HOBSON and CHESS: TROPHIC INTERACTIONS 



dividuals collected later judged to be the largest 

 and smallest in the group). In the moderate cur- 

 rent, with the giant kelp lying over at about 25° 

 (attempts to measure the current in this habitat 

 proved unsatisfactory owing to complex eddy 

 systems), 10 individuals (selected haphazardly) 

 plucked at plankters 50 to 73 {x = 58) times during 

 1-min periods. One hour later, when there was no 

 discernible current, 10 individuals in a similar, if 

 not the same aggregation, each plucked at 

 plankters 30 to 51 (.f = 39) times during 1-min 

 periods. Probably there is an optimum current 

 speed beyond which the fish find the increasing 

 diflSculty of maintaining station outweighs the 

 advantage of added food. We lack data on feeding 

 rates, but have noted that in strong currents 

 blacksmiths abandon the open places within the 

 forest, where they had been dispersed and feed- 

 ing, and concentrate in dense numbers close in the 

 lee of individual kelp columns. 



Of 41 adults (92-145 mm, .r = 118) collected from 

 aggregations in the water column throughout the 

 study area during the afternoons, 36 were full of 

 food, much of it fresh. The other 5, all collected 

 during midafternoon along the margin of the 

 forest bordering the inshore edge of the Dictijop- 

 teris field, were empty. All prey taken by these 

 blacksmiths, ranked below, are forms we have 

 collected in the water column during the day. 



1. LARVACEANS (100: 448.1: 57.5) 



most of them Oikopleura spp. 



2. CALANOID AND CYCLOPOID COPEPODS (100: 256.3: 



33.7) 



calanoids, including Calanus pacificus, Acartia tonsa, 

 Labidocera sp., and Rhincalanus nasutits (100: 2.53.6: 32.4); 

 cyclopoids, including Conjcaeus sp., Oncaea sp., and Oith- 

 ona sp. (72: 2.7: 1.3). 



3. FISH EGGS (69: 17.9: 4.1) 



unidentified. 



4. CLADOCERANS (75: 24.6: 2.5) 



Evadne spp. 

 .5. CARIDEAN LARVAE (33: 1.4: 0.9) 

 unidentified. 



6. EUPHAUSID LARVAE (33: .5.1: 0.6) 



calyptopis stage. 



7. CIRRIPEDIAN LARVAE (33: 3.5: 0.6) 



cypris stage. 



8. BRYOZOAN LARVAE (17: 1.2: 0.2) 



cyphonautes. 



9. CHAETOGNATHS(3:0.1:0.5) 



unidentified. 



10. REPTANTIAN ZOEA (3: 0.4: 0.1) 



unidentified. 



11. HARPACTICOID COPEPODS (3: <0.1: <0.1) 



Microsetella sp. A. 



12. FISHES (3: <0.1:<0.1) 



unidentified larvae. 



13. ISOPODS(3:<:0.1:<0.1) 

 gnathiid juvenile. 



In feeding so heavily on larvaceans, their major 

 prey, these adult blacksmiths differ from other 

 species treated in this report. Significantly, how- 

 ever, larvaceans are also major prey of other 

 species of Chromis elsewhere, e.g., in Hawaii 

 (Hobson 1974) and in the West Indies (Randall 

 1967). Probably larvaceans are important food of 

 the blacksmith throughout its range, even though 

 they have gone unreported in previous food-habit 

 studies of this species. Larvaceans are difficult to 

 recognize, especially if digestion is advanced or 

 preservation faulty, and this may account for 

 them going unreported. 



Because juvenile blacksmiths were in feeding 

 aggregations distinct from those of the adults, we 

 analyzed their gut contents separately. Of 14 

 juveniles (16-47 mm, .f = 34) collected from ag- 

 gregations during the afternoon, all were full of 

 food, much of it fresh. All prey, ranked below, are 

 forms that we have collected from the water 

 column during the day. 



1. CALANOID AND CYCLOPOID COPEPODS (100: 394.4: 



.54.1) 



calanoids, including Acartia tonxa (100: .366.2: 50.5); cy- 

 clopoids, including Corijcaeua sp. and Oncaea sp. (93: 28.2: 

 3.6). 



2. LARVACEANS (93: 48.9: 26.4) 



most of them Oikopleura spp. 



3. CLADOCERANS (100: 108.8: 12.5) 



Evadne spp. 



4. CIRRIPEDIAN LARVAE (79: 63.6: 4.6) 



cypris stage. 



5. BRYOZOAN LARVAE (79: 63.6: 4.6) 



cyphonautes. 



6. HARPACTICOID COPEPODS (.57: 6.4: 0.8) 



MicroKetella sp. A. 



7. FISH EGGS (43: 2.1: 0.7) 



unidentified. 



Diff"erences in the diet between juvenile and 

 adult blacksmiths can be related to the sizes of the 

 various organisms in the water column. Most prey 

 of the juveniles were less than 0.5 mm long. 

 Compared to the prey of adults these included 

 more cladocerans, copepods, and larvae of barna- 

 cles and bryozoans, but fewer larvaceans and fish 

 eggs (there were no larvaceans in the smallest 

 blacksmith, 16 mm long, and no fish eggs in all six 

 <34 mm). 



During the day, the heaviest concentrations of 

 adult blacksmiths in the vicinity of the study area 

 were at the mouth of the cove seaward of the kelp 



593 



