INTERTIDAL FEEDING AND REFUGING 

 BY CUNNERS, TAUTOGOLABRUS 

 ADSPERSUS (LABRIDAE). 



The cunner, Tautogolabrus adspersus, is the north- 

 ernmost representative of the wrasses (Labridae) in 

 the western North Atlantic (Bigelow and Schroeder 

 1953). While its food habits have been studied (011a 

 etal. 1975; Shumway and Stickney 1975), there is on- 

 ly one report of feeding behavior (011a et al. 1975). 

 During the summer of 1981, 1 studied the foraging 

 behavior of cunners in an intertidal habitat and found 

 a feeding pattern which may be an adaptation to 

 predators. 



Methods 



The study site was a rocky intertidal area at 

 Scituate, about 25 km south of Boston, Mass. 

 Boulder density ranged from 1 to 4 per m 2 in the area. 

 Observation sites were completely exposed at low 

 water and submerged to a depth of 1.8 m during high 

 tides. Water temperature was 15°-21°C during the 

 study period. 



From 5 July to 21 August 1981, 1 spent 57 h in un- 

 derwater observations, primarily at three refuges in 

 the intertidal zone where cunners were seen during 

 each high tide. On three occasions I followed the ini- 

 tial occupancy (first fish in) and final desertion (last 

 fish to leave) of the largest refuge during flood and 

 ebb tides, respectively. On 9 d I followed 69 feeding 

 forays by individual fish from the three refuges, 

 recording the estimated size of the fish, the duration 

 of the foray, the number of feeding acts per foray 

 (these consisted of pecks at items on the bottom or 

 rare rises to strike at objects in the water column), the 

 prey at which the feeding was directed, the maximum 

 distance ventured from the refuge on a foray, and 

 fidelity of return to the starting point. On two oc- 

 casions the same fish was followed on consecutive 

 forays (two and three forays each, respectively); 

 otherwise, each observation was made on different 

 fish. A 9 m X 9 m grid marked every 30 cm was laid 

 out over the bottom to help in determining the dis- 

 tance covered per foray and to aid in the estimation of 

 fish size. 



Other underwater observations included three high 

 and two low tide dives with scuba on the seaward side 

 of the intertidal zone in areas which remained per- 

 manently submerged (minimum depth 2 m), and two 

 100 m snorkel transects swum over the intertidal 

 area during the high tide for 9 d (18 transects total). 

 The dives were to determine qualitatively if a portion 

 of the cunner population stayed in the subtidal region 



throughout the tidal cycle (high tide dives were made 

 after confirming the refuges were occupied). The 

 subtidal habitat consisted of patches of small (20 cm 

 diameter), substrate-embedded rocks, interspersed 

 with sand bar areas. These rocks were typically 

 covered with Irish moss, Chondrus crispus. Larger 

 rocks (>2 m diameter) were scattered within the area 

 and provided shelter for cunners in hollows and mac- 

 roalgal (principally Laminaria sp.) growth. Water 

 depth was 2-6 m. The snorkel transects served to 

 identify other fish species, which moved into the in- 

 tertidal zone with high tide, especially potential pre- 

 dators of cunners. 



Available foods for cunners in the rock range were 

 measured by randomly placing a 0.25 m 2 hoop in the 

 area where the fish fed during low tide and by enu- 

 merating the animals found on the surface within 

 the hoop. 



Finally, the species and the number of potential 

 bird predators in the study area during high tide 

 were recorded. 



Results 



Cunner activity in the intertidal zone centered 

 about the refuges where 2-8 fish were seen at a given 

 time. The refuges were boulders from 0.5 to 1.0 m in 

 diameter with hollows underneath in which the cun- 

 ners remained when not foraging. Similar boulders 

 with hollows were present in the study area but were 

 not used as refuges. The only difference I noticed 

 between the occupied and unoccupied sites was that 

 the occupied refuges had two entrances while unoc- 

 cupied hollows had only one. Both entrances were 

 used in the occupied sites. 



Cunners moved into the refuges an average of 107.2 

 min(SE = 64; range 64-150; n = 17) before peak high 

 tide. Water depth at the time the refuges were first 

 occupied was 126 cm (SD = 5; n = 3). The fish ap- 

 proached singly from a seaward direction. Feeding 

 forays could start immediately, but up to 20 min 

 might pass between first occupancy and the observa- 

 tion of the first foraging activity. 



On the average, forays by 10-20 cm fish lasted 106 s, 

 included 6 feeding acts, and took fish 3.7 m from the 

 refuge (n = 69; Table 1). The cunners showed a 

 remarkable fidelity to their refuges. In 69 observed 



TABLE 1.— Feeding behavior of cunners in the intertidal zone. 



426 



FISHERY BULLETIN: VOL. 81. NO. 2, 1983. 



