QUIFDRNIA
FISH-GAME

"CONSERVATION OF WILDLIFE THROUGH EDUCATION"

California Fish and Game is a journal devoted to the conservation of wild-
life. If its contents are reproduced elsewhere, the authors and the California
Department of Fish and Game would appreciate being acknowledged.

Subscriptions may be obtained at the rate of $5 per year by placing an
order with the California Department of Fish and Game, 1416 Ninth Street,
Sacramento, California 95814. Money orders and checks should be made out
to California Department of Fish and Game. Inquiries regarding paid sub-
scriptions should be directed to the Editor.

Complimentary subscriptions are granted, on a limited basis, to libraries,
scientific and educational institutions, conservation agencies, and on exchange.
Complimentary subscriptions must be renewed annually by returning the post-
card enclosed with each October issue.

Please direct correspondence to:

Perry L. Herrgesell, Ph.D., Editor
California Fish and Game
1416 Ninth Street
Sacramento, California 95814

u

D

VOLUME 70

JULY 1984

NUMBER 3

Published Quarterly by

STATE OF CALIFORNIA

THE RESOURCES AGENCY

DEPARTMENT OF FISH AND GAME

—LDA—

130 CALIFORNIA FISH AND GAME

STATE OF CALIFORNIA
GEORGE DEUKMEJIAN, Governor

THE RESOURCES AGENCY
GORDON VAN VLECK, Secretary for Resources

FISH AND GAME COMMISSION

WILLIAM A. BURKE, Ed.D,, President

Brentwood

BRIAN J. KAHN, Vice President ABEL C. GALLETTI, Member

Santa Rosa Los Angeles

NORMAN B. LIVERMORE, JR., Member ALBERT C. TAUCHER, Member

San Rafael Long Beach

DEPARTMENT OF FISH AND GAME
JACK C. PARNELL, Director

1416 9th Street
Sacramento 95814

CALIFORNIA FISH AND GAME

Editorial Staff

Editorial staff for this issue consisted of the following:

Wildlife William E. Grenfell, Jr.

Marine Resources Robert N. Lea, Ph.D.

Inland Fisheries Jack Hanson

Striped Bass Donald E. Stevens

Editor-in-Chief Perry L. Herrgesell, Ph.D.

CONTENTS 131

Page

Distributions of Some Common Decapod Crustaceans and a

Pycnogonid from the Contintental Shelf of Northern

California Mary K. Wicksten 132

Diurnal Activity and Habitat Use by Feral Pigs on Santa Cruz

Island, California Dirk Van Vuren 140

Environmental Conditions and Fish Faunas in Low Elevation

Rivers on the Irrigated San Joaquin Valley Floor,

California Michael K. Saiki 145

Relocation of Original Territories by Displaced Black-and-yel-

low Rockfish, Sebastes chrysomelas, from Carmel Bay,

California Leon E. hiallacher 158

Surgical Implantation of a Radiotelemetry Device in Wild

Black Bears, Ursus americanus David A. Jessup

and Donald B. Koch 163

Seasonal Variation in the Diet of the Threespine Stickleback,
Gasterosteus aculeatus, in Contra Costa County,
California Randal J. Snyder 167

Attacks on Divers by White Sharks in Chile Alfredo Cea Egana

and John E. McCosker 173

Notes

Ecological Status of Striped Bass, Morone saxatilis, in Upper

Newport Bay, California Michael H. Horn, Larry G. Allen

and F. Dennis Hagner 180
Food Habits of Black-tailed Deer, Odocoileus hemionus

columbianus, in Trinity County, California John G. Kie,

Timothy S. Burton, John W. Menke,

and William E. Grenfell, Jr. 183

Additional Notes on the Repeat Spawning by Pacific

Lamprey John H. Michael, Jr. 186

A Northern Fur Seal, Callorhinus ursinus, Found in the Sacra-
mento-San Joaquin Delta L. A. Dierauf 189

Exteme Mercury Concentrations of a Striped Bass, Morone
saxatilis, With a Known Residence Time in Lahontan Reser-
voir, Nevada James J. Cooper and Steven Vigg 190

ERRATUM

Vondracek, Bruce, Larry R. Brown, and Joseph J. Cech. 1982. Comparison of
age, growth, and feeding of the Tahoe sucker from Sierra Nevada
streams and a reservoir. Calif. Fish Game, 68(1): 36-46.
Page 41 , Figure 3. The equations shown on Figure 3 should be changed to read
as follows: Stampede Reservoir, LOG W = -3.576 + 2.487 LOG L; Sage-
hen Creek, LOG W = -4.314 + 2.787 LOG L; Little Truckee River, LOG
W = -4.083 + 2.677 LOG L.

132 CALIFORNIA FISH AND CAME

Calif. Fish and Came 70 ( 3 ) : 1 32—1 39 1 984

DISTRIBUTIONS OF SOME COMMON DECAPOD

CRUSTACEANS AND A PYCNOGONID FROM THE

CONTINENTAL SHELF OF NORTHERN CALIFORNIA^

MARY K, WICKSTEN
Department of Biology
Texas A&M University

College Station, Texas
77843

Ten decapods and the pycnogonid Nymphon pixellae were taken at more than
three stations by the R. V. SEARCHER by otter trawling. Collections were made from
off Trinidad Head to off Bodega Bay. Pandalus danae ind Oregon/a graci/is occurred
at 92 m or less on mixed bottoms. Crabs of the genus Oncer usually lived at similar
depths on sand. At 37-92 m on sand, Pagurus armatus, Crangon spinosissima, and
Spirontocaris lamellicornis were collected. Pandalus jordani usually was taken on
mud deeper than 92 m. Spirontocaris liolmesiwidiS found primarily deeper than 130
m. Crangon alasliensis and N. pixellae were taken throughout the depth range sam-
pled, 15-185 m. These records agree with other published records of decapods taken
in the area. Sicyonia ingentis And Cancer anthonyi, common on the continental shelf
of southern California, have not been trawled off northern California.

INTRODUCTION

The soft bottoms of the continental shelf off northern California have long
been known for yielding rich harvests of decapod crustaceans. The major com-
meaaaal species of the area are the Dungeness crab, Cancer magister, and the
ocean shrimp, Pandalus jordani.

The most extensive study of benthic decapods in the area between San
Francisco and Oregon was that of the steamer Albatross in January 1912-April
1913. Dredging operations included some areas just outside the Golden Gate,
but were concentrated in San Francisco Bay (Schmitt 1921 ). Goodwin (1952)
reported on collections made by the N. B. SCOFIELD using a beam trawl off
Humboldt County, Fort Bragg, and Drakes Bay. A few specimens of decapods
taken by commercial fishermen in the area have been donated to the collections
of the California Academy of Sciences (CAS) and the Allan Hancock Founda-
tion (AHF), University of Southern California. In 1971, the R. V. SEARCHER,
under the auspices of the Los Angeles County Museum of Natural History,
collected specimens by otter trawling at 35 stations from 12.5 miles north of
Trinidad Head, Humboldt County to off Tomales Point, Marin County.

It can be very difficult to collect marine specimens off northern California.
Strong winds, rough seas, and harsh weather prevent regular sampling. Except
for the two commercial species, decapod crustaceans are not sufficiently com-
mon to obtain in quantities suitable for statistical analysis. Finer definitions of the
distributions of the species discussed here await more detailed sampling pro-
grams.

' Accepted for publication July 1983.

PYCNOCONID AND DECAPOD DISTRIBUTIONS 133

METHODS

At the Allan Hancock Foundation, I sorted and identified to species the
crustaceans and pycnogonid taken by the R. V. SEARCHER. These specimens
were collected with an otter trawl towed for 20 minutes at various stations
(Figure 1 and Table 1 ). Further information on the stations of the SEARCHER
is available through the Allan Hancock Foundation (Crustacea Section) and the
Los Angeles County Museum of Natural History ( Division of Ichthyology) . I also
examined at the California Academy of Sciences other specimens taken by the
SEARCHER off Marin County.

TABLE 1. List of R. V. SEARCHER Stations

HUMBOLDT COUNTY:
August 5, 1971:

173
176
177
178
179:
180

12.5 mi. N. of Trinidad Harbor, 55 m
12 mi. NNW of Trinidad Harbor, 110 m
12 mi. NW of Trinidad Harbor, 129 m
10.5 mi. NW of Trinidad Harbor, 148 m
10 mi. NW and W of Trinidad Harbor, 166 m
9.5 mi. W of Trinidad Harbor, 185 m

August 6, 1971:

184: 8 mi. NNW of Cape Mendocino, 65 m
185: 8 mi. NW of Cape Mendocino, 83 m
186: 5.5 mi. SW of Pt. Delgada, 138 m
187: 5 mi. SW of Pt. Delgada, 92 m
188: 2 mi. SW of Pt. Delgada, 46 m

)uly 29, 1971:

159: 1 mi. W of Pt. Delgada, 24 m

August 7, 1971:

189: 1 mi. S of Shelter Cove, 24 m
190: 2 mi. S of Shelter Cove, 28 m
191: 3 mi. S of Shelter Cove, 31 m
192: 5 mi. S of Shelter Cove, 37 m

MENDOCINO COUNTY:

August 7, 1971:

193: 9 mi. NW of Fort Bragg, 138 m
194: 5 mi. N of Fort Bragg, 92 m
195; 4.5 mi. N of Fort Bragg, 74 m
196: 6 mi. N of Fort Bragg, 55 m
197: 7 mi. N of Fort Bragg, 37 m

August 8, 1971:

199: 6 mi. S of Fort Bragg, 102 m
200: 8.5 mi. SW of Fort Bragg, 138 m
201: 12 mi. SW of Fort Bragg, 185 m

August 9, 1971:

205: 1 mi. S of Arena Cove Buoy, 55 m
207: Cualala Point, depth not recorded
208: 12 mi. S of Pt. Arena, 92 m or less

134 CALIFORNIA FISH AND CAME

TABLE 1. Continued

July TJ, 1971:

152: 1.5 mi. from Arena Cove Pt., 55 m

153: 1.25 mi. abeam of Arena Cove Buoy, depth not recorded

SONOMA-MARIN COUNTIES:

August 10, 1971:

209: 0.5 mi. SSW of Mile Rocks, 19 m

210: 1 mi. SW of Gull Rock, 28 m

211: 1.5 mi. W of Duncan's Landing, 37 m

213: 1 mi. W of Bodega Head, 55 m

August 20, 1971:

A: 1.45 mi. NNW of Tomales Pt., 31^3 m

B: 0.25 mi. N of Tomales Pt., 15-18 m

C: 1 mi. S of Bodega Head, 55 m

D: 3 mi. N X W of Tomales Pt., 74 m

Note: the stations of August 20 were not assigned numbers.
I have given them letter designations.

Records of the ten decapods and one pycnogonid taken at three or more
stations of the SEARCHER are published here for the first time. Range extensions
of other, less common species taken in the area are published elsewhere (Haig
and Wicksten 1975; Wicksten 1976, 1984). The records of the species taken by
the SEARCHER were compared with records given by Schmitt (1921) and
Goodwin (1952) for the coast of northern California, Butler (1980) for British
Columbia, Word and Charwat (1976) and Wicksten (1980) for the continental
shelf of southern California.

RESULTS

The SEARCHER collected 21 species of decapods off northern California by
trawling. Of these, only six shrimp, one hermit crab, and three brachyuran crabs
occurred at three or more stations ( Table 2 ) . These species also varied by depth
(Table 3).

Crangon alaskensis was by far the most common decapod taken by the
SEARCHER. Up to 175 (at station 159) were collected in a single tow (Table
2). A total of 29 C alaskensis from 14 stations were infected with bopyroid
isopods.

Pagurus armatus was the only anomuran common in trawls. The shells it
inhabited were covered by a hydroid, Hydractinia sp.

DISCUSSION

From the records of the specimens taken by the SEARCHER, nine out of
eleven common species tend to be more common at certain ranges of depth.
Pandalus danae, Oregonia gracilis, and the two species of Cancer (jordani diud
magister) were only collected at depths of 92 m or less. Spirontocaris lamellicor-
nis, Crangon spinosissima, and Pagurus armatus were most abundant at 46-92
m, although P. armatus a\so was taken at shallower and greater depths. P.jordani
and 5. holmesi v^ere taken mostly at more than 92 m. These depth ranges are
consistent with previous records of the species, although all have been reported

PYCNOCONID AND DECAPOD DISTRIBUTIONS

135

at greater and lesser depths (Schmitt 1921, Goodwin 1952, Garth 1958,
McLaughlin 1974, Butler 1980, Garth and Abbott 1980). Crangon alaskensis\NdiS
taken at depths from less than 19 m to 185 m. Nymphon pixellae a\so occurred
at a wide range of depths.

FIGURE 1. Station locations.

136 CALIFORNIA FISH AND CAME

TABLE 2. Species by Station

CLASS CRUSTACEA, ORDER DECAPODA:

Section Caridea:

Crangon alaskensis Lockington

Stations and number of specimens per station: 152 (28), 159 (175), 173 (85), 176 (13), 177 (25),

178 (39), 179 (6), 180 (20), 184 (20), 185 (30), 186 (20), 187 (11), 188 (25), 189 (37), 190 (109),

191 (95), 192 (82), 193 (5), 194 (70), 195 (47), 196 (94), 197 (16), 199 (47), 201 (19), 205 (44),

207 (20), 208 (32), 209 (44), 210 (166), 211 (35).

Crangon spinosissima Rathbun

187 (1), 194 (8), 195 (1), 199 (7), 205 (1), 213 (9).

Pandalus danae Stimpson

189 (28), 192 (2), 213 (23), A (3), D (2).

Pandalus Jordan! Rathbun

176 (24), 177 (15), 178 (37), 179 (28), 180 (33), 186 (28), 187 (1), 193 (1), 201 (4), 207 (1).

Spirontocaris holmesi Holthuis
178 (2), 179 (5), 201 (8).

Spirontocaris lamellicornis ( Dana )

152 (1), 173 (4), 184 (4), 185 (1), 186 (1), 195 (1).

Section Anomura:
Pagurus armatus (Dana)

152 (1), 173 (23), 179 (1), 184 (6), 185 (1), 192 (2), 195 (8), 196 (3), 197 (1),201 (5), 205 (24),

207 (7), 208 (1), 209 (1).

Section Brachyura:
Cancer jordani Rathbun

196 (1), 209 (2), 211 (1).

Cancer magister Dana

208 (3), 209 (2), 210 (2).

Oregonia gracilis Dana

153 (1), 159 (1), 196 (3), 197 (1), 210 (2), 211 (4), A (1), D (1).

CLASS PYCNOGONIDA:

Nymphon pixellae Scott

186 (8), 187 (49), 190 (1), 195 (5), 197 (2).

Substrate is important in the distribution of benthic decapods. P. danae lives
on bottoms of sand and gravel or shell (Schmitt 1921, Butler 1980). O. gracilis
occurs on various substrates, including rocks (Garth 1958). Cancer magister
lives buried in sand (Garth and Abbott 1980). Crangon alasl<ensis, C spinosis-
sima, and Pagurus armatus have been reported from sandy bottoms (Schmitt
1921, Butler 1980). 5. lamellicornis has been found on bottoms of mud or sand
(Butler 1980). P. jordani Wwqs on bottoms of green mud (Goodwin 1952, Pruter
and Harry 1952, Butler 1980). There are no previous reports of the substrate
preferences of S. holmesi, C jordani, and N. pixellae. Judging from their occur-
rences by depth, however, one might guess that they inhabit mud, sand, and
sand or mud, respectively.

