CALIFORNIA FISH-" GAME California Fish and Game is a journal devoted to the conservation and understanding of fish and v/ildlife. If its contents are reproduced elsev^here, the authors and the California Department of Fish and Game v/ould appreciate being acknovv'ledged. Subscriptions may be obtained at the rate of $10 peryear by placing an order with the California Department of Fish and Game, 2201 Garden Road, Monte- rey, CA 93940. Money orders and checks should be made out to California Fish and Game. Inquiries regarding paid subscriptions should be directed to the Editor. Complimentary subscriptions are granted on an exchange basis. Please direct correspondence to: Robert N. Lea, Ph.D., Editor-in-Chief California Fish and Game 2201 Garden Road Monterey, CA 93940 u 1 ] VOLUME 74 JANUARY 1988 NUMBER 1 Published Quarterly by STATE OF CALIFORNIA THE RESOURCES AGENCY DEPARTMENT OF FISH AND GAME —LDA— STATE OF CALIFORNIA GEORGE DEUKMEJIAN, Governor THE RESOURCES AGENCY GORDON VAN VLECK, Secretary for Resources FISH AND GAME COMMISSION ALBERT C. TAUCHER, President Long Beach ROBERT A. BRYANT, Vice President JOHN A. MURDY III, Member Yuba City Newport Beach BENJAMIN F. BIAGGINI, Member E. M. McCRACKEN, JR., Member San Francisco Carmichael HAROLD C. CRIBBS Executive Secretary DEPARTMENT OF FISH AND GAME PETE BONTADELLI, Director 1416 9th Street Sacramento 95814 CALIFORNIA FISH AND GAME Editorial Staff Editorial staff for this issue: f\^arine Resources Peter L Haaker, Paul N. Reilly, and John P. Scholl Sturgeon Donald E. Stevens Wildlife Bruce E. Deuel and William E. Grenfell, Jr. Editor-in-Chief Robert N. Lea, Ph.D. CONTENTS Identification of Mammal Tracks from Sooted Track Stations in the Pacific Northwest Cathy A. Taylor and Martin G. Raphael 4 Feeding Ecology of Ten Species of Rockfishes (Scorpaenidae) from the Gulf of Alaska Richard J. Rosenthal, Victoria Moran-O'Connell, and Margaret C. Murphy 16 Growth of Young-Of-The-Year and Juvenile Pacific Herring from San Francisco Bay, California Paul N. Reilly 38 Diet of Juvenile and Subadult White Sturgeon in the Lower Columbia River and Its Estuary William D. Muir, Robert L. Emmett, and Robert J. McConnell 49 Herd Segregation in Harbor Seals at Point Reyes, California Sarah G. Allen, Christine A. Ribic, and Janet E. Kjelmyr 55 NOTES Record of the Chameleon Goby, Tridentiger trigonocephlus, from the Sacramento-San Joaquin Delta Paul F. Raquel 60 Additional Notes on Migrational Distribution of Northern Pintail Banded in California Warren C. Rienecker 61 BOOK REVIEWS 64 4 CALIFORNIA FISH AND GAME Calif. Fish and Came 74 ( 1 ) : 4-1 5 1 988 IDENTIFICATION OF MAMMAL TRACKS FROM SOOTED TRACK STATIONS IN THE PACIFIC NORTHWEST' CATHY A. TAYLOR Pacific Southwest Forest and Range Experiment Station 1700 Bayview Drive Areata, California 95521 and MARTIN C. RAPHAEL Rocky Mountain Forest and Range Experiment Station 222 South 22nd Street Laramie, Wyoming 82070 Sooted aluminum tracking stations were used to investigate distribution patterns of medium and large mammals in Douglas-fir forest in northwestern California. Track stations consisted of two adjacent aluminum sheets covered by a thin layer of kerosene soot with a central bait. Stations were located at 166 sites and monitored for eight-day periods between 1 July and 15 September in 1981, 1982, and 1983. Tracks from 23 species were preserved with transparent tape and up to seven measurements were recorded. We found track impressions on the hard surface to be much smaller and more detailed than those in soil or snow; available measure- ments and drawings in field guides were useless for identification purposes. Therefore we present a standard method of measurement and a key to distinguish the tracks of medium to large sized mammals from sooted track stations in the Pacific Northwest. INTRODUCTION Wildife biologists have used various tracking techniques to assess mammal populations. The most common method is to attract animals to a baited station where tracks are detected in fine soil or snow (Cook 1949; Wood 1959; Linhart and Knowlton 1975; Lindzey, Thompson, and Hodges 1977). However, this technique is infeasible in areas where soils are rocky and fine soil cannot be easily transported (Barrett 1983). A sooted surface was first used by Mayer (1957) to track small mammals. Further modifications were made by Justice (1961 ) and Lord et al. (1970) to investigate small mammal populations. Barrett (1983) expanded the use of a sooted aluminum surface to determine the distribution of martens, Martes americana, and other carnivores in the Sierra Nevada. This method was used on a larger scale in the present study to investigate distribution patterns of medium and large mammals in Douglas-fir forests in northwestern California (Raphael and Barrett 1984). Tracks on a hard surface provided by aluminum track plates differ markedly in size and shape from those in softer substrates such as snow or fine soil, appearing much smaller and providing much more detail. As a result, measurements and drawings in available field guides are unreliable and confusing. Increasing numbers of researchers are using this method and a standardization of the technique is required. Therefore we present a standard method of measurement (Figure 1, Table 1 ) and a key to distinguish the tracks of medium to large sized mammals from aluminum track plates in the Pacific Northwest. The list of mammals Accepted for publication April 1987. IDENTIFICATION OF MAMMAL TRACKS 5 presented here is incomplete, as field work was restricted to forested areas of Northwestern California. We recommend that additional tracks and measure- ments be made available as they are identified. B J FIGURE 1. Standard measurements taken on all distinct tracks were as follows: A — Longest vertical line drawn from distal edge of foremost toe to back edge of palm pad; B — Horizontal line measuring widest spread of toes; C — Height of the palm pads; D — Width of palm pads; E — Vertical distance from foremost toe to back edge of heel pad; F — Distance from foremost toe to end of heel mark; C — Distance from foremost claw to back of palm pad. TABLE 1. Means and 95% Confidence Intervals ^ for Track Measurements (mm) \ FORE F007 HIND FOOT Species X N SD 95% C.I. X N SD 95% a. DIdelphis virginiana A 20.5 2 0.7 20-21 21.0 1 B 35.5 2 0.7 35-36 40.0 1 C 10.0 2 1.4 9-11 8.0 1 D 16.0 2 1.4 15-17 21.0 1 Tamias spp A 13.3 58 1.6 12.9-13.7 14.5 41 1.1 14.2-14.9 B 14.7 58 2.0 14.1-15.2 16.6 40 2.2 15.8-17.3 C 5.4 58 1.0 5.1-5.6 6.0 41 0.8 5.&-6.3 D 6.5 58 0.8 6.3-6.7 8.3 41 1.0 8.0-5.7 E 18.1 56 2.1 17.5-18.6 F 17.7 3 1.2 17-19 CALIFORNIA FISH AND CAME TABLE 1. Means and 95% Confidence Intervals' for Track Measurements (mm) I— Continued Species Spermophilus beecheyi. Spermophilus lateralis . Sciurus griseus Tamiasciurus douglasii. Claucomys sabrinus. Neotoma fuscipes. Erethizon dorsatum Canis latrans Urocyon cinereoargenteus . Ursus americanus . Bassariscus astutus . FORE FOOT HIND FOOT X N SD 95% C.I. X N SD 95% C.I A 20.9 44 1.8 20.3-21.4 24.2 42 2.1 23.5-24.8 B 19.0 44 2.6 18.2-19.8 23.8 42 4.1 22,5-25.1 C 8.0 44 1.1 7.7-8,3 9.7 42 1.4 9,3-10.2 D 10.4 44 1.5 10.0-10.9 14.1 42 1.9 13.5-14.6 E 28.5 44 2.7 27.6-29.3 F 37,9 9 4,0 34,8^1.0 G 27.9 7 3.8 24.4-31.3 30,9 12 4,2 28.2-33.6 A 14.7 6 2.7 11.9-17,5 16.3 6 1,9 15.3-17.4 B 13.3 6 2.4 10.8-15.9 14.5 6 1.9 12.5-16.5 C 5.8 6 1.2 4.6-7.1 6.3 6 1.4 4.9-7.8 D 6.5 6 1.4 5.1-8.0 8.7 6 1.5 7.1-10.3 E 20.4 5 1.8 18.2-22.7 A 30.6 20 1.8 29.7-31.4 35.1 16 2.3 33.9-36,3 B 26.6 20 2.8 25.3-27.9 35.4 16 6,4 32.0-38.8 C 11.4 20 1.7 10.6-12.2 13.9 16 2.1 12.8-15.1 D 13.6 20 1.7 12.7-14,4 19.3 16 2.1 17.8-20.8 E 42.2 19 2.1 41.1-43.2 F 58.0 2 1.4 57-59 G 41.0 2 1.4 40-42 A 20.5 63 2.9 19,7-21.2 21,7 47 2.2 21.1-22.3 B 20.2 63 2.8 19.5-20.9 23,9 47 4,1 22,7-25,1 C 7,7 63 1.1 7,5-8,0 8,5 47 1,4 8.1-8,9 D 9.2 63 1.1 8,9-9,5 11,8 47 1,7 11.3-12.3 E 26.9 59 3.9 25.9-27,9 F 30.8 4 4,0 26-35 G 26.2 5 5.0 22-33 27.3 10 3,7 24.^30.0 A 14.2 49 1.4 13.8-14,6 14.3 69 1.1 14.1-14.6 6 12.9 49 2.3 12.2-13.5 15,1 69 2.6 14.5-15.7 C 5.0 49 0.7 4.8-5.2 6,0 69 1.1 5.8-6.3 D 6.6 49 0.7 6.4-6.8 9.4 69 1.0 9,1-9,6 E 19.5 48 1.3 19.1-19.9 F 31.3 12 3.3 29,3-33,4 G 17.0 1 16.3 4 1,5 14-17 A 12.6 29 1.3 12.1-13.1 14.9 24 1.6 14.2-15.6 B 15.1 29 1.7 14.4-15.7 16.7 24 2.7 15.^17.9 C 5.9 29 0,6 5.7-6.1 7,3 24 1,3 6.7-7.8 D 7.5 29 0,9 7.1-7.8 8,0 24 1,1 7.5-8.4 E 16.9 29 2.6 15.9-17.9 21,0 19 3,1 19.4-22.5 C 47.7 3 2.1 46-50 D 35.7 3 2.3 33-37 A 67.0 B 50.0 C 36.0 D 31.0 G 71.0 A 34.7 52 2.7 34.0-35.5 B 33.8 52 3.2 32.0-34.7 C 12.3 52 2.3 11.7-13.0 D 18.7 52 2.2 18.1-19.3 A 100.6 56 9.9 97.9-103,2 110,0 22 21.2 100.6-119.3 B 105.3 56 8.8 102,9-107,6 99.4 22 12.4 93.9-104.9 C 49,1 50 8,9 46,5-56.4 62.6 20 20,7 52.^-72.2 D 88,9 50 11,1 85,8-92.1 83,9 20 15,3 76.8-91.0 A 24.0 23 1.7 23.3-24.8 B 21.6 23 2.8 20.4-22.8 C 12.4 23 1.6 11.7-13.1 IDENTIFICATION OF MAMMAL TRACKS TABLE 1. Means and 95% Confidence Intervals ^ for Track Measurements (mm) I— Continued Species Procyon lotor Martes americana . Martes pennant/'. Muste/a erminea . Mustela frenata Muste/a vison. Spilogale gracilis Mephitis mephitis . Felis concoior. Felis rufus. ' Actual ranges were used when ^ Codes for track measurements FORE FOOT HIND F007 X N SD 95% CI X N SD 95% C.I D 14.0 23 1.9 13.1-14.8 A 45.8 5 4.3 40-52 48.8 4 1.5 47-50 B 36.6 5 3.5 33^2 37.3 4 5.5 32^5 C 24.2 5 1.6 22-26 26.5 4 1.7 24-28 D 25.4 5 1.1 24-27 25.3 4 2.8 22-28 E 67.0 3 7.9 61-76 A 32.5 8 4.5 29.5-35.5 32.8 4 2.2 30-35 B 30.8 8 5.3 27.2-34.4 31.8 4 4.2 29-38 C 12.4 8 2.6 10.7-14.1 13.0 4 1.4 12-15 D 16.0 8 3.8 13.5-18.5 18.3 4 2.6 16-22 E 45.1 3 8.7 35-50 A 45.9 14 5.1 42.9^8.8 46.7 16 3.5 44.8^8.6 B 44.4 14 4.9 41.6-47.2 42.9 16 4.9 40.3^5.6 C 20.5 14 5.0 17.6-23.4 19.6 16 3.5 17.7-21.4 D 25.4 14 4.3 22.9-27.8 26.6 16 4.1 24.4-28.8 E 72.5 2 5.0 70-75 58.0 5 7.0 49.4-66.7 A 11.0 6 1.7 9.6-12.4 11.0 2 0 11-11 B 13.5 6 2.0 11.2-15.8 16.0 2 1.0 15-17 C 4.2 6 1.1 3.4-5.0 5.0 2 0 5-5 D 5.2 6 1.6 4.0-6.5 5.0 2 0 5-5 E 19.5 3 1.5 18-21 A 15.4 5 0.5 15-16 14.5 2 0.7 14-15 B 14.6 5 0.5 14-15 20.5 2 0.7 20-21 C 6.4 5 0.5 b-7 4.5 2 0.7 4-5 D 8.4 5 0.9 7-9 8.0 2 0 8^ E 19.3 3 0.6 19-20 A 26.9 8 2.7 25.1-28.7 28.5 4 2.1 26-31 B 29.5 8 2.3 28.0-31.0 28.3 4 2.1 2^30 C 12.0 8 2.1 10.5-13.5 10.3 4 1.7 8-12 D 13.5 8 2.1 12.0-15.0 14.5 4 1.3 13-16 E 40.9 8 2.2 39.4-42.4 A 17.9 33 2.0 17.2-18.6 19.1 36 2.0 18.4-19.8 B 16.9 32 2.4 16.1-17.8 17.8 36 2.1 17.1-18.5 C 7.6 33 1.4 7.1-8.1 8.4 36 1.3 7.9-8.8 D 10.6 33 2.0 9.9-11.3 12.1 36 2.2 11.3-12.8 E 23.7 20 2.8 22.4-24.9 25.7 32 2.4 24.9-26.6 A 24.3 7 1.6 22.8-25.8 29.3 4 4.4 25-33 B 23.9 7 2.7 21.4-26.3 25,8 4 4.5 22-31 C 11.6 7 1.3 10.4-12.8 15.5 4 3.5 12-19 D 17.9 7 2.0 16.0-19.7 19.7 4 4.4 15-24 E 36.0 2 1.4 35-37 40.3 4 6.1 34^6 A 77.5 2 17.7 65-90 B 74.5 2 7.8 69-80 C 40.5 2 7.8 35^6 D 50.5 2 0.7 50-51 A 37.0 2 0 37-37 B 38.5 2 0.7 38-39 C 14.5 2 0.7 14-15 D , 19.5 2 0.7 19-20 samp le size n < 5. follow those outlined in Fig. 1 8 CALIFORNIA FISH AND GAME METHODS AND MATERIALS Track stations consisted of two adjacent 814x407x0.6 mm aluminum sheets covered by a thin layer of kerosene soot and baited with a can of tuna pet food (described by Barrett 1983). We checked each of 135 stations for an 8-day period between 1 July and 15 September during 1981, and 166 stations during the same period in 1982 and 1983. Tracks were preserved in the field by firmly pressing transparent tape over them, then transferring the tape to a white data sheet. We recorded up to seven measurements for each track; distinguishing front and hind tracks. Tracks of 23 species were collected. All track stations were located in the Klamath Mountains in Humboldt, Trinity, and Siskiyou counties in northwestern California. All forest stands were dominated by Douglas-fir, Pseudotsuga menziesii, in association with tanoak, Lithocarpus densiflora, and Pacific madrone, Arbutus menziesii. KEY TO MAMMAL TRACKS ON SOOTED ALUMINUM TRACKING STATIONS IN THE PACIFIC NORTHWEST I. FOUR TOES ON FORE FEET (FF) AND HIND FEET (HF) (Figure 2) A. CANIDAE: General shape is oval with the toes approximately ^2 the size of the palm pad. Latter bi- or tri-lobed on posterior border, uni-lobed on anterior end. Fore foot slightly larger than similarly shaped hind foot, toes spread more widely on the forefoot. 1. Canis latrans (Coyote): Large dog track, greater than 40 mm in length and width. Claw marks present. 2. Urocyon cinereoargenteus (Gray fox): Small dog track in which the claws do not register. Palm pad small, anterior end rarely extends to a line drawn at halfway point on the posterior toes. Posterior end lacks protruding lobes found in bobcat track. B. FELIDAE: Tracks generally round or slightly oval. Palm pad is larger than that of the Canidae in relation to toes, which are approximately Va the size of the palm print. Tri-lobed on posterior end and bi-lobed on anterior end, the palm pad extends to or past halfway point on posterior toe prints. Claws do not register, prints of fore and hind feet are similar. 1. Felis concolor (Cougar): Large cat track greater than 60 mm in length and width. 2. Felis rufus (Bobcat): Small round track, about 38 mm in length and width. Posterior end of palm pad exhibits rounded, protruding lobes. II. FOUR TOES ON FORE FEET, FIVE TOES ON HIND FEET (Figure 3) A. SCIURIDAE: Squirrels of the Pacific Northwest exhibit following pattern: fore track has four toe pads, followed by three palm pads and two heel pads. Hind track has five toe pads followed by four palm pads in an arc. 1. Tamias spp. (Chipmunk): Small squirrel track, similar in size to Glaucomys sabrinus and Spermophilus lateralis. Species of chipmunk present must be determined through trapping. FF: Central palm pad appears as one large pad while outer palm pads and heel pad are roughly kidney-shaped. HF: Palm pads are kidney-shaped and are arranged in an exaggerated crescent shape. Toes irregularly spaced. IDENTIFICATION OF MAMMAL TRACKS 9 2. Glaucomys sabrinus (Northern flying squirrel): Size is similar to Tamias spp. and Spermophilus lateralis. FF: Central palm pad is irregularly shaped or oval and outer pads are oval. Inner heel pad is twice the size of outer pad. HF: Toe and palm pads are distinctly oval. Four palm pads occur in a smooth gradual arc and are evenly spaced. Three inner toes are evenly spaced in a tight linear array. 3. Spermophilus lateralis (Golden-mantled ground squirrel): Similar to previous two species. FF: Palm and heel pads are more obviously kidney-shaped than in previous two species and the central pad is T-shaped. HF: Palm pads are triangular in shape, the third palm pad extending forward of the second in a lopsided arc. 4. Spermophilus beecheyi (California ground squirrel): Similar in size to Tamiasciurus douglasii. FF: Palm pads larger than toe pads; irregularly shaped. HF: Palm pads larger than toe pads; irregularly shaped. 5. Tamiasciurus douglasii (Douglas' squirrel): Medium-sized squirrel track, similar in size to Spermophilus beecheyi. FF: Palm pads oval to round, similar in size to toe and heel pads. HF: Palm pads and toe pads oval, similar in size to toe pads. 6. Sciurus griseus (Western gray squirrel): Largest squirrel track. FF: Palm pads irregularly shaped. HF: Palm pads irregularly shaped. May register a long narrow heel pad on interior side of foot. This is the only squirrel to exhibit this auxiliary pad. B. CRICETIDAE: Neotoma fuscipes (Dusky-footed woodrat): Toes leave a distinct figure-eight pattern. FF: Central pad of three palm pads is T-shaped while outer pads are exaggerated kidney-shaped. Three small, round heel pads occur in a row just posterior to palm pads. HF: Three palm pads and three heel pads occur in groups of three, then two, then one. Central palm pad is T-shaped while all others are kidney-shaped. FIVE TOES ON FORE AND HIND FEET (Figure 4) Members of the families Didelphidae, Erethizontidae, Ursidae, Procyonidae, and Mustelidae exhibit this pattern. A. Didelphis virginiana (Virginia Opossum): This odd track is easily distinguished. FF: Five toes form a half circle around exaggerated crescent-shaped palm pad. HF: First toe is widely separated from the other toes, points below palm pad on inner side of foot. B. Erethizon dorsatum (Porcupine): This odd track is immediately recog- nizable. Toes rarely are evident; large, oval palm pads have a pebbled texture. Fore and hind prints are similar. TO CALIFORNIA FISH AND GAME C. Ursus americanus (Black bear): Largest track in the Pacific northwest. Toes form gentle arc over large palm pad. FF: One large crescent-shaped palm pad. HF: One large, elongate palm pad. D. Bassariscus astutus (Ringtail): Fore and hind feet leave similar impres- sions. Palm pad registers one large pad; very small second pad may appear posteriorly near the small first toe. E. Procyon lotor (Raccoon): Five elongated toe impressions. FF: The palm pad is thick and wide and usually appears as three to five large pads. HF: Large palm pad is crescent shaped, may appear as a multi-lobed pad or four distinct pads. Smaller pad appears posterior to the palm pad on the outer side of the foot. F. Martes pennant/ (Fisher): Large weasel-like track greater than 40 mm in length and width. Toe prints are circular to oval, palm pad leaves a crescent-shaped impression. FF: Small secondary palm pad occurs below first toe. A wide, thin heel pad lies posterior to the crescent-shaped palm pad. HF: Similar to forefoot; no heel pad. G. Martes americana (Marten): Similar to the fisher track, a male marten's track may overlap in size with a female fisher track. Marten track generally more hairy than fisher track, small first toe may not leave an impression. The palm pads register as three distinct pads. H. Mustela vison (Mink), Mustela frenata (Long-tailed weasel), Mustela erminea (Ermine): These tracks exhibit the same pattern and may overlap in size. Five toes usually are evident though the first toe is smaller than the others. There are three palm pads, and heel pads rarely appear on forefoot track, resulting in similar fore and hind tracks. I. Spilogale gracilis (Western spotted skunk): FF: Three to four pads, central pad heart-shaped. Two small heel pads register posterior to the palm pads. Claw marks occur well past the toes. HF: Typically show four palm pads and two heel pads. Larger palm pad is heart-shaped while other pads are oblong and more elongate than in the fore foot. Claw marks appear close to the toes. J. Mephitis mephitis (Striped skunk): FF: Five toes are oblong, large palm pad is wider than long. Elongated claws always leave scratch marks well past toes. HF: Large palm pad is adjoined by one or two round heel pads. IDENTIFICATION OF MAMMAL TRACKS 11 A.I. \ \ A. 2. .