sign in sign up
<<
Home , Buldir Island, Delarof Islands, Neocalanus cristatus, Neocalanus plumchrus

The Condor91165-72 0 TheCooper Ornithological Society 1989

FOOD HABITS OF THE AT , ’

ROBERT H. DAY Institute of Marine Sciences, 200 ONeill’ Building, University of Alaska, Fairbanks, AK 997751080

G. VERNON BYRD Alaska Maritime National Wildrife Refuge, Unit, P.O. Box 5251, Adak, AK 98791

Abstract. We studied the food habits of Whiskered Auklets ( pygmaea) at Buldir Island, western Aleutian Islands, Alaska, during the summer of 1976. At Buldir, thesebirds fed primarily in and near convergent tidal fronts in the passesbetween three islets just offshore; feeding occurred throughout the day. Whiskered Auklets ate at least 10 different speciesof zooplanktonic prey, of which the plumchrus was the most important. Next in importance were chaetognaths(probably Sagitta elegans), the amphipod Parathemisto pacifica, megalopae of the crab Erimacrus isenbeckii, and an unidentified amphipod. Pteropods (probably Limacina helicina), larval cephalopods,and a larval fish occurred in trace amounts. During most of the summer, Whiskered Auklets were nearly monophagouson N. plumchrus; we do not consider a difference in prey during late incu- bation/early chick rearing to representa biologically-significanttrend. We collected Least (A. pusilla), Crested (A. cristatella), and Parakeet (Cyclorrhynchus psittacula) auklets during late chick rearing, to compare their food habits with those of Whiskered Auklets. At this time, both Least and Whiskered auklets were monophagouson smaller N. plumchrus, whereas Crested and Parakeet auklets mainly ate larger N. cristatus and P. pacifica; Parakeet Auklets ate the widest diversity of zooplankton. Selection of prey by the four auklet speciesappeared to occur in relation to size, with the smaller auklets eating smaller zooplankters. Key words: WhiskeredAuklet;food habits: oceanic zooplankton; tidalfronts; ; ; Parakeet Auklet; competition.

INTRODUCTION 7 g; range = 102-138 g; n = 60), is slightly larger The Whiskered Auklet (Aethia pygmaea) is one than the Least Auklet and considerably smaller of four plankton-feeding auklet speciesthat are than Crested and Parakeet auklets. confined to the Bering Seaand the Seaof Okhotsk Although the four auklet speciesexhibit broad (AOU 1983). This species exhibits broad eco- ecologicaloverlap, thesedifferences in size result logical overlap with the other three auklet species in fine-scale partitioning of nesting habitat and (Least Auklet, A. pusilla; CrestedAuklet, A. cris- food resources.For example, all four speciesnest tatella; and Parakeet Auklet, Cyclorrhynchus in rock crevices (primarily in talus slopes), but psittacula) in nesting and in feeding: all generally the crevicesused are of different sizes,with crev- nest in rock crevices and feed on zooplankton. ice size proportional to the size of the bird (e.g., The four speciesdo differ in size, however, in Least Auklets use the smallest crevices, Crested that the smallest member of the group (Least Auklets use the largest;Bedard 1969c, Knudtson Auklet) weighs only 84 g (SD = 7 g; range = 72- and Byrd 1982). Partitioning of food resources 98 g; n = 457) whereas the largest members of also has been observed for Least, Crested, and the group (Crested Auklet and Parakeet Auklet) Parakeet auklets (Bedard 1969b, Hunt et al. weigh 264 g (SD = 19 g; range = 195-330 g; n 198 l), but practically nothing hasbeen published = 192) and 258 g (SD = 19 g; range = 215-292 about the food habits of the Whiskered Auklet g; II = 42), respectively(Knudtson and Byrd 1982; (Vermeer et al. 1987). Day and Byrd, unpubl. data). The Whiskered BCdard (1969a) predicted that the small dif- Auklet, which weighs an average of 121 g (SD = ference in size between Least and Whiskered auklets should cause extensive overlap in food habits. He further suggestedthat this (hypothe- I Received28 March 1988.Final acceptance28 Sep- sized) competition caused the paucity of Whis- tember 1988. kered Auklets in the Aleutian Islands, where Least