PYCNOCONID AND DECAPOD DISTRIBUTIONS

137

a
O

1/)
a

Q.

e
o

3

UJ

efi

<

I

t

T
u-i

s

O 1^

r^ rr^ a^ O^

•<r >— ^£>

<^ LTl

o-i CO ro rs(

5

^ !-sl

■— oo

^

1

I

LT)

(T>

^

1 —

<-M

•8

»"-

1

1^

1

T3

C

•q-

,

^

rsi

■—

§

5

^

OS

-3

1 — rsi

O^

CO

^

rsi

l\

a

i

U~, Tf

■^

1

5

r\

3

LA

2P
•so"

LO U-l

^

LTl

1

5

l3^

60

o-^

1

1-^

0

1

rsi

1

0-.

Tl

5

§

.2

.1

^ Hj C
c: c:

• i

c: c: g
Qc oc S-

€ 3 ^

si i

5 -^

[B rrj

15. ,T! ,Ti ,c; vi ^
Ot ^o L) 'o O <

") ^ isr*
S3 J S

.!5

I

"a;

(3

c: is

2
I-

■i; Q.
^T3

^

5

T3

o
u

a;

O
c

E S

a

oi

o

z

138 CALIFORNIA FISH AND CAME

The eleven common species taken by the SEARCHER seem to form three
groups. The first, the shallowest group, contains species of mixed bottoms [P.
danae and O. gracilis) and sandy bottoms (the two species of Cancer). At
intermediate depths a second group lives on sand, containing 5. lamellicornis,
P. armatus, and C. spinosissima, P. jordani and 5. Iiolmesi form a deep group,
living on mud. Crangon alasl<ensis and N. pixellae occur throughout the entire
range of depths.

All of the eleven common species mentioned here occur north of California,
ranging as far as Alaska or British Columbia ( Scott 1 91 2, Schmitt 1 921 , McLaugh-
lin 1974, Butler 1980). To the south, however, three species reach their southern
limits. O. gracilis has not been reported south of Monterey Bay, California (Garth
1958). Cancer magister has been reported once from Santa Barbara, but the
southernmost sizable population is in the area of Avila to Morro Bay, San Luis
Obispo County (Garth and Abbott 1980). S. lamellicornis has been reported
once from Point Conception and once from Santa Monica Bay (Standing 1981,
Word 1983). Crangon alaskensis is common on the shelf of both northern and
southern California. The other seven species range into southern California, but
have not been reported as common in trawls (Word and Charwat 1976, Wicks-
ten 1980).

Two decapods common off southern California in trawls were not collected
on the continental shelf of northern California. The peneid prawn, Sicyonia
ingentis, has not been collected north of Santa Barbara County (AHF unpubl.
data). The yellow crab. Cancer anthonyi, ranges north to Humboldt Bay, Califor-
nia, but has been said to be "uncommon north of San Pedro" (Garth and Abbott
1980). From records I have seen, this crab seems to inhabit sheltered waters
north of San Pedro. The crab has been collected at King Harbor, Redondo Beach
(Straughan 1978); Princeton Harbor, San Mateo County (MKW, unpubl. field
notes). Fort Baker, San Francisco Bay and Bodega Bay (AHF unpubl. data) and
Humboldt Bay (Willis 1968). Both the crab and the prawn probably prefer
warmer, more protected waters than the common species of the northern coast.

ACKNOWLEDGMENTS

I thank J. Garth, J. Haig, and R. Brusca for their help while I worked at the Allan
Hancock Foundation and D. Chivers and W. Lee for assistance at the California
Academy of Sciences. D. Gotshall of the California Department of Fish and
Game gave helpful suggestions on the original manuscript. C. Swift, Los Angeles
County Museum of Natural History, donated specimens and provided data from
the collections of the SEARCHER \o the Allan Hancock Foundation. The opera-
tions of the SEARCHER were supported by the Janss Foundation, Thousand
Oaks, California.

LITERATURE CITED

Butler, T.H. 1980. Shrimps of the Pacific coast of Canada. Canad. Bull. Fish. Aquat. Sci. 202: 280 p.

Garth, ).S. 1958. Brachyura of the Pacific coast of America: Oxyrhyncha. Allan Hancock Pac. Exped. 21. Part 1,

text, 499 p.
Garth, J.S. and DP. Abbott. 1980. Brachyura: the true crabs, p. 594-630. In Morris, R.H., D.P. Abbott, and E.G.

Haderlie, eds. intertidal invertebrates of California. Stanford Univ. Press, Stanford, Calif.
Goodwin, D.G. 1952. Crustacea collected during the 1950 bottom-fish investigations of the M.V. N.B. Scofield.

Calif. Fish Game, 38(2): 163-181.

PYCNOCONID AND DECAPOD DISTRIBUTIONS 139

Haig, ). and M.K. Wicksten. 1975. First records and range extensions of crabs in California waters. Bull. So. Calif.

Acad. Sci., 74(3): 100-104.
McLaughlin, P. 1974. The hermit crabs (Crustacea: Decapoda, Paguridea) of northwestern North America. Zool.

Verhand. Leiden, 130: 396 p.

Pruter, A.T. and G.Y. Harry. 1952. Results of preliminary shrimp explorations off the Oregon coast. Fish. Comm.
Oregon Res. Briefs, 4(1): 12-24.

Schmitt, W.L. 1921. The marine decapod Crustacea of California. Univ. Calif. Publ. Zool. 23: 1^70.

Scott, F.M. 1912. On a species of Nymphon Uom ihe North Pacific. Ann. Mag. Nat. Hist., ser. 8, 10(56): 206-209.

Standing, J.D. 1981. Occurrences of shrimps (Natantia: Penaeidea and Caridea) in central California and Oregon.
Proc. Biol. Soc. Wash., 94(3): 774-786.

Straughan, D. 1978. Report on field biology of subtidal rocky shores, /n Influence of power generating facilities
on southern California coastal waters, phase 4: study of enrivonmental conditions in King Harbor April 1,
1977-March 31, 1978. So. Calif. Edison Res. and Develop. Ser. 78-RD-108. 157 p.

Wicksten, M.K. 1976. First record of Argis levior (Rathbun) from California. Bull. So. Calif. Acad. Sci., 75(1 ): 56.

1980. Mainland and insular assemblages of benthic decapod crustaceans of southern California, p.

357-367. In Powers, D.M. ed. The California islands: proceedings of a multidisciplinary symposium. Santa

Barbara: Santa Barbara Mus. Nat. Hist.
1984. New records and a new species of hippolytid shrimp from the northeastern Pacific. Crustaceana.

(In press).

Willis, M. 1968. Northern range extension for the yellow crab. Cancer anthonyi. Calif. Fish Game, 54(3): 217.

Word, ).Q. 1983. Spirontocaris lamellicornis (Dana, 1852), new to the fauna of southern California (Decapoda:
Hippolytidae). Calif. Fish Game, 69 (1): 58-60.

Word, ).Q. and D. Charwat. 1976. Invertebrates of southern California coastal waters. II. Natantia. El Segundo,

Calif.: So. Calif, Coastal Water Res. Project. 238 p.

140 CALIFORNIA FISH AND CAME

Calif. Fish and Came 70(2,): 140- 1 44 1 984

DIURNAL ACTIVITY AND HABITAT USE BY FERAL PIGS
ON SANTA CRUZ ISLAND, CALIFORNIA '

DIRK VAN VUREN

Department of Systematics and Ecology

University of Kansas

Lawrence, KS 66045

Feral pigs, Sus scrofa, on Santa Cruz Island were active mornings and evenings
during fall and spring and were active at midday during winter. Canyon bottoms
were frequented in fall and early winter, with a shift to ridgetops during late winter
and spring. Pigs occurred most often in chaparral and oak woodland in fall, then
moved to chaparral-grassland and grassland from midwinter through spring. Activity
and distribution of pigs were related to seasonal changes in temperature and food
availability, and to availability of escape cover.

INTRODUCTION

Feral pigs were introduced to California in 1769 and have become the second
nnost abundant big game species in the state (Barrett 1978). Although popular
as a game animal, pigs can cause severe damage to natural ecosystems and
agricultural lands (Wood and Barrett 1979). Despite their importance, feral pigs
in California have received relatively little study. Published descriptions of diur-
nal activity and habitat use are largely qualitative ( Pine and Gerdes 1 973, Barrett
1978, Mansfield 1978, Barrett and Pine 1981 ); only Barrett (1982) has reported
quantitative data on habitat use by pigs. My objective was to evaluate seasonal
patterns of diurnal activity, distribution on slopes, and use of plant communities
by feral pigs on Santa Cruz Island, California. This information will help identify
those areas where damage by feral pigs is most likely to occur.

STUDY AREA

Santa Cruz Island is about 25,000 ha in size, 40 km offshore of Santa Barbara,
Santa Barbara County. The climate is the Mediterranean type characteristic of
much of California, with hot, dry summers and cool, wet winters. Most of the
52 cm average annual precipitation falls as rain between November and April.
Free water was available in most canyons on the island throughout the study.
Topography of most of the island is rugged, with a maximum elevation of 750
m. Plant communities were described by Philbrick and FHaller (1977) and
mapped by Minnich (1980), who reported that 84% of the island supported
grassland, chaparral, and oak woodland communities. Feral pigs have inhabited
Santa Cruz Island since at least the 1920's ( Rogers 1929), and possibly since the
mid-1 800's (Brumbaugh 1980). In 1964, McKnight (1964:44) reported "several
thousand" on Santa Cruz, Santa Rosa, and Santa Catalina islands. Population size
on Santa Cruz Island at the time of my study was unknown.

METHODS

I obtained data on feral pigs through visual surveys, by foot or from a vehicle,

conducted almost daily from 24 October 1979 through 19 May 1980. Because

of limited road access and other constraints, survey effort was not distributed

uniformly over the island; however, about two-thirds of the island was surveyed

' Accepted for publication October 1983.

FERAL PICS ON SANTA CRUZ ISLAND 141

at least once. For each pig observed, I recorded time of day (PST), location on
canyon slope, and plant community. Approximate slope location was deter-
mined by visually dividing into thirds (upper, middle, lower) the slope on which
each pig stood, and assigning the pig to the appropriate division. Plant commu-
nity was designated according to dominant vegetation within 30 m of each pig.
Community classifications were similar to those of Philbrick and Haller (1977)
and Minnich (1980), with one addition; areas dominated by a mixture of cha-
parral shrubs and open grassy areas were classified as chaparral-grassland. Pref-
erence by pigs for each plant community was evaluated with a preference index
(Van Dyne and Heady 1965), which was calculated by dividing percent cover
on the island (Minnich 1980) into percentage of pigs observed in the commu-
nity. Minnich (1980) included chaparral-grassland areas in the grassland com-
munity, and consequently a single preference index was calculated for both
communities. Preference was indicated by an index greater than 1.0.

RESULTS AND DISCUSSION
Diurnal Activity

A total of 701 pig observations was recorded during 699 hours of field work.
The number of pigs recorded per hour of observation exhibited a bimodal
pattern during most of the study, with peaks during morning and evening hours
( Figure 1 ) . An exception to this pattern occurred during December and January,
when a unimodal pattern was evident. Average number of pigs observed per
hour was greatest during February and March, and least during April and May.

Pig activity probably was influenced by seasonal changes in temperature and
food availability (Barrett 1978). Mild temperatures in fall (October-November)
were associated with a morning and evening activity pattern. Pigs became more
active at midday during the short, cool, often rainy days of early winter ( Decem-
ber-January). The return of drier, milder weather in late winter (February-
March), particularly March, coincided with a return to morning and evening
activity. Pigs were observed more often during this period, probably because
they had moved into more open habitats to forage and hence were more visible.
Frequency of pig observations declined 70% between late winter and spring
(April-May), probably because pigs shifted to nocturnal activity as conditions
became warm and dry.

Habitat Use
In October, most pigs were at midslope or in canyon bottoms (Figure 2). By
December and January most observations were in canyon bottoms. During
February, March, and April, pig activity shifted toward ridgetops, and in May
midslopes were used most frequently. Changes in pig distribution on canyon
slopes probably were related to a combination of weather conditions and food
availability. Santa Cruz Island was very dry from October through mid-Decem-
ber, when pigs generally avoided ridgetops; canyon bottoms were much cooler
and water was available in most. Winter storms in January and February brought
moisture and growth of herbaceous vegetation, and coincided with movement
of pigs to ridgetops. Very little rain fell after March, and most grasses and forbs
had dried by May, with a concurrent downslope movement by pigs. Pine and

142

CALIFORNIA FISH AND CAME

Gerdes (1973) observed that pigs in Monterey County remained near damp
creek bottoms during the dry season and frequented ridgetops in winter and
spring.

6-8 8-10 10-12 12-14 14-16 16-18 18-20
Time of Day (2-hr intervals)

FIGURE 1. Seasonal variation in diurnal activity of feral pigs on Santa Cruz Island.

Evaluation of plant community use and preference by pigs must be considered
in light of bias in observations of pigs; survey effort was not distributed equally
among plant communities, and pigs were not equally visible in different com-
munities. Chaparral was the community sampled most intensively in proportion
to its availability, but it was also the community in which pigs were least visible,
tending to cancel some of the bias. Moreover, bias in survey effort was relatively
consistent from month to month, and hence comparisons among months proba-
bly reflected trends in actual pig distribution.

Most pigs observed in October and November were in chaparral, grassland,
or oak woodland (Table 1 ). In December and January, occurrence in chaparral
and oak woodland declined, while occurrence in grassland increased. Beginning
in February and March, and continuing through April and May, few pigs were
observed in chaparral as they apparently shifted to more open habitats, appear-
ing most often in chaparral-grassland and grassland.

FERAL PICS ON SANTA CRUZ ISLAND

143

100t-

CO

(D

n

-— ^

o

U)

>

L

c

Q)

o

0)

c
o

J3
O

-•-^

fO

n

en

^—

0_

-f—

en

*^-

n

Q

^p

en

o~~

Q.

Oct

Nov

Dec

Jan

Feb

Mar

Apr

May

FIGURE 2. Distribution of feral pigs on canyon slopes on Santa Cruz Island.

TABLE 1. Bimonthly Percentages of Feral Pigs Recorded in Plant Communities on Santa
Cruz Island. Preference Indices are in Parentheses.

Oct-Nov Dec-Jan Feb-Mar Apr-May All months

Pig observations 151

Chaparral 45 (1.5)

Grassland' 21 (0.7)

Chaparral-grassland 1 1

Oak woodland 23 (3.3)

Other 0

147

289

114

701

36(1.2)

2 (0.1)

5 (0.2)

19(0.6)

34(1.0)

18(1.7)

27 (1.8)

23 (1.3)

14

61

54

39

11 (1.6)

9(1.3)

10(1.4)

12 (1.7)

5

10

4

7

' Preference indices calculated from combined grassland and chaparral-grassland data.

Pigs preferred chaparral and oak woodland in October and November (Table
1). Preference for grassland (includes both grassland and chaparral-grassland
communities) increased in subsequent months, and by April and May grassland
was the most preferred community, while chaparral was least preferred. Oak
woodland remained a preferred community throughout the study. Barrett and
Pine (1981) reported that habitats which included oak woodland were most
preferred by feral pigs in San Benito County, Mansfield (1978) noted that wild
pigs in Monterey County depended on oak grassland sites for foraging, and
Barrett (1982) found that feral pigs preferred oak thicket vegetation in Tehama
County.

