cover. During the time mullet 

 <50 mm SL are found in the intertidal estuarine 

 region in Hawaii, seawater temperatures are min- 

 imal and increasing, and average cloud cover is 

 seasonally maximal. During the late winter- 

 spring, the lowest seasonal seawater tempera- 

 tures occur in the tropical-temperate Northern 

 Hemisphere, and increase until maximum levels 

 are reached about September. 



The average maximum amount of cloud cover, 

 which occurs in Hawaii during April (when young 



Figure 4.— The relative abundance of mullet <50 mm SL in the 

 Hawaiian estuarine intertidal tide pools compared with environ- 

 mental data collected (seawater temperature, rainfall, and sky 

 cover) and predicted (tides). The monthly means (cormected by 

 horizontal lines), ranges (vertical lines), and standard deviations 

 (vertical boxes) were derived from data for the 12-yr period, 

 1962-73. Mullet abundance data were from field observations 

 and collections in 1972-73. Sky cover data were derived from 

 monthly average values. Sky cover and rainfall data were taken 

 from Climatological Data, Hawaii, U.S. Weather Bureau, 

 NOAA; tidal information from Tide Tables, West Coast of North 

 and South America including the Hawaiian Islands, National 

 Ocean Survey, NOAA; seawater temperature data were col- 

 lected by the National Marine Fisheries Service, Honolulu. 



308 