[651 66 ROBERT H. DAY AND G. VERNON BYRD

Auklets are present, and the abundance of Whis- bers and total weights that were ~0.1% for a kered Auklets in the Kuril Islands, where Least given taxon were rounded to 0.1% in the data Auklets are absent. We thus conductedthis study summaries. Becauseof the small sample sizesof to describethe prey eaten by the Whiskered Auk- auklets, data from both sexeswere combined for let throughout its breeding season and to com- the data summaries. Data from stomachs and pare the foods of Whiskered Auklets with those sublingual pouchesalso were combined, because of Least, Crested,and Parakeetauklets. The study the foods were essentially identical. was conducted at Buldir Island (52”24’N, Food samples from five of the seven Crested 175”56’E), Alaska, where all four speciesof auk- Auklets were too large for analyses of each in- lets breed (Knudtson and Byrd 1982, Byrd et al. dividual food item; these five stomachs con- 1983, Byrd and Day 1986). tained an estimated 400-2,400 prey items. For thesestomachs, we used a Stemple pipette (com- METHODS monly used for subsampling zooplankton sam- The at-sea distribution of Whiskered Auklets was ples) to take two subsamplestotaling 5.0-13.2% studied in the vicinity of Buldir Island during of the total volume of food items. Identifications, May-August 1985 and 1986 (M. A. Spindler, numbers, and weights were determined for these U.S. Fish and Wildlife Service, unpubl. manu- subsamples, then total numbers and weights of script). Data were collected at a total of 269 sea- each taxon were estimated for each of the sub- bird transect stations. These line transectswere sampled birds from the subsample data. 300 m wide by a distance that was covered in Only those birds containing food were consid- either 10 min or 15 min of time while the boat ered in the data summaries; out of 40 auklets was moving forward in a straight line at a known collected, only one had an empty stomach (a speed. Later, data were stratified into “near- Whiskered Auklet on 16 July). For each taxon, shore” (< 3 nm [5.5 km] from shore) and “off- we calculated the three frequently-reported mea- shore” (~3 nm [5.5 km] from shore) transects. suresof food habits: percent frequency of occur- Thus, offshore transects occurred anywhere in rence (F), percent of the total number of indi- the 70-nm-wide (130-km-wide) passesto the east viduals of all taxa combined (N), and percent of and west of Buldir (Fig. 1). the total weight of all taxa combined (W). We We collected Whiskered Auklets at Buldir Is- then calculated an Index of Relative Importance land from a small boat during late egg laying/ (IRI) with an equation modified from Pinkas early incubation (26 May; six birds), late incu- (1971): bation/early chick rearing (l-2 July; six birds), IRI = F(N + W). and middle and late chick rearing (15-l 6 July; seven birds; and 4 August; six birds) in 1976; we IRI values range from 0.02 [i.e., 0.1% x (0.1% also have included data from one bird collected + O.l%)]to20,000[i.e., 100% x (lOO%+ lOO%)], on 3 August 1974. For comparison of food habits with larger numbers representingfood items that during the chick-rearing period in 1976, we col- are relatively “important” in the birds’ diets. IRI lected four Parakeet Auklets (5 August), three values combine the three main attributes that Least Auklets (15 and 3 1 July), and seven Crest- frequently are discussed separately in feeding ed Auklets (5 August). All birds were collected studies and thus provide an overall analysis of within 1.0 km of the north shore of the island. the relative importance of prey taxa. Stomachs were preserved in buffered formalin and later were transferred to 10% isopropanol RESULTS for sorting and identification. In the laboratory, we sorted and identified all FORAGING HABITAT prey items in each stomachto the lowest possible At Buldir Island, Whiskered Auklets fed pri- taxon, counted the number of recognizable in- marily in and near convergent tidal fronts that dividuals of each taxon, and weighed the tissue formed in and near the passesbetween three is- of each taxon to the nearest 1 mg. When only lets just off the northwestern comer of the island fragments of a taxon were encountered, the num- (Fig. 1). All birds were collected in these tidal ber of individuals of that taxon was recorded as fronts, and 25 of 26 contained food. Whiskered 1 +, and the number used for that bird in the Auklets apparently feed diurnally, for we en- data summaries was 1. Percentagesof total num- countered them during every visit to the tide rips FEEDING OF WHISKERED AUKLETS 67