Distribution of pigs among plant communities probably was related to
changes in food availability. Preferences for chaparral and oak woodland in
October and November coincided with acorn production; oaks were an impor-
tant component of both communities, and acorns were a major food of feral pigs

144 CALIFORNIA FISH AND CAME

in California (Pine and Gerdes 1973, Barrett 1978). The shift in activity to
chaparral-grassland during February occurred shortly after the first major storm
of the season in early January brought new growth of grasses and forbs. Green
herbaceous vegetation dominated spring diets of pigs elsewhere in California
(Pine and Gerdes 1973, Barrett 1978). Absence of a concurrent shift by pigs into
grasslands, where herbaceous vegetation also was available, probably was inhib-
ited by lack of cover in open grasslands. Pig hunters were present on the island
throughout the study, and Pine and Gerdes (1973) reported that hunted pigs
sought brush for escape cover. Barrett (1978) observed that pigs in Tehama
County moved into more open habitats in winter and spring to graze.

Results of this study illustrate and emphasize the importance of weather and
food in the activity and distribution of feral pigs. Pigs are highly adaptable,
adjusting quickly to seasonal changes in temperature and the availability of two
of their most important foods, acorns and green herbaceous vegetation.

Rooting by wild pigs can cause severe damage to vegetation (Wood and
Barrett 1979). On Santa Cruz Island, concentration of pigs in canyon bottoms
during late fall and early winter indicates a high potential for damage to stream-
side vegetation and water sources. Chaparral and oak woodland communities
were highly preferred during fall. These communities are particularly rich in plant
species endemic to the Channel Islands (Philbrick and Haller 1977), and hence
the presence of concentrated pig activity constitutes a serious threat to vegeta-
tion of the island.

ACKNOWLEDGMENTS

I thank The Nature Conservancy for financial support, and Santa Cruz Island
Company for their cooperation during field work.

LITERATURE CITED

Barrett, R. H. 1978. The feral hog on the Dye Creek Ranch, California. Hilgardia, 46: 283-355.

1982. Habitat preferences of feral hogs, deer, and cattle on a Sierra Foothill range. J. Range Manage.,

35: 342-346.

, and D. S. Pine. 1981. History and status of wild pigs, Sus scrofa, in San Benito County, California. Calif.

Fish Came, 67: 105-117.

Brumbaugh, R, W. 1 980. Recent geomorphic and vegetal dynamics on Santa Cruz Island, California. Pages 1 39-1 58
in D. M. Power, ed. The California Islands: proceedings of a multidisciplinary symposium. Santa Barbara
Museum of Natural History, Santa Barbara, Calif. 787 pp.

Mansfield, T. M. 1978. Wild pig management on a California public hunting area. Pages 187-201 in D. L. Koch,
ed. Cal-Neva Wildlife Transactions 1978. 274 pp.

McKnight, T. 1964. Feral livestock in Anglo-America. Univ. Calif. Publ. Ceogr., 16: 1-78.

Minnich, R. A. 1980. Vegetation of Santa Cruz and Santa Catalina Islands. Pages 123-137 in D. M. Power, ed. The

California Islands: proceedings of a multidisciplinary symposium. Santa Barbara Museum of Natural History,

Santa Barbara, Calif. 787 pp.

Philbrick, R. N., and). R. Haller. 1977. The southern California islands. Pages 893-906 />7 M. G. Barbour and J. Major,

eds. Terrestrial vegetation of California. John Wiley and Sons, New York. 1002 pp.
Pine, D. S., and C. L. Gerdes. 1973. Wild pigs in Monterey County, California. Calif. Fish Game, 59: 126-137.

Rogers, D. B. 1929. Prehistoric man of the Santa Barbara coast. Santa Barbara Museum of Natural History, Santa
Barbara, Calif. 452 pp.

Van Dyne, G. M., and H. F. Heady. 1965. Botanical composition of sheep and cattle diets on a mature annual
range. Hilgardia, 36: 465^92.

Wood, G. W., and R. H. Barrett. 1979. Status of wild pigs in the United States. Wildl. Soc. Bull., 7: 237-246.

SAN JOAQUIN VALLEY FISH FAUNAS 145

Calif. Fish and Game 70 ( 3 ) : 145—157 1 984

ENVIRONMENTAL CONDITIONS AND FISH FAUNAS IN

LOW ELEVATION RIVERS ON THE IRRIGATED SAN

JOAQUIN VALLEY FLOOR, CALIFORNIA^

MICHAEL K. SAIKI

U.S. Fish and Wildlife Service

Columbia National Fisheries Research Laboratory

Field Research Station — Dixon

6924 Tremont Road

Dixon, CA 95620

Physical and chemical characteristics of the lower San Joaquin and Merced rivers
showed longitudinal (upstream to downstream) patterns related primarily to irriga-
tion activities. Turbidity, total alkalinity, conductivity, and concentrations of mac-
ronutrients (NH3-N, NOj + NOj-N, total N, ortho-PO^, and total P) increased
conspicuously at downstream sampling sites. The fish faunas also differed at up-
stream and downstream sites. Species captured primarily at upstream sites were
sculpins, Cottus spp.; green sunfish, Lepomis cyanellus; redear sunfish, L. microlo-
pA//5; Sacramento squawfish, Ptychochei/usgrandis; hardhead, Mylopharodon cono-
cephalus; and threespine stickleback, Gasterosteus aculeatus; species that seemed
to be restricted to downstream sites were inland silverside, Menidia bery//ina; white
crappie, Pomoxis annularis; threadfin shad, Dorosoma petenense; fathead minnow,
Pimepha/es prome/as; spWttaW, Pogonichthys macro/epidotus; Sacramento blackfish,
Orthodon microlepidotus; tule perch, Hysterocarpus traski; and striped bass. Mo-
rone saxatilis. Published information on the general habitat requirements of fishes
indicates that environmental changes are probably influencing the composition and
abundance of fish faunas. Further research is needed to isolate the direct and indirect
effects of environmental contaminants on fish, with particular emphasis on identify-
ing the toxic components of irrigation return flows.

INTRODUCTION

Dam construction, water withdrawals, contamination by municipal and agri-
cultural wastes, and other man-related activities have contributed toward
changes in the fish communities of the San Joaquin Valley (Moyle 1976, Moyle
and Nichols 1973, U.S. Bureau of Reclamation 1978), but the full extent of these
changes is unknown. Recent studies of fish communities in the Valley have
focused either on Sierra Nevada foothill streams and reservoirs above 90 m
elevation (e.g., Moyle and Nichols 1973, 1974) or on the Sacramento-San
Joaquin Delta (e.g.. Turner and Kelley 1966). The fishes of low-elevation rivers
on the San Joaquin Valley floor that connect these two major habitat types have
not been investigated.

Irrigated agriculture is the dominant land use on the San Joaquin Valley floor
(California Department of Water Resources 1960, 1969). During the irrigation
season (usually March to October), return flows from irrigated fields can com-
pose most of the discharge in low elevation rivers (California Department of
Water Resources 1960, U.S. Bureau of Reclamation 1978). Knowledge of
changes in the river environment associated with irrigated agriculture and at-
tendant changes in the fish faunas may contribute toward a better understanding
of potential ecological impacts from existing and future irrigation drainage
projects.

' Accepted for publication October 1983.
2—78391

146 CALIFORNIA FISH AND CAME

This paper reports the results of a survey of general environmental conditions
and fishes in low elevation rivers on the San Joaquin Valley floor. Specific
objectives of the survey were to: (i) compare the physicochemical environ-
ment of rivers from locations at the head of the Valley floor (i.e., above most
irrigated croplands) with those within the irrigated areas; (ii) determine the
occurrence and relative abundance of fishes at each of the locations; and (iii)
relate the measured environmental conditions to the broad habitat requirements
of fishes.

STUDY AREA

The study area included the San Joaquin River and two of its tributaries, the
Merced River and Salt Slough ( Figure 1 ) . Six sampling sites were established on
the San Joaquin River, two on the Merced River, and one on Salt Slough, at
elevations ranging from 3 to 79 m above mean sea level. Both the San Joaquin
and Merced rivers originate in the Sierra Nevada at elevations exceeding 3,500
m. Their initial water supplies are derived from snowmelt and rainfall; at lower
elevations, natural seepage and irrigation return flows make up progressively
greater proportions of their flows. Salt Slough originates on the Valley floor, and
derives most of its discharge from natural seepage and irrigation return flows.

MATERIALS AND METHODS

Field sampling extended from July 1 980 to November 1 981 . At each sampling
site, I monitored physicochemical variables likely to change from irrigation-
related activities (e.g., water diversions, return flows, etc.) and/or to affect fish
distribution. The following measurements were made monthly: water tempera-
ture, turbidity, discharge (data obtained from the U.S. Geological Survey, the
U.S. Bureau of Reclamation, and the California Department of Water Re-
sources), dissolved oxygen, pFH, total alkalinity, and conductivity. At 3-month
intervals, I also measured macronutrients and the particle size distribution of
bottom sediments.

Fish were collected monthly from July 1980 to October 1981 at two sites
(SJR-1 and SJR-4;see Figure 1 for explanation of abbreviations). At all other sites,
collections were made once every 3 months during the same period, except that
collections were terminated in October 1 980 at SJR-3 and started in January 1 981
at SJR-6. Samples were obtained with two seines, one 5.5 m long by 2.4 m deep
with 9.5-mm square mesh, and one 30.5 m long by 1.8 m deep with 12.7-mm
square mesh. Fish were also collected by backpack electrofishing at SJR-1, MR-1,
and MR-2; other sites were too turbid and salty for effective electrofishing.

Captured fish were identified and counted, and the fish were returned to the
water. Species similarity indices for pairs of sampling sites were calculated by
using Jaccard's formula, S| = Tii/(T, + T, - T;,), where T; and T, are the numbers
of species at sites i and j, respectively, and T,, is the number of species at both
sites (FHorwitz 1978). Fish counts for each gear type were also assigned relative
abundance ratings according to the following criteria: rare, 0.1 to 1.0% of the
mean catch at the sample site; common, 1.1 to 7.0%; and abundant, > 7.0%.
Failure to collect a species in an area did not necessarily mean it was absent but
strongly suggested that it was very rare. Certain species (e.g., lampreys, Lampe-
tra spp.; and sculpins) may have been inadequately sampled by the various
gears because these fishes can burrow into the substrate. Also, fast swimmers
(e.g., adult striped bass and chinook salmon, Oncorhynchus tshawytscha) and

SAN JOAQUIN VALLEY FISH FAUNAS

147

fish that inhabited deep, swift-flowing channels were probably underrepresented
in the catches.

CALIFORNIA

Enlarged Area

Sampling Sites

® SJR-

SJR-2
(3) SJR-3
@ SJR-4

SJR-5

SJR-6
® MR- I

MR-2
(D SS

I I Nonirrigated lands
[ I Irrigated lands
^1 Reservoirs

FIGURE 1. Locations of sampling sites and irrigated croplands in the study area. Names of sam-
pling sites are as follows: SJR-1, San Joaquin River near Fort Washington; SJR-2, San
Joaquin River at Firebaugh; SJR-3, San Joaquin River at Lander Avenue; SJR-4, San
Joaquin River at Fremont Ford State Recreational Area; SJR-5, San Joaquin River near
Crows Landing; SJR-6, San Joaquin River at South County Park; MR-1, Merced River
below FHighway 59; MR-2, Merced River at George J. Hatfield State Recreational Area;
and SS, Salt Slough below Lander Avenue.

RESULTS
Physicochemical Environment

With the exception of temperature, physicochemical characteristics showed
no consistent temporal patterns either within or between sampling sites during
this study. Water temperatures were lowest during December-January and high-
est during July-September. Nearly all variables showed longitudinal (i.e., up-
stream to downstream) patterns.

Water temperatures in the San Joaquin and Merced rivers were lowest at
upstream sites (Table 1 ). In the San Joaquin River, mean temperatures reached
a maximum at SJR-3, then steadily declined from SJR-4 to SJR-6. The declining
temperatures coincided longitudinally with inflows of lower temperature water
from Salt Slough and the Merced River.

Turbidities were lowest at upstream sites (Table 1), and increased rapidly
downstream (less than threefold in the Merced River at MR-2, but over seven-
fold in the San Joaquin River at SJR-4). The progressively higher turbidities

Turbidity

Discharge

(NTU'S)^

(m'/s)^

3.2 ± 2.1

4.5 ± 7.3

(0.6- 8.6)

(1.0-49.3)

18.6 ± 14.9

7.1 ± 5.1

(5.8 - 74.7)

(0.2 - 14.2)

21.3 ± 8.9

2.5 ± 2.3

(6.0 - 39.8)

(0.6- 8.9)

25.0 ± 10.1

8.3 ± 3.4

(9.4 - 60.3)

(3.5 - 16.7)

20.1 ± 9.4

26.0 ± 10.7

(5.2 - 39.0)

(11.0-83.0)

17.5 ± 8.9

57.4 ± 22,1

(6.9 - 39.5)

(31,4-92.8)

1.7 ± 0.6

6,5 ± 4,2

(0.5 - 3.2)

(1,4- 19.1)

5.7 ± 6.6

9,8 ± 4,1

(0.7 - 23.5)

(4,0-21.7)

27.0 ± 8.7

5.2 ± 2.5

(8.3 - 45.0)

(1.9- 10.1)

148 CALIFORNIA FISH AND GAME

measured between SJR-1 and SJR-4 were associated with turbid discharges from
the Delta-Mendota Canal (see Figure 1 for location) and several agricultural
drains, including Salt Slough. Turbidity in the San Joaquin River decreased below
SJR-4, coinciding with inflows of clearer waters from tributary streams such as
the Merced River.

TABLE 1. Water Temperature, Turbidity, and Discharge in the San Joaquin and Merced
Rivers, and Salt Slough (mean ± SD; ranges in parentheses).

Temperature
Site' CCf

S)R-1 15.6 ± 4.6

(6.8 - 24.4)
S)R-2 18,2 ± 5,4

(9,7 - 27.4)
SJR-3 19.0 ± 6.3

(7.5 - 30.4)
SIR-4 18.0 ± 6.4

(7.1 - 30.0)
S)R-5 17,6 ± 5,6

(7,9 - 27.0)
S)R-6 16,8 ± 5,1

(8.3 - 25.5)
MR-1 14.3 ± 3.5

(8.5 - 21.0)
MR-2 18.0 ± 5.8

(8.5 - 27.5)
SS 18.5 ± 5.7

(7.8 - 27.5)

* See Figure 1 for locations.

"^ Based on unweighted monthly data from July 1980 to November 1981 adjusted for a 1-year interval,

* Based on unweighted monthly data from July 1980 to September 1981 adjusted for a 1-year interval.

River discharge was relatively low at upstream sites (Table 1), with higher
flows usually recorded downstream. In the San Joaquin River, discharge in-
creased at SJR-2 mainly due to inflow from the Delta-Mendota Canal. However,
discharge was reduced at SJR-3 because of diversions for irrigation. Between
SJR-3 and SJR-6, mean discharge increased with inputs from Salt Slough, the
Merced River, and other tributaries.

Mean dissolved oxygen concentrations in the San Joaquin River (Table 2)
reached a maximum at SJR-3 (11.9 mg/l, 132% saturation), then gradually
declined downstream to SJR-6 (8.0 mg/l, 85% saturation). Mean dissolved
oxygen levels in the Merced River were relatively high at both MR-1 and MR-2
(9.8 mg/l and 97% saturation, and 9.3 mg/l and 100% saturation, respectively).
Oxygen concentrations averaged 8.1 mg/l (92% saturation) in Salt Slough.