•». B.1. B.2. FIGURE 2. Mammals with four toes on fore feet and hind feet. A.I. Canis latrans A.2. Urocyon cinereoargenteus B.I. Felis conco/or B.2. Fe/is rufus. 12 CALIFORNIA FISH AND CAME A.I. ' !.«.*' • \ 1 •ft / A.2.'« A.4. ^ " A. 5. V A. 3. •♦ " A. 6. 19, B.I. f.f. h.f. FICURE 3. Mammals with four toes on fore feet and five toes on hind feet. A.I. Tamias spp. A.2. Claucomys sabrinus A.3. Spermophilus lateralis A.4. Spermophilus beecheyi A.5. Tamiasciurus douglasii Kh.Sciurus griseus B.I. Neotoma fuscipes. Tracks A.2. through A.6. exhibit the same pattern as that of A.I. (f.f. = fore feet; h.f. = hind feet). IDENTIFICATION OF MAMMAL TRACKS 13 A. % 4^ , « > ' l.f. I.h. B. C. l.f. D. 1^ 1 l.f I.h, FIGURE 4. Mammals with five toes on fore and hind feet. A. Didelphis virginiana B. Erethizon dorsatum C. Ursus americanus D. Bassariscus astutus E. Procyon lotor. (continued) 14 CALIFORNIA FISH AND GAME F. ^ l.h. H.I. • H.2. 1. 4r I.f. J. / .^» .... •••• Ww r.h. H.3. « l.h. FIGURE 4. (continued) F. Martes pennantiG. Martes americana H.I. Mustela vison H.2. Mustela frenata H.3. Mustela erminea I. Spilogale gracilis J. Mephitis mephitis. (I.f. = left fore; l.h. = left hind; r.f. = right fore; r.h. = right hind). IDENTIFICATION OF MAMMAL TRACKS 15 ACKNOWLEDGMENTS This research was part of a cooperative agreement between the University of California, Berkeley (UCB) and the U.S.D.A. Forest Service (Raphael and Barrett 1984), and was also supported by UC Agricultural Experiment Station Project AES 3501 MS. We would like to thank P. Barrett, J. Brack, C. Brown, E. Hesky, L. Jones, R. LeValley, K. Rosenberg, and H. Welsh for collection of field data, and K. Aubry, S. Martin, and W. Zielinski for providing additional measurements. LITERATURE CITED Barrett, R. H. 1983. Smoked aluminum track plots for determining furbearer distribution and relative abundance. Calif. Fish and Came, 69(3):188-190. Cook, A. H. 1949. Furbearer investigations. New York State Conservation Dept. PR Proj. 1-R, Suppl. G. Final Rep., 57 p. Justice, K. E. 1961. A new method of measuring home ranges of small mammals. J. Mammal., 42:462-470. Lindzey, F.G., S. K. Thompson, and J. I. Hodges. 1977. Scent station index of black bear abundance. J. Wildl. Manage., 41:151-153. Linhart, S.B., and F. F. Knowlton. 1975. Determining the relative abundance of coyotes by scent station lines. Wildl. Soc. Bull., 3:119-124. Lord, R. D., A. M. Vilches, j. I. Maiztegui, and C. A. Soldini. 1970. The tracking board: a relative census technique for studying rodents. J. Mammal., 51 :828-829. Mayer, M. V. 1957. A method for determining the activity of burrowing mammals. J. Mammal., 38:531. Raphael, M.G., and R. H. Barrett. 1984. Diversity and abundance of wildlife in late successional Douglas-fir forests. In New Forests for a Changbing World. Soc. of Am. Foresters, Proc. Pp. 352-360. Wood, J. E. 1959. Relative estimates of fox populations levels. J. Wildl. Manage., 23:53-63. 15 CALIFORNIA FISH AND GAME Calif. Fish and Came 74 ( 1 ) : 1 6-37 1 988 FEEDING ECOLOGY OF TEN SPECIES OF ROCKFISHES (SCORPAENIDAE) FROM THE GULF OF ALASKA' RICHARD J. ROSENTHAL P.O. Box 1044 Cardiff by the Sea, California 92007 VICTORIA MORAN-O'CONNELL Alaska Department of Fish and Game Commercial Fisheries Division 304 Lake St., Room 103 Sitka, Alaska 99835 and MARGARET C. MURPHY National Marine Fisheries Service Auke Bay Laboratory P.O. Box 210155 Auke Bay, Alaska 99821 Summer diets of ten species of rockfish collected in the inshore waters of southeastern Alaska during 1980-62 are presented and compared with studies of rockfish food habits from other regions of the Pacific coast. All species utilized a variety of food and usually capitalized on the most accessible prey types. The bottom-dwelling species were most dependent on detrital based food sources, whereas schooling or pelagic rockfish consumed substantial quantities of zooplankton and fish. A number of common food items (e.g., crabs, shrimps, brittle stars, and fish) were shared by the bottom-dwellers. Pacific sand lance, a key component in the inshore forage base, was the dominant food of the more pelagic black, yellowtail and widow rockfishes. Two other pelagic schoolers, the Puget Sound and dusky rockfishes ate significant amounts of pelagic Crustacea and gelatinous zooplankton. Dietary overlaps for these common pelagic and demersal rockfish species were strong during the three summers of observation. Geographic variations in diet suggested that rockfish were capable of substituting prey as long as these foods are of the same general size and type. INTRODUCTION The assemblage of reef-dwelling fishes that inhabit the exposed waters of the Gulf of Alaska adjacent to the Alexander Archipelago (Figure 1) are visually dominated by rockfish of the genus Sebastes (Carlson and Straty 1981; Rosenthal et al. 1982). Rockfish landings dominate commercial bottomfish catches off California, Oregon and Washington (Gunderson and Lenarz 1980). This group of fish is also important to west coast sport or party-boat fleets (Miller and Geibel 1973). Over the past few decades increased fishing effort has resulted in stock depletions, and produced changes in the size structure of rockfish populations (Love 1980). Despite their commercial value, relatively little information is available in the literature on the foods used by rockfish. Information on the types of prey consumed by this group is critical to understanding the functional role of Sebastes in the nearshore system. ' Accepted for publication July 1987. FEEDING ECOLOGY OF ROCKFISHES 17 97"'40'N- sr'oo'N- 56»00'N 57°40'N 57°00 N -56°00N I36°40'W ISeOQO'W I35°00 W I34°00'W FIGURE 1. Primary study region in the outside waters of southeast Alaska. We present data on the summer food habits of ten species of rockfish found off the west coast of southeast Alaska. Trophic information on co-occurring species was obtained simultaneously to provide preliminary information on niche breadth and dietary overlap. Differences in rockfish diet along the west coast of North America has been linked to the availability of certain kinds of prey and to changes in species composition of the Sebastes community at different latitudes. As most of the shallow exposed waters of the Gulf of Alaska remain in a relatively pristine state, with large stocks of inshore rockfish (Rosenthal et al. 1982), we were able to study fish foraging behavior prior to intense exploitation. 18 CALIFORNIA FISH AND GAME MATERIALS AND METHODS Samples of rockfish were obtained during the summers of 1980-82 while conducting resource assessment studies in the outside waters of southeast Alaska. Automatic electric jigging machines on the 47-ft SEARCHER were the major fishing gear (Rosenthal 1982). Additional specimens were taken with hand spears while scuba diving. All fish collections and samples were taken at depths between 25 and 110 m. Rockfish were identified to species, weighed and measured (fork length) to the nearest centimeter. The stomach contents of each fish were removed at the time of capture and either preserved in 10 percent buffered formalin or examined fresh. When a rockfish regurgitated food at the sea surface, the contents were also recorded. Preserved stomachs were rinsed in fresh water, and the percent fullness was estimated before the contents were removed. Food items were examined under dissecting microscope and identified to the lowest possible taxonomic level. The number, stage of digestion and percent volume of the various prey taxa were recorded for each stomach. Percent volume was used to estimate the amount of a prey item in a stomach when counts were not possible due to advanced stage of digestion. All data were recorded using the NOAA, NODC (1981) taxonomic code. The data were analyzed at the Computer Center of the NMFS Auke Bay Laboratory. We used the program ECO-INDEX to estimate index values (Vadopich and Hoover 1981). Frequency of occurrence was determined for each species on the basis of percentage of samples in which a specific prey type occurred. The percent total count was arrived at by dividing total number of individuals in a prey taxa by total number of individuals counted in all prey groups. A third parameter recorded was percent volume. These three measures were used in calculating an Index of Relative Importance (IRI) as described by Pinkas et al. (1971). Prey categories were ranked according to the computed IRI values and the percentages of the total IRI. Prey categories with the highest numerical rank were considered major food items in the diets of these fish. In this way, more representative prey are not dominated by numerically rare but high biomass prey, or by numerically abundant prey which may contribute little to the total biomass of prey. The ten rockfish species were classified as pelagic or demersal for compar- ison of dietary indices. Classification was based on the position and location of these species in the water column as observed by scuba divers in the nearshore waters of southeast Alaska. Pelagic or schooling species included black, yellowtail, dusky, widow, and Puget Sound rockfishes. China, yelloweye, quillback, copper, and tiger rockfishes were considered demersal or bottom- dwelling species. Niche breadth for the 10 species of rockfishes was measured with three indices: Levin's niche breadth (Hurlbert 1978), Shannon-Weaver niche breadth, and Simpson's niche breadth (Petraitis 1979) (Appendix 1 ). The three indices were used to indicate the degree to which a species was a dietary generalist or specialist. FEEDING ECOLOGY OF ROCKFISHES 19 Dietary overlap for schooling and demersal species was measured with three overlap indices: Schoener's niche overlap, Levin's directional overlap, and Lloyd's directional overlap (Hurlbert 1978) (Appendix 2). Schoener's niche overlap is a measure of "dietary similarity"; similarity in resources consumed, not similarity in consumer electivity. Levin's directional overlap measures the degree species X impinges on species Y and vice versa. Lloyd's directional overlap was re-expressed in percentages to simplify its interpretation: species X or Y consumes * percent of the total volume of dietary items that species X and Y have in common. The overlap indices used do not account for variation in the food resources because data on the availability of food resources in the environment were not obtained. The availability of food resources was therefore assumed equal. Overlap indices measure the degree that two species share a group of common resources or utilize the same parts of the environment. Overlap measures are functions of the proportion of food type / in the diet of consumer j. These proportions depend on the availability of food resources and the consumers' electivity. Differences in resource abundance between environments can yield misleading conclusions when comparing species inhabiting different commu- nities. Similar species may not bear any resemblance if they occur in considerably different environments. Dissimilar species may appear quite similar because of species-environment interactions (Lawlor 1980). Dietary indices for schooling and demersal species were therefore examined separately. RESULTS A total of 1,030 rockfish, representing ten different species was captured and examined for food items. Prey of various stages of digestion and size were found in 635 of the specimens (Figure 2). All collections were made during daylight between 0800 and 2000 hrs. Black Rockfish, Sebastes melanops Black rockfish was observed feeding just beneath the sea surface to depths of at least 40 m. Fish was a key component of the summer diet, and occurred in 98% of the samples (total prey volume of 90%; Table 1 ). Pacific sand lance, Ammodytes hexapterus, (Figure 2) was the most important forage species, and accounted for 67% of the total prey volume and occurred in 60% of the 142 stomachs that contained food (Table 1). Juvenile sablefish, Anoplopoma fimbria, was eaten by 8.5% of the specimens, and accounted for 9.1% of the total percent volume. Other fish identified from stomach contents included Pacific herring, Clupea harengus pallasi; juvenile greenling, f-fexagrammus spp.; juvenile salmon, Oncorhynchus spp., and Puget Sound rockfish, Sebastes emphaeus. Unidentified fish remains accounted for another 10.8% of total percent volume. Zooplankton was also present in the food samples. Crustacea, particularly brachyuran crab larvae (frequency of occurrence 14.8%), and mysids (5.6%) were the predominant invertebrate groups. Other zooplanktors in the diet included pteropods, particularly Limacina sp.; gammarid amphipods; caridean shrimps and thalaceans (Table 1 ). In all, 38 different prey taxa occurred in the summer diet of 5. melanops. 20 CALIFORNIA FISH AND GAME 20mm FIGURE 2. Representative prey items in the summer diets of ten species of rockfishes collected in the inshore waters of southeastern Alaska. A. Pacific sand lance; B. brachyuran crab; C. calanoid copepod; D. brittle star; E. Puget Sound rockfish; and F. caridean shrimp. The importance of fish in the diet of the black rockfish has been substantiated by other investigators. For exannple, Rosenthal (1983) found that 51% of the black rockfish captured in Prince William Sound, Alaska, had eaten Pacific sand FEEDING ECOLOGY OF ROCKFISHES 21 lance. Leaman (1980) observed that 35% of the black rockfish taken in the outside waters of southern British Columbia had consumed herring. Additional studies by Moulton (1977) in northern Puget Sound concluded that 53% of the 5. melanops examined had fish remains in their stomachs. Further south off Oregon, Steiner (1978) found that smelt {Allosmerus spp.), herring, anchovy (Engraulis mordax), and juvenile rockfishes comprised 71.4% of the total volume of black rockfish stomachs. TABLE 1. Food of the Black Rockfish. % Freq Prey occur Phaeophyta 0.007 Laminariales 0.007 Nereocystis luetkeana 0.007 Siphonophora 0.007 Beroidae 0.007 Gastropoda 0.007 Limacinidae 0.021 Cymnosomata (Clione) 0.007 Crustacean remains 0.028 Barnacle cirri 0.007 Mysidacea 0.021 Mysidae 0.035 Tanaidacea 0.007 Idoteidae 0.007 Isopoda 0.007 Idotea fewkesi 0.007 Gammaridae 0.042 Hyperiidae 0.021 Caprellidae 0.007 Caridea 0.021 Anomuran 0.007 Paguridae 0.007 Brachyura (larvae) 0.148 Salpidae 0.007 Fritillariidae 0.007 Fish remains 0.239 Clupea harengus 0.014 Juvenile salmonidae 0.021 Sebastes emphaeus 0.007 Juvenile Hexagrammidae 0.007 Hexagrammos decagrammus . . 0.007 Anoplopoma fimbria 0.085 Ammodytes hexapterus 0.599 Celatinuous zooplankton, unid.. 0.014 Spines, unid 0.007 Scale, unid 0.042 Stone 0.007 Organic matter 0.085 Total 1.000 Total % Total Total Average %Total Relative count count volume volume volume importance 0 0.0000 20.00 0.14 0.0018 0.1244 0 0.0000 20.00 0.14 0.0018 0.1244 1 0.0003 80.00 0.56 0.0071 0.5202 1 0.0003 49.50 0.35 0.0044 0.3304 1 0.0003 10.00 0.07 0.0009 0.0846 1 0.0003 0.50 0.00 0.0000 0.0255 296 0.0940 40.50 0.29 0.0036 20.6148 200 0.0635 10.00 0.07 0.0009 4.5349 3 0.0010 114.50 0.81 0.0101 3.1182 1 0.0003 1.00 0.01 0.0001 0.0286 489 0.1553 110.25 0.78 0.0097 34.8653 1315 0.4176 283.45 2.00 0.0250 155.8585 1 0.0003 0.25 0.00 0.0000 0.0239 1 0.0003 37.50 0.26 0.0033 0.2556 1 0.0003 1.00 0.01 0.0001 0.0286 1 0.0003 2.00 0.01 0.0002 0.0348 16 0.0051 22.95 0.16 0.0020 3.0037 5 0.0016 0.85 0.01 0.0001 0.3513 4 0.0013 2.50 0.01 0.0001 0.1050 10 0.0032 9.00 0.06 0.0008 0.8389 4 0.0013 1.00 0.01 0.0001 0.0957 3 0.0010 0.50 0.00 0.0000 0.0702 183 0.0581 112.15 0.79 0.0099 100.5976 0 0.0000 15.00 0.11 0.0013 0.0933 5 0.0016 3.00 0.02 0.0003 0.1305 10 0.0032 1223.25 8.61 0.1081 266.3982 2 0.0006 100.00 0.70 0.0088 1.3339 4 0.0012 132.50 0.93 0.0117 1.2853 1 0.0003 40.00 0.28 0.0035 0.2713 1 0.0003 75.00 0.53 0.0066 0.4890 1 0.0003 40.00 0.28 0.0035 0.2713 28 0.0089 1034.00 7.28 0.0914 84.7222 544 0.1728 7584.75 53.41 0.6702 5045.7208 0 0.0000 33.75 0.24 0.0030 0.4200 2 0.0006 1.50 0.01 0.0001 0.0541 11 0.0035 5.75 0.04 0.0005 1.6907 1 0.0003 5.00 0.04 0.0004 0.0535 2 0.0006 94.60 0.67 0.0084 7.6004 3149 1.0000 11317.50 79.70 1.0000 5736.1698 142 = Total number of stomachs 38 — Number of prey taxa 22 CALIFORNIA FISH AND GAME Yellowtail Rockfish, Sebastes flavidus Yellowtail rockfish was found in offshore waters near the surface to depths of 110 m. Fish was the most important prey in the summer diets of 5. flavidus. Pacific sand lance occurred in 44.6% of the 112 stomachs (Table 2). The IRI for this food category was 2,645.8, and the percent total volume was 47.9. Juvenile sablefish; larval gadids; unidentified greenlings and rockfish were also present. Unidentified fish remains accounted for another 1 1.6% of the total volume, and occurred in 23.0 of the specimens. Frequency of occurrence and percent total volume was also significant for brachyuran crab larvae (36.6% and 7.9%, respectively), mysids (16.1% and 10.8%), pteropods (LImaclna and Cllone), and caridean shrimp (Table 2). TABLE 2. Food of the Yellowtail Rockfish. % Freq Prey occur Polychaete 0.009 Gymnosomata (Clione) 0.009 Limacina 0.072 Crustacean remains 0.1 1 6 Mysidacea 0.009 Mysidae 0.152 Amphipoda 0.009 Gammaridae 0.071 Hyperiidae 0.036 Caridea 0.080 Anomuran (larvae) 0.009 Paguridae ( larvae) 0.009 Brachyura (larvae) 0.366 Salpidae 0.018 Fish remains 0.232 Gadid Larvae 0.063 Sebastes sp 0.009 Hexagrammidae 0.009 Anoplopoma fimbria 0.045 Ammodytes hexapterus 0.446 Gelatinous Zooplankton, unid. . 