FIGURE 1. Bathymetryin the vicinity of Buldir Island,Alaska. Shading delineates the primaryforaging area of WhiskeredAuklets.

(09:00-22:00 local time); we did not determine Aleutians, Day and B. E. Lawhead estimated ap- if nocturnal feeding also occurs. proximately 10,000 Whiskered Auklets feeding Whiskered Auklets were seen on nine (36.0%) in standing waves l-2 m high in Kagamil Strait of 25 transects in nearshore (~5.5 km (Islands of Four Mountains) on 2 June 1978, and from shore) waters, but significantly fewer (x2 = G. A. Putney (pers. comm.) saw “thousands” the 19.36; df = 1; P < 0.001) were seen in offshore following day in tide rips between and (~5.5 km from shore) waters around Buldir ( 19 Chagulak islands (Islands of Four Mountains). [7.8%] of 244 transects;Spindler, unpubl.). Fur- In the central Aleutians, E. P. Knudtson and Day ther, thoserecords from offshorewaters occurred saw approximately 200 birds feeding in tide rips only early in the breeding season, during May off Ugidak Island () on 30-3 1 July and June. We interpret these results to indicate 1977. In the western Aleutians, T. J. Early (U.S. that this speciesis primarily a nearshore feeder Fish and Wildlife Service; pers. comm.) saw ap- during the breeding season. proximately 750 birds feeding in a tidal front off This pattern of feeding at tidal fronts, “tide the southwesterncomer of Island (Rat Is- rips,” or standing waves also occurs elsewhere lands) on 10 August 1978. in the Aleutians. For example, Byrd and Gibson FOOD HABITS-GENERAL (1980) found most Whiskered Auklets near tide rips, usually within 16 km of land, during their Whiskered Auklets at Buldir ate prey of at least surveys throughout the Aleutians. In the eastern 10 species(Table 1). Zooplanktonic 68 ROBERT H. DAY AND G. VERNON BYRD were the most important prey, with chaeto- Auklets again exhibited near-monophagy on N. gnaths, molluscs, and larval fishes next in de- plumchrus (88.1% of total IRI). A further 8.7% creasing order of importance. formed of the total IRI consistedof copepod fragments, a majority of the food, with Neocalanus plum- probably of the same species.The remaining 3.2% chrusalone constituting at least 87.8% of the total of the food consistedof an unidentified euphau- IRI. In contrast, N. cristatusrarely was eaten by siid, megalopae of E. isenbeckii, frag- Whiskered Auklets (0.2% of total IRI), being re- ments, and unidentified pteropods (probably L. corded only twice. Chaetognaths were next in helicina), cephalopods,and chaetognaths. importance (0.2% of total IRI), with most indi- In early August (late chick rearing), Whiskered viduals probably Sagitta elegans(the most com- Auklets moved even closer to monophagy than mon epipelagicchaetognath in this area-see Bieri they had been earlier. Neocalanus plumchrus 1959; Fager and McGowan 1963; Kotori 1976; constituted 90.4% of the total IRI, N. cristatus Cooney 198 1, 1987). The hyperiid amphipod was present in trace amounts, and copepod frag- Parathemistopacifica and megalopae of the crab ments (probably N. plumchrus) formed another Erimacrus isenbeckii were recorded in trace 9.3% of the total. Thus, 99.8% of the food con- amounts. Pteropods, probably Limacina heli- sisted of copepods,with essentially all probably cina (the shellshad dissolved; also seeFager and being N. plumchrus. The remaining 0.2% of the McGowan 1963; Cooney 1981, 1987), were the food consisted of unidentified chaetognaths, a primary molluscan prey. Larval cephalopodsand larval fish, crustaceanfragments, and tis- a larval fish were too fragmented to be identified. sue.