Mean hydrogen-ion concentrations (pH's) at all sampling sites ranged from
7.8 to 8.3, with no apparent longitudinal pattern (Table 2). Highest pH was
measured at SJR-3, and coincided with supersaturated oxygen conditions.

Total alkalinity concentrations increased considerably from upstream to
downstream sites in both the San Joaquin and Merced rivers (Table 2). In the
San Joaquin River, alkalinity reached a maximum at SJR-3. The alkalinity at Salt
Slough was intermediate between concentrations measured at SJR-3 and SJR-4.

Conductivities were lowest at upstream sites, and increased downstream
(Table 2). In the San Joaquin River, mean conductivity increased more than
30-fold between SJR-1 and SJR-4, then decreased by nearly one-half at SJR-6.
Conductivities in the Merced River increased only fourfold from MR-1 to MR-2.
Conductivity at Salt Slough was usually higher than elsewhere.

SAN JOAQUIN VALLEY FISH FAUNAS

149

1/1

00

c
19

Q
I/)

+1

c

Of

E

y

^°^
s "^^ ^

S

s

<->

§

CO

'a-

O

o

o

o

0^

CO

O

-^

O

r^

8

OO

^—

-—

0^

LTi

rsj

(-^

rsi

CNI

•—

0^

vX)

LT,

rs

ir\

(-N^

+1

1

+1

1

+ 1

1

+ 1

1

+ 1

1

+ 1

1

+ 1

1

+ 1

1

+ 1

1

r^

T—

r^

s

r\

o

,

O

(T^

in

•~D

O

0^

LTl

0^

LO

t^

CO

in

m

rO

o

0^

rsi

■-D

O

vX>

rsi

■q-

r-i

(T>

1-^

■^

LO

"— '

•^

<-o

r^

r^

t--^

r^

l-sl

CO

m

■ — '

' —

~ '

a^

^

3
-a

V

u
o

I/)

c

>

T5 >0

^

I +1

I +1

. . o o
T— OO vD r^

1 — LO r^ o>

OO

rsi 1 — Ln

rsi rsi (^J

+ 1 I +1

O — ^ — OO

■^ m rn u-i OO LO r^i
1 — rsi T— r\ rsj rsi

+ 1

+ 1

I +1 I

r^i^^rsiLris^rs^O^r^^Or^rsi
^DOrr,-^' — h^r-ji — Lnrsi|s^r-j

J2

3

ro
U

HI
J3

O

c

■o

0)

u

w
0)

■o
c

3:

^ .— o c-s- t-^ o

CO CO CO o6 CO od

CO ' 1-^ ' t< ' r< ' i<

LO

•^

o^

OO

CO

CO

CO

1

'=? 1

l-r,

r\

1

CO 1

CO '

o

r^

o

r\

t~^

CO

CO

I

o
l<

I

CO

1^

I

LTl
l<

I

t\

3

o

X

O

3

o-

c

W1

'>

u

3
"D
C
O

u

■o
c

<

COrsjrorOi-nLn'^ LOO^Ors|^x>CT''f1"

I

^ ^ +1

> -5 ^

§ o _ _ _ _

f^ rs) vX> r^ CO ' —

I +1

rsi . —

^- n ^

I

.— p

CO uS

+ 1 I +1 I +1 I +1 I +1 I
0^a^m■^CT^0^O^-DC0Ln

^or^' — o-io^^ocosOco^oo^r\

0^
O

+ 1 I +1
ro 0^ <—

o-I ^

'9

^

<

^^

»«

o

1-

^

I

*

Q.

4

C

'^

u

ec

>-

X ^-v

ved
thes

0 F

ir> i>

« ^

•- iv

O o.

—

rsi

ro

■^

LO

^

LO

LO

LO

LO

' — rsi

>

HI
■D

sS: i\
.5. a;

■- ^

<^ >-

»— -D

•- "D

01 m

l!

o "
" m

§1

CT^ ^

'~ T3
>- 0^

3 E

1^
o c

^O

>■ E
■£ 2

o ^ a

3 a; =5
o| "

-<- 5

^ C
3

2^ CI

Li. T3 c
(Dm

CO

O)

150 CALIFORNIA FISH AND CAME

Mean macronutrient concentrations (NH3 — N, NO34-NO2 — N, total N, or-
tho-P04, total P) typically increased, upstream to downstream (Table 3). In the
San Joaquin River, macronutrient concentrations increased rapidly from SJR-1 to
SJR-4, decreased at SJR-5, then increased slightly at SJR-6. (Macronutrient data
for SJR-3 were omitted from this analysis because collections were made only
in July and October 1980.) In the Merced River, concentrations were highest
downstream at M R-2. Concentrations of NO3 + NO2 — N and total N were high-
er at Salt Slough than at other sampling sites.

Sediments of upstream sites contained a higher percentage of large diameter
particles (i.e., pebbles and rocks) than downstream sites (Table 4). Sand, gravel,
and pebbles made up most of the sediments at SJR-1 and MR-1; silt and sand
were dominant further downstream.

Distribution and Abundance of Fishes

A total of 43,605 fish representing 33 species (excluding hybrid sunfish, Lepo-
mis) and 14 families was collected by combined gear types during this study
(Table 5) . In the San Joaquin River, the number of species captured at each site
ranged from 17 at SJR-3 to 26 at SJR-5 (Table 5). Native species ranged from
two at SJR-3 to six at both SJR-1 and SJR-5. In the Merced River, 20 species were
caught at MR-1 and 22 species at MR-2. Native fishes totalled seven at MR-1 and
six at MR-2. Salt Slough contained 18 species, only one of which was native.

The total numbers of species captured at each of the nine sites were not
significantly different (X^ = 3.55, d.f. = 8, P > 0.05). Furthermore, the propor-
tions of native to introduced species were not significantly different when com-
pared among the various sites (X^ = 8.16, d.f. = 8, P > 0.05); however, native
species generally comprised a larger percentage of the total taxa at the two
upstream sites (30% at SJR-1, 35% at MR-1 ) than at the seven downstream sites
(6% at Salt Slough, 12% at SJR-3, 16% at SJR-2, 17% at SJR-4, 19% at SJR-6,
23% at SJR-5, 27% at MR-2).

The fish communities at the various sites were compared by calculating Jac-
card's species similarity index. If indices > 0.65 are arbitrarily assumed to be
significant, then 15 pairs of similar sites were observed (Table 6). The data
showed that the two upstream sites (SJR-1 and MR-1 ) contain a higher propor-
tion of the same species than when each is compared with downstream sites.
Furthermore, except for MR-2, the downstream sites shared more species
among themselves than with upstream sites. The fish community at MR-2,
however, was not significantly similar to communities from either the upstream
or other downstream sites.

Several species were captured primarily or exclusively at upstream sites (e.g.,
sculpins, green sunfish, redear sunfish, Sacramento squawfish, hardhead, and
threespine stickleback); other species were captured only at downstream sites
(e.g., inland silverside, white crappie, threadfin shad, fathead minnow, splittail,
Sacramento blackfish, tule perch, and striped bass) (Table 5). Anadromous
Chinook salmon were observed from October through April only in the Merced
River.

Although catch statistics were influenced by gear selectivity, general trends in
the data (Table 5) indicated that the most abundant species at SJR-1 and MR-1
were Sacramento sucker, Catostomus occidentalis; green sunfish; bluegill, Lepo-
mis macrochirus; redear sunfish; largemouth bass, Micropterus salmoides; scul-
pins; and Sacramento squawfish. Hardhead was abundant at MR-1 but rare at

SAN JOAQUIN VALLEY FISH FAUNAS

151

+1

U-)

u-l

0^

^

r^

l-O

(T^

is^

_

O

rsi

rsi

in

f\

, —

rsi

,

r^

rs|

,

CM

,

s

•X>

ul

rn

O

r—

, —

1 —

sr>

s

u

— ^

^_

■^

in

, —

rn

(T.

t~~.

r^

rsi

m

•^

CO

1 —

-^

m

rsi

rs|

rn

< —

1 —

.'a

+1

1

+ 1

1

+1

1

+1

1

+ 1

1

+ 1

1

+ 1

1

+1

1

^5

<

■^

Ln

■^

a^

a^

0^

rsi

rsi

1 —

rn

sD

■^

sD

CO

CO

^J

^o

,

in

,_

in

m

rn

s

■^

sD

t-)

CO

o>

r^

o

a^

ro

(~s|

CO

rsi

m

■q-

Ov

m

T

en

CO

rsi

vD

■ — -

1 —

rsi

1 —

1 —

^—

rsi

1 —

■ — '

' —

■ — ■

rs|

■ — ■

C

V

k

m

o.

_

_

_

c

cT

sO

u-i

m

CO

■^

rsi

rsi

■q-

m

rsi

O

in

rsi

r\

CO

rn

Tf

r-l

^

CO

sO

<T>

TT

t-1

r^

o

sr>

m

t~^

u-i

1 —

in

<i.

-^

rsi

•rr

r-j

CTN

I\

in

TT

m

■^

rn

CO

in

BO

c

S

1 —

rn

rsi

rsi

r—

rsi

-^

+1

sD

1

+ 1

,

+ 1
0^

,

+ 1
in

1
O

+ 1
in

,

+ 1

1
o

+ 1
<-n

1

+1

1

rs

T-

m

,

no

r^

rn

o

sO

sD

rsi

rn

in

o

O

rsi

LO

a^

im

o

- — '

m

1 —

sD

sD

|s^

CO

rj

1-^

rsi

in

sD

1 —

rsi

■ ■

1 —

1 —

■ — -

I —

■ — ■

Q

~

(/>

s

E

m

in

_

r\

m

CO o

in

CO

0^

rsi

r^

_

0^

O

in

rr~

rr>

m

m

rsi

o CO

, —

f->

1 —

?

r—

s

, —

rsi

r-i

sO

ro

o

OO sD

rsi

CO

o

in

rs|

CO

>

^

Ln

rN

<-o

1-^

o o

(T.

»X>

■^

<-N|

rs|

(T-

in

sD

BC

3
O

1

s

+ 1

1
'J-

+ 1

T

+ 1
rsi

rsi

1 +1

in in

rsi

0^

+ 1
in

rsi

1
in

+ 1

s

+ 1
OO

1
o

+ 1

CO

rsi

1

m

M-

Ln

?

a^

O sO

O

0^

rsi

?

sO

in

rsi

m

1 —

/>

o

^^

<^

CT^

rn r^

r->

in

sD

CO

rsi

O

"S-

r-i

m

0^

I\

CO

sD in

Ln

r-s

o_

■"

■ — '

in

^

0^

rsi^

c

>

u

•a
c

rs

'5
a-

_o

c

c
o

c

V

u

c
o

u

c

3

c
o

k

u

n

m

eo

<

I

in o ^ ^ ■—

rn ^ rsi cjs •— *

•— o -^ M- c:>

r^ r-) 1 — O 1"^

+ 1

I +1 I +1
CT^ CT^ rn sD
t' CO

— o

CO

00

^ °2
rq"

I +1

rs| Ln

0-> rsi

CO .—

rs| ir^

1^. rsi

sO O
O^ m

r-l

I

0^

IT- -^ —

rs! CO r-^ O

1 — 1 — Lo r^

rsT

+ 1
rsi -^

rn ■^'
i\ rsi

rn sD

I +!

|s^ r\

sd ^'

1 — m
O

1 +1

p o

^ OO

Ln rsi

O-I CO

CO •—

LO 0^

1 +1

Ln r\

^ CO

s£>
rsT

I

■^ a^ •—

•3- ^ CTs

r\r\-«^Orsirs|s^Lnrs|-^< —

sD <r^ r-t

m c^ ' —

cy^ T^ a^ r^

r\ rn r^ t\ r^ -^

+ 1

+ 1

I +1
hs. r-^ oq

^ sD sD LO

+ 1
l~^

I +1

Ln CO

I +!

sD v^
rsi ■^

rn

rsi

I

I— T— uni^'— M-^ — ■ •— — ■—

I +1

o^ sO oq

i< rsi Tf

f^i Ln CO

■Si

,—

rsi

^

Ln

sD

^7

rsi

c^

Qi

q:

C^

c*:

2

-2

LO

LO

LO

ly-i

U~l

c
g

U
O

3

BO

3

_o

f8

C

01

>

■o

it

■o
c

3

n

c

01

c

3

V

.S
(^

E
(/>

V
BO
rs

>
<

^

CO •>»■
CO t<

CO •— p

r-i O O

rsj sD rs< p p
rsj hs.' o O O

■-^ 0^ p
Tf' rn •<?

r^ <—

in ^r>

^

<r>

0^

rsi

o

o

^

w

in

r-t
0^

O

o

o

^

_

-^

_

"S-

o

rsi

CO

CO

o

o

>!■

CO

r^

Ln

_

o

^
^

m

in
0^

O

o
V

o

•^ .— rn ^o o o

'A'

Lo rn rsi

— o
o o

V

t^ CO .—

o ■— ^

o o
o

.N

c5

o

V

Ln
o

o^ rn c:

J I ->

<i Ln ^

Ln
rsj (T^
O ^-

- c ™ -Q
;y^ u^ IsJ Q.

r-v
A

00 .

I ^
S O

§>!

OJ —

^ .9P

152

CALIFORNIA FISH AND GAME

V

19

it

"5.
E

lU

Q

-a
c

>

u

■o
c

3
o"

at u
-c uj

iZ g^

5 ^

c ea

3 ^
^ 01
< C

> (/>
Q£ ..

01 O
(J u.

3"°

0 3

CO

<

^

^

w

■^

^
w

O o °^_0 =^.00000
< o 0.: c±f ^ Q Q- of ac of

O O O O U O O U' o <
< o U Qtf < < o < n ^

O < <u<uuuoO
o o o o < < o ^; o '-^"'
o ^~ do < "^ "^^ o o

<

<

u
<

<

o
o
o

o

O

d

o
d

d

-o

n, > "=

<

<

o
d
d

o
d

d
d

<

d

'-{<0000^00<

^ '^ d d d d < d d ^

"-{ *-i 1^. o "^^ o o o o o
d d ^. d o d d d d' <->"■
d <=^''^ d°^ d d d d d

O 0=:^ Di U O O O O < O

o Q o d d d d d d d
o o t^ < o d d d < d

Qi Q < C^ Q Qi Di oi"

u Qi u d d '-^"^ '^'' d'

<

o

d

u u Qi^ O t^. O °1 O '^^ <
ac cx^ c^ C) ^ n *^~ Qi ^~ <

Qi otf < Qi Q u i^~ < U

o

^^ M. °=l O °1 O =>1 O O <

Cji c^

Od

■ U

O O

O' ^ d d' ocT o" u oi' d

d ^~d< '^^ ""' '^ "~

" O d' C^" Di

< o QioiQioOOOOO"-'

< o o ^ °^ o d d o d d ^

Q c£ < < o" o d u

(J- Qi < u o" < ^ ^

.-Qi<uo<0=^
;- Qi <- < Q- <- Q u

r O < < r> < ^- <-*"

u

<

O

d
d

o

d

Qi Qi- Qi Q U cti" < o" d °^

6 d d o d" d d d d d
d d o o < d d d "^ d

=1

- ^

.5
I

I

<T3 ID

■ •=i -o

O OJ ;^

fi ITS c

c (J (—

E
o

tn O i-T
O 03

ID

u

I' _

Ol TO

o = ._

Ol

13

o

II

'J ti a;

- — ■ m

a 3

c O 5 cc ^

u

P i; ■^ "D X

U

c n3 i^ ^.
o; ■?: T3 ^ E

c
o

■I
I

c

.1^

5

c

5^

(-0

-3 -§- c 2 E -

-^ a;

"2 3

ro O

X u

S3 E

o
d

o
d
d

o
d
d

<

O

uuQiO^OOOO<

Qi

u

O

O Di Qi Q Qi Q- (J Q- Q U

o

of

o

O

d

d

o
d
d

O- 03
U^ U^ U^

JO

O (D
^ -S
E '"^

2 2

'-> .^

cl

U

kj

SAN JOAQUIN VALLEY FISH FAUNAS

153

o

o

u u o °^.