0.018 Organic matter 0.054 Total 1.000 Total %Total Total Average %Total Relative count count volume volume volume importance 1 0.0006 20.00 0.18 0.0023 0.2661 260 0.0830 50.00 0.45 0.0059 7.9371 833 0.2660 462.50 4.13 0.0541 199.7238 22 0.0070 81.25 0.73 0.0095 19.1988 0 0.0000 1.00 0.01 0.0001 0.0104 1015 0.3242 921.50 8.23 0.1079 655.8381 2 0,0006 50.00 0.45 0.0059 0.5798 20 0.0064 8.75 0.08 0.0010 5.2945 5 0.0016 6.50 0.06 0.0008 0.8422 15 0.0048 169.50 1.51 0.0198 19.7989 53 0.0169 4.00 0.04 0.0005 1.5532 8 0.0026 8.00 0.07 0.0009 0.3118 464 0.1482 676.00 6.04 0.0792 832.2721 3 0.0010 50.00 0.45 0.0059 1.2166 26 0.0083 991.75 8.85 0.1161 288.8647 35 0.0112 207.50 1.85 0.0243 22.1725 1 0.0003 20.00 0.18 0.0023 0.2376 1 0.0003 60.00 0.54 0.0070 0.6558 10 0.0032 497.00 4.44 0.0582 27.4065 356 0.1137 4090.25 36,52 0.4790 2645.77540 0 0.0000 27.00 0.24 0.0032 0.5646 0 0.0000 137.50 1.23 0.0161 8.6254 3131 1.0000 8540.00 76.25 1.0000 4739.1459 112 = Total number of stomachs 22 = Number of prey taxa Moulton (1977) found fish remains in 52.5% of the stomachs from Puget Sound. However, megalops and zoeal crab larvae occurred most frequently (66.5%). Euphausiids, mysids and chaetognaths comprised the other major food categories. Pereyra et al. (1969) found lanternfish, sergestid shrimps and euphausiids in the stomachs of yellowtail rockfish taken while trawling at depths of 58 to 70 m, in the Columbia River Canyon off Washington. These studies in Washington and observations made off southeast Alaska indicate that 5. flavidus forage throughout the water column, with fish being the principal food item. FEEDING ECOLOGY OF ROCKFISHES 23 Dusky Rockfish, Sebastes ciliatus Dusky rockfish pick and seize food items that are either drifting or suspended in the water column. Off southeast Alaska, the diet was heavily comprised of invertebrate zooplankton (Table 3). Brachyuran crab larvae, both megalops and zoea, were the most important food items taken during summer (Figure 2). Crab larvae occurred in 34% of the 44 stomachs containing food, and comprised 75.6% of the total number of prey counted. Total volume of crab larvae was 25.4% with an IRI of 3,442.9. Salps, hyperiid and gammarid amphipods, calanoid copepods, crustacean remains, jellyfish and pteropods were also consumed. Fish were rarely eaten. For example, the frequency of sand lance was only 6.8% in these samples. Simenstad et al. (1977) reported that mysids, amphipods, copepods, polychaete worms and sand lance were found in the stomachs of 5. ciliatus captured in the shallow subtidal zone off Amchitka Island. Northeast of the Aleutian Chain at latitude 60°N in Prince William Sound, Rosenthal (1983) recorded the major prey groups from stomach samples of dusky rockfish to be copepods, pteropods, larval fishes, mysids, chaetognaths and tomopterid worms. TABLE 3. Food of the Dusky Rockfish. % Freq Prey occur Scyphozoan medusae 0.045 Hoplonemertea 0.023 Polychaeta 0.068 Limacina 0.045 Crustacea remains 0.1 14 Copepoda 0.045 Mysidacea 0.023 Mysidae 0.045 Amphipoda 0.023 Gammaridae 0.1 14 Hyperiidae 0.1 1 4 Caprellidae 0.023 Caridea 0.023 Anomuran (larvae) 0.023 Brachyura (larvae) 0.341 Salpidae 0.114 Fish remains 0.091 Cadid larvae 0.045 Ammodytes hexapterus 0.068 Gelatinous zooplankton unid. . . 0.273 Scale, unid 0.045 Organic material 0.228 Total 1.000 Total %Total Total Average %Total Relative count count volume volume volume importance 1 0.0058 21.50 0.49 0.0209 12.1308 1 0.0058 0.00 0.00 0,0000 1.3214 1 0.0058 1.75 0.04 0.0017 5.1225 3 0.0087 5.25 0.12 0.0051 6.2809 2 0.0058 29.25 0.66 0.0284 38.8773 4 0.0116 25.75 0.59 0.0250 16.6490 1 0.0029 2.00 0.05 0.0019 1.1020 2 0.0058 0.70 0.02 0.0007 2.9516 2 0.0058 0.25 0.01 0.0002 1.3765 5 0.0145 15.25 0.35 0.0148 33.3417 12 0.0349 11.00 0.25 0.0107 51.7765 1 0.0029 1.25 0.03 0.0012 0.9365 1 0.0029 1.25 0.03 0.0012 0.9365 2 0.0058 0.50 0.01 0,0005 1.4317 260 0.7558 261.75 5.95 0.2541 3442.9778 22 0.0640 30.30 0.69 0.0294 106.1034 1 0.0029 145.00 3.30 0.1408 130.6215 4 0.0116 77.00 1.75 0.0748 39.2660 8 0.0233 172.00 3.91 0.1670 129.7133 4 0.0116 122.25 2.78 0.1187 355.4106 4 0.0116 10.50 0.24 0.0102 9.9191 0 0.0000 95.50 2.17 0.0928 175.5295 344 1.0000 1030.00 23.41 1.0000 4563.7762 44 - Total number of stomachs 22 = Number of prey taxa Widow Rockfish, Sebastes entomelas Widow rockfish stomachs contained substantial numbers of fish, crustaceans and gelatinous plankton (Table 4). A total of 23 different prey taxa were found in the 27 stomachs containing food. Fish accounted for 78.8% of the total 24 CALIFORNIA FISH AND GAME volume. A major prey item both by number and percent volume was sand lance with an IRI of 1014.2 (Table 4). Most of the sand lance were eaten whole. Juvenile sablefish and walleye pollock, Theragra cha/cogramma, also occurred in these same stomach samples. TABLE 4. Food of the Widow Rockfish. % Freq Prey occur Caprellidae 0.037 Cnidaria 0.074 Physonectid colony 0.037 Aequorea aequorea 0.037 Polychaete 0.037 Limacinidae 0.074 Octopus sp 0.037 Crustacean remains 0.1 1 1 Copepoda 0.037 Mysidae 0.111 Cammaridae 0.074 Hyperiidae 0.148 Caridea 0.037 Brachyura (larvae) 0.111 Salpidae 0.111 Fish remains 0.185 Theragra cha/cogramma 0.074 Anoplopoma fimbria 0.037 Ammodytes hexapterus 0.259 Gelatinous zooplankton, unid. . 0.185 Gill, unid 0.037 Eggs, unid 0.037 Organic material 0.022 Total 1.000 Tota/ % Total Total Average % Total Relative count count volume volume volume importance 0 0.0000 1.25 0.05 0.0006 0,2055 0.0044 25.00 0.93 0.01 1 1 11,4693 0.0000 22.50 0.83 0.0100 3,6992 0.0132 75.00 2.78 0.0333 17,2039 0,0000 7.50 0.28 0.0033 1,2331 0.0219 5.25 0.19 0.0023 17,9706 0.0044 10.00 0.37 0.0044 3,2685 0.0044 92.75 3.44 0.0412 50,6199 0.0044 0.25 0.01 0.0001 1.6655 0.0132 3.00 0.11 0.0013 16.0996 0.0132 1.75 0.06 0.0008 10.3220 87 0.3816 9.00 0.33 0.0040 571.2208 0.0000 7.50 0.28 0.0033 1.2331 0.0175 2.75 0.10 0.0012 20.8495 0.0307 53.50 1.98 0.0237 60.5006 0.0175 1016.00 37,63 0.4510 867.6818 0.0351 109.00 4.04 0.0484 61.8319 0.0044 74.25 2.75 0.0330 13.8317 31 0.1360 575.00 21.30 0,2552 1014,2443 66 0.2895 95.50 3.54 0.0424 614,5673 0.0044 0.50 0.02 0.0002 1 ,7066 0.0044 0.30 0.01 0,0001 1,6738 0 0.0000 65.20 2.41 0,0289 64.3165 228 1.0000 2252.75 83.44 1,0000 3427.4148 27 = Total number of stomachs 23 = Number of prey taxa The frequency of Crustacea zooplankton (62.9%) was also high. Hyperiid amphipods were the numerical dominant from this group. Others included brachyuran crab larvae, mysids, gammaridean amphipods and caridean shrimps. Gelatinous plankton — a general category containing groups such as Aequorea (a cnidarian), physonectid colonies, salps, and scyphozoan medusae or "jellyfish" — occurred frequently enough to have a combined IRI of 707.4 (Table 4). Gunderson (pers. comm.) has found that widow rockfish aggregate off bottom in large schools during nocturnal hours, but disperse to feed during daylight. Widow rockfish feed in the daytime off southeast Alaska, as a high percentage of specimens captured during this time period had fresh food in their stomachs. Puget Sound Rockfish, Sebastes emphaeus Puget Sound rockfish school off bottom in the water column during daylight hours and feed on a variety of small zooplankters. Seventeen different prey types were identified from the stomach contents of 19 specimens that retained FEEDING ECOLOGY OF ROCKFISHES 25 food in their stomachs after capture (Table 5). Calanoid copepods dominated these samples both in number, frequency and percent volume (Figure 2). The IRI for copepods alone was 3206.8. Other important prey included gammarid amphipods, unidentified gelatinous plankton, physonectid (siphonophore) fragments, pteropods, salps, mysids and fish eggs. TABLE 5. Food of the Puget Sound Rockfish. % Freq Prey occur Physonectid colony 0.053 Calliostoma annulatum 0.053 Margaritas sp 0.053 Maigarites puppillus 0.053 Crustacean remains 0.105 Cladocera 0.053 Ostracoda 0.105 Copepoda 0.368 Mysidae 0.053 Gammaridae 0.211 Caprellidae 0.21 1 Caridea 0.053 Salpidae 0.053 Fish remains 0.053 Gelatinous zooplankton, unid. . 0.263 Eggs, unid 0.053 Organic material 0.158 Total 1.000 19 = Total number of stomachs 17 = Number of prey taxa Total % Total Total A verage %Total Relative count count volume volume volume importance 1 0.0083 22.50 1.18 0.0244 17.1763 1 0.0083 25.00 1.32 0.0271 18.6014 1 0.0083 10.00 0.53 0.0108 10.0504 1 0.0083 15.00 0.79 0.0162 12.9007 1 0.0083 8.00 0.42 0.0087 17.8205 1 0.0083 30.00 1.58 0.0325 21.4518 4 0.0331 6.25 0.33 0.0068 41.9236 70 0.5785 269.50 14.18 0.2919 3206.7958 1 0.0083 7.50 0.39 0.0081 8.6252 11 0.0909 57.50 3.03 0.0623 322.5033 3 0.0248 196.75 10.35 0.2131 341.2779 1 0.0083 100.00 5.26 0.1083 61.3566 2 0.0165 37.50 1.97 0.0406 30.0770 2 0.0165 5.00 0.26 0.0054 11.5498 16 0.1322 101.25 5.33 0.1097 636.5746 5 0.0413 2.50 0.13 0.0027 23.1738 0 0.0000 29.00 1.53 0.0314 49.5959 121 1.0000 923.25 48.59 1.0000 4831.4546 China Rockfish, Sebastes nebulosus Forty-four different prey taxa were found in the stomachs of 59 China rockfish (Table 6). Although a wide variety of food items were taken as prey, only seven groups occurred in 10% or more of the samples. These included: brittle star, Ophiopholis aculeata (Figure 2); rock crab. Cancer oregonensis; decorator crab, Pugettia gracilis; brachyuran crab larvae; caridean shrimps; hermit crabs and fish remains. Most observed feeding was directed toward the bottom, as China rockfish forage on sedentary invertebrates comprising the living turf. Brittle star, Op/iiopho/is aculeata, was the dominant food item in terms of frequency, occurring in 78% of the stomach samples with a total volume of 38.5% (Table 6). Brittle stars were usually eaten whole, and as many as 65 O. aculeata have been recorded from a single stomach sample. Crustacea was the next most important group both by number and volume with a combined frequency of occurrence of 89%. Fish were only occasionally found in these samples. These data from southeast Alaska agree with the findings of Rosenthal (1983) for China rockfish taken in the exposed water of the northeastern Gulf of Alaska where brittle stars occurred in 94.7% of the stomachs sampled. Rosenthal also found that 5. nebulosus consumed decorator crabs; small rock crabs and caridean shrimps. China rockfish, feeding off the coast of Oregon, consumed 26 CALIFORNIA FISH AND CAME Crustacea, particularly decorator or spicier crabs and Cancer oregonensis (Steiner 1978). TABLE 6. Food of the China Rockfish. % Freq Total Prey occur count Rhodophyta 0.017 2 A/g/aophenia sp 0.017 0 Rhynchocoela 0.017 1 Ha/iotis kamtschatkana 0.034 8 Homalopoma luridum 0.01 7 1 Velutina sp 0.034 2 Polyplacophora 0.034 2 Cyanoplax dentiens 0.01 7 1 Balanus crenatus 0.017 1 Mysidaceae 0.017 3 Mysidacea 0.017 37 Cammaridae 0.017 0 Caridea 0.153 10 Hippolytidae 0.017 1 Heptacarpus sp 0.017 1 Panda/us spp 0.017 3 Crab, unid 0.017 1 Paguridae 0.034 4 Pagurus spp 0.068 5 Rhinolithodes wosnessenskii .. 0.017 1 Acantholithodes hispidus 0.034 2 Phyllolithodes papillosus 0.017 1 Cryptolithoides sp 0.01 7 1 Cryptolithodes typicas 0.01 7 1 Petrolisthes eriomerus 0.01 7 1 Petrolithes cinctipes 0.01 7 1 Brachyura 0.203 22 Hyas lyratus 0.051 4 Pugettia sp 0.034 3 Pugettia richii 0.034 2 Pugettia gracilis 0.119 10 Scyra acutifrons 0.051 4 Cancridae 0.068 3 Cancer oregonensis 0.305 33 Ophiuroidae 0.051 31 Ophiopholis aculeata 0.780 220 Boltenia villosa 0.01 7 1 Fish remains 0.102 4 Hemilepidotus jordani 0.017 1 Ammodytes hexapterus 0.068 3 Shell, unid 0.017 0 Scale, unid 0.017 0 Stone 0.051 3 Organic material 0.068 \ Total 1.000 436 %Total count 0.0046 0.0000 0.0023 0.0183 0.0023 0.0046 0.0046 0.0023 0.0023 0.0069 0.0849 0.0000 0.0229 0.0023 0.0023 0.0069 0.0023 0.0092 0.0115 0.0023 0.0046 0.0023 0.0023 0.0023 0.0023 0.0023 0.0505 0.0092 0.0069 0.0046 0.0229 0.0092 0.0069 0.0757 0.0711 0.5046 0.0023 0.0092 0.0023 0.0069 0.0000 0.0000 0.0069 0.0023 Total volume 7.50 2.00 1.25 88.75 1.00 5.00 35.00 2.50 7.50 2.50 3.75 1.25 39.25 7.50 12.50 20.00 5.00 23.50 11.75 70.00 170.00 15.00 80.00 33.75 15.00 0.00 82.50 38.75 150.00 22.50 203.75 57.50 11.75 312.50 109.00 1229.75 5.00 90.75 8.00 50.00 5.00 0.50 125.00 34.00 1.0000 3197.50 A verage volume 0.13 0.03 0.02 1.50 0.02 0.08 0.59 0.04 0,13 0.04 0.06 0.02 0.67 0.13 0.21 0.34 0.08 0.40 0.20 1.19 2.88 0.25 1.36 0.57 0.25 0.00 1.40 0.66 2.54 0.38 3.45 0.97 0.20 5.30 1.85 20.84 0.08 1.54 0.14 0.85 0.08 0.01 2.12 0.58 54.19 % Total volume 0.0023 0.0006 0.0004 0.0278 0.0003 0.0016 0.0109 0,0008 0.0023 0.0008 0.0012 0.0004 0.0123 0.0023 0.0039 0.0063 0.0016 0.0073 0.0037 0,0219 0.0532 0.0047 0.0250 0.0106 0,0047 0,0000 0,0258 0.0121 0.0469 0.0070 0,0637 0,0180 0,0037 0.0977 0,0341 0,3846 0.0016 0.0284 0.0025 0.0156 0,0016 0.0002 0,0391 0,0106 Relative importance 1,1750 0,1060 0,4550 15,6287 0,4417 2,0850 5.2655 0,5213 0.7863 1.2987 14,5822 0,0662 53,7117 0.7863 1.0513 2.2264 0,6538 5,6013 10.2662 4,0993 19,5775 1,1836 4.6293 2,1777 1,1839 0,3887 155,1053 10,8270 18,2347 3,9403 102,8139 13.8087 7,1563 529,0800 53,4865 6392,6249 0,6538 38,1924 0,8128 15,2664 0,2650 0,0265 23.3765 8,7640 1,0000 8064.3842 59 = Total number of stomachs 44 = Number of prey taxa Yelloweye Rockfish, Sebastes ruberrimus The yelloweye rockfish is a large predatory reef fish that usually feeds close to the bottom. It has been observed underwater capturing smaller reef fishes FEEDING ECOLOGY OF ROCKFISHES 27 with rapid bursts of speed and agility. Stomach samples were examined from 48 specimens. Of these, 33 regurgitated prey at the surface after capture. The analysis (Table 7), however, is only for 15 specimens with stomachs and food items completely intact. Fishes were the major food item in their diet. Rockfish prey included Puget Sound rockfish; quillback rockfish, 5. maliger; rosethorn rockfish, 5. helvomaculatus; redstripe rockfish, 5. proriger; and juvenile yellow- eye rockfish. Other prey included juvenile gadids, sand lance, herring, and lumpsucker (family: Cyclopteridae). Puget Sound rockfish was the most important prey both by number and volume in the regurgitated samples (Figure 2). As many as three adult Puget Sound rockfish have been found in a single yelloweye rockfish stomach. FHerring and other unidentified fish remains comprised 27 and 30% of the total volume, respectively (Table 7). Caridean shrimp; a lithodid crab, Acantholithodes hispidis; C. oregonensis; green sea urchin, Strongylocentrotus droebachiensis; gastropod snails, and lingcod, Ophlodon elongatus, eggs were also consumed by yelloweye rockfish. TABLE 7. Food of the Yelloweye Rockfish. % Freq Total Prey occur count Fish remains 0.067 0 Caridea 0.067 4 Panda/us spp 0.133 2 Panda/us tridens 0.067 1 Brachyura 0.067 0 Fish remains 0.333 7 Clupea harengus 0.333 5 Gadid larvae 0.067 4 Sebastes ipp 0.133 0 Sebastes emphaeus 0.133 2 Cyclopteridae 0.067 1 Ammodytes hexapterus 0.133 40 Total 1.000 66 %Total count 0,0000 0.0606 0.0303 0,0152 0.0000 0.1061 0.0758 0,0606 0,0000 0,0303 0,0152 0,6061 Total Average %Total Relative volume volume importance volume 30,00 4.00 30.00 25.00 1.00 380,00 350,00 50,00 125.00 80.00 60,00 140,00 1,0000 1275,00 2,00 0.27 2.00 1.67 0.07 25.33 23.33 3,33 8.33 5.33 4,00 9,33 85.00 0,0235 0,0031 0.0235 0.0196 0.0008 0.2980 0.2745 0.0392 0.0980 0.0627 0.0471 0,1098 15,6863 42,4955 71.7766 23.1729 0.5229 1346,9994 1167.5579 66.5478 130,7190 124.0642 41,4736 954,4860 1,0000 3985,5021 15 = Total number of stomachs 12 = Number of prey taxa Rosenthal (1983) found juvenile flounder (family: Pleuronectidae); snail, Calliostoma ligatum; lithodid crabs, Placetron wosnessenskii; and fish remains in the stomachs of yelloweye rockfish taken in Prince William Sound. The major food items in terms of percent biomass off Oregon (Steiner 1978) included cancroid crabs; cottids, Artedius spp.; righteye flounders; adult rockfish; and pandalid shrimps. Quillback Rockfish, Sebastes maliger Quillback rockfish is another solitary reef-dwelling fish that lives close to, or on the bottom. Sixty-seven of the specimens examined had food present in their stomachs. A list of 45 identifiable prey taxa is presented in Table 8. Crustacea dominated these prey groups both by number, frequency and volume. Caridean shrimps; cancroid crabs; decorator crabs of the genera Pugettia, Oregonias, Scyra and Chorilla; hermit crabs; mysids and sphaeromatid isopods were the major Crustacea in terms of IRI. Caridean shrimps occurred in 37.4% of the samples. 28 CALIFORNIA FISH AND CAME TABLE 8. Food of the Quillback Rockfish. % Freq Prey occur Sertularella 0.015 Polychaete 0.030 Bivalve shell 0.015 Musculus sp 0.01 5 Pectinidae 0.015 Octopus sp 0.015 Crustacean remains 0.075 Mysidae 0.075 Tanaidacea 0.01 5 Sphaeromatidae 0.090 Gammaridae 0.075 Caprellidae 0.015 Thysanoessa raschii 0.01 5 Caridae 0.299 Panda/us spp 0.060 Crangonidae 0.01 5 Crab, unid 0.119 Anomuran 0.01 5 Pagurus spp 0.060 Acantholithodes hispidus 0.030 Cryptolitholdes sitchensis 0.01 5 Rhinolithodes wosnessenskii . . . 0.01 5 Brachyura (larvae) 0.015 Oregon/a gracilis 0.030 Pugettia spp 0.030 Pugettia gracilis 0.075 Scyra acutifrons 0.01 5 Chorilla longipes 0.030 Cancridae 0.045 Cancer branneri 0.01 5 Cancer oregonensis 0.104 Ectopocta 0.030 Microporina borealis 0.01 5 Myriozoum 0.01 5 Ophiopholis aculeata 0.045 Holothuroidea 0.01 5 Boltenia villosa 0.015 Salpidae 0.030 Fish remains 0.209 Clupea harengus 0.030 Sebastes emphaeus 0.030 Ammodytes hexapterus 0. 