FOOD HABITS-TEMPORAL TRENDS FOOD HABITS-THE AUKLETS COMPARED In late May (late egg laying/early incubation of Food habits of Least, Crested, and Parakeet auk- these birds; Knudtson and Byrd 1982), Whis- lets at Buldir are compared with those of Whis- kered Auklets were eating primarily N. plum- kered Auklets in Table 2. These data illustrate chrus(62.0% of total IRI). An additional 36% of the major prey eaten by thesezooplankton-feed- the total IRI consisted of copepod fragments, ers at Buldir. making copepods (probably all N. plumchrus) Least Auklets fed primarily on N. plumchrus, almost 98% of the total foods at this time. About with an unidentified amphipod, euphausiid frag- 2% of the food consisted of unidentified chae- ments, and one pteropod (L. helicina) recorded tognaths, with crustaceanfragments the remain- in trace amounts. Crested Auklets fed primarily ing prey. on N. cristatus and secondarily on P. pact&s. In early July (late incubation/early chick rear- Neocalanusplumchrus were relatively unimpor- ing), Whiskered Auklets appeared to have tant, as were unidentified copepodsand euphau- switched to eating a variety of foods. The am- siid and crustaceanfragments. Parakeet Auklets phipod P. pactjicawas the dominant food (33.8% ate a wider diversity of foods, with at least 11 of total IRI), with N. cristatussecond (30.1% of taxa recorded from the four stomachs. Neoca- total IRI) and N. plumchrus third in importance lanus cristatus,P. pact&a, and E. isenbeckiiwere (25.2% of total IRI). The remaining 11% con- the primary prey (in decreasingamounts). Neo- sistedof copepodfragments, an unidentified am- plumchrus constituted only 2% of the phipod, amphipod fragments, an unidentified total, and the remaining prey consistedof a gam- chaetognath, crustacean fragments, and uniden- marid amphipod, the commensal hyperiid am- tified animal tissue.We doubt that this difference phipod Hyperia medusarum(see Bedard 1969b), is biologically significant, however, because the another hyperiid, H. macropa, euphausiids, amount of food in this sample of stomachswas shrimp larvae of two species,and an unidentified relatively small: the number of individuals rep- chaetognath. resentedonly 2.9% and the massrepresented only At Buldir, Neocalanus copepodsclearly were 5.4% of the totals for the entire sample of stom- the primary food source of all four species of achs. Hence, P. pactjica and N. cristatus were auklets (Table 2). The two smaller auklets took inflated in importance during this sampling pe- primarily the smaller N. plumchrus, whereasthe riod, whereasthey clearly were unimportant dur- two larger auklets took primarily the larger N. ing the summer as a whole (Table 1). cristatus.In addition, the two largerauklet species In mid-July (middle chick rearing), Whiskered ate significant amounts of the large P. paczjica

70 ROBERT H. DAY AND G. VERNON BYRD

TABLE 2. Importance of selectedprey taxa to four speciesof aukletsat Buldir Island, Alaska, during late July/ early August 1976. Values are expressedas percentagesof the total Index of Relative Importance (IRI).

TaXOn LeastAuk&% WhiskeredAukIet CrestedAuklet ParakeetAuklet

Arthropods Copepoda (total)’ 86.6 99.8 65.8 58.2 Neocalanusplumchrus 86.6 90.4 1.4 2.0 N. cristatus 0 co.1 64.3 56.2 Unidentified copepod fragments 0 9.3 0.1 0 Amphipoda (total)’ 0.1 co.1 33.4 23.2 Parathemistopa&a 0 0 33.4 21.5 Euphausiacea(total)l 0.1 0 0.1 0.9 Decapoda (total)l 0 0 0 16.2 Caridea (total)l 0 0 0 2.8 Brachyura Erimacrus isenbeckii 0 0 0 13.4 Unidentified crustaceanfragments 13.1 co.1 0.7 3.1 Mollusca Gastropoda Pteropoda (total)l 0.1 0 0 0 Cephalopoda (total)’ 0 0 0 0 Chaetognatha Chaetognatha(total)l 0 0.1 0 0.1