Cd LJ

o

o

o

<

o
d

o
d

d

u o O ^,

d d d Q
^ d d o

o d o o o
o =^~ d °=^ d

o
d
d

o
d
d

O O '^'^ o °^, °=^-
o d d d o ^

d

d
d

d

d
of

u

d
d

o
d

<

d
<

d

O
<

on

d

d

u

d

o
d

-a
c
m

1

CO
o

o
d

cn<QC^^
Di Qi oi" d

o

o
d

<

o
d

Q.

5

O

d

Qi < Qi Ci:^

C^" Ctf c^" Q

O

d

<

o
d

Q.

o
d

d d d °<

d d d o

d

O

d

<

o
d

■a
u

O

o
d

o
u

an
on

en

en

o
d

<

o
d

m
O

c

IT.
CO

d
d

0/

l_) !=«: C^ O

d °< d o
d^=^~d

■^?; ty^ n3

2

:3

i5 ?

o
d
d

d

J-

I

u

d
d

_™ i^ _n

(^ fTj LA

C .C i nj u
™ o .B-T3

U
-= c

<

d

o
d
d

O

E

u.^St|l|||

OJ a> OJ O ro

D_ Q_ D_ Q- CA)

O
O

U

-a ^

Q. O
u

>= £
II ^

>

E "

II 5'

c
0.9

oj u
-O o

E-

5 o

■?^

'^ c C-

II ^^

< " H

:/ > ij

O) w _

O Oj o

U cj: h-

154 CALIFORNIA FISH AND GAME

TABLE 6. Jaccard's Species Similarity Indices for Fish from Selected Sites in the San Joaquin
and Merced Rivers, and Salt Slough.* '

SjR- SIR- SIR- SIR- SIR- SIR- MR- MR- SS
12 3 4 5 6 12

SJR-1 0.50 0.32 0.52 0.53 0.41 0.74 0.56 0.41

SJR-2 0.68 0.96 0.82 0.77 0.45 0.62 0.72

SJR-3 0.71 0.65 0.81 0.32 0.56 0.75

SJR-4 0.85 0.80 0.47 0.64 0.75

S)R-5 0.81 0.48 0.60 0.69

SJR-6 0.37 0.59 0.70

MR-1 0.62 0.36

MR-2 0.54

SS

' See Figure 1 for locations.

^ Calculations exclude hybrid sunfish.

5JR-1. At MR-2, inland silverside, bluegill, redear sunfish, largemouth bass, black
crappie, Pomoxis nigromaculatus, threadfin shad, and mosquitofish, Cambusia
affinis, were abundant (Table 5). The rennaining locations (i.e., SJR-2, SJR-3,
SJR-4, SJR-5, SJR-6, and Salt Slough) showed some variability in the relative
abundance of fishes. In general, the most abundant species at downstream sites
were inland silverside, bluegill, largemouth bass, black crappie, threadfin shad,
fathead minnow, and mosquitofish (Table 5). Sacramento blackfish was com-
mon or abundant at all downstream sites except SJR-2, where it was rarely
captured. All Centrarchidae were rare or never captured at Salt Slough even
though several species (e.g., bluegill, largemouth bass, black crappie, and redear
sunfish) were common or abundant at one or more of the other downstream
sites.

DISCUSSION

Environmental modifications resulting from drainage of agricultural lands may
influence aquatic biota (Campbell and Whitley 1970, Dance and Hynes 1980,
Hill 1976, Marsh and Waters 1980, Mitchell 1975), including fish (Luey and
Adelman 1980). The present study documented several physical and chemical
changes at sampling sites on the irrigated San Joaquin Valley floor that generally
coincided with changes in fish species composition and relative abundance of
individual species.

Except for reduced discharge resulting from storage and diversion of water at
reservoirs in the Sierra Nevada foothills (e.g., Millerton Lake on the San Joaquin
River; Lake McClure on the Merced River), environmental conditions at the
farthest upstream sampling sites (SJR-1 and MR-1 ) were relatively unaffected by
irrigation activities. Turbidity, total alkalinity, conductivity, and macronutrient
concentrations were usually low, and dissolved oxygen concentrations were
always near saturation.

According to the Jaccard index, fish communities at the two upstream sites
were more similar to each other than to communities at downstream sites.
Native fishes typical of upstream sites (e.g., Sacramento sucker, Sacramento
squawfish, hardhead, threespine stickleback) are usually most abundant in clear,
relatively undisturbed streams of the Sierra Nevada foothills, where introduced
fishes are rare or absent ( Moyle 1 976, Moyle and Nichols 1 973 ) . Hardhead and
threespine stickleback were captured only from upstream sites during this study,
even though Moyle (1973) indicated that their original ranges included the

SAN JOAQUIN VALLEY FISH FAUNAS 155

Sierra foothills, Valley floor, and Sacramento-San Joaquin Delta. Moyle et al.
(1982) suggested that high turbidity may restrict both squawfish and threespine
stickleback from low-elevation sites in the Sacramento-San Joaquin drainage
system. The most abundant introduced species at upstream sites (i.e., redear
sunfish, green sunfish, bluegill, largemouth bass) prefer clear water with aquatic
vegetation, and with bottoms of silt, sand, or gravel (Carlander 1977, Moyle
1976, Trautman 1957), although green sunfish are also abundant in turbid, mud-
dy-bottomed environments that have been extensively modified by human dis-
turbances (Moyle 1976, Moyle and Nichols 1973).

In the San Joaquin River, most of the water stored at Millerton Lake is trans-
ported southward to Kern County by the Friant-Kern Canal, with a small amount
moved north to Madera County by the Madera Canal. Water users downriver
in Fresno, Merced, and Stanislaus counties now receive a replacement supply
of poorer quality water (i.e., higher in turbidity and conductivity) from the
Delta-Mendota Canal (which transports water from the Sacramento-San Joa-
quin Delta near Tracy, San Joaquin County) at Mendota Pool (California Depart-
ment of Water Resources 1969), located about 15 km above SJR-2. Water in the
Delta-Mendota Canal is initially derived mainly from the Sacramento River
basin, but receives irrigation return flows and storm runoff at numerous points
along its 190-km length (California Department of Water Resources 1960). Even
with this supplemental water, withdrawals during the irrigation season still occa-
sionally eliminate surface flows in the portions of the lower San Joaquin River
between Mendota Pool and the confluence with the Merced River (Hazel et al.
1976).

Irrigation return flows compensate for some of the river water withdrawn for
irrigation below Mendota Pool. However, return flows usually have physical and
chemical characteristics much different from the supply water (Hotes and Pear-
son 1977; Law and Skogerboe 1972; Tanji etal. ]977a, 1977/? ). The turbid, saline,
and nutrient-rich water environment that characterizes Salt Slough results from
its function as a drain for irrigation return flows. In downstream areas of the San
Joaquin River (e.g., SJR-3 and 4) where return flows from Salt Slough and other
point and nonpoint sources constitute the greatest proportion of total river flow,
physical and chemical conditions deviate the most from initial conditions re-
corded upstream at SJR-1.

The influence of irrigation activities (including return flows) is much less
pronounced in the Merced River than in the San Joaquin River. In a previous
investigation by the U.S. Geological Survey, Sorenson and Hoffman (1981)
concluded that even though some chemical changes occur in the lower reaches
from return flows, water in the Merced River is generally of high quality. Inflow
of high quality water from the Merced River into the San Joaquin River is
reflected in slightly improved conditions (i.e., lower temperature, pH, conduc-
tivity, turbidity, and total alkalinity and lower macronutrient concentrations) at
SJR-5 when compared to SJR-4.

The modified environmental conditions documented at downstream sites are
paralleled by considerable changes in the fish faunas. With the exception of
MR-2, the Jaccard index showed that downstream sites shared more species
among themselves than with upstream sites. Native fishes captured only at
downstream sites (i.e., hitch, Lavinia eA'///ca6/o'«3; Sacramento blackfish; splittail;
tule perch) are most abundant in the Sacramento-San Joaquin Delta and tributar-
ies immediately upstream from the Delta (Moyle 1976); however, these popula-

1 56 CALIFORNIA FISH AND GAME

tions may be slowly declining because they are now absent from localities where
they were formerly present (Moyle 1973, 1976). According to Moyle (1976),
the herbivorous filter-feeding habits of Sacramento blackfish, coupled with their
ability to survive in warm, turbid habitats, probably has allowed them to main-
tain relatively large populations despite changes in their environment. Hitch and
splittail also seem to be persisting under the pressures of environmental changes
and introductions of exotic predators and competitors. Tule perch, though, are
disappearing from streams with reduced flows, increased turbidity, heavy pollu-
tion, or reduced cover (Moyle 1976).

Introduced species (i.e., inland silverside, bluegill, largemouth bass, black
crappie, threadfin shad, fathead minnow, mosquitofish) were especially abun-
dant at downstream sites. Although these fishes can tolerate turbid, brackish,
nutrient-rich environments, several species (i.e., bluegill, largemouth bass, black
crappie) prefer clear, fresh water with abundant beds of submerged aquatic
vegetation (Carlander 1977, Moyle 1976, Trautman 1957). Other species (e.g.,
fathead minnow, mosquitofish) are able to produce large populations over a
broad range of environmental conditions as long as predatory fishes and com-
petitors are rare or absent (Moyle 1976, Trautman 1957). Fathead minnow
seemed to be especially abundant in Salt Slough where piscivorous largemouth
bass and black crappie were rarely caught.

The downstream environment at MR-2 was degraded somewhat from condi-
tions at MR-1, but not to the extent exhibited by downstream sites on the San
Joaquin River and Salt Slough. The site at MR-2 also contained a fish fauna
composed of species characteristic of both upstream and downstream sites. The
most abundant species (i.e., bluegill and mosquitofish) were also generally
abundant at other locations. Sacramento sucker, redear sunfish, and green sun-
fish (typical of upstream sites), and inland silverside, threadfin shad, and Sacra-
mento blackfish (typical of downstream sites) were all captured at MR-2. These
data suggest that MR-2 represents a transitional habitat containing environmental
and fish faunal elements of both upstream and downstream sites.

This study shows that irrigation activities have modified the physical and
chemical environment of low elevation rivers on the San Joaquin Valley floor,
and that the modifications have probably influenced the composition and abun-
dance of fish faunas. However, there is still a great need for further investigations
that identify the toxic components of irrigation return flows and measure their
effects on the survival, growth, reproduction, and behavior of fishes. Inasmuch
as food availability, predation, competition, and other biological factors can
influence fishes, these interactions must also be studied. Such research would
help to establish more clearly the magnitude of impacts of return flows on fishes
and enable more objective decisions to be made on the merits of future irrigation
drainage projects.

ACKNOWLEDGMENTS
I thank C. and J. Finch, D. Oliver, F. Freitas, the Flintkote Company, and park
personnel at George J. Hatfield State Recreational Area for allowing me access
to sampling sites; the U.S. Geological Survey, the U.S. Bureau of Reclamation,
and the California Department of Water Resources for furnishing unpublished
hydrological data; D. Cacela, T. Takagi, R. Nakamoto, D. Hartmann, and E.
McClary for assistance in the field; C. Schmitt for advice and assistance in

SAN JOAQUIN VALLEY FISH FAUNAS 157

analyzing the data; and S. Finger, R. Hothem, J. Hunn, T. P. Lowe, P. Moyle, and
C. Schmitt for critically reviewing early drafts of the manuscript and providing
many useful suggestions.

LITERATURE CITED

California Department of Water Resources. 1960. Lower San Joaquin Valley water quality investigation. Calif. Dep.

Water Resources, Bull., 89; 1-189.
1969. Lower San Joaquin River water quality investigation. Calif. Dep. Water Resources, Bull., 143-5:

1-207.

Campbell, R. S., and J. R. Whitley. 1970. Effects of agricultural pollutants on recreational uses of surface waters.

Pages 331-343 in T. L. Willrich and G. E. Smith, eds. Agricultural practices and water quality. Iowa State Univ.

Press, Ames.
Carlander, K. D, 1977. Handbook of freshwater fishery biology, volume two. Iowa State Univ., Ames, Iowa. 431

P-
Dance, K. W., and H. B. N. Hynes. 1980. Some effects of agricultural land use on stream insect communities.

Environ. Pollut., Ser. A, 22: 19-28.

Hazel, C, S. Herrera, H. Rectenwald, and |. Ives. 1976. Assessment of effects of altered stream flow characteristics
on fish and wildlife. Part B: California case studies. U.S. Fish Wildl. Serv., FWS/OBS— 76/34: 1-606.

Hill, A. R. 1976. The environmental impacts of agricultural land drainage. J. Environ. Manage., 4: 251-274.

Horwitz, R. J. 1978. Temporal variability patterns and the distributional patterns of stream fishes. Ecol. Monogr.,

48: 307-321.
Hotes, F. L., and E. A. Pearson. 1977. Effects of irrigation on water quality. Pages 127-158 in E. B. Worthington,

ed. Arid land irrigation in developing countries: Environmental problems and effects. Pergamon Press, New

York.

Kinney, E. C. 1973. Average or mean pH. Prog. Fish-Cult., 35(2): 93.

Law, J. P., Jr., and C. V. Skogerboe. 1972. Potential for controlling quality of irrigation return flows. J. Environ. Qua!.,

1: 140-146.
Luey, J. E., and I. R. Adelman. 1980. Downstream natural areas as refuges for fish in drainage-development

watersheds. Am. Fish. Soc, Trans., 109: 332-335.

Marsh, P. C, and T. F. Waters. 1980. Effects of agricultural drainage development on benthic invertebrates m
undisturbed downstream reaches. Am. Fish. Soc, Trans., 109: 213-223.

Mitchell, S. F. 1975. Some effects of agricultural development and fluctuations in water level on the phytoplankton
productivity and zooplankton of a New Zealand reservoir. Freshw. Biol., 5: 547-562.

Moyle, P. B. 1973. Recent changes in the fish fauna of the San Joaquin River system. Cal-Neva Wildl. Trans., 1973:

60-63.
1976. Inland fishes of California. Univ. of California Press, Berkeley. 405 p.

Moyle, P. B., and R. D. Nichols. 1973. Ecology of some native and introduced fishes on the Sierra Nevada foothills
in Central California. Copeia, 1973(3): 478-490.

1974. Decline of the native fish fauna of the Sierra Nevada foothills. Central California. Am. Midi. Nat.,

9: 72-83.

Moyle, P. B., J. J. Smith, R. A. Daniels, T. L. Taylor, D. G. Price, and D. M. Baltz. 1982. Distribution and ecology
of stream fishes of the Sacramento-San Joaquin drainage system, California. Univ. Calif. Publ. Zool. 1 15: 1-256.

Sorenson, S. K., and R. j. Hoffman. 1981. Water-quality assessment of the Merced River, California, in the 1977
water year. U.S. Geol. Surv., Water Resources Invest., 80-75: 1-31.