1 49 Shell, unid 0.075 Stone 0.060 Organic material 0.030 Total 1.000 Total %Total Total Average %Total Relative count count volume volume volume importance 0 0.0000 1.00 0.01 0.0003 0.0421 2 0.0041 13.00 0.19 0.0037 2.3083 1 0.0020 2.50 0.04 0.0007 0.4086 2 0.0020 3.75 0.06 0.0011 0.4613 0 0.0000 10.00 0.15 0.0028 0.4211 1 0.0020 85.00 1.27 0.0240 3.8826 3 0.0061 35.00 0.52 0.0099 11.9195 140 0.2846 183.75 2.74 0.0518 251.0406 1 0.0020 1.50 0.02 0.0004 0.3665 10 0.0203 108.50 1.62 0.0306 45.6147 10 0.0203 18.50 0.28 0.0052 19.0631 5 0.0102 3.75 0.06 0.0011 1 .6747 88 0.1789 100.00 1.49 0.0282 30.9067 37 0.0752 273.45 4.08 0.0771 454.7822 5 0.0102 95.00 1.42 0.0268 22.0687 0 0.0000 5.00 0.07 0.0014 0.2105 5 0.0102 121.00 1.81 0.0341 52.8961 0 0.0000 25.00 0.37 0.0071 1.0527 20 0.0407 74.25 1.11 0.0209 36.7753 2 0.0041 110.00 1.64 0.0310 10.4775 1 0.0020 0.00 0.00 0.0000 0.3034 1 0.0020 40.00 0.60 0.0113 1.9877 2 0.0041 10.00 0.15 0.0028 1.0278 1 0.0020 51.25 0.76 0.0145 4.9229 2 0.0041 45.00 0.67 0.0127 5.0033 8 0.0163 116.25 1.74 0.0328 36.6104 1 0.0020 0.00 0.00 0.0000 0.3034 2 0.0041 80.00 1.19 0.0226 7.9509 8 0.0163 148.75 2.22 0.0420 26.0719 1 0.0020 30.00 0.45 0.0850 1.5666 12 0.0244 261.25 3.90 0.0737 102.4894 1 0.0020 2.25 0.03 0.0006 0.7962 0 0.0000 3.75 0.06 0.0011 0.1579 0 0.0000 11.25 0.17 0.0032 0.4737 27 0.0549 102.50 1.53 0.0289 37.5208 1 0.0020 30.00 0.45 0.0085 1.5666 0 0.0000 1.00 0.01 0.0003 0.0421 5 0.0102 25.00 0.37 0.0071 5.1391 7 0.0142 361.00 5.39 0.1018 242.5490 2 0.0041 170.00 2.54 0.0480 15.5306 2 0.0041 110.00 1.64 0.0310 10.4775 70 0.1423 602.50 8.99 0.1700 466.0604 3 0.0061 9.75 0.15 0.0028 6.6032 3 0.0061 15.50 0.23 0.0044 6.2511 1 0.0020 47.50 0.71 0.0134 4.6071 492 1.0000 3544.45 52.90 1.0000 1932.3860 67 = Total number of stomachs 45 = Number of prey taxa Quillback rockfish are not only skilled at foraging for sedentary or slow- moving invertebrates, but are also adept at capturing small fish from the water column. Sand lance and fish remains had an IRI of 466.1 and 242.5, respectively. Sand lance occurred in 15% of the specimens. Herring and small rockfish were also present in these samples. FEEDING ECOLOGY OF ROCKFISHES 29 Dock shrimp, Pandalus danae; brachyuran crabs; gammarid amphipods; euphausiids and calanoid copepods were the principal components of the diet of quillback rockfish taken from an artificial reef in Puget Sound, Washington (Hueckel and Stayton 1982). The results of their study concur with our dietary observations for 5. maliger off southeast Alaska, despite the wide differences in habitats and geographic location. Copper Rockfish, Sebastes caurinus Copper rockfish usually forage close to the bottom, although some individ- uals were observed to catch small fish that hovered in the water column. Twenty-five specimens were collected and examined for food items. All of the sampled stomachs contained some type of food, with 18 different prey taxa represented (Table 9). The diet consisted heavily of Crustacea and small fish. Eleven percent of the identifiable food items were shrimp, and another 23% were fish. Sand lance occurred in 32% of the samples, and this accounted for 32.8% of the total volume. The IRI value of sand lance was 1681.2 (Table 9). Puget Sound rockfish and juvenile greenlings were also preyed upon by copper rockfish. Other noteworthy prey include mysids (IRI 516.5) and cancroid crabs such as Cancer oregonensis and C. branneri. TABLE 9. Food of the Copper Rockfish. Prey Octopodidae Crustacean remains (A) . Balanidae Mysidae Rocinela belliceps Caridea Pandalus spp Brachyura (larvae) Cancridae Cancer branneri Cancer oregonensis Microporina boreaiis Fish remains Sebastes emphaeus Juvenile Hexagrammidae. Ammodytes hexapterus. . Shell Stone Total % Freq Total %Total Total Average % Total Relative occur count count volume volume volume importance 0.040 1 0.0052 30.00 1.20 0.0249 12.0207 0.120 1 0.0052 20.50 0.82 0.0170 26.6114 0.040 1 0.0052 3.75 0.15 0.0031 3.3161 0.080 107 0.5544 110.00 4.40 0.0912 516.4767 0.040 1 0.0052 15.00 0.60 0.0124 7.0466 0.240 20 0.1036 67.50 2.70 0.0560 383.0052 0.080 2 0.0104 30.00 1.20 0.0249 28.1865 0.080 2 0.0104 13.00 0.52 0.0108 16.9119 0.120 3 0.0155 67.50 2.70 0.0560 85.8031 0.040 1 0.0052 75.00 3.00 0.0622 26.9430 0.080 4 0.0207 81.25 3.25 0.0674 70.4663 0.040 0 0.0000 2.00 0.08 0.0017 0.6632 0.120 1 0.0052 42.50 1.70 0.0352 48.4974 0.080 4 0,0207 170.00 6.80 0.1409 129.3264 0.040 2 0.0104 50.00 2.00 0.0415 20.7254 0.320 38 0.1969 396.25 15.85 0.3285 1681.2435 0.120 2 0.0104 4.75 0.19 0.0039 17.1606 0.120 3 0.0155 27.25 1.09 0.0226 45.7617 1.000 193 1.0000 1206.25 48.25 1.0000 3120.1658 25 = Total number of stomachs 18 = Number of prey taxa Fish, shrimp and crab were the major components of the diet of 5. caurinus captured from Puget Sound (Moulton 1977, Patten 1973). Off northern California, brachyuran crabs, gammarid amphipods, caprellids, caridean shrimps and fish were the most frequently eaten food items. In that study, copper rockfish were examined from an artificial reef in Humboldt Bay (Prince and Gotshall 1976). Based on previous findings and information collected by us off southeast Alaska, copper rockfish feed heavily on crabs, shrimps and fish. This preference held despite major differences in habitat and distance between the aforementioned locations. 30 CALIFORNIA FISH AND GAME Tiger Rockfish, Sebastes nigrocinctus Tiger rockfish forage by picking food items from the sea floor. A total of 19 different prey taxa were found in the stomachs of 16 tiger rockfish (Table 10). Caridean shrimp (Figure 2) was the most important prey group in terms of frequency of occurrence (50%), percent total volume (19.5%) and number (Table 10). The IRI value for this group alone was 3144.4. Stomach samples also contained substantial amounts of brachyuran crabs, particularly rock crab, C. oregonensis, and decorator crabs of the genus Pugettia, Oregonia, and Scyra. Other important foods included gammarid amphipods, hermit crabs and small fishes such as herring and juvenile rockfish. Despite the occurrence of fish in the stomachs, tiger rockfish seemed to be more adept at feeding on macroinver- tebrates associated with the bottom than they were at catching fish from the water column. This was substantiated by diving observations and from samples taken from commercial fishing boats based out of Sitka during 1980-82. TABLE 10. Food of the Tiger Rockfish. % Freq Prey occur Nereidae 0.063 Cammaridae 0.063 Caridea 0.500 Pandalidae 0.063 Pandalus montagui. 0.063 Pandalus danae 0.063 Paguridae 0.063 Pagurus sp 0.125 Brachyura 0.063 Oregonia gracilis 0.063 Pugettia gracilis 0.063 Scyra acutifrons 0.063 Cancer oregonensis 0.063 Ophiopholis aculeata 0.063 Fish remains 0.188 Clupea harengus 0.063 Sebastes sp 0.063 Celatenous plankton 0.063 Organic matter 0.125 Total 1.000 Total %Total Total Average %Total Relative count count volume volume volume importance 0 0.0000 23.75 1.48 0.0403 25.1589 0.0755 15.00 0.94 0.0254 63.0596 23 0.4340 115.00 7.19 0.1949 3144.3876 0.0377 10.00 0.63 0.0169 34.1781 0.0189 25.00 1.56 0.0424 38.2755 0.0566 0.25 0.02 0.0004 35.6422 0.0189 10.00 0.63 0.0169 22.3857 0.0377 48.50 3.03 0.0822 149.9240 0.0189 1.25 0.08 0.0021 13.1166 0.0189 5.00 0.31 0.0085 17.0891 0.0189 2.50 0.16 0.0042 14.4408 0.0566 10.00 0.63 0.0169 45.9706 0.0377 22.50 1.41 0.0381 47.4197 0.0755 18.75 1.17 0.0318 67.0321 0.0189 42.50 2.66 0.0720 170.4409 0.0189 40.00 2.50 0.0678 54.1653 0.0189 80.00 5.00 0.1356 96.5382 0.0000 10.00 0.63 0.0169 10.5932 0.0377 110.00 6.88 0.1864 280.2207 53 1.0000 590.00 36.88 1.0000 4330.0388 16 = Total number of stomachs 19 = Number of prey taxa Niche Breadth Niche breadths for pelagic and demersal species of rockfishes are presented in Table 11. All three indices exhibit similar niche breadth trends for each species; however, the Shannon-Weaver index results in relatively wider niche breadth estimates than Simpson's and Levin's indices. The relative niche widths are determined by expressing niche breadth values as percent of the maximum value. Species with relatively high index values are dietary generalists as opposed to specialists. FEEDING ECOLOGY OF ROCKFISHES 31 TABLE 11. Niche Breadth Values ^ for 10 Species of Rockfishes Collected During Summer Months in Coastal Waters of Southeast Alaska. Shanon- Total No. Total No. Levin's Maxi- Weaver Maxi- Simpson's Maxi- of of Prey Niche mum Niche mum Niche mum Stomachs Taxa Breadth^ Value Breadth^ Value Breadth^ Value Pelagic Species Black rockfish 142 38 .053 1 .594 1,60 2.125 38 Dusky rockfish 44 22 .305 1 .968 1.36 7.020 22 Yellowtail rockfish 112 22 .162 1 .800 1.36 3.723 22 Widow rockfish 27 23 .156 1 .765 1.36 3.594 23 Puget Sound rockfish 19 17 .359 1 .970 1.25 6.465 17 Demersal Species Yelloweye rockfish 15 12 .427 1 .829 1.08 5.127 12 Quillback rockfish 67 45 .329 1 1.334 1.65 14.815 45 China rockfish 59 44 .131 1 1.095 1.64 5.745 44 Copper rockfish 25 18 .356 1 .990 1.25 6.408 18 Tiger rockfish 16 19 .456 1 1.054 1.28 8.662 19 ' See Appendix 1 . ^Hurlbert 1978. ^Petraitis 1979. Pelagic species have more specialized diets than demersal species. Black rockfish are the most specialized of pelagic species, yet consume the greatest number of dietary items. In contrast, Puget Sound rockfish are most generalized of pelagic species but utilize the least number of dietary items. Dusky, yellowtail, and widow rockfishes ate equal numbers of prey taxa. Dusky rockfish are dietary generalists compared to yellowtail and widow rockfishes. Demersal species are intermediate between dietary specialists and dietary generalists, with the exception of China rockfish. China rockfish are the most specialized of demersal species, although they consumer 44 different prey taxa. Quillback rockfish also consume a large number of prey taxa (45) but are more dietary generalists. Yelloweye rockfish feed on the fewest number of prey taxa of the demersal species. Copper and tiger rockfishes consume similar numbers of prey taxa. However copper rockfish are more specialized than tiger rockfish, which is the most generalized of demersal rockfish. Dietary Overlap Among pelagic rockfish species, the Puget Sound rockfish had the lowest dietary overlap values (Tables 12-14). Puget Sound rockfish are much smaller than black, dusky, yellowtail, and widow rockfishes and therefore are restricted to a smaller size of dietary items. Puget Sound rockfish show little overlap with other rockfishes in the food resources they consume. Black rockfish is dietarily the most specialized pelagic species. Black rockfish consume a greater percentage of the dietary items they have in common with dusky, yellowtail, and widow rockfishes, and consequently show a greater degree of impingement on these species than they do on black rockfish. Food resources common to the black rockfish and each of the other pelagic species are more important in the specialized diet of the black rockfish. 32 CALIFORNIA FISH AND CAME TABLE 12. Schoener's Niche Overlap (Huribert 1978). "Dietary Similarity" Between Pelagic Rockfishes X and Y During the Summer Months in Coastal Waters of Southeast Alaska. Y X Black Dusky Yellowtail Widow Paget Sound Black .314 .716 .430 .042 Dusky 314 .428 .452 .227 Yellowtail 716 .428 .451 .068 Widow 430 .452 .451 .125 Puget Sound 042 .227 .068 .125 TABLE 13. Levin's Directional Overlap (Huribert 1978). The Degree Pelagic Rockfish X Impinges on Pelagic Rockfish Y During the Summer Months in Coastal Waters of Southeast Alaska. Y_ X Black Dusky Yellowtail Widow Puget Sound Black .920 1.276 .804 .012 Dusky 279 .449 .417 .171 Yellowtail 729 .846 .640 .032 Widow 476 .815 .663 .065 Puget Sound 004 .186 .018 .036 TABLE 14. Lloyd's Directional Overlap (Huribert 1978). Pelagic Rockfish X Consumes * Percent of the Total Volume of Dietary Items That Pelagic Rockfishes X and Y have in Common During the Summer Months in Coastal Waters of Southeast Alaska. Y_ X Black Dusky Yellowtail Widow Puget Sound Black 91.7 57.0 83.4 92.5 Dusky 8.3 10.8 31.4 52.7 Yellowtail 43.0 89.2 79.1 90.2 Widow 16.6 68.6 20.9 70.9 Puget Sound 7.5 47.3 9.8 29.1 Yellowtail and widow rockfishes both consume a greater percentage of the food items they have in common with dusky rockfish and impinge on dusky rockfish to a greater degree than dusky rockfish impinge on yellowtail or widow rockfishes. The food resources common to yellowtail and dusky rockfishes and widow and dusky rockfishes are more important in the diets of the yellowtail and widow rockfishes. Examination of Tables 15-17 shows several patterns in overlap measures for demersal species pairs. Quillback rockfish consume a greater percentage of the total volume of common food resources than any of the other demersal species. All the demersal species show the highest degree of impingement on quillback rockfish. Quillback rockfish are dietary specialists compared to yelloweye, tiger, or copper rockfishes. The China rockfish is more of a dietary specialist than the quillback rockfish; however, each consumes approximately 50% of the total volume of each species diet are in common items. China rockfish have their greatest dietary similarity with, and degree of impingement on, quillback rockfish. Conversely, quillback rockfish have their lowest dietary similarity with, and degree of impingement on, China rockfish. A change in the common resources consumed by either the quillback rockfish or China rockfish would have a greater effect on the China rockfish. FEEDING ECOLOGY OF ROCKFISHES 33 TABLE 15. Schoener's Niche Overlap (Hurlbert 1978). "Dietary Similarity" Between Demersal Rock- fishes X and Y During the Summer Months in Coastal Waters of Southeast Alaska. y X Quillback Yelloweye China Tiger Copper Quillback .318 .271 .319 .465 Yelloweye .318 .054 .261 .234 China .271 .054 .156 .165 Tiger .319 .261 .156 .131 Copper .465 .234 .165 .131 TABLE 16. Levin's Directional Overlap (Hurlbert 1978). The degree Demersal Rockfish X Impinges on Demersal Rockfish Y During the Summer Months in Coastal Waters of Southeast Alaska. K X Quillbacl< Yelloweye China Tiger Copper Quillback .333 .172 .294 .508 Yelloweye .963 .060 .475 .360 China .444 .053 .203 .094 Tiger .503 .281 .135 .103 Copper 1.175 .288 .085 .139 TABLE 17. Lloyd's Directional Overlap (Hurlbert 1978). Demersal Rockfish X Consumes • Percent of the Total Volume of Dietary Items that Demersal Rockfishes X and Y Have in Common During the Summer Months in Coastal Waters of Southeast Alaska. Y_ X Quillback Yelloweye China Tiger Copper Quillback 73.5 52.6 85.7 74.6 Yelloweye 26.5 28.5 68.4 51.4 China 47.4 71.5 84.4 72.6 Tiger 14.3 31.6 15.6 32.8 Copper 25.4 48.6 27.4 67.2 The X, Y species pairs — yelloweye and tiger rockfishes, yelloweye and copper rockfishes, China and tiger rockfishes, and the copper and tiger rockfishes — exhibit similar patterns in percentage of total volume consumed and degree of impingement. Species X consumes a greater percentage of the total volume of dietary items that species X and Y have in common. The degree of impingement is greater for species X than species Y. The food resources common to the X, Y species pairs are of greater importance in the diet of species X. DISCUSSION The rockfish species complex has radiated into a number of different habitats and depth strata in the northeast Pacific ocean. The success displayed by this group from Baja California to Prince William Sound, Alaska, is exemplified by a highly opportunistic foraging behavior. Rockfishes utilize a broad spectrum of food types during summer months in shallow waters of the Gulf of Alaska. Of the ten species of Sebastes examined in this study, most demonstrated a remarkable flexibility in diet, while capitalizing on the most accessible types of prey. Stomach contents of these ten species included 128 identifiable prey taxas or groups. 34 CALIFORNIA FISH AND GAME Demersal species are more dependent on detrital sources of food than pelagic species which consume substantial quantities of zooplankton and fish. Plankton and forage fish production in the nearshore waters of Alaska during spring and summer contribute vast amounts of energy to higher trophic levels. A cornerstone of the forage base is the Pacific sand lance, termed the "anchovy of the north" because of its widespread distribution and importance as a forage species. A planktivore, the Pacific sand lance eats mostly drifting organisms (e.g., copepods, amphipods, mysids, etc.), and in doing so provides a means of "energy transfer between zooplankton and the piscivorous fishes" (Simenstad et al. 1977). Sand lance is a major food item in the summer diets of black, yellowtail and widow rockfishes. However, dusky and Puget Sound rockfishes depart from the fish eating trend in that they feed more on small Crustacea and gelatinous zooplankton. Black, yellowtail, and dusky rockfishes co-occur over several depth strata in the eastern Gulf of Alaska. Black rockfish decrease in abundance with depth while yellowtail and dusky rockfishes increase in abundance with depth (Rosenthal et al. 1982, Field 1984). Black and yellowtail rockfishes are dietary specialists that feed most heavily on fish. A surplus of food resources common to both black and yellowtail rockfishes may be responsible for their extensive overlap (Pianka 1978). In comparison, dusky rockfish are dietary generalists and consume primarily invertebrate zookplankton. Resource partitioning between yellowtail and dusky rockfishes, habitat partitioning by black and yellowtail rockfishes, and an abundant supply of some food resources appear to be contributing factors in the nearshore distribution of black, yellowtail, and dusky rockfishes. Puget Sound rockfish occur with subadult black and yellowtail rockfishes in the shallow sublittoral zone. The Puget Sound rockfish is a generalist, despite being confined to picking tiny epipelagic Crustacea from the water column due to its small size. It shows little dietary overlap with the other pelagic rockfish species. Like the ubiquitous Pacific sand lance, the Puget Sound rockfish is an important converter of plankton in the nearshore system, for it is eaten by a number of reef-dwelling rockfish. Demersal species with the exception of the China rockfish are intermediate between dietary specialists and generalists. China rockfish exhibit a strong preference for substrate-oriented prey, particularly brittle stars. This specializa- tion in diet is probably important in reducing competition with other bottom- dwelling rockfish. Yelloweye rockfish increase in abundance with depth and feed primarily on fish. Quillback rockfish is most abundant at intermediate depths and consumes a wide variety of Crustacea and small fishes. Quillback rockfish consume a greater percentage of the dietary items that are shared with yelloweye rockfish; however, the degree of impingement on yelloweye rockfish by quillback rockfish is relatively low. The consumption of common food resources may have a minimal effect on the co-occurrence of these two species. The dietary overlaps for pelagic and demersal rockfishes in the Gulf of Alaska were strong during summer when food resources are known to be high. This appears to be characteristic of the group as Love and Ebeling (1978) also found considerable overlap in the diet of California rockfishes particularly during periods (summer-fall) when prey were abundant. Most rockfishes are oppor- FEEDING ECOLOGY OF ROCKFISHES 35 tunists that take advantage of the most abundant prey available to them in the nearshore zone. Geographic variations in diet suggest that substitution of prey is possible as long as the foods are of the same general size and type, and the species stocks similar. ACKNOWLEDGMENTS We would like to thank L. Field, L. Haldorson, M. LaRiviere, J. Underwood, J. Race, T. Nelson, and D. Myer for their field assistance. J. Field helped in sorting and specimen identifications. J. R. Chess and E. Hobson joined us for a portion of the 1980 summer field work. P. Rigby, B. Bracken and R. Morrison of the ADF&G Groundfish Division also participated in various stages of the program. P. DeSloover, librarian, Alaska Department of Fish and Game, facilitated numerous requests for literature searches and reprints. E. Landingham of the National Marine Fisheries Service, Auke Bay Laboratory, prepared the illustrations. The laboratory facilities at Sheldon Jackson College (Sitka); Alaska Department of Fish and Game (Sitka) and the College of Fisheries, University of Washington (Seattle) provided invaluable storage and laboratory space. The National Marine Fisheries Service, Northwest and Alaska Fisheries Center Resource Ecology and Fisheries Management Division, Resource Assessment and Conservation Ecology Division, and the Auke Bay Laboratory provided computer services and resources. This research was supported by grants from the State of Alaska Bottomfish task force. Office of the Governor (Hon. J. Hammond), Department of Fish and Game, the Alaska Council on Science and Technology and the National Marine Fisheries Service Resource Ecology and Fisheries Management Division. The encouragement and support provided throughout by P. Rigby is keenly appreciated. LITERATURE CITED Carlson, R. H., and R. R. Straty, 1981. Habitat and nursery grounds of pacific rockfish, Sebastes spp., in rocky coastal areas of southeastern Alaska. Mar. Fish. Rev. 43 (7): 13-19. Field, L. J. 1984. Bathymetric patterns of distribution and growth in three species of nearshore rockfish from the southeastern Gulf of Alaska. M.S. Thesis. University of Washington, Seattle, Washington. 