I Total values representthe sumsof IRIS of speciesconstituting that higher-leveltaxon (e.g.,the IRI of total Copepodais the sum of individual IRIS of all copepodspecies); they do not representa new IRI calculatedfor all taxa togetherwthin that higher-leveltaxon. and the large E. isenbeckii. Euphausiids, ptero- mation and upwelling. These convergent fronts pods, cephalopods,and chaetognathswere eaten are generated by the rapid flow of large volumes in trace amounts by all four species. of water through “bottlenecks” (i.e., passesbe- tween islands or in passesbetween small islets). DISCUSSION For example, at a tidally-generated convergent All available information indicates that the front in the English Channel, Pingree et al. (1974) Whiskered Auklet is adapted for feeding in near- found concentrations of zooplanktonic crusta- shore, convergent tidal fronts. For example, in ceans (copepods, euphausiids, caridean larvae, wide island passes(e.g., Pass)far from and decapod zoeae and megalopae) to be up to nesting islands, only single Whiskered Auklets 75 times higher within the front than on either are seen (Day and Byrd, pers. observ.). The pref- side of it. In addition, some fronts near Buldir erence of this speciesfor feeding near shore ap- are generated by the flow of water acrosstopo- parently extendsto the winter, for Stejneger(1885) graphic shallows or shoals (e.g., the reefs), also commented on the number of birds seenjust off causing upwelling (Owen 198 1, Wolanski and the beach in the (U.S.S.R.) Hamner 1988). during December and January. Both Buldir’s location and the feeding data We believe that these convergent tidal fronts presentedhere indicate that all foragingby Whis- are the dominant force affecting the distribution kered Auklets there takes place within the ocean- and abundance of prey for auklets (especially ic zooplanktonic community. There is a poorly- Whiskered Auklets) around Buldir Island and developed coastal community of zooplankton at elsewhere in the Aleutian Islands; further, these Buldir, probably because of the narrow conti- fronts are stable in space and highly repeatable nental shelf around the island (Fig. 1). In addi- in time (with each tidal cycle). In the Aleutians, tion, the two Neocalanus copepods,the hyperiid twice-daily tides between the Pacific Ocean and P. pacijica, and the pteropod L. helicina, are the provide the energy for frontal for- dominant members of this epipelagic oceanic FEEDING OF WHISKERED AUKLETS 7 1 community (Fager and McGowan 1963; Cooney pae that were about 10 mm in length.) If (as we 1981, 1986, 1987). believe) turbulence from tidal mixing determines Few other data on food habits of Whiskered the vertical distribution of the main zooplankton Auklets are available for comparison with our speciesaround Buldir, the four seabird species results. Cottam and Knappen (1939) found that have to be selecting preferred prey on the basis three of five birds from the Aleutians during of size out of a mixture of possible prey items. summer contained copepods exclusively. The BCdard (1969a) predicted that the small dif- fourth stomach contained a mixture of amphi- ference in size between Least and Whiskered pods, isopods, copepods,a scorpaenid fish, and auklets should cause extensive overlap in food a trace of a spider; the fifth stomach contained habits, but comparative data on food habits from what appeared to be crustacean fragments and areas where the two species coexist were not mollusc eggs.The only data on winter prey are available at that time. He further suggestedthat from seven adults at the Commander Islands in this (hypothesized) competition causedthe