Tan|i, K. K., J. W. Biggar, R. J. Miller, W. O. Pruitt, and G. L. Horner. 1977<3. Evaluation of surface irrigation return
flows in the Central Valley of California. Pages 167-173 in\. P. Law, jr., and G. V. Skogerboe, eds. Proceedings
of national conference: Irrigation return flow quality management. Fort Collins, Colorado.

Tanji, K. K., M. M. Iqbal, A. F. Quek, R. M. Van De Pol, L. P. Wagenet, R. Fujii, R. |. SchagI, and D. A. Prewitt.
19776, Surface irrigation return flows vary. Calif. Agric, 31: 30-31.

Trautman, M, B. 1957. The fishes of Ohio. Ohio State Univ. Press. 683 p.

Turner, J. L., and D. W. Kelley. 1966. Ecological studies of the Sacramento-San Joaquin Delta. Part II: Fishes of the
Delta. Calif. Dep. Fish Game, Fish Bull., 136: 1-168.

U.S. Bureau of Reclamation. 1978. Fish and wildlife: Problems, opportunities, and solution: Total water manage-
ment study for the Central Valley basin, California. U.S. Bur. Reclamation, Mid-Pacific Region, Working
Document No. 12. Sacramento, California. 59 p.

158 CALIFORNIA FISH AND CAME

Calif. Fish and Came 70 ( 3 ) ; 1 58—1 62 1 984

RELOCATION OF ORIGINAL TERRITORIES BY DISPLACED

BLACK-AND-YELLOW ROCKFISH, SEBASTES

CHRYSOMELAS, FROM CARMEL BAY, CALIFORNIA '

LEON E. HALLACHER^

Department of Zoology

University of California

Berkeley, California 94720

The black-and-yellow rockfish, Sebastes chrysomelas, is a perennially territorial
species that normally exhibits limited movement away from its territory. A total of
thirty individuals were transported away from their territories during four transloca-
tion experiments. A high percentage (66%-71%) of fish returned to the site of
capture within two weeks, from distances up to 50 meters. Fish moved 1.5 kilometers
from their territories did not return, even after 4 months. Sensory parameters used
by this species for territory relocation are unknown. Homing behavior in a normally
sedentary species seems paradoxical, and this species may be more mobile than
previously thought.

INTRODUCTION

Reports of manipulations used to induce homing behavior and studies of
natural homing behavior among salmonid fishes are too numerous to list.
However, the homing phenomenon has not been so widely studied among other
groups of fishes. Williams (1957) discusses possible homing behavior by fishes
inhabiting the tidal zones of California. Green ( 1 971 ) reports that a high percent-
age of the tidepool sculpin, Oligocottus maculosus, return to specific tidepools
with the incoming tide after leaving with the outgoing tide. Clarke (1970) in his
work on the garibaldi, Hysypops rubicunda, observed that adult males returned
to their territories after moving off them to forage.

Homing by fishes of the genus Sebastes has been examined by a few workers.
Miller and Geibel (1973) reported seasonal movements by the blue rockfish, 5.
mystlnus, and suggested that local populations may return to specific kelp for-
ests. Carlson and Haight (1972) demonstrated that the yellowtail rockfish,
5. flavidus, can return to its home site over distances of 22.5 km even after 3
months in captivity.

Both 5. myst/nus and 5. flavidus are nonterritorial, aggregating species. Would
a territorial species of rockfish, normally exhibiting limited movement away from
its territory, home if displaced from that territory? This study was designed to
determine if the black-and-yellow rockfish, 5. chrysomelas, a territorial species,
could locate its home site if artificially displaced, and to ascertain if individual
fish would reoccupy the specific territory from which they were removed.

MATERIALS AND METHODS
Translocation manipulations were carried out between July 1975 and Decem-
ber 1976 in Carmel Bay, California. Fish were captured underwater by a diver
dangling a baited hook in front of them. They were tagged with plastic floy tags
and the holes from which fish were removed were marked with numbered lead
markers. Fish were measured, placed in a canvas bag, and moved to the release

' Accepted for publication )uly 1983.

^ Present address; Biology Discipline, Natural Sciences Division, University of Hawaii, Hilo, Hawaii 96720

BLACK-AND-YELLOW ROCKFISH RELOCATION 159

site. The fabric weave of the canvas bag used for transport of fish was tight,
preventing the fish from seeing out during the translocation process, and proba-
bly inhibiting water currents through the bag. The entire translocation process
was done using Scuba without bringing fish to the surface, except on one
long-distance translocation where fish were brought to the surface and trans-
ported by boat. Gas bladder damage to fish brought to the surface was reduced
by inserting a hypodermic needle into the gas bladder so that gas could escape.

All manipulations were conducted within inshore Macrocystis pyrfera kelp
forests. Habitats of capture and release were similar, consisting of level rocky
bottom with numerous holes and intersticies. Depth of capture and release for
each fish were within 1 m of each other. Areas chosen for removal of fish were
approximately 25 m^ and ranged in depth from 7 to 12 m.

Three short translocations were conducted over the distances of 25, 40, and
50 m. Nine fishes, which were not measured but were adults, were displaced
on the 25 m translocation, and the home and release sites were checked 3 times
over 21 days (Table 1 ). Seven fishes, ranging in size from 180-300 mm tl, were
displaced on the 50 m translocation, and the home and release sites were
checked 7 times over 326 days (Table 1 ). Seven fishes, ranging in size from
200-280 mm tl, were displaced on the 40 m translocation and the home and
release sites were checked 6 times over 43 days (Table 1 ). In the 25 and 50 m
translocations, none of the territorial residents in the release area were removed.
In the 40 m translocation, all territorial fish occupying the release area were
removed by spearing before translocated fish were released. Eight 5. chrysome-
las, ranging in size from 180-290 mm TL, were removed. The gopher rockfish,
5. carnatus, which is also territorial, was not seen on the release site.

On the long-distance translocation, seven fishes ranging in size from 200-305
mm TL, were moved to a different kelp forest within Carmel Bay. The direct
distance separating the home site (Carmel Point — Butterfly House) and the
release area (Point Lobos State Reserve) was approximately 1.5 km, while I
estimate the distance by coastline to be about 10 km. The home and release sites
were checked five times over 123 days.

RESULTS

A high percentage of translocated black-and-yellow rockfish returned to their
home sites after being moved 25, 40 and 50 m (Figure 1 ). Of nine individuals
moved 25 m, a total of six were sighted on the 25 m^ grid area after 13 days,
with three being seen at least one time in the holes from which they were
captured (Table 1 ) . Three of the individuals translocated 25 m were not resight-
ed, either on the home site or in the release area. Five of the seven fish moved
40 m were resighted on the home site after 8 days, with all five being seen at
least one time in their original holes. On the 50 m translocation, five of the seven
fish moved were resighted on the home site after 11 days, with all five being
found in their original holes (Table 1 ). On both the 40 m and 50 m transloca-
tions, two of the fish were not resighted.

None of the seven individuals moved 1 .5 km were observed on the home site
during 5 checks of the area over a period of 3 months (Figure 1). As in the
shorter translocations, each time the home site was checked on this long-
distance translocation, the release area was also inspected. None of the tagged
fish were seen.

160 CALIFORNIA FISH AND CAME

TABLE 1. Resightings of Previously Translocated Fish on Their Original Home Sites Are
Listed. Day Values Represent Time Elapsed Between the Site Check and the
Translocation Date. Meter Designations Represent the Distance That a Resightied
Fish Was From Its Original Hole. Distances for Individuals Seen in Their Original
Holes Are Listed As 0 m. Individuals Not Sighted During a Particular Check of the
Home Site Are Denoted by Dashes ( — ).

25m Translocation
29 August 1975

Check 1 Check 2 Check 3

Specimen No. Size 6 days 13 days 21 days

1 — — 1.2 m 1.8 m

2 - - 1.5 m -

4 — — 3.5 m —

7 300mm* Om 2.1m —

8 300mm* Om 1.8 m —

9 180mm* Om — —

* Size estimated.

40m Translocation
24 August 1976

Check 1 Check 2 Check 3 Check 4 Checks Checks

Specimen No. Size 1 day 7 days 8 days 9 days 13 days 43 days

1 250mm - 0.3 m Om - - 1.2 m

2 250mm — _____

3 200mm ______

4 280mm — — ' 0 m — — —

5 270mm 1.8 m 3.5 m 3.5 m 3.5 m 3.5 m 3.5 m

6 240mm 0.3 m — - - - 0.3 m

7 280mm Om — — — Om Om

50m Translocation
12 September 1975

Check 1 Check 2 Checks Check 4 Checks Checks Check 7

Specimen No. Size 7 days 10 days 11 davs 12 days 306 days 310 days 326 days

1 315mm — OmOm— — — —

2 270mm 0.9 m 0 m 0 m 0 m — — 2.0 m

3 215mm — — Om— — — —

4 300mm - — OmOm— — —

5 300mm Om Om Om — 0.3 m Om —

6 180mm _______

7 250mm _______

The removal of territorial residents from the 40 m translocation release area
did not alter homing behavior. The same percentage of fish returned from this
release area, as from the 50 m release area where resident territorial fish were
not removed.

The time required for displaced fish to return to the home site appeared to
be short. Minutes after their release on the 40 m translocation, two fish were
observed about half way between the release area and the home site. Both were
swimming close to the bottom in the direction of the area from which they were
removed. .

BLACK-AND-YELLOW ROCKFISH RELOCATION

161

80

70

60

FISH RETURNED TO HOME SITE
FISH RETURNED TO ORIGINAL HOLE

Z
IX

I-
m
a.

\-
z

UJ

u

oc

LU
Q.

50-

40-

30'

20-

10.

25M
n=9

40M
n=7

DISTANCES MOVED

1.5=km
n=7

FIGURE 1.

Return of 5. chrysome/asto home sites (open bars) and original holes within home sites
(hatched bars) after their translocation to release areas 25m, 40m, 50m, and 1.5 km
away from the home site. Columns represent percentage of n fish translocated that
returned.

CONCLUSION

The black-and-yellow rockfish, 5. chrysomelas, is a territorial species, proba-
bly occupying territories on a perennial basis (Hallacher 1977; Larson 1980^,
b). Observations suggest that this species normally remains closely associated
with its territory, although Larson (1980a) observed individuals, which he re-
ferred to as "commuters", that moved up to 20 m on a daily basis. The results
of this investigation indicate that a high percentage of 5. chrysomelas were able
to return to their territories when displaced distances of 25 to 50 m, although
none were able to return to their home site after being displaced 1.5 km.

The results of the successful translocations raise the interesting question of
why a species, apparently perennially sedentary in habit, possesses the ability
to home over distances of at least 50 m; distances which exceed, by a factor
of two, previously reported movements by this species. Black-and-yellow rock-
fish may, in fact, be more mobile than has been reported, which could result
in their familiarity with features of the bottom near their territories, and could
explain their short-range homing ability. However, a general knowledge of sur-

-[62 CALIFORNIA FISH AND CAME

rounding bottom topography may be retained from recruitment, and life as a
nonterritory holder. This, too, could account for short-range homing in a species
normally exhibiting limited extraterritorial movement.

The use of physical parameters other than visual knowledge of local topogra-
phy may be involved in successful homing by this species. If so, the parameters
utilized in relocating the home site are unknown. The use of stimuli such as
thermal, or rheotactic clues seems unlikely due to the homogeneous nature of
the kelp forest in regard to those parameters. Perhaps territorial residents deline-
ate their home ranges with chemical markers, but this, too, seems unlikely.
Identification of the mechanisms utilized by the black-and-yellow rockfish to
home, as well as the adaptive significance of this seemingly paradoxical behavior
remains to be discovered.

ACKNOWLEDGMENTS
N. VanValkenberg, D. Moe, and D. Roberts assisted with the translocations.
Equipment was provided by the University of California Bodega Marine Labora-
tory and by the Department of Zoology, University of California at Berkeley. K.
McKaye and R. Colwell reviewed the manuscript and provided valuable sugges-
tions. Anonymous reviewers were also very helpful. The manuscript was typed
by G. Inouye and Y. Higa.

LITERATURE CITED

Carlson, H. R. and R. E. Haight. 1972. Evidence for a home site and homing of adult yellowtail rockfish, Sebastes

flavidus. Can., Fish. Res. Bd., J., 29(7): 1011-1014.
Clarke, T. A. 1970. Territorial behavior and population dynamics of a pomacentrid fish, the garibaldi, Hypsypops

rubicunda. Ecol. Monogr., 40(2): 189-212.
Green, J. M. 1971. High tide movements and homing behavior of the tidepool scuplin Ollgocottus maculosus. Can,,

Fish. Res. Bd., J., 28: 383-389.
Hallacher, L. E. 1977. Patterns of space and food use by inshore rockfishes (Scorpaenidae, Sebastes) of Carmel

Bay, California. Dissertation, University of California, Berkeley, California. 115 p.
Larson, R. J. 1980a. Territorial behavior of the black-and-yellow rockfish and gopher rockfish (Scorpaenidae,

Sebastes). Marine Biology, 58: 111-122.
Larson, R. ). 19806. The influence of territoriality on adult density in two rockfishes of the genus Sebastes. Marine

Biology, 58: 123-132.
Miller, D. J. and J. ). Ceibel. 1973. Summary of blue rockfish and lingcod life histories; a reef ecology study; and

giant kelp, Macrocystis pyrifera, experiments in Monterey Bay, California. California Department of Fish and

Came, Fish Bull. (158) 137 p.
Williams, C. C. 1957. Homing behavior of California rocky shore fishes. Univ. Calif. Publ. Zool., 59: 249-284.

RADIOTELEMETRIC IMPLANTATION IN BLACK BEARS 163

Calif. Fish and Game 70(Z): 1 63— 1 66 1 984

SURGICAL IMPLANTATION OF A RADIOTELEMETRY
DEVICE IN WILD BLACK BEARS, URSUS AMERICANUS^

DAVID A. JESSUP

California Department of Fish and Came

1701 Nimbus Road, Suite D

Rancho Cordova, California 95670

AND

DONALD B. KOCH

California Department of Fish and Game

P.O. Box 47

Yountville, CA 94559

Ten free-roaming black bears, Ursus americanus, were implanted surgically with
radiotelemetry devices. The devices were implanted subcutaneously, parallel with
the spine, in the cervical-scapular region. Three of the 10 implants were rejected and
recovered. One implant was recovered when the bear was killed on a depredation
permit. The fate of the remaining six implants is uncertain. Mechanical failure,
conspecific interactions, and surgical failure are factors that affected the success of
this study. Poaching also may have been a factor.

The advent of radiotelemetry in the late 1960's enabled biologists for the first
time to determine movements, habitat utilization patterns, and mortality factors
of black bears.

Generally bears are fitted with conventional radiotelemetry collars on all age
and sex classes of a typical population. Adult black bears, especially the males,
often have neck girths larger than their maximum head girths resulting in slippage
and/or removal of the telemetry collar. Because they are still growing, subadult
bears can only be fitted with expandable radiotelemetry collars which often are
removed easily. Also, the elastic in these collars deteriorates and breaks within
six months, a time period too short to follow bears from den emergence in spring
to den entry in winter (Koch 1983).

A radiotelemetry device that is surgically implanted can be used on all mem-
bers of a bear population. Implantable telemetry devices have the added advan-
tage of generating physiologic data such as heart rate, body temperature, and
respiration rate.

Prior use of radiotelemetry implants in black bears has met with mixed suc-
cess. The body location chosen to implant the device appears to be the critical
factor. Implanting the device in areas subject to licking and/or scratching have
failed in the past. Intra-abdominal implants have resulted in a greatly reduced
transmission range (Mike Philo, Institute of Arctic Biology, University of Alaska,
Fairbanks, pers. commun.). Due to these problems, we chose to experiment
with a subcutaneous location in the cervical and scapular region.