88 p. Cunderson, D. R., and W. H. Lenarz. 1980. Cooperative survey of rockfish and whiting resources off California, Washington, and Oregon 1977: Introduction. Mar. Fish. Rev. 42: 1. Hueckel, C. J., and R. L. Stayton. 1982. Fish foraging on an artificial reef in Puget Sound, Washington, Mar. Fish, Rev. 44 (6-7): 38--M. Hurlbert, S. L. 1978. The measurement of niche overlap and some relatives. Ecology 59: 67-77. Lawlor, L. R. 1980. Overlap similarity and competition coefficients. Ecology 61: 245-251. Leaman, B. M. 1980. The ecology of fishes in British Columbia kelp beds I. Barkley Sound nerocyctis beds. Fisheries Development Report No. 22. Marine Resources Branch, Ministry of Environment, Province of British Columbia. Love, M. S. 1980. Isolation of olive rockfish, Sebastes serranoides, populations off southern California. Fish. Bull. U.S., 77 (4): 975-983. Love, M. S. and A. W. Ebeling. 1978. Food and habitat of three switch-feeding fishes in the kelp forests off Santa Barbara, California. Fish. Bull. U.S., 76: 257-271. Miller, D. J. and J. J. Geibel. 1973. Summary of blue rockfish and lingcod life histories; a reef ecology study, and giant kelp, Macrocyctis pyrifera, experiments in Monterey Bay, California. Calif. Dept. Fish Game, Fish Bull. 158: 137 p. Moulton, L. L. 1977. An ecological analysis of fishes inhabiting the rocky nearshore regions of northern Puget Sound, Washington, Ph.D. Thesis. University of Washington, Seattle, Washington. NODC Taxonomic Code. 1981. Third Edition. Prepared for National Oceanic and Atmospheric Administration, National Oceanographic Data Center. 82 p. Patten, B. G. 1973. Biological information on copper rockfish in Puget Sound, Washington. Trans. Am. Fish. Soc. 102 (2): 412^16. 36 CALIFORNIA FISH AND CAME Pereyra, W. T., W. G. Pearcy, and F. E. Carver, Jr. 1969. Sebastodes flavidus, a shelf rockfish feeding on mesopelagic fauna, with consideration of the ecological implications. J. Fish. Res. Board Can. 26 (8): 2211-2215. Petraitis, P. S. 1979. Likelihood measures of niche breadth and overlap. Ecology 60: 703-710. Pianka, E. R. 1978. Evolutionary ecology. Harper and Row, New York. 397 p. Pinkas, L., M. S. Oliphant, and I. L. K. Iverson. 1971 . Food habits of albacore, bluefin tuna, and bonito in California wafers. Calif. Dept. Fish Game, Fish. Bull. 152: 105 p. Prince, E. D. and D. W. Gotshall. 1976. Food of the copper rockfish, Sebastes cauhnus Richardson, associated with an artificial reef in south Humboldt Bay, California. Calif. Dept. Fish Game, Fish Bull. 62 (4): 274-285. Rosenthal, R. ). 1983. Shallow water fish assemblages in the northeastern Gulf of Alaska: habitat evaluation, sfjecies composition, abundance, spatial distribution and trophic interactions. Pages 451-540. In Environ- mental Assessment of the Alaskan Continental Shelf. Final Reports of Principal Investigators. Volume 17. Biological Studies. U.S. Dept. of Commerce. NOAA/NMFS. Rosenthal, R. J., L. Haldorson, L. J. Field, V. Moran-O'Connell, M. LaRivere, ). LInderwood, and M. C. Murphy. 1982. Inshore and shallow offshore bottomfish resources in the southeastern Gulf of Alaska (1981-1982). Alaska Dept. Fish Came, Juneau, Alaska. 166 p. Simenstad, C. A., J. S. Isakson and R. E. Nakatani. 1977. Marine fish communities of Amchitka Island, Alaska. In M. L. Merritt and R. C. Fuller, ed. The Environment of Amchitka Island, Alaska. U.S. W.R.D.A. TID 267-12: 451^92. Steiner, R. E. 1978. Food habits and species composition of neritic reef fishes off Depoe Bay, Oregon. M.S. Thesis. Oregon State University, Corvallis, Oregon. 59 p. Vadopich, D. S., and J. J. Hoover. 1981. A computer program for integrated feeding ecology analyses. Bull of Marine Sci. 31: 922-925. APPENDIX 1 Niche Breadth Levin's niche breadth (Hurlbert 1978): B = 1/n 2 p^.2 i Shannon-Weaver niche breadth (Petraitis 1979): r H / = - 2 p,y log p,y j = l Simpson's niche breadth (Petraitis 1979): r B, = 1/2 p^.. j = l p,y — Percent total volume of prey / in predator /. n,r = Number of prey taxa utilized by predator /. FEEDING ECOLOGY OF ROCKFISHES 37 APPENDIX 2 Dietary Overlap Schoener's niche overlap (Hurlbert 1978): C,, = 1 - V2 (2 I P., - ?yi I ) i Levin's directional overlap (Hurlbert 1978): ax(y) = 2 (P,, P^,) / 1 P,,2 I i ay(x) = 2 (P^,- P^,) / 2 P^,-2 i i Lloyd's directional overlap (Hurlbert 1978): Zx{y) = 2 (X,Y,/X) i Zy(x) = 2 (Y,X,/Y) X = predator 1 y = predator 2 / = prey Px/ = X , /X = Percent total volume of prey / in predator x. Pyi — Y , /Y = Percent total volume of prey / in predator y. X ; = Total volume of prey / in predator x stomachs. X = Total volume of predator x stomachs. Y / = Total volume of prey / in predator y stomachs. Y = Total volume of predator y stomachs. 38 CALIFORNIA FISH AND GAME Calif. Fish and Came 74 ( 1 ) : 38-48 1 988 GROWTH OF YOUNG-OF-THE-YEAR AND JUVENILE PACIFIC HERRING FROM SAN FRANCISCO BAY, CALIFORNIA' PAUL N. REILLY California Department of Fish and Came Marine Resources Division 411 Burgess Drive Menio Park, Calif. 94025 Young-of-the-year (YOY) and juvenile (1+ yr old) Pacific herring, Clupea harengus pallasi, were sampled from April to June 1983 and 1986 and from April to July 1984 and 1985 by midwater trawl in San Francisco Bay, California. The majority of YOY herring ranged from 40 to 70 mm body length (BL) during these months while most juveniles ranged from 105 to 150 mm bl. Fifteen-meter bay temperatures during hatching and early development were directly related to YOY growth rate. El Nino conditions during 1982 and 1983 adversely affected growth of the 1982 yr class as juveniles in the ocean. INTRODUCTION The Pacific herring, Clupea harenqus pallasi, spawning population in San Francisco Bay supports a roe fishery that has yielded an average of approxi- mately 6000 tons of whole fish during the last 12 seasons. The California Department of Fish and Game (CDFG) manages the fishery by quota based on spawn escapement estimates and commercial catch from the previous season. Life history parameters, such as age and growth rates, are essential for proper management of this valuable resource. Little is known about young-of-the-year (YOY) and juvenile herring in California. Hubbs (1925) used length frequencies and scales to determine age of young herring from San Francisco Bay in 1922 and 1923. Miller and Schmidtke (1956) sampled yearling herring from the Monterey Bay lampara fishery in 1947, 1948, and 1952 and aged them by scales. More recently, Spratt (1981 ) reported average length of YOY and juvenile herring caught incidentally in the winter Monterey Bay anchovy fishery and in the summer bait fishery. Herring spawn in San Francisco Bay from late October to March and YOY herring remain in the bay until mid-summer. The following spring many juvenile herring return to the bay and mix with the next year class of YOYs. CDFC's Bay-Delta Project, Stockton has provided an opportunity to study growth of YOY and juvenile herring from San Francisco Bay within and between years. From 1983 to 1986 intensive midwater trawling was conducted during spring and summer primarily to sample out-migrant salmon smolts. Length measure- ments of the incidental catch of herring, primarily YOYs, have provided a wealth of data to follow growth. METHODS All San Francisco Bay samples were obtained with a midwater trawl with a 10-ft by 30-ft mouth opening. Towing was conducted throughout each month from April to June or July with the exception of April 1984 (mid- to late April ' Accepted for publication July 1987. GROWTH OF PACIFIC HERRING 39 only) and July 1985 (late July only). The net was towed at 3 to 7 knots between the surface and 30 ft (Charles Armor, CDFG, Stockton, pers. commun.). All tows were conducted in daytime along one standard transect between the Golden Gate Bridge and Alcatraz Island (Figure 1). A maximum of 30-50 randomly selected herring was measured from each of three general overlap- ping length classes from each tow: (i) small herring were usually less than 90 mm; (ii) medium herring ranged from 80 to 130 mm; (iii) large herring were usually greater than 120 mm. All herring were counted from small catches; large catches were subsampled by a volumetric aliquot and total number of herring captured was calculated. Fork length (fl) was measured in the field by Bay-Delta Project personnel and subsequently converted to body length (bl) using the regression: BL = 0.62668 + 0.93059 fl, N = 123, r = 0.99 BL is the standard measurement for herring in California and is the distance from the snout tip to the end of the pigment underneath the last column of scales on the caudal peduncle. 122°35' 122°30^ Gulf of the Faral lones FIGURE 1. Location of Pacific herring sampling area. 40 CALIFORNIA FISH AND CAME On an approximately monthly basis from April to July, subsamples of herring were frozen for later age and growth analysis. A correction factor of 1.021 was applied to bl of thawed herring to account for shrinkage (Reilly and Moore 1983). Adult herring (age 2+ yr) were included in these samples. Overlap occurred in length frequency distributions of 1+ and 2+ yr old herring. The largest 1 + yr old fish each year was used as the cutoff on length frequency histograms for juveniles. Number of 1 + yr old fish in each 2-mm bl interval within the overlap range were calculated based on the proportion of 1 + yr old fish among those aged by otoliths. Rarely did ages overlap between YOY and juvenile fish for a particular month. When this did occur, a cutoff point between year classes was determined from the distribution of otolith ages. In a separate study, San Francisco Bay temperature profiles between the surface and 25 m at 5-m intervals were determined approximately three times per week, usually at high tide, from November to March at Peninsula Point (Figure 1 ), using a Martek Mark VI Water Quality Analyzer (Reilly and Moore 1982, 1983, 1984, 1985, 1986). RESULTS Total Catch During four years of trawl sampling, approximately 1.37 million YOY herring wore caught and about 2% were measured (Table 1). Approximately 14,175 juvenile herring were caught and about 52% were measured. Thus, relative numbers of YOY and juvenile herring in length frequency histograms for a particular year are not indicative of year class strength. The analysis reported here is based only on samples from April to June in 1983 and 1986 and from April to July in 1984 and 1985. TABLE 1. Total Catch of YOY and Juvenile Herring from Midwater Trawls in San Francisco Bay, April to August, 1983 fo 1986. Part 1. YOY 1983 1984 1985 1986 Number Total Number Total Number Total Number Total Month of tows catch of tows catch of tows catch of tows catch April 67 10,245 50 14,530 90 122,785 87 50,705 May 180 67,985 110 74,536 230 255,255 88 80,475 )une " 150 16,995 110 132,505 70 157,240 162 340,040 July 30 28 150 34,175 49 13,350 August 40 16 30 78 - - - Total 467 95,269 450 255,824 439 548,630 337 471,220 Total measured for study 3,563 6,685 8,980 7,320 Part 2. Juveniles Total Number Total measured Year of tows catch for study 1983 467 1,538 875 1984 450 2,374 1,966 1985 439 7,146 2,581 1986 337 3,119 1,883 GROWTH OF PACIFIC HERRING 41 Length Frequencies Length frequency histograms during all 4 yr show a distinct separation between the majority of YOY and juvenile herring (Figures 2 and 3); few herring are in the 80-100 mm bl range. In 1983 and 1986 a slight amount of overlap occurred between year classes, while in 1984 and 1985 a wide separation existed. In general, from April to July most YOY herring in San Francisco Bay ranged from 30 to 90 mm bl, with the peak in length frequency distribution at 40 to 70 mm and monthly mean bl from 45 to 68 mm (Table 2). juveniles ranged from 90 to 160 mm bl with peak distribution from 105 to 150 mm and monthly mean bl from 108 to 147 mm (Table 2). Each season, approximately 90% of spawning occurred within a 2-month period ranging from mid-December to mid-March (Spratt 1983, 1984, 1985, 1986a). Thus, the majority of YOY herring from April to July will be from 1 to 7 months old, while juvenile herring will be 13 to 19 months old. TABLE 2. Mean bl and Standard Deviation of YOY and Juvenile Herring Samples by Month, Spring and Summer 1983 to 1986, from San Francisco Bay. Part 1. Age 0+ Year Month 1983 May June 1984 April May June July 1 985 April May June July 1986 April May June Part 2. Age 1-1- Vear Month 1983 April May June 1984 April May June July 1985 April May )une 1986 April May June Standard N Mean bl deviation 2327 52.4 6.8 1236 63.8 8.9 675 51.0 6.6 1818 54.0 6.3 2322 62.9 7.2 1870 68.2 5.6 1781 46.7 8.5 4452 44.7 7.4 1699 49.5 5.7 958 60.3 3.1 1689 57.2 6.6 1813 54.2 10.3 3818 59.2 11.7 Standard N Mean bl deviation 104 107.2 12.0 408 109.6 11.8 363 121.7 10.6 53 125.7 6.8 463 130.9 7.0 777 133.4 7.1 673 137.2 6.8 268 138,8 9.4 2138 146.8 6.8 175 146.2 8.4 252 108.2 10.4 458 119.1 14.5 1173 135.4 8.5 Usually from nine to 13 herring schools enter San Francisco Bay and spawn at approximately 2-wk intervals from late October to mid-March; however, only from four to seven schools are large (greater than 1000 tons) and these 42 CALIFORNIA FISH AND GAME BODY LENGTH (mm] 100- 75- 25- APR 84 N = 675 N = 53 1 ""''^l 1 T 1 1 1 1 1 1 1 10 -5 300- 200- 100- MAY 84 1 r'^"^ N = 1818 rL N=463 1 1 1 1 1 '~^-\ 1 1 r >■ (J z ID 400- gjooH liJ 200- "- 100 00 00 00 00-1 JUL 84 N-1870 JUN 84 -^ 1 n N=2322 ^ a N = 777 I ■ - 1 T 1 1 1 1 1 "^ 1 1 1 1 75 50 25 100 75 50 25 — I 1 r""^ I \ 1 — ="T \ 1 ' 1 1 — ~r 20 30 40 SO 60 70 80 90 100 110 120 130 140 ISO 160 BODY LENGTH (mm) FIGURE 2. Number of YOY and juvenile Pacific herring (2-mm intervals) from midwater trawl samples, San Francisco Bay, April-June 1983 and April-July 1984. GROWTH OF PACIFIC HERRING 200- 100- JUL 85 1 r- 20 30 40 N = 958 -^ "T~~r I I I 1 1 1 1 1 1 r 50 60 70 80 90 100 110 120 130 140 150 160 BODY LENGTH (mm) "T I I 1 r "0 70 80 90 BODY LENGTH (mm) FIGURE 3. Number of YOV and juvenile Pacific herring (2-nnm intervals) from midwater trawl samples, San Francisco Bay, April-July 1985 and April-June 1986. 44 CALIFORNIA FISH AND CAME contribute the bulk of the total spawning biomass ( Reilly and Moore 1 983, 1 984, 1985, 1986; Spratt 1983, 1984, 1985, 1986a). Larvae from each spawn can be considered a cohort and would be responsible for the occurrence of multiple modes in the length frequency histograms. The shift in median bl for YOY herring from month to month cannot always be interpreted as growth, such as from April to May 1986, due to the appearance of different cohorts in the sampling area. Most YOY herring leave the bay by mid-summer, and, based on the large number of herring collected, the sampling location appears to be along the emigration route. An approximation of average growth can be obtained from examining the frequency distribution of the peak farthest left in the histograms during June of each year (Figure 4). This most likely represents the cohort from the last major spawn each year which occurred from early February to mid-March. Thus, herring may grow to 50-60 mm bl in 3 to 4 months. -150 100 -50 BODY LENGTH (mm) FIGURE 4. Number of YOY and juvenile Pacific herring (2-mm intervals) from midwater trawl samples, San Francisco Bay, June, 1983 to 1986. GROWTH OF PACIFIC HERRING 45 Single modes from large samples, such as in May, June, and July 1985 (Figure 3), reflect the growth of the major YOY cohort or cohorts. In May, the mode was 42-43 mm, in June 50-51 mm, and in July 60-61 mm, a growth rate of approximately 10 mm per month. Seasonal Differences in Growth Length distribution of juveniles from 1986 changed dramatically from April to June, with the majority of 1 + yr old fish ranging from 90 to 130 mm BL in April (Figure 3). Medium-sized herring were all but absent in June, when most juveniles ranged from 120 to 150 mm bl. The smaller juveniles had a definite annulus on the otolith and most likely represent a slow growing cohort from a late spawn during the 1984-85 season. The presence of the annulus precludes the possibility of their origin from early spawns in the 1985-86 season. Samples in June probably are more representative of the growth of the majority of the year class; in 1986 the ratio of juveniles caught greater than 120 mm to those less than 120 mm was 5.1:1. Juveniles in 1983 and 1986 averaged less in bl and had a wider length distribution than those in 1984 and 1985 (Figures 2 and 3). A comparison of growth between years is best seen, however, in samples from June of each year ( Figure 4) . Sample size is large and this is usually the last month of high relative abundance of YOY and juvenile herring in the bay before their departure. In 1985 and 1986 YOY modes were less than those of 1983 and 1984. A noticeable lack of YOY herring greater than 70 mm was characteristic of June and July 1 985 (Figure 3). Since major spawns occurred at approximately the same time each year (more than 50% of all spawning occurred from January 5 to February 7), conditions for growth in the bay must have been more favorable in 1983 and 1984 than in 1985 and 1986. An examination of mean 15-m temperatures from profiles at Peninsula Pt. in San Francisco Bay from December to February (the period of hatching and early development for the majority of larvae) indicates that temperature may have a direct influence on YOY growth in the bay (Table 3). In 1983 and 1984, temperatures were relatively warm, while in 1985 mean temperature