pau- December and January (Stejneger 1885). All but city of Whiskered Auklets in the Aleutian Is- one of thesebirds were filled with gammarid and lands, where Least Auklets are abundant, and unidentified amphipods; the remaining one con- the abundance of Whiskered Auklets in the Kuril tained a small gastropod (a pteropod?) and the Islands, where LeastAuklets are absent.Our small remains of a decapod. samples from the Aleutians indicate extensive Neocalanus copepodsare seasonally-migrating overlap in prey between these two auklet species speciesthat are most abundant in surface layers at Buldir, thus lending support to Btdard’s hy- from spring to early fall, whereupon they migrate pothesis of interspecific competition. Interspe- to depths and reproduce in midwinter, using cific competition for food may help shape the stored lipids for egg production (Heinrich 1962, distribution and abundance of Whiskered Auk- Fulton 1973, Miller et al. 1984, Cooney 1987). lets, but additional studiesof food habits and the Whiskered Auklets apparently become nearly- prey baseare required to test Bedard’s hypothesis monophagous on Neocalanus copepods during adequately. summer, when the latter are abundant near the ocean’s surface,but then probably switch to eat- ACKNOWLEDGMENTS ing amphipods and other zooplanktonic crusta- The specimenswere collectedwhile we were employees ceans during winter, when the Neocalanus are of the Aleutian Islands Unit of the Alaska Maritime living 600-1,000 m below the ocean’s surface. National Wildlife Refuge. We thank E. P. Knudtson, At St. Lawrence Island, Least and Crested auk- T. G. Tobish, and D. W. Woolington for assistancein lets took gammarid amphipods in number only the field, E. P. Hoberg and M. A. Spindler for the seabirdtransects, and Capt. G. A. Putney and the crew when their “preferred” prey (copepodsand eu- of the R/V Aleutian Tern for logistical support. C. I. phausiids, respectively) were not common (Be- Winiecki and K. 0. Coyle of the Institute of Marine dard 1969b). Further, when chicks of these two Sciencesdid most of the sorting and all of the identi- specieswere fed only gammarids, they consis- fications. This paper was written while Day was an tently lost weight and died within 4 days. Hence, Angus Gavin Memorial Fellow and a ResourceFellow at the University of Alaska, Fairbanks. We thank F. there may be a nutritional basis for the apparent H. Fay, S. M. Murphy, D. M. Troy, K. Vermeer, and preference for copepods. an anonymous reviewer for constructive criticism of The comparison of the food habits of the four this manuscript. This is Contribution No. 704 of the species in late July/early August indicates that Institute of Marine Sciences, University of Alaska, Fairbanks, Alaska. interspecific differences in prey eaten resulted from size-related differencesin food preferences, LITERATURE CITED rather than from differences in the availability AMERICANORNITHOLOGISTS UNION.’ 1983. Check- of certain prey. That is, the two smaller auklets list of North American birds. 6th ed. American concentrated on N. plumchrus (about 5 mm in Ornithologists’ Union, Washington, DC. length), the smaller of the two large oceanic co- B~D.DARD,J. 1969a. Adaptive radiation in Alcidae. Ibis pepods, whereas the two larger auklets concen- 111:189-198. B~DARD,J. 1969b. Feeding of the Least, Crested,and trated on the larger N. cristatus (about 10 mm Parakeet auklets around St. Lawrence Island, in length) and P. paczjica about 10 mm in length). Alaska. Can. J. Zool. 47: 1025-1050. (Parakeet Auklets also took E. isenbeckii megalo- BBDARD,J. 1969~. The nesting of the Crested, Least, 72 ROBERT H. DAY AND G. VERNON BYRD