Ten radiotelemetry transmitters designed for surgical implantation were pur-
chased from Telonics, Inc. (932 E. Impala Ave., Mesa, Arizona 85204). The
implants each weighed 150 g and were shaped like a small, fat cigar 12.5 cm by
3.5 cm in diameter. The implants were coated with two layers of moisture-
impervious resin and paraffin. They transmitted in the 159 MHz range with an

Accepted for publication November 1983. Supported by Federal Aid in Wildlife Restoration Project W52R.

164 CALIFORNIA FISH AND CAME

effected radiated power of 250 mW. All of the transmitters were designed to
double their pulse rates when motionless for five or more hours (mortality
mode). Additionally, three of the transmitters were designed to vary pulse rate
with body temperature, unless overridden by the mortality mode. Prior to im-
plantation, the transmitters had an effective transmission range of 5-8 km. In an
attempt to increase this range, the transmitter antenna length was increased by
wrapping 00 surgical stainless steel wire around the center of the implant three
times. To assure tissue compatibility, the implants were dipped three times in
medical grade silicone (Silastic, Dow Chemical Co., P.O. Box 6851 1, Indianapo-
lis, Indiana 46268) that had been thinned with petroleum ether for easier han-
dling. When dry, the implants were gas sterilized. Four mock-up transmitters
with the exact weight, dimensions, and external coatings were prepared for use
in laboratory tests to determine the most effective surgical technique and loca-
tion. Results from four separate implants on two captive bears determined
surgical technique later used for ten black bears in the French Meadows area
of the Tahoe National Forest.

Bears were captured in culvert traps and immobilized with a 0.02 mg/kg of
body weight injection of etorphine ( M-99, Drummer Laboratories, 1 1 1 Leuning
St. South, Hackensack, NJ 07706). When the bears vital signs were stable the
implant site was prepared for aseptic surgery. A 15-cm incision was made along
the dorsal mid-line over the last two cervical vertebrae. Subcutaneous tissues
were bluntly separated and the dermal shield was incised. The insertion and
aponeurosis of the trapezius muscles were located and incised, exposing the
rhomboideus muscles. A bed for the implant was prepared between the left and
right rhomboideus muscles and the implant slipped into place. As a prophylactic
measure, one million units of potassium penicillin were injected around the
implant. The stainless steel antenna wires of the device were secured to deep
connective tissue and then to one another. The trapezius aponeurosis and the
dermal shield were each closed by simple continuous suture pattern of #1
dexon. The skin was closed with a simple interrupted suture pattern of #1
dexon. One million units each of benzathine and procain penicillin (Flo-Cillin,
Bristol Laboratories, Syracuse, New York 13201 ) were administered intramuscu-
larly. After the implant was tested for inplace function, a 0.04 mg/kg of body
weight injection of diprenorphine (M50-50, Drummer Laboratories) was given
intravenously to reverse the affects of etorphine. Recovery was marked by
several rapid respirations, a brief groggy period, then fairly rapid well-coordinat-
ed flight. Recovery took from 2-5 minutes.

The healing process was evaluated in only two bears. Bear L8828 was held
in captivity after surgery for approximately two months prior to his release in the
study area. The healing process was rapid and hair regrowth made it difficult to
visually detect the implant one month later. When this animal was killed on a
depredation permit 14 months after the implant surgery, the transmitter was
accidentally discovered when the bear was skinned. Bear #4890 was recap-
tured 15 days after surgery. FHealing was complete and there was no evidence
of post-surgical infection.

The experimental use of radiotelemetry implants during this study was not
successful (Table 1 ). Three of the telemetric devices were either sloughed due
to surgical failure, removed by the bear itself, or by another bear. One transmitter
was recovered when an implanted bear was killed on a depredation permit. The

RADIOTELEMETRIC IMPLANTATION IN BLACK BEARS

165

remaining six implants were not recovered. These implants functioned for a short
time. They either failed, or the bears left the study area.

TABLE 1. Summary of Results for Ten Radiotelemetry Implants in Free-Roaming
Black Bears, Tahoe National Forest, California.

Bear # Sex /Age

4894 F/5

4895 M/2

4897 F/2

4896 F/5

4892 M/5

4890 F/7

4888 M/5

4891 F/7

4893 M/3

L8878 M/5

Number Oi

Implant Date

Last Location

Locations

1 July 1979

28 Aug 1979

23

13 July 1979 28 Sept 1979

16

16 July 1979

7 Aug 1979

16)ulyl979

20 Aug 1979

11 July 1979

21 Aug 1979

1 July 1979 5 Sept 1979

6 June 1979
10 July 1979
12)uly 1979

28 Sept 1979

30 luly 1979

5 Oct 1979

15

10|unel978 23 July 1979

Comments
Transmitter rejected; recovered
in area of heavy bear use. Mated
with bear #4892.
Transmitter functioning sporadi-
cally from 29 Aug 1979. Trans-
mitter failure or illegal kill.
Suspected transmitter failure or
illegal kill.

Transmitter rejected: recovered
in area of low bear use.
Suspected transmitter failure or
illegal kill. Mated with bear
#4894.

Bear recaptured 15 days after*
implant surgery; implant area
completely healed, but transmit-
ter rejected and recovered.
Transmitter functioning sporadi-
cally, and not recovered.
Suspected transmitter failure or
illegal kill.

Transmitter shifted to mortality
mode; however, bear moved
several times after that. Trans-
mitter not recovered.
Testbear;trapped 15 April 1978,
released 10 June 1978. Shot on
depredation permit 28 July 1979.
Implant recovered.

The three recovered transmitters all had teeth marks the size, depth, and
spread suggesting the canine teeth of a large canivore, probably ursine. These
transmitters were free of blood or tissue and one had lost its Silastic coating.
There was no evidence of a carcass or a bear kill in the area surrounding the
recovered transmitters. It is possible that these three transmitters were removed
during breeding activities. Bear #4894 mated with bear #4892 five days after
her implant surgery. Bears #4890 and #4896 were adult females in estrus at the
time of their implant surgery, so it is likely that they mated after surgery. The
possibility of surgical failure in bear #4890 is unlikely as excellent healing was
observed 15 days after surgery.

The fate of six unrecovered transmitters remains unknown. Mechanical failure
is suspected as two of the implants functioned sporadically before contact was
lost. The transmitter in bear #4893 shifted to mortality mode while the bear
continued to move throughout the study area. The increase in pulse rate when
the transmitter is in mortality mode will greatly reduce the battery life of the
transmitter, and this may explain the loss of contact.

166 CALIFORNIA FISH AND CAME

Illegal take of black bears is a serious problem throughout California ( Piekieiek
and Burton 1975, Sitton 1982, Koch 1983) and the possibility that transmitters
were removed from the study area because of illegal hunting cannot be ruled
out. Natural movements out of the study area by implanted bears are not likely
as little dispersion was observed by animals fitted with conventional radio-
telemetry collars in the same area (Koch 1983). Also, extensive efforts, via
fixed-wing aircraft, to locate implanted bears outside of the study area were
unsuccessful.

The lack of success in this study should not be attributed solely to the surgical
procedure. Mechanical failure, and conspecific activity also contributed to the
failure of this experiment. In addition, suspected illegal hunting may have been
a factor.

The authors would like to thank C. Jenner, M. Philo, and G. Esra for consulta-
tion, and A. Brody, T. Burton, B. Cole, B. Grenfell, and R. Imai for their assistance
on this study, and S. Records for her clerical contribution.

LITERATURE CITED

Koch, D. B. 1983. Population, home range, and denning characteristics of black bears in Placer County, California.
Thesis. California State Univ., Sacramento. 71 pp.

Piekieiek, W., and T. S. Burton. 1975. A black bear population study in northern California. Calif. Fish Came,
61(1): 4-25.

Sitton, L. S. 1982. The black bear in California. Calif. Dept. of Fish and Game, Sacramento. 85 pp.

SEASONAL VARIATION IN STICKLEBACK DIET 167

Calif. Fish and Came 70 ( 3 ) : 1 67—1 72 1 984

SEASONAL VARIATION IN THE DIET OF THE THREESPINE

STICKLEBACK, GASTEROSTEUS ACULEATUS, IN

CONTRA COSTA COUNTY, CALIFORNIA ^

RANDAL J. SNYDER 2
University of California
Berkeley, California

The major food items of threespine stickleback, Casterosteus aculeatus, in Contra
Costa County were insects (41.8%), Crustacea (27.8%), and earthworms (9.6%). The
relative importance of these items changed seasonally, although chironomid larvae
formed a large part of the diet throughout most of the year. The percentage of fish
with empty stomachs also varied seasonally, with the largest percentage occurring
in fall and winter. The existence of schools of females and nonbreeding males may
account for conflicting reports concerning differences in feeding behavior between
the sexes during the breeding season. These schools may also account for the appar-
ent correlation between food shortage and an increase in nest raiding.

INTRODUCTION

The threespine stickleback has a wide distribution across parts of Asia, Europe,
and North America (Wootton 1976). In California, it is found in streams along
the northern coast and throughout the Central Valley (Bell 1976). Sticklebacks
have a large impact on California's waterways, consuming eggs and young of
commercially important species and serving as food for predatory fishes and
birds (Moyle 1976). The remarkable success of this species is due to a variety
of characteristics that enable it to survive in diverse environments. These charac-
teristics include a tolerance for a broad range of salinities and temperatures,
morphological and behavioral adaptations to deter predators, and a high degree
of care of the young by the male.

Another factor contributing to the success of sticklebacks is opportunistic
feeding behavior. Their diet consists of a wide range of items including the larvae
and pupae of insects, crustaceans, molluscs, oligochaetes, fish eggs, plant materi-
al, and detritus (Wootton 1976). A study of seasonal variation in the diet of
stream-dwelling sticklebacks in England showed that the relative importance of
food items changed greatly with the season: Crustacea were important during the
summer, while molluscs and oligochaetes were consumed mainly in winter
(Hynes 1950). Chironomid larvae were an exception, forming a major part of
the diet in all months.

According to Schoener (1971), differences in feeding behavior should be
expected between the sexes. Females generally maximize their reproductive
output by increasing the energy spent on egg production and by choosing an
optimal mate. In contrast, males generally maximize their reproductive output
by choosing appropriate behaviors, such as mating as often as possible or by
providing care for their young. The result is often lower energy requirements for
males during the breeding season. In sticklebacks, there are conflicting reports
regarding feeding differences between males and females. Worgan and Fitz-
gerald (1981 ) reported that male sticklebacks feed less frequently than females

^ Accepted for publication October 1983

^ Mr. Snyder's current address: Department of Zoology, University of California, Davis, California 95616

168 CALIFORNIA FISH AND CAME

during the breeding season, but Hynes (1950) found no significant differences
in feeding habits between the sexes.

Sticklebacks of both sexes commonly raid nests, although males are more
often involved than females (Kynard 1978). The widespread occurrence of this
behavior among breeding males suggests that it plays an important role in their
reproductive process. Nest raiding may increase the attractiveness of the raider
to females (Li and Owings 1978, Rohwer 1978), and raiding males may increase
their reproductive output by fertilizing the eggs in a rival's nest (Assem 1967).
There is evidence that the frequency of nest raiding increases in both sexes if
there is a shortage of food (Adalsteinsson 1979). This correlation between the
supply of food and the frequency of nest raiding has not been reconciled with
theories of nest raiding as a reproductive strategy.

This study provides information on the diet of California sticklebacks and adds
to our knowledge of nest raiding and differences in feeding behavior between
the sexes.

STUDY AREA

All specimens were collected from San Pablo Creek, a freshwater drainage in
Contra Costa County. Four study sites wre chosen using ease of access and
presence of sticklebacks as criteria. They were "Garden," "Park," "Fordham,"
and "Appian" and were approximately .8, 1.8, 3.5, and 6.0 km from the San
Francisco Bay, respectively. The width of the creek ranged from 2.0-3.0 m and
the depth from 0.5-1.5 m. The flows increased greatly during periods of heavy
rain and flooding occurred at most locations. Water temperature ranged from
9°C in January to 18°C in August. The creek contained sparse vegetation except
for emergent grasses {Phalaris) and ground ivy {Hedera) near the banks.

METHODS AND MATERIALS

Specimens were collected with a long-handled dipnet monthly from February
1982 to January 1983. They were anesthetized in 0.5% phenoxyethanol, and
within minutes transferred to a 10% formalin solution. After 3^ days they were
rinsed and placed in 50% isopropyl alcohol.

For each specimen the following were recorded: body weight, standard length
(SL), number of lateral plates, body depth (greatest value), and length of pelvic
and dorsal spines. These features separate subspecies of Gasterosteus. Sex was
determined by the size and color of the gonads: ovaries were large and yellow
or white, testes were smaller and gray or black. To assure accurate plate counts
and correct determination of sex, only specimens over 30 mm were used.

The composition of the diet was determined by the point system outlined by
Hynes (1950). The contents of each stomach were examined within 2 wk
following capture and identified as far as possible under a stereoscope. Insects
were grouped into pupal, larval, or adult stages. The relative frequency of each
food type was estimated visually, and assigned points based on their relative
abundance. The points for each category were summed over all individuals to
obtain the total diet composition.

RESULTS
The sticklebacks collected from San Pablo Creek were Gasterosteus aculeatus
microcephalus, characterized by low plate counts and lack of a caudal keel ( Bell
1976). Across ail seasons and sites, 41.8% of the diet was insects (mainly

SEASONAL VARIATION IN STICKLEBACK DIET 169

chironomid larvae), 27.8% was Crustacea (mainly ostracods), and 9.6% was
earthworms {Lumbricidae). Stomachs were empty in 20.7% of the specimens
examined (Table 1 ).

TABLE 1. Composition of the Diet of Threespine Stickleback from San Pablo Creek, Contra
Costa County, California

All Sites
Food Item Appian Fordham Park Garden Combined

Crustacea

Ostracoda 11.5 12.2 4.5 7.5 9.0

Other Crustacea 14.9 22.3 21.5 17.0 18.8

Insects

Pupae 5.7 14.4 9.8 6.7 9.2

Chironomid larvae 16.8 18.1 18.0 28.5 20.6

Other larvae 9.7 4.7 3.2 3.5 5.2

Diptera adults 2.4 2.2 1.3 2.9 2.2

Other adults 5.4 5.1 3.2 4.5 4.6

Nematodes .3 .5 1.2 .1 .5

Annelids

Earthworms 13.6 4,7 11.0 9.1 9.6

Other Annelids .8 .4 .8 .3 .5

Fish eggs 2.0 3.0 2.2 2.3 2.4

Plant Material 2.6 1.8 12.6 9,7 6.6

Animal Material 14.3 10.6 10.7 7.9 10.8

Stomachs examined 167 183 161 175 686

Empty stomachs 33 45 34 30 142

Percent empty stomachs 19.8 24.6 21.1 17.1 20.7

The diet of the sticklebacks showed pronounced seasonal variation (Figure 1;
Table 2 ) . Earthworms were the most common item eaten from February to April,
but were absent from the diet from May to October. Insects, especially chirono-
mid larvae, were consumed throughout the year and were the major component
of the diet from May to August. Crustaceans were also eaten in most months,
and were preyed upon heavily from September to December.

The percentage of fish with empty stomachs varied seasonally, with the high-
est percentage in winter and the lowest in spring. During the breeding season
from February to August, the number of empty stomachs among males (29 out
of 134) was significantly higher than among females (24 out of 212) (X ^ = 5.97,
d.f. = 1; P < .05).