was relatively cold. Temperatures during the 1981-82 spawning season were unusually low; thus the 1982 year class hatched during a period of unfavorable temperatures for growth. Surface temperatures are more affected by ambient air temperatures and show greater variation than 1 5-m temperatures (Table 3 ) . The relationship between surface temperature and YOY growth is not as clear as that of 15-m temperature. Length frequency distributions of juveniles reflect growth conditions during part of their first year, in the bay, and during the remainder of their first and part of their second year, in the ocean, before they return to the bay in spring. The magnitude of growth differences among the four year classes of juveniles is greater than that of YOY fish (Figure 4). The 1982 year class, shown as juveniles in June 1983, exhibited relatively poor growth. These herring were undoubtedly influenced by El Nifio conditions in the eastern Pacific during 1982 and 1983 (Wooster and Fluharty 1985), as were all adult herring from central California (Spratt, in press). The additional factor of lower temperatures in the bay in 1982 46 CALIFORNIA FISH AND CAME resulted in the relatively small mean bl of this year class as juveniles in 1983. In contrast, the 1984 year class, shown as juveniles in June 1985, experienced relatively warm bay temperatures and normal oceanic conditions, resulting in exceptionally good growth. The growth of the 1983 year class is most likely average for San Francisco Bay herring. Normal oceanic conditions probably aided the 1985 year class after a slow start in San Francisco Bay, but the presence of juveniles in April and May 1986 between 90 and 120 mm probably reflects the earlier influence of poor growth in the bay. TABLE 3. Mean and Range of 15-m and Surface Temperature, December to February, at Peninsula Pt., San Francisco Bay and mean bl for YOY herring, June. 15-m temperature (C) Mean bl Range Mean YOY Herring year class affected Dec-Feb Dec-Feb June 1982 9.5-10.4 10.1 1983 10.0-13.8 12.5 63.8 1984 11.4-13.6 12.3 62.9 1985 10.2-12.7 11.2 49.5 1986 9.7-13.2 11.7 59.2 Surface temperature (C) 1982 8.4-11.4 9.6 1983 9.6-13.4 11.4 63.8 1984 10.0-13.3 11.7 62.9 1985 9.3-12.6 11.0 49.5 1986 9.4-13.5 11.5 59.2 DISCUSSION Hubbs (1925) found YOY herring in San Francisco Bay to range from 36 to 71 mm (mode 47 mm) in early June 1923. A small amount of herring collected in the bay from October to December 1922 ranged from 63 to 90 mm (mode 75-78 mm ) . The distribution of San Francisco Bay's YOY herring during fall and winter is unknown; however, in October 1982 Reilly and Moore (1983) collected 400 YOY herring 1 n.m. west of San Francisco in the Gulf of the Farallones. They ranged from 63 to 82 mm bl and averaged 70 mm. This is consistent with the slower growth rate observed for this year class in general. Spratt (1981 ) reported that YOY herring in Monterey Bay reach 90 mm bl by September. Hubbs (1925) also found "yearling" herring, 79 to 115 mm (mode 91-92 mm), in mid-April in San Francisco Bay. These were most likely from a cohort hatched late in the previous season. Spratt ( 1 981 ) found that Monterey Bay herring averaged 1 1 3 mm as 1 -yr olds in January. Miller and Schmidtke (1956) found one annulus on juvenile herring from Monterey Bay in June 1948; these fish averaged 145 mm bl, similar to those from San Francisco Bay in 1985. Estimated lengths for 1-yr old Pacific herring in other states include 112-120 mm for Alaska (Barton 1979) and 120-134 mm for Washington (Gonyea 1985, Chapman, Katz and Erickson 1941). Growth rates observed in this study are consistent with those of 2-yr old herring sampled from the San Francisco Bay fishery. Spratt (1986b, in press) GROWTH OF PACIFIC HERRING 47 found 2-yr olds in the 1983-84 season (1982 yr class) to average only 152 mm BL, due to El Nino effects, while the 1983 and 1984 yr classes averaged 163 and 162 mm BL, respectively. Other studies report average length of 2-yr olds to be 157-166 mm in Humboldt Bay, California (Rabin and Barnhart 1986) and 160 mm in Washington (Gonyea 1985). However, isolated stocks of Pacific herring exist with much lower growth rates. Barton and Steinhoff (1980) report that herring from Kotzebue, Alaska reach only 126 mm by their second yr. A "Fidalgo" stock of herring from Washington averages only 139 mm as 2-yr olds (Buchanan 1985). Lower growth rates may also be characteristic of a local population near Fort Bragg in northern California. Reilly and Moore (CDFG, Menio Park, unpub. data) found all of 15 herring aged in the range 160-169 mm BL to be 3 or 4 yr old. Growth regulatory mechanisms are often classified as either density indepen- dent or density dependent. Density independent factors are usually environ- mental, such as temperature and salinity, while density dependent factors include food supply, cannibalism, and year class size. Pacific herring growth may reflect both types of mechanisms during their first 2 yr. A direct relationship between bay temperature and growth of YOY herring, as measured by mean bl in June, was found for the period 1983 to 1986. Moores and Winters (1982) found evidence that temperature regulated growth of Atlantic herring during their first year. Anthony and Fogarty (1985) found that temperature during the first growing season of Atlantic herring was positively correlated with mean length at age 2 yr, while no correlation was found between the latter and temperature during the second growing season. Although the El Nino of 1982-83 was characterized by above normal temperatures for coastal California waters, growth of herring was inhibited. This infers that the direct physiological link between elevated temperatures and increased metabolic rate (causing increased growth for YOY herring in the bay) was superseded by other factors, such as a reduction of food supply in the ocean. Winters, Wheeler and Dalley (1986) found that growth of juvenile Atlantic herring was not correlated with temperature but was inversely correlated with year class size (which could indirectly affect food supply through competition). The 1982 year class of herring from San Francisco Bay was more than twice as abundant as 2-yr olds than each of the next three year classes (Reilly, unpublished data). The 1982 year class, as 1 -|- yr olds, had the smallest mean bl in June of the four year classes studied; this year class also experienced El Nino conditions during its first season in the ocean. Other workers have found density dependent relationships for growth of juvenile Atlantic herring (Anthony 1971, Hubold 1978). Haist and Stocker (1985) concluded that Pacific herring in British Columbia exhibit density dependent growth as juveniles moderated by environmental factors, and they found an optimum temperature range for growth. Results from this study support the fact that growth is controlled by a complex interraction between environmental factors and population size. ACKNOWLEDGMENTS This study could not have been completed without the cooperation and assistance of CDFG's Bay-Delta Project, particularly Charles Armor, Susan Hatfield, Caryla Larsen, and Pat Coulston. 48 CALIFORNIA FISH AND GAME LITERATURE CITED Anthony, V.C. 1971. The density deptendence of growth of the Atlantic herring in Maine. Rapp. P.V. Reun. Cons. Int. Explor. Mer 160:19&-205. and M.J. Fogarty. 1985. Environmental effects on recruitment, growth, and vulnerability of Atlantic herring (Clupea harengus harengus) in the Gulf of Maine region. Can. ). Fish. Aquat. Sci. 42 (suppl. 1):15&-173. Barton, L.H. 1979. Assessment of spawning herring and capelin stocks at selected coastal areas in the eastern Bering Sea. Third Prog. Rep. No. Pac. Fish. Man. Coun. Contract 78-5. 44 p. and D.L. Steinhoff. 1980. Assessment of spawning herring and capjelin stocks at selected coastal areas in the eastern Bering Sea. Final Report No. Pac. Fish. Man. Coun. Contract 78-5. 63 p. Buchanan, Kurt D. 1985. The general purpose herring fishery 1957-1983. Washington Dept. Fisheries Tech. Rep. 85. 72 p. Chapman, W.M., M. Katz, and D.W. Erickson. 1941 . The races of herring in the state of Washington. Wash. Dept. Fisheries Biological Rept. No. 38A. 36 p. Conyea, Gary P. 1985. Abundance of Pacific herring in central and southern Puget Sound. State of Washington Dept. of Fisheries Tech. Rep. No. 87. 77 p. Haist, V. and M. Stocker. 1985. Growth and maturation of Pacific herring (Clupea harengus pallasi) in the Strait of Georgia. Can. ). Fish. Aquat. Sci., 42 (suppl. 1): 138-146. Hubbs, Carl L. 1925. Racial and seasonal variation in the Pacific herring, California sardine and California anchovy. Calif. Fish and Game Commission Fish Bulletin No. 8. 23 p. Hubold, H. 1978. Variation in growth rate and maturity of herring in the northern North Sea in the years 1955-1973. Rapp. P.V. Reun. Cons. Int. Explor. Mer 172:154-163. Miller, Daniel ). and John Schmidtke. 1956. Report on the distribution and abundance of Pacific herring (Clupea pallasi) along the coast of central and southern California. Calif. Fish and Game 42(7):163-186. Moores, J. A. and G.H. Winters. 1982. Growth patterns in a Newfoundland Atlantic herring (Clupea harengus harengus) stock. Can. ). Fish. Aquat. Sci. 39(3):454-461. Rabin, Douglas |. and Roger A. Barnhart. 1986. Population characteristics of Pacific herring, Clupea harengus pallasi, in Humboldt Bay, California. Calif. Fish and Game 72(1):4-16. Reilly, Paul N. and Thomas O. Moore. 1986. Pacific herring, Clupea harengus pallasi, studies in San Francisco Bay, central and northern California, and Washington. March 1985 to May 1986. Calif. Dept. Fish and Game, Mar. Resources Admin. Rep. No. 86-6. 1985. Pacific herring, Clupea harengus pallasi, studies in San Francisco Bay and the Gulf of the Farallones, June 1984 to March 1985. Calif. Dept. Fish and Came, Mar. Resources Admin. Rep. No. 85-4. 73 P- 1984. Pacific herring, Clupea harengus pallasi, studies in San Francisco Bay, Monterey Bay, and the Gulf of the Farallones, May 1983 to March 1984. Calif. Dept. Fish and Game. Mar. Resources Admin. Rep. No. 84-3. 67 p. 1983. Pacific herring, Clupea harengus pallasi, studies in San Francisco Bay, Monterey Bay, and the Gulf of the Farallones, July 1982 to March 1983. Calif. Dept. Fish and Game, Mar. Resources Admin. Rep. No. 83-5. 49 p. 1982. Pacific herring, Clupea harengus pallasi, studies in San Francisco Bay, December 1981 to March 1982, Calif. Dept. Fish and Came, Mar. Resources Admin. Rep. No. 82-8. 43 p. Spratt, Jerome D. 1987. Variation in the growth rate of Pacific herring from San Francisco Bay, California. Calif. Fish and Game, 73(3): 132-138. 1986a. Biomass estimates of Pacific herring, Clupea harengus pallasi, in California from the 1985-86 spawning ground surveys. Calif. Dept. Fish and Game, Mar. Resources Admin. Rep. No. 86-3. 28 p. 1986b. Biological characteristics of the catch from the 1985-86 Pacific herring, Clupea harengus pallasi, roe fishery in California. Calif. Dept. Fish and Game, Mar. Resources Admin. Rep. No. 86-5. 15 p. 1985. Biomass estimates of Pacific herring, Clupea harengus pallasi, in California from the 1984-85 spawning ground surveys. Calif. Dept. Fish and Game, Mar. Resources Admin. Rep. No. 85-2. 27 p. 1984. Biomass estimates of Pacific herring, Clupea harengus pallasi, in California from the 1983-84 spawning ground surveys. Calif. Dept. Fish and Game, Mar. Resources Admin. Rep. No. 84-2. 29 p. 1983. Biomass estimates of Pacific herring, Clupea harengus pallasi, in California from the 1982-83 spawning ground surveys. Calif. Dept. Fish and Game, Mar. Resources Admin. Rep. No. 83-3. 23 p. 1981. Status of the Pacific herring, Clupea harengus pallasi, resource in California 1972 to 1980. Calif. Dept. Fish and Came, Fish Bull. 171. 107 p. Winters, C.H., J. P. Wheeler, and E.L. Dalley. 1986. Survival of a herring stock subjected to a catastrophic event and fluctuating environmental conditions. Jour, du Conseil 43(1);26-42. Wooster, Warren S. and David L. Fluharty (eds) . 1985. El Nino North: Nino effects in the eastern subarctic Pacific Ocean. Washington Sea Grant Program. University of Washington. Seattle. 312 p. DIET OF WHITE STURGEON 49 Calif. Fish and Game 74 ( 1 ) : 49-54 1 988 DIET OF JUVENILE AND SUBADULT WHITE STURGEON IN THE LOWER COLUMBIA RIVER AND ITS ESTUARY' WILLIAM D. MUIR ROBERT L. EMMETT and ROBERT). McCONNELL Coastal Zone and Estuarine Studies Division Northwest and Alaska Fisheries Center National Oceanic and Atmospheric Administration 2725 Montlake Boulevard East Seattle, Washington 981 1 2 This note describes the diet of white sturgeon, Acipenser transmontanus, from the lower Columbia River and its estuary (Washington and Oregon). Samples were collected from 1980 to 1983. White sturgeon were captured by bottom trawl, purse seine, and hook and line and ranged from 5 to 129 cm in total length. Diet varied with length. The amphipod Corophium salmonis was primary prey for white sturgeon < 80 cm long but not white sturgeon > 80 cm. Larger white sturgeon consumed fish and large invertebrates. In general, diet diversity increased with length. INTRODUCTION This note provides information on food habits of juvenile and subadult white sturgeon, Acipenser transmontanus, in the lower Columbia River and its estuary. White sturgeon occur in large rivers and their estuaries and in marine waters on the Pacific coast of North America from the Aleutian Islands of Alaska to Monterey, California, (Scott and Crossman 1973). They are anadromous, although in some rivers are landlocked by hydroelectric dams (Haynes and Gray 1981). The Columbia River has the highest commercial and sport catches of white sturgeon on the Pacific coast; the commercial catch peaked in 1892 at more than 5 million pounds. This peak was followed by a rapid decline in landings due to overfishing; by 1899 less than 100,000 pounds were landed (Craig and Hacker 1940). Commercial and sport fishermen have recently targeted on white sturgeon because of their availability and the decline of Pacific salmon, Oncorhynchus spp. In 1969 only 13,000 white sturgeon were captured (total sport and commercial) in the lower Columbia River below Bonneville Dam; in 1984 over 59,000 were captured (King 1985). Despite the monetary value of white sturgeon, limited life history information is available, especially for juveniles. MATERIALS AND METHODS A total of 174 white sturgeon were collected; 120 were obtained between River Kilometers (RKm) 3 and 62 during a comprehensive study of Columbia River estuarine fishes in 1 980-1 981 ( McConnell et al. 1 983 ) . The sampling gears were an 8-m semiballoon shrimp trawl, 200-m purse seine, and hook and line (Table 1). An additional 54 white sturgeon were collected in July-September ' Accepted for publication August 1987. 50 CALIFORNIA FISH AND GAME 1983 using a 5-m semiballoon shrimp trawl between RKm 47 and 161 (Figure 1). TABLE 1. Number of White Sturgeon Captured Using Various Gear Types in the Columbia River and Its Estuary. Sampling gear Month Trawl Purse seine Hook and line Total Jan 2 2 -' 4 Feb 6 0-6 Mar 5 0-5 Apr 2 0-2 May 11 2 - 13 Jun 3 8 39 50 Jul 17 4 - 21 Aug 27 5 - 32 Sep 30 0 - 30 Oct 12-3 Nov 2 1-3 Dec _5 _0 ^ _5 Total Ill 24 39 174 ' Indicates no collections were made. WASHINGTON FICLJRE 1. The lower Columbia River and its estuary. DIET OF WHITE STURGEON 51 Stomachs were initially preserved in the field with a buffered 8% formalde- hyde solution, and transferred to vials containing 70% ethyl alcohol in the laboratory. Stomach and esophageal contents were examined with a 10X binocular dissecting microscope. Food organisms were identified to the lowest practical taxon (usually species), blotted, air-dried for 10 min, and weighed to the nearest 0.0001 g. All data were stratified by fish length to examine the effect of size on white sturgeon feeding habits. White sturgeon <80 cm total length (tl) were stratified by 20-cm size increments; those >80 cm were combined into one group. Percent weight, percent number and percent frequency of occurrence were calculated for each prey. RESULTS The amphipod Corophium salmonis was the dominant food item for white sturgeon <80 cm (Figure 2). Other invertebrates, including the mysid Neomysis mercedis, the decapod shrimp Crangon franciscorum, and the isopod Saduria entomon were also utilized by smaller white sturgeon. Fish became increasingly important as prey for white sturgeon >60 cm; especially the northern anchovy, Engraulis mordax, which was the dominant food item by weight for white sturgeon > 80 cm. White sturgeon > 60 cm consumed a wide variety of prey species, including large invertebrates and fishes. The white sturgeon examined ranged in length from 5 to 129 cm (tl). It is important to note that 94% of the white sturgeon < 60 cm were captured above RKm 27, whereas only 7% of the white sturgeon > 60 cm were captured above RKm 27. Therefore, differences in diet may partially reflect different areas of capture. The estuary appears to be a particularly important feeding area for white sturgeon >80 cm. From 1979 to 1984, the estuary (RKm 1-61 ) was the most productive sport fishing area; in 1984, 55% (21,900) of the total white sturgeon sport catch occurred here (King 1985). Our results generally agree with similar studies in the Sacramento-San Joaquin estuary, California (Schreiber 1962; Radtke 1966; and McKechnie and Fenner 1971) and the Fraser River, British Columbia (Semakula 1963; and Semakula and Larkin 1968). These studies also showed that small white sturgeon ate small invertebrates, and that larger white sturgeon ate fish and large invertebrates. Specific differences in food habits between our and other studies probably relate to the occurrence and abundance of prey in the different drainages. For example, eulachon, Thaleichthys pacificus, was seasonally dominant in the Fraser River study, and was also a seasonally important prey in larger white sturgeon in the Columbia River (Dees 1961 ). We did not find many eulachon in Columbia River white sturgeon, but collected few stomachs in winter when eulachon are abundant. Our study documents the diet of juvenile and subadult white sturgeon in the Columbia River system. This is valuable because of the growing interest in white sturgeon by sport and commercial fishermen in the Columbia River. Accurate sturgeon life history information is essential for resource managers. Small 52 CALIFORNIA FISH AND CAME White sturgeon < 20 cm total length White sturgeon > 20 cm and < 40 cm total length 75 50 25 0 25 50 75 Corophium salmonis n = 32 (4 empty a E C C s a> Q. - Xj^ = 14.4 cm Neomysis mercedis *^ c •^; (J ^ 100' 100 71 Percent frequency of occurrence lUU r C( yrophium sail nonis n = 41 (1 emptyl x^ " 31.5 cm 75 - 1 fc « 1 c 50 - OJ ■5 ■*- 40 cm and < 60 cm total length E 100 Corophium salmonii n = 30 (2 emptyl X. = 46 3 cm 75 50 25 - Neomysis mercedis Saduria entomon Crangon franciscorum 0 1 ;:;x:::::::::::::::::::::::::::::l:::::X::::|:X:: 25 50 •y 75 1 1 1 1 89 75 29 18 Percent frequency of occurrence FICURE 2. Food of different size groups of white sturgeon, Acipenser transmontanus, in the lower Columbia River. Food is represented by % composition by number and weight and by frequency of occurrence. Prey items that had number and weight comprising less than 3.0% are not shown. DIET OF WHITE STURGEON 53 White sturgeon > 60 cm and < 80 cm total length 60 40 23 17 10 7 Percent frequency of occurrence White sturgeon > 80 cm total length 100 r 75 50 25 0 25 50 1;; 75 * 100 ■■ 38 (8 empty) = 94.4 cm ■0 E £ nj (J &P J I 1_1_U 43 27 23 17 1010 773 Percent frequency of occurrence FIGURE 2. (continued) Sturgeon, as well as juvenile salmonids and many other species (McCabe et al. 1983), feed extensively on Corophium salmonis in the Columbia River estuary. Although benthic invertebrates in the Columbia River estuary have been well studied (Durkin and Emmett 1980; Higley et al. 1983), very little information is available on the benthic invertebrates of the Columbia River upstream from the estuary to Bonneville Dam (RKm 233). We believe a more comprehensive study to determine the life history, distribution, and abundance of white sturgeon and C. salmonis in the Columbia River is essential for the proper management of white sturgeon resources of the lower Columbia River. 