and Parakeetauklets on St. LawrenceIsland, Alas- HUNT, G. L., JR., B. BURGESON,AND G. A. SANGER. ka. Condor 71:386-398. 1981. Feeding ecology of seabirdsof the eastern BIERI, R. 1959. The distribution of the planktonic Bering Sea, p.629-647. In D. W. Hood and J. A. Chaetognathain the Pacific and their relationship Calder leds.1.The easternBering Sea shelf: ocean- to water masses.Limnol. Oceanogr.4: l-28. ograph; and resources.Vol. 2.NOAA, Office of BYRI), G. V., AND R. H. DAY. 1986. The avifauna of Marine Pollution Assessment,Juneau, AK. Buldir Island, Aleutian Islands, Alaska. Arctic 39: KNUDTSON,E. P., AND G. V. BYRD. 1982. Breeding 109-l 18. biology of Crested, Least, and Whiskered auklets BYRD,G. V., R. H. DAY, ANDE. P. KNUDTSON. 1983. on Buldir Island, Alaska. Condor 84: 197-202. Patterns of colony attendance and censusingof KOTORI,M. 1976. The biology of Chaetognathain auklets at Buldir Island, Alaska. Condor 85:274- the Bering Sea and the northern North Pacific 280. Ocean, with emphasis on Sagitta elegans.Mem. BYRD,G. V., AND D. D. GIBSON. 1980. Distribution Fat. Fish. Hokkaido Univ. 23:95-183. and population statusof Whiskered Auklet in the MILLER,C. B., B. W. FROST,H. P. BATCHELDER,M. J. Aleutian Islands,Alaska. West. Birds 11:135-l 40. CLEMONS,AND R. E. CONWAY. 1984. Life his- COONEY,R. T. 1981. Bering Sea zooplankton and toriesof largegrazing copepods in a subarcticocean micronekton communities with emphasis on an- gyre:Neocalanus plumchrus, Neocalanus cristatus, nual production, p. 947-974. In D. W. Hood and and Eucalanus bungii in the northeast Pacific. J. A. Calder leds.1.The eastern Berina Sea shelf: Progr. Oceanogr. 13:201-243. oceanographyand ’resources. Vol. 2. NOAA, Of- OWEN, R. I98 1. Fronts and eddies in the sea: mech- fice of Marine Pollution Assessment,Juneau, AK. anisms,interactions, and biologicaleffects, p. 197- COONEY,R. T. 1986. The seasonaloccurrence ofNeo- 233. In A. R. Longhurst led.1.Analvsis of marine calanuscristatus, Neocalanus plumchrus, and Eu- ecosystems.Academic P;ess;‘New York. calanusbungii over the shelf of the northern Gulf PINGREE,R. D., G. R. FORSTER,AND G. K. MORRISON. of Alaska. Cont. Shelf Res. 5:541-553. 1974. Turbulent convergent tidal fronts. J. Mar. COONEY,R. T. 1987. Zooplankton, p. 285-303. In Biol. Assoc. U.K. 541469-179. D. W. Hood and S. T. Zimmerman [eds.], The PINKAS,L. 1971. Food habits study, p. 5-10. In L. Gulf of Alaska: physical environment and biolog- Pinkas, M. S. Oliphant, and I. L. K. Iverson [eds.], ical resources.Mineral Management Servicepubl. Food habits of albacore,bluefin tuna, and bonito no.: OCS study, MMS 86zOO95. Government in California waters. Calif. Dep. Fish and Game, Printing Office, Washington. DC. Fish Bull. 152. COT~AM,C.: AND P. KNAPPEN. 1939. Food of some STWNEGER,L. 1885. Results of ornithologicalexplo- uncommon North American birds. Auk 56:138- rations in the Commander Islands and in Kamt- 169. schatka.U.S. Natl. Mus. Bull. No. 29:1-382. FAGER,E. W., AND J. A. MCGOWAN. 1963. Zooplank- VERMEER,K., S. G. SEALY,AND G. A. SANGER. 1987. ton speciesgroups in the North Pacific. Science Feeding ecology of Alcidae in the eastern North 140:453-460. Pacific Ocean, p. 189-227. In J. P. Croxall [ed.], FULTON,J. D. 1973. Some aspectsof the life history :feeding biology and role in marine eco- of Calanus plumchrusin the Strait of Georgia. J. systems. Cambridge Univ. Press, Cambridge, Fish. Res. Board Can. 30:81 l-8 15. United Kingdom. HEINRICH,A. K. 1962. The life history of plankton WOLANSKI,E., AND W. M. HAMNER. 1988. Topo- and seasonal cycles of plankton in the graphicallycontrolled fronts in the oceanand their oceans.J. Cons. Int. Explor. Mer 27:15-24. biological influence. Science24 1:177-l 8 1.