Evidence of intraspecific egg predation was found in samples collected from
May to August. The number of males that had consumed eggs (13 out of 80)
was significantly larger that the number of females (6 out of 128) (X ^ = 6.60,
d.f. = 1; P < .05). The males involved were over 39 mm sl, the females over
48 mm SL. All of the eggs consumed were fertilized, and eyed embryos were
visible.

DISCUSSION

The seasonal variation in the diet of sticklebacks in San Pablo Creek was
similar to that elsewhere (Hynes 1950, Walkey 1967). Chironomid larvae were
eaten in large numbers throughout the year, while the importance of other items

170

CALIFORNIA FISH AND CAME

varied seasonally. The appearance of earthworms in the diet during the winter
may have been a result of high water levels washing the worms out of the soil.
A similar occurrence involving terrestrial arthropods was mentioned by Hynes
(1950). Seasonal variation in the diet is a reflection of changes in the actual
numbers of food items as well as changes in their relative availability.

My results and those of others (Semler 1971, Kynard 1978) indicate that egg
predation is more common among male sticklebacks than among females.
Breeding males may consume eggs obtained in successful raiding attempts, and
there is evidence that males may eat some of their own brood (Rohwer 1978).
Other instances of egg predation are probably due to attacks on nests by schools
of females and nonbreeding males. These raider packs have been observed to
destroy nests and to eat the eggs (Kynard 1978).

0)

o

3

o
o
O

70-

60'

50'

40'

30'

2 0'

10-

I

±

±

±

Insects

Earthworms
A

A .A — ^ A -A A--

T ' 1 ' i ' 1 • 1 ' 1 r—

Feb Apr Jun Aug Oct Dec

Month!

FICURE 1. Seasonal variation in the consumption of major food groups by threespine stickleback
from San Pablo Creek, Contra Costa County, California.

The correlation between food supply and frequency of nest raiding (Adal-
steinsson 1979) may be due to increased attacks by raider packs when food is
scarce. It is likely that the purpose of these attacks is to obtain food and their
frequency would be influenced greatly by food supply. In contrast, nest raiding
in breeding males is closely linked to the reproductive process, and therefore not
affected by supply of food.

SEASONAL VARIATION IN STICKLEBACK DIET

171

e

3

o
U

•~D

r^

OS

CO rsi On

c:

LiS

O

O

^ - rvi

■5

<

1-^

■*

e

<-S| Tj- OO

rvi i~^ vO

e
o
U

u

::>.

^

s

■^ a^ r-^
m CM

CT^ un O
O O O

C
(/>

E

o

u
(/)

c
'a

(A
U

V

>•
JO
in

a

3

o

u

■D
O
O

^
^

eq LTi o
(->j t< o

.5

o -r

CO —

o

^

h^

CO

0

C:

O

r^

0

.§

-^

0.4

U1

.^w

i^

o

g.

1

,_

P

,—

^

0

•—

c

1>

be

<b

<^

03

, —

0

5

rvi

l<

0

^

^

s

CO LO

0^

> ■— ',0 LO

^ O CO

CO en m

m •— rsi

0-1 O O

,0

Tj- rj- CM

1^ CO O
06 >X> O

rs| LT)

O
O

a
E

3
VI - — -

c -o

o &•

u.S

£i

** o
c w

'" (/)

C V

o .■=

rt _

'%3.

> r,

19 E
§5

4* t )

eq

t-^

r-^

v£> Ln CO

;P

!-si

^

rsi

^ - 0

ro

rO

t-M

I

03

c

1-5 >-

E
o

a.

E

t Q. O

2 E S

172 CALIFORNIA FISH AND CAME

My results support the observation of Worgan and Fitzgerald (1981) that
breeding males feed less often than females. Insufficient data were collected in
the present study to provide a comparison between feeding behavior of breeding
and nonbreeding males. However, the observations of Kynard (1978) strongly
suggest that nonbreeding males continue to feed throughout the breeding sea-
son, in contrast to the infrequent feeding in breeding males. This difference in
feeding behavior accounts for the results of Hynes (1950), which show no
significant differences in feeding behavior between the sexes. If the samples
collected by Hynes contained a large proportion of nonbreeding males, then the
infrequent feeding of breeding males might have been overlooked.

Although much is known about the general feeding habits of sticklebacks,
there is more to be learned about their feeding behavior during the breeding
season. Controlled experiments in the laboratory could test the hypothesis that
nest raiding in breeding males is independent of the food supply. In addition, a
careful analysis of specimens collected from the field might give further support
to the claim that breeding males feed less frequently than females and nonbreed-
ing males.

ACKNOWLEDGMENTS

My thanks to G. Barlow for his guidance and encouragement throughout this
project. I also wish to express my appreciation to R. Kaplan, G. Olesen, and T.
Hopkins for their help and advice during this study. The University of California,
Berkeley, Department of Zoology graciously supplied the materials and supplies
used in this project.

LITERATURE CITED

Adalsteinsson, H. 1979. Size and food of Arctic char Salvelinus alplnus and stickleback Casterosteus aculeatus in

Lake Myvatn. Oikos, 32: 228-231.
Assem, ), Van den. 1967. Territory in the three-spined stickleback, Casterosteus aculeatus L. An experimental study

in intraspecific competition. Behaviour Suppl., 16: 1-164.

Bell, M. A. 1976. Evolution and phenotypic diversity in Casterosteus aculeatus superspecies on the Pacific coast
of North America. Systematic Zoology, 25(3): 211-227.

Hynes, H. B. N. 1950. The food of fresh-water sticklebacks (Casterosteus aculeatus and Pygosteus pungitius) , with
a review of methods used in studies of the food of fishes. |. Anim. EcoL, 19: 36-57.

Kynard, B. E. 1978. Breeding behavior of a lacustrine population of threespined sticklebacks (Casterosteus aculea-
tus L.). Behaviour, 67: 178-207.

Li, S. K., and D. H. Owings. 1978. Sexual selection in the three-spmed stickleback: II. Nest raiding during the
courtship phase. Behaviour, 64: 298-304.

Moyle, P. B. 1976. Inland Fishes of California. Univ. Calif. Press, Berkeley, viii + 405p.

Rohwer, S. 1978. Parent cannibalism of offspring and egg raiding as a courtship strategy. Amer. Natur., 112: 429-
440.

Schoener, T. W. 1971, Theory of feeding strategies. Annual Reviews of Ecology and Systematics, 11: 369-404.

Semler, D. E. 1971. Some aspects of adaptation in a polymorphism for breeding colours in the threespine stick-
leback (Casterosteus aculeatus) . ]. ZooL, Lond., 165: 291-302.

Walkey, M. 1967. The ecology of Neolchinorynchus rutili (Muller). J. Parasit., 53: 795-801.

Wootton, R. ). 1976. The Biology of the Sticklebacks. Academic Press, London, x + 387p.

Worgan, J. P. and C. J. Fitzgerald. 1981. Diel activity and diet of three sympatric sticklebacks in tidal salt marsh
pools. Can ). ZooL, 59(12): 2375-2379.

i

WHITE SHARK ATTACKS IN CHILE 173

Calif. Fish and Came 70 ( 3 ) : 1 73— 1 79 1 984

ATTACKS ON DIVERS BY WHITE SHARKS IN CHILE ^

ALFREDO CEA ENCANA

Universidad del Norte

Coquimbo, Chile

AND

JOHN E. McCOSKER

California Academy of Sciences

San Francisco, California 94118, USA

Three attacks upon divers in Chile are reported, analyzed, and illustrated. Attacks
at El Panul on 29 September 1963 and Pichidangui on 5 January 1980 were fatal and
probably caused by the white shark, Carcharodon carcharias. The third, a non-fatal
attack, was at Totoralillo and probably caused by a white shark or mako shark, Isurus
oxyrinchus. Shark attack and diver behavior are compared to those in North America.
Suggestions concerning diver safety are presented.

INTRODUCTION

The white shark, Carcharodon carcharias, is responsible for the majority of
shark attacks upon humans worldwide in temperate coastal waters (McCosker
1 981 ) . Three recent attacks on divers off the Chilean coast which resulted in two
fatalities appear to have involved white sharks. The senior author investigated
those attacks, visited the sites, and interviewed a victim and eyewitnesses.
Thorough medical study, when possible, was made. In California and Oregon,
the white shark is implicated in at least 40 attacks on humans and at least five
fatalities since 1926 (Miller and Collier 1981; and R. N. Lea unpublished data).
The similarities of the attacks on Chilean divers to those on divers in the
northeastern Pacific has allowed us to make an analysis and comparison of white
shark behavior in widely separated areas.

The Chilean coastline is over 3200 km long, reaching from subtropical waters
in the north to subantarctic waters at the tip of the continent ( Figure 1 ) . Where
accessible from the shore or by small boat, much of the coastline is exploited
for its shallow water marine resources. These include the edible sea urchins, the
large gastropod called "loco" (Concholepas concholepas) which is much like
the abalone {Haliotis spp.) in terms of its ecology and desirability by humans,
and an increasing amount of spearfishing. Since shallow inshore areas of most
of the accessible coastline have been heavily collected, diving for these re-
sources is now increasingly necessary. Spearfishing is becoming popular, both
recreationally and commercially. At any given time of the day, hundreds of
divers are working the Chilean coastline, primarily in the central part of Chile
from 29° to 37° S lat. Divers customarily wear full rubber wet suits and many of
them dive with the aid of hookah to depths as great as 40 m. Such divers are
usually aided by an airline tender and a boatman. Scuba diving was, until
recently, relatively unknown in Chile.

We further note that many places are found along the Chilean coast which
serve as haulout sites for pinnipeds. Like similar North American locations, these
rocky headlands and associated offshore reefs provide habitat for exploitable
marine resources. In that white sharks feed, as adults, primarily upon pinnipeds

Accepted for publication September 1983.

174

CALIFORNIA FISH AND CAME

and are located in the vicinity of marine mammal populations, it is perhaps
surprising that more shark attacks on man have not been reported from this
coast. The following case histories are the first, to our knowledge, which have
occurred in Chile.

30

32

c-i-slager

FIGURE 1. Shark attack locations along the Chilean coast.

ATTACK 1. BAHIA EL PANUL

El Panul is a small coastal indentation about 12 km south of Coquimbo. The
boulder-strewn shoreline has sea lion haulout areas within 800 m of the attack
site. The attack occurred on 29 September 1963 at ca. 1100 h on a clear and
sunny day. Crisologo Urizar C, a sport diver wearing a .64 cm black nylon-lined
wet suit with a yellow weight belt and blue "Cressi Rondine" fins, mask, and
snorkel, was spearfishing with a partner over a sand and kelp-covered bottom.
At the time of the attack, Urizar was about 50 m offshore in close proximity to
a semi-submerged rock formation in the center of the bay. A gentle current,
moving in a SE direction, was noticed. The divers had been spearing adult
sheephead wrasse, Semicossyphus maculatus, for 45 min. and had attached
them to a stringer on their weight belts.

Immediately prior to the attack, the diver's partner saw Urizar signal that he
was about to dive for a fish on a rock formation. The witness then hyperventilat-
ed, began to descend, and noticed a large "gray shark," about 1 m below the

WHITE SHARK ATTACKS IN CHILE 175

surface and 3 m away from him, with the victim's blue metallic speargun trailing
from the shark's mouth by the spearline. He also noticed a white object in or
near the mouth which he was unable to identify. The witness recovered one of
the victim's floating fins, and then swam to shore. He climbed to a vantage point
and observed a 2 m surface patch of bloodstained water next to the rock
formation with a flock of gulls at the surface. The shark had returned to the attack
site with its dorsal and caudal fins plainly visible; the shark then submerged.
Within two hours of the attack, an underwater search was carried out but no
remains of the victim were found. The victim's wet suit jacket, severely gashed,
washed ashore. A line was set on the morning of 30 September to catch the
shark; two large hooks, each baited with 2 kg of horsemeat, were suspended at
the attack site to a depth of 2 m. The lines were attached to a 24 m tether strung
between two 120 I drums, from which bloody horsemeat was dripping. The line
was anchored at each end by a 50 kg concrete block. The following morning
the line had been torn away and the drums were found heavily battered. The
shark was not sighted again.

From discussions with the witness and an examination of the wet suit remains,
it appears that the shark was at least 4 m long. The size and the attack behav-
ior would suggest that it was either a white shark or a tiger shark, Caleocerdo
cuvier. The water temperature at the site was not measured, but is usually
12°-14°C at that time. The temperature is appropriate for Carcharodon and is
very similar to that of California waters north of Pt. Conception. The presence
of pinnipeds in the immediate vicinity also increases the likelihood that the
locality would be inhabited by Carcharodon.

ATTACK 2. PUNTA NEGRA

Jose Larenas Miranda was fatally attacked on 5 January 1980 at Punta Negra,

7 km S of Pichidangui. This attack was recorded in a Valparaiso newspaper
account (Garcia and Inzunza 1980) and by Balbontin and Reyes (1981); we
herein analyze and supplement their data. Miranda was hookah diving for locos
along a steep rocky shoreline. The sea was calm, the day windless and overcast,
water temperature was approximately 16°C, and the water visibility was about

8 m. The victim was diving from a 7 m wood boat and was wearing a complete
black neoprene wet suit with black fins and mask. He carried a collecting bag
in his left hand and a pry bar in his right. At 1 100 h he had completed a 35 min
dive during which he had speared fish and collected locos. He moved to a
second site which was within 150 m of a sea lion haulout. During several days
preceding the attack other divers had seen parts of sea lion bodies on the bottom
near the attack site; the cause of their death (possibly human induced) was not
obvious.

At the attack site, the boat was 20 m from shore, over 1 m of water. The diver
entered the water, swam 5 m from the boat and descended obliquely. After
about one minute, the diver's forward progress stopped and the bubbles stopped
rising. Pressure began to build in the accumulator tank and then a mass of bloody
bubbles ascended. Immediately thereafter, the head and anterior trunk region
of an immense shark broke the surface about 3 m from the boat. From the shark's
jaws were visible the victim's decapitated upper torso and the left arm and
shoulder (Figure 2). The shark then dislodged the remains, which was hanging
from its mouth by the shards of the wet suit, and briefly submerged in a circling

176

CALIFORNIA FISH AND GAME

pattern. The shark soon reappeared with its head above water and its mouth
open. Nearby boats rapidly came to the site as the shark continued to circle the
body and occasionally mouth it. Motormen of the approaching boats tried to
drive off the shark by revving their motors. This partially succeeded and the men
in the victim's boat were able to recover the body, now exsanguinated. As they
were doing this, the shark attempted to snatch away the body, passing close by
and biting its right arm. The shark moved off about 15 m from the boat and then
proceeded to attack the boat at high velocity with its head out of the water and
its mouth open. The boatman managed to turn the boat obliquely to avoid the
collision and the impact was taken on the left stern, lifting the boat from the
water and turning it 90°. The boatman started the motor in an attempt to escape
but the shark then violently bumped the prow with side thrusts of its head. The
shark's snout was even with the prow of the boat and its tail was seen to extend
about 1 m beyond the stern. It was seen to be as broad as the boat's beam (1.4
m). The boatman guided the boat to shoal water and the shark followed and
became grounded. The crew escaped before the shark freed itself. The shark was
briefly sighted the next day after which it was not seen again. The shark was
described by witnesses to be brown-gray dorsally and white ventrally. The
dorsal fin was broadly triangular and the tail lobes