54 CALIFORNIA FISH AND GAME ACKNOWLEDGMENTS We wish to express our thanks to personnel at the Hammond, Oregon, Field Station of the National Marine Fisheries Service's Northwest and Alaska Fisheries Center who helped collect data or assisted in the preparation of this paper. J. Peacock and his staff prepared our text figures. G. McCabe Jr., T. Blahm, and G. Monan reviewed the manuscript. Data collections for this study were in part supported by Columbia River Estuary Data Development Program funds. LITERATURE CITED Craig, ). A., and R. L. Hacker. 1940. The history and development of the Fisheries of the Columbia River. U.S. Fish Wildl. Serv., Fish. Bull., 32:133-216 Dees, L. T. 1961. Sturgeons. U.S. Fish Wildl. Serv., Fish. Leaf!. 526. 8 p. Durkin, ). T., and R. L. Emmett. 1980. Benthic invertebrates, water quality, and substrate texture in Baker Bay, Youngs Bay, and adjacent areas of the Columbia River estuary. Northwest and Alaska Fisheries Center, National Marine Fisheries Service, NOAA, 2725 Montlake Boulevard East, Seattle, WA 98112. 44 p. Haynes, J. M., and R. H. Gray. 1981, Diel and seasonal movements of white sturgeon, Acipenser transmontanus, in the mid-Columbia River. U.S. Natl. Mar. Fish. Sen/., Fish. Bull., 79:367-370.- Higley, D. L., S. L. Wilson, K. K. Jones, and R. L. Holton. 1983. Distribution and community structure of benthic infauna in channel and protected flat habitats of the Columbia River estuary. School of Oceanography, Oregon State University, Corvallis, OR 97331. 81 p. King, S. D. 1985. The 1984 lower Columbia River recreational fisheries: Bonneville to Astoria. Oregon Dept. of Fish and Wildl., Fish Division, Clackamas, Oregon. 47 p. McCabe, C. T. Jr., W. D. Muir, R. L. Emmett, and ). T. Durkin. 1983. Interrelationships between juvenile salmonids and nonsalmonid fish in the Columbia River estuary. U.S. Natl. Mar. Fish Serv., Fish. Bull., 81:815-826. McConnell, R., T. Blahm, C. McCabe, Jr., T. Clocksin, T. Coley, J. Durkin, R. Emmett, and W. Muir. 1983. Columbia River Estuary Data Development Program data report: salmonid and non-salmonid fish. Four volumes, various pagination. Hammond Field Station, Northwest and Alaska Fisheries Center, National Marine Fisheries Service, NOAA, P.O. Box 155, Hammond, OR 97121. McKechnie, R. )., and R. B. Fenner. 1971. Food habits of white sturgeon, Acipenser transmontanus, in San Pablo and Suisun Bays, California. Calif. Fish Came, 57:209-212. Radtke, L. D. 1966. Distribution of smelt, juvenile sturgeon, and starry flounder in the Sacramento-San Joaquin Delta with observations on the food of sturgeon, p. 115-129. In J. L. Turner and D. W. Kelley, compilers. Ecological studies of the Sacramento-San Joaquin Delta. Part 2, Fishes of the Delta. Calif. Dept. Fish Game, Fish Bull. 136. 168 p. Schreiber, M. R. 1962. Observations on the food habits of juvenile white sturgeon. Calif. Fish Game, 48:79-80. Scott, W. B., and E. J. Grossman. 1973. Freshwater fishes of Canada. Bull. Fish. Res. Board Can. 184:966 p. Semakula, S. N. 1963. The age and growth of the white sturgeon (Acipenser transmontanus Richardson) of the Eraser River, British Columbia, Canada. M. S. Thesis, Univ. British Columbia, Vancouver, B.C. 115 p. Semakula, S. N., and P. A. Larkin. 1968. Age, growth, food, and yield of the white sturgeon (Acipenser transmontanus) of the Fraser River, British Columbia. J. Fish. Res. Board Can. 25:2589-2602. HERD SEGREGATION IN HARBOR SEALS 55 Calif. Fish and Game 74 ( 1 ) : 55-59 1 988 HERD SEGREGATION IN HARBOR SEALS AT POINT REYES, CALIFORNIA' SARAH G. ALLEN Point Reyes Bird Observatory 4490 Shoreline Hwy. Stinson Beach, CA 94970 CHRISTINE A. RIBIC Center for Quantitative Science University of Washington Seattle, WA 98195 JANET E. KJELMYR Point Reyes Bird Observatory 4490 Shoreline Hwy. Stinson Beach, CA 94970 A mixed class aggregation of harbor seals resides at Double Point, California. Seals are present year round but are more abundant during the March through June breeding season. Seals were segregated by sex and age class within the herd, and patterns changed during the season. Early in the 1984 and 1985 breeding seasons the herd was equally composed of males, females, and immatures, but mid-way was composed of mostly females and pups. Late in the season more males and immatures were present. We hypothesize that changes in segregation patterns are related to intolerance of males and immatures by pregnant females and females with pups. INTRODUCTION The harbor seal, Phoca vitulina, is gregarious when hauled out on shore. Scheffer and Slipp's (1944) contention that harbor seal aggregations are composed of mixed ages and sexes, has been modified by observations that groups can be composed almost exclusively of one class; for example, nursery (Knudtson 1977, Naito and Konno 1979, Brown and Mate 1983, and Slater and Markowitz 1983), bachelor (Knudtson 1977), and juvenile aggregations (Payne and Schneider 1984). These variations in group structure have been correlated to availability of haul-out space and reproductive and feeding strategies. Little is known about spatial segregation within mixed aggregations. We studied such segregation during the breeding season with two objectives: (i) to identify herd composition; (ii) and to determine whether the relative space-use pattern of various age/sex categories changed as the season progressed. METHODS Double Point is a coastal habitat named for its concave, steep, cliff-backed beach enclosed by two jagged promontories and is located in Marin County, California (lat 37° 56' N, long 122° 46' W). We censused a mixed sex/age group of seals at south Bolsa Beach, a pocket beach at Double Point, nine times during the 1984 and 12 times during the 1985 breeding seasons. Observations were conducted during midday, low to medium low tides when maximum numbers of seals were assumed to be present (Stewart 1984, Allen et al. 1985). South ' Accepted for publication October 1987. 56 CALIFORNIA FISH AND CAME Bolsa Beach contained 75-100% of the total seals present at Double Point. We divided the breeding season into Early (March to mid-April), Middle (mid-April to mid-May), and Late (mid-May to mid-June) periods. These periods were based on the reproductive status of the majority of females and on the age of the pups (early = mostly pregnant females and few pups, middle = majority of females with pups, late — older and weaned pups). We censused each period three times in 1984 and four times in 1985. Censuses were done a week apart and were conducted by one observer from the cliff 30 m above the haul-out. We classified seals as adult male, adult female, immature, and pup. Immature seals were distinguished from adults based on relative size. Seals mature at ages 3-5 (Boulva and McLaren 1979) so some overlap would be expected between the immature and other categories. Pups were easily distinguished by their small size and bright silver pelage at a time when adults and immatures are in their pre-molt condition. Females were further categorized as pregnant, non- pregnant, or with pup. Males were distinguished from females by the penal opening on the ventrum. Because adult male and female seals are almost identical in size and seals often lay on their ventrum, we were not able to sex all adult animals but were able to classify an average of 71% of the herd. We assumed that males and females were equally likely to lay on their ventrum. We calculated the average percent representation of each sex/age category and standard deviation for each period. To determine relative spatial position, we divided the herd into quadrants from the south to the north end of the herd (A south, B, C, D north). Each quadrant equaled one quarter of the total number of seals hauled out on south Bolsa Beach on a given day; throughout the season, quadrant A was bordered by tide pools and B,C,D by deep water. We tested the null hypothesis for independence of position with sex and age class for each breeding period for each season with a Chi-square test of independence and analysis of residuals (Bishop, Fienberg and Holland 1975, Fienberg 1980). RESULTS The prevalence of each sex/age class changed as the breeding season progressed in both 1984 and 1985 (Table 1). During the early period, males, females, and immatures were equally abundant, and pups began to appear; during the middle period the herd was composed primarily of females and pups; and during the late period females and pups still were most abundant, although in 1985 the percentage of males began to increase then. The level of spatial sex/age segregation also changed over the course of the season. The null hypothesis, independence of position in the herd with respect to sex and age class, was rejected for all three periods for both years (Table 2). Analysis of residuals indicated that in the early period most females hauled out in quadrant D, most immatures were in quadrant A, and males tended to be in quadrants B and C. In the middle period females and pups were uniformly present in all quadrants, fewer males were in quadrant D and immatures tended to be absent from quadrant D. In the late period most females were present in quadrant D, more males were present in quadrant B or C, and immatures and HERD SEGREGATION IN HARBOR SEALS 57 weaned pups tended to be in quadrant A or B. For all three periods except late 1984 there was a strong tendency for immatures to be in the extreme opposite quadrant (A) of the herd from females (D). TABLE 1. Percent Composition by Sex and Age Class at Double Point, California. 1984 1985 Early Male Female .... pregnant. with pup non-preg Immature . . Pup Count Middle Male Female pregnant. . with pup . non-preg . Immature . . . Pup Count Late Male Female pregnant. . with pup . non-preg . Immature . . . Pup Count 1* SD X SD n = 3 n = 4 31 2 22 9 32 2 33 9 (61) - (50) (19) - (39) (20) - (11) 31 4 27 15 7 5 16 18 170 90 249 65 12 5 8 3 40 3 44 3 (15) - (3) (83) - (91) (2) - (6) 10 2 5 1 38 8 44 2 296 103 299 46 13 5 21 11 38 7 44 10 (1) - (0.2) (62) - (51) (37) - (49) 17 11 10 5 32 8 25 7 267 56 294 101 * X is the average for each period, SD is the standard deviation, and n is the number of censuses. Percentages of pregnant females, females with pups, and non-pregnant females are taken with respect to total females. Females of different reproductive status (pregnant or with pup) also tended to segregate within the herd (Table 1 ). Pregnant females seen primarily in the early period were mostly present in quadrant D whereas those with pups were present uniformly. During the middle and late periods, females with pups were still seen evenly distributed throughout all quadrants, and females that were neither pregnant nor with pup were primarily counted in quadrant C or D. 58 CALIFORNIA FISH AND CAME TABLE 2. Numbers of Seals by Sex and Age Class at Double Point, for Each Quadrant and Period Within the 1984 and 1985 Breeding Seasons. / '984 1985 Sex/ Age Class Quadrant Quadrant Period A B C D A B C D Early . . Male/adult 33* 49 42 32 62 66 61 33* Female/adult 35 41 33 50 ♦ 48' 65 103 141 • Immature 75* 36 32 21 • 118* 59 56 32* Pup 13 9 5 5 22 29 47 52 X^ = 42 p < 0.005 X^ = 140 p < 0.005 Middle .. Male/adult 22 33 31 13* 26 28 35 9* Female/adult 87 83 86 98 105 125 128 163 Immature 37* 16 18 13 32* 10 10 3* Pup 97 90 80 83 126 125 126 144 X^ = 29 p < 0.005 X^ = 62 p < 0.005 Late . . Male/adult 26 13* 28' 18 54' 103* 50 34 Female/adult 66 56 54 77' 140 101 112 126* Immature 37 26 36' 23 110' 19* 24 20 Pup 76 61 37* 46 129 64 47 41 X^ = 24 p < 0.005 X^ = 141 p < 0.005 * The standardized residuals and Freeman-Tukey deviates for these cells are significant at «^= 0.05. DISCUSSION Segregation by haul-out location has been observed by others; however, these observations often occurred in estuaries or bays v^here ample haul-out space is available. On a rugged coastal site such as Double Point, haul-out space is limited, and seals form a mixed aggregation in one location. During the breeding season, the mixed aggregation of seals at Double Point is not random, but instead is structured by sex, age, and female reproductive status. We hypothesize that segregation on land is in part related to intolerance of males and immatures by females. The degree of segregation, which changed over the season, may be related to changes in the biological condition of females (parturition, weaning, and estrous). Segregation appeared most pronounced during the middle period when the majority of females were with pups; such females are particularly aggressive and will threaten nearby seals. The numbers of males and immatures were greatly reduced in all quadrants at that time, and females with pups were evenly distributed throughout all quadrants. Many advantages exist for females to form sub-clusters within a herd while on shore. They are able to form and maintain tighter bonds with their offspring and to rest and nurse without interference from sexually active males. At the same time, they are able to retain the benefits of gregariousness (i.e. protection from predators, and proximity to males during the mating period). For segregation to persist into the late breeding period is not unexpected because mating occurs primarily in the water. While on shore, females in estrous rest and generally rebuff males (Allen 1985). Some researchers have stated that topography and/or male social interac- tions dictate the spacing patterns of harbor seals on shore (Sullivan 1982, Slater HERD SEGREGATION IN HARBOR SEALS 59 and Markowitz 1983). At Double Point, topography may have partially influenced herd segregation. Ample haul-out space was available at all but the highest of tides; however, ready access to deep water was limited to quadrants B, C, and D. Our observations indicated that in the early breeding period, females, the majority of which are pregnant, may choose quadrant D over the others because there is more space for birthing. Later, females nursing pups shift to quadrants A and B because of proximity to tide pools which are ideal "nursery pens" providing warm, calm water. During very low tides we consistently observed females with pups in these pools. In the late period, immatures and weaned pups may have been relegated to quadrant A because females, most probably in estrous, would be intolerant to them. Most males were in quadrant B with ready access to deep water and departing females. Our observations provided no evidence for or against a male dominance hierarchy. A few males were present in quadrant D which could be interpreted as active exclusion of other males; these males, however, were often clustered rather than dispersed within the quadrant. In conclusion, our results indicate that spatial segregation occurs within a mixed aggregation of harbor seals and that the relative space-use pattern of the different age/sex categories changed over the course of the breeding season as the reproductive status of females changed. We hypothesize that segregation is related to female intolerance of males and immatures. Proof that females are actively excluding other categories would require quantification of interactions between females of different reproductive status and males and immatures. ACKNOWLEDGMENTS Support for this study was contributed by the Point Reyes National Seashore, the Sanctuary Programs Office of the National Oceanic and Atmospheric Administration, the Point Reyes Bird Observatory, and D. Miller. We are also grateful to D. Ainley and three anonymous reviewers for reviewing this manuscript. This is Point Reyes Bird Observatory publication number 322. LITERATURE CITED Allen, S. G. 1985. Mating behavior in the harbor seal. Mar. Mammal Sci., 1:84-87. Bishop, Y. M. M., S. E. Fienberg and P. W. Holland. 1975. Discrete multivariate analysis. The MIT Press, Cambridge, Mass. Boulva. J. and I. A. McLaren. 1979. Biology of the Harbor seal, Phoca vitulina, in eastern Canada. Fish. Res. Board Can., Bull. No. 200 24p. Brown, R. F. and B. R. Mate. 1983. Abundance, movements, and feeding habits of harbor seals, Phoca vitulina, at Netarts and Tillamook Bays, Oregon. Fish. Bull., 81:291-301. Fienberg, S. E. 1980. The analysis of cross-classified categorical data, 2nd ed. The MIT Press, Cambridge, Mass. Knudtson, P. M. 1977. Observations on the breeding behavior of the harbor seal in Humboldt Bay, California. Calif. Fish and Game, 63(1 ): 66-70. Naito, Y. and S. Konno. 1979. The post-breeding distributions of ice-breeding harbor seals, Phoca vitulina, and Ribbon seal, Phoca fasciata, in the southern sea of Okhotsk. Sci. Rep. Whales Res. Inst., 31:105-119. Payne, P. M. and D. C. Schneider. 1984. Yearly changes in abundance of harbor seals, Phoca vitulina, , at a winter haul-out site in Massachusetts. Fish, Bull., 82:440-442. Scheffer, V. B. and J. W. Slipp. 1944. The harbor seal In Washington State. Am. Midi. Nat., 32(2):373-416. Slater, L. M. and H. Markowitz. 1983. Spring population trends in Phoca vitulina richardi in two central California coastal areas. Calif. Fish and Game 69(4):217-226. Stewart, B. S. 1984. Diurnal hauling patterns of harbor seals at San Miguel Island, California. J. Wild. Manage., 48:1459-1461. Sullivan, R. M. 1982. Agonistic behavior and dominance relationships in the harbor seal, Phoca vitulina. J. Mammal., 63:554-569. 50 CALIFORNIA FISH AND GAME Calif. Fish and Came 74 ( 1 ) : 60-63 1 988 NOTES RECORD OF THE CHAMELEON GOBY, TRIDENTIGER TRIGONOCEPHALUS, FROM THE SACRAMENTO-SAN JOAQUIN DELTA On 5 March 1987 a chameleon goby, Tridentiger trigonocephalus (Gill), was collected during routine fish salvage operations at the John E. Skinner Delta Fish Protective Facility near Byron, Contra Costa County, California, approximately 130 km upstream from the Golden Gate Bridge (Skinner 1974). This Facility is located at the head of the California Aqueduct ahead of the FHarvey O. Banks Delta Pumping Plant v^hich exports water from the Sacramento-San Joaquin Delta via Old River. This species has been observed in Los Angeles Harbor and San Francisco Bay (Miller and Lea 1972) and most recently in Suisun Bay (P. B. Moyle and K. Hieb, pers. comm.), however, this is the first record of occurrence this far upriver in the Sacramento-San Joaquin Delta. The surface water temperature and salinity during the time of collection was approximately 12.5°C and 3.2 0/00. Identification of the specimen was confirmed by K. Hieb, California Depart- ment of Fish and Game, Stockton. The specimen was 91 mm TL and weighed 11.4 g. Identification was based on the following characteristics: black spot at base of caudal fin; dark bar followed by a yellow arch on base of pectoral fin; teeth tricuspid; fin ray counts D Vl + I, 12 and A I, 11; midlateral scales 50 (see Miller and Lea 1972). The specimen is being retained in the collection of California Department of Fish and Game, Bay-Delta Project, Stockton, Califor- nia. The chameleon goby was inadvertently introduced from the Orient into San Francisco Bay (Miller and Lea 1972). The most likely mechanism of introduc- tion involves the transport of eggs laid on fouling organisms of ships hulls in the Orient; minimum ship transit time from the Orient, average incubation time of eggs and seawater along shipping routes are favorable for egg development and hatching upon arrival to the west coast of the United States (Haaker 1979). It is a shallow-water form that has not been collected in fresh water in California (Moyle 1976) until this occurrence. Larry Neyman, California Department of Water Resources, Delta Fish Protective Facility Operator, deserves special commendation for his alertness in recognizing this specimen. I would like to thank Kathy Hieb for verifying the identification of the specimen. LITERATURE CITED Haaker, P. L. 1979. Two Asiatic gobiid fishes, Tridentiger trigonoceptialus and Acanthogobius flavimanus, in Southern California. Bull. Southern California Acad. Sci. 78(1):56-61. Moyle, P. B. 1976. Inland Fishes of California. Univ. Calif. Press. 405 p. Miller, D. )., and R. N. Lea 1972. Guide to Coastal Marine Fishes of California. Calif. Dept. Fish & Game, Fish. Bull. 157:235 p. Skinner, J., E. 1974. A Functional Evaluation of a Large Louver Screen Installation and Fish Facilities Research on California Water Diversion Projects. Pages 225-249 in L. D. Jensen, Ed. Proceedings of the Second Entrainment and Intake Screening Workshop. The Johns Hopkins University Cooling Water Research Project, Report No. 15, Baltimore, MD. NOTES 5-| — Paul F. Raquel, Bay-Delta Project, California Department of Fish and Game, 4001 N. Wilson Way, Stockton, California 95205. Accepted for publication August 1987. ADDITIONAL NOTES ON MIGRATIONAL DISTRIBUTION OF NORTHERN PINTAIL BANDED IN CALIFORNIA During 1948-79, the California Department of Fish and Game banded 245,174 northern pintail, Anas acuta, mostly on seven banding stations located through- out California. Analysis of approximately 18,000 band recoveries from these pintails resulted in a comprehensive study of their migration patterns and distribution (Rienecker 1987). The objective of the present study is to document and evaluate pintail band recovery areas, emphasizing those out-of- state. Only indirect band recoveries from the following banding stations were used: Klamath Basin National Wildlife Refuges, Honey Lake Wildlife Area including Mountain Meadows, South San Francisco Bay, Suisun Marsh, Gray Lodge Wildlife Area, Imperial Valley and Los Banos Wildlife Area. Indirect recoveries are bands recovered in subsequent band recovery years (May 1 through April 30) following the year of banding. RECOVERY AREAS California About 79% (N = 13,784) of all indirect band recoveries were recovered in California. The Sacramento Valley ( 23.4%, N =4,096) and the San Francisco Bay-Delta (24.3%, N =4,258) were the two most important recovery areas, followed by the San Joaquin Valley (14.8%, N = 2,586), Imperial Valley (8.2%, N = 1,439) and northeastern California (5.9%, N = 1,029). Washington and Oregon About 1% (N = 174) of recoveries were from Washington, with three quarters of them from Western Washington. Many were recovered in Decem- ber and January, which suggests that they had either wintered there or were at some point in a round-robin migration. Oregon accounted for over 2% (N=409) of recoveries with nearly all from the coastal and southern (Klamath Basin) regions. Idaho, Nevada, Utah and Arizona Less than 1% (N=41 ) of the total recoveries from California bandings were from Idaho, which suggests that there is little stopover by pintails in Idaho during the hunting season. About 1% (N = 182) of recoveries were from Nevada, mostly Western Nevada. The majority of these birds had been banded in Imperial, San Joaquin and Sacramento Valleys. This suggests that Nevada is within the migration routes to both the Central Valley and the Imperial Valley. Utah was an important recovery area, as nearly 3.5% (N = 601) of all 52 CALIFORNIA FISH AND GAME recoveries were taken there with most in the Bear River Marshes of Northern Utah. Many of these pintails were banded in the Imperial Valley. Most of the harvest in Utah took place during October and November, and by January few pintails were left in the state. Less than 1% (N = 66) of recoveries were from Arizona, most of which were from birds banded in Imperial Valley. Few pintails banded north of Imperial Valley migrate to or stop over in Arizona. Alaska Alaska, with less than 1% (N = 142) of recoveries, was not an important recovery area for pintails banded in California, but was an important breeding area for pintails wintering in California (Bellrose 1976). Band recoveries (N = 798) from pintails banded in Alaska showed 41% were taken in California (Patuxent Bird Banding Laboratory, unpubl. data). Major recovery areas for California-banded pintails were the Yukon-Kuskokwim Delta and Cook Inlet. About half of the recoveries were obtained during the hunting season, mostly September, and half during spring and summer months. Half of the spring and summer recoveries, mostly during May, were from shot birds. During the spring, pintails sustain the greatest harvest among ducks due to their relative abundance throughout the Yukon-Kuskokwim Delta (Klein 1966). Canada Almost 4.4% ( N = 767 ) of recoveries were from Canada, mostly from Alberta (2.3%, N=406). Most of the Alberta bands were recovered during September and October, the remainder during spring and summer. Approximately 1% (N = 204) of recoveries were from Saskatchewan, about half of which were recovered in September and October and half during the spring and summer months. Most spring and summer recoveries were from birds found dead from unknown causes and from miscellaneous causes, e.g., hit by farm machinery, entanglement in fishing gear, caught due to injury, steel traps or by raptors. All other areas in Canada combined produced less than 1% (N = 157) of recoveries. Mexico Recoveries from Mexico comprised about 3% (N = 513) of the total and were recovered during all months of the year, but primarily December and January. The west coast of Mexico was the main recovery area, accounting for over half of the recoveries, most of which were from Imperial Valley bandings. Other Flyways From pintail bandings in California, over 3% (N-551) of recoveries were from the Central Flyway and about 1% (N = 141 ) from the Mississippi Flyway. Approximately half of Central Flyway recoveries were from Los Banos and Imperial Valley bandings. Most of the recoveries from the northern half of the Central Flyway States were obtained in October and November, and from the southern half in December and January. Over 2% (N = 385) of recoveries were from the Central Flyway wintering grounds, mainly Texas. Recoveries from the Mississippi Flyway were found among all of the states in the Flyway except Ohio. Louisiana, the major wintering area on the Flyway, accounted for over half of the recoveries. NOTES 63 USSR Less than 1% (N = 110) of recoveries were from the USSR, and all but two were from the north and south coasts of northeastern Siberia. Two recoveries were from farther west, one of which was from a Gray Lodge banded pintail recovered near Lugansk, north of the Black Sea. This is the longest migration on record, half way around the world, for a pintail banded in California. Over half of the Siberian recoveries were from birds shot during May. Henny (1973) suggested that a portion of the pintails wintering in the western United States annually nest in Siberia. Pacific Islands and Eastern Asia In addition to 11 pintails banded in Northern California and retrapped in Hawaii, 9 were recovered at points south and west of the Hawaiian Islands. Three of these distant recoveries were direct recoveries of 58 days from Suisun Marsh to the Line Islands, 82 days from the Klamath Basin National Wildlife Refuges to the Tokelau Islands and 91 days from the Klamath Basin National Wildlife Refuges to the Cook Islands. Medeiros (1958) suggested that pintails fly west on the tradewinds from the California coast and return on the westerlies which reach North America in an extended area from Crescent City, California to the Gulf of Alaska. The pintail is the only duck common to California that migrates this far from its breeding grounds. Band recoveries of pintails from areas outside the normal pattern of breeding in Alaska and Canada and wintering in California add to our knowledge of pintail migrations and distributions. This knowledge is, however, valued more as a curiosity than as a management tool, although it does indicate the pintail's capacity to find and adapt to new habitats. Rienecker (1987) states that the vast majority of pintails banded in California migrate from staging areas in south- eastern Alberta, southwestern Saskatchewan and Alaska to the Central Valley of California, where they spend the winter. All other routes are of minor importance. ACKNOWLEDGMENTS Thanks is given to members of the California Department of Fish and Game and the U.S. Fish and Wildlife Service who banded pintails. Thanks is also given to M. Miller, J. Bartonek and D. Gilmer of the U.S. Fish and Wildlife Service for review of the manuscript. This study was supported by Federal Aid in Wildlife Restoration Project W-30-R, "Waterfowl Studies". LITERATURE CITED Bellrose, F. C. 1976. Ducks, geese and swans of North America. Stackpole Books, Harrisburg, PA. 544 p. Henny, C. J. 1973. Drought displaced movement of North American pintails in Siberia. J. Wildl. Manage., 37(1 ): 23-29. Klein, D. R. 1966. Waterfowl in the economy of the Eskimos on the Yukon-Kuskokwim Delta, Alaska. Arctic, 19(4): 319-336. Medeiros, J. S. 1958. Present status of migratory waterfowl in Hawaii. J. Wildl. Manage., 22(2): 109-117. Rienecker, W. C. 1987. Migration and distribution of pintails banded in California. Fish and Came, 73(3); 139-155. — Warren C. Rienecker, California Department of Fish and Game, 1416 Ninth Street, Sacramento, CA, 95814. Accepted for publication August 1987. M CALIFORNIA FISH AND GAME BOOK REVIEWS Seashore Discoveries By Wesley M. Farmer. Available from: Dr. Wesley M. Farmer, P.O. Box 1323, Santee, Calif. 92071, [no date]; 124 p. $14.95 soft cover. This apparently recent field guide is primarily designed for the novice, occasional, or casual visitor to the seashore. Two hundred and thirty-five species are covered with 240 line drawings and four photographs. The primary emphasis is on southern California and San Diego county in particular. As one might expect, invertebrates make up the bulk of the subject matter with eleven fish, four birds, two mammals, and only ten algae included. The drawings are well detailed, but more photographs would be helpful for species closely resembling others or those with distinctive coloration. Most of the text consists of brief yet concise descriptions, ranges, distributions, and in some cases, notes on ecology or natural history for each organism. I noticed a few discrepancies, which seem mostly editorial in nature, such as describing Phyllospadix as both an alga and a flowering plant, green abalone commonly being 10 inches long, the maximum size of red sea urchins being only 100 mm, and some confusion about the relative sizes of Pisaster giganteus and ochraceus. An encouraging feature is the emphasis on conservation of our intertidal resources. A typical nature walk in the rocky intertidal is described, and a creel census (although dated) of tidepool visitors is presented which documents the considerable impact on intertidal populations that even casual collecting can have. This would be a useful book for anyone seeking to learn about or introduce someone to the seashore life along the southern California coast. — David O. Parker Mammals in Hawaii, Second Edition. A synopsis and notational bibliography. By P. Quentin Tomich. Bishop Museum Press, Honolulu, Hawaii, 1986; 375 p. $42.95 plus $1.00 for handling and shipping. As with the 1969 edition, this book offers excellent species accounts with extensive details and bibliographic references documenting 22 Hawaiian mammals both historically and naturally. About half the book is dedicated to useful references for each species; while the text of individual species accounts is written in a scholarly, easily understood style for the scientist or the lay person. The accounts are packed with pertinent information such as dates of arrival for introduced species, range in Hawaii, habitat requirements, conflicts with other species. . . . For some of the species accounts, it is somewhat distracting that the author has obviously updated the first edition (to include recent publications) by adding paragraphs to the 1969 text instead of rewritting the original account. Many of the species accounts (especially marine mammals) are well illustrated with 27 black and white photographs which further enhance a skillfully constructed synopsis. The author has provided a great deal of information about each animal and references that allow more complete investigation as desired. The annotated bibliography is very well done. Each reference contains important notes about the reference, thus creating a readable and useful bibliography. The marine mammal descriptions are excellent and highlight the book. The author provides additional comments on the environmental perils of introduced species and other aspects of mammals in Hawaii that make the book interesting and a valuable asset. — Doyle A. Ha nan 88 77262 INSTRUCTIONS TO AUTHORS EDITORIAL POLICY California Fish and Game is a technical, professional, and educational journal devoted to the conservation and understanding of fish and v/ildlife. Original nnanu- scripts submitted for consideration should deal with the Califoriiia flora and fauna or provide information of direct interest and benefit to California researchers. Authors may submit an original plus two copies, each, of manuscript, tables, and figures at any time. MANUSCRIPTS: Authors should refer to the CBE Style Manual (Fifth Edition) and a recent issue of California Fish and Game for general guidance in preparing their manuscripts. Some major points are given below. 1. Typing — All material submitted, including headings, footnotes, and literature cited must be typewritten doublespaced, on white paper. Papers shorter than 10 typewritten pages, including tables, should follow the format for notes. 2. Citations — All citations should follow the name-and-year system. The "library style" will be followed in listing literature cited. 3. Abstracts — Every article must be introduced by a concise abstract. Indent the abstract at each margin to identify it. 4. Abbreviations and numerals — Use approved abbreviations as listed in the CBE Style Manual. In all other cases spell out the entire word. TABLES: Each table should be typewritten with the heading margin left justified. Tables should be numbered consecutively beginning with "1" and placed together in the manuscript following the Literature Cited section. Do not double space tables. See a recent issue of California Fish and Game for format. FIGURES: Consider proportions of figures in relation to the page size of California Fish and Game. The usable printed page is 1 1 7 by 1 91 mm. This must be considered in planning a full page figure, for the figure with its caption cannot exceed these limits. Photographs should be submitted on glossy paper with strong contrasts. All figures should be identified with the author's name in the upper left corner and the figure numbers in the upper right corner. Markings on figures should be made with a blue china marking pencil. Figure captions must be typed on a separate sheet headed by the title of the paper and the author's name. PROOFS: Galley proofs will be sent to authors approximately 60 days before publi- cation. The author has the ultimate responsibility for the content of the paper and is expected to check the galley proof carefully. PAGE CHARGES AND REPRINTS: All authors will be charged $35 per page for publication and will be billed before publication of manuscripts. Reprints may be ordered through the editor at the time the proof is submitted. Authors will receive a reprint charge schedule along with the galley proof. a Z z H z Ul H ^5 5 o =: z ^ -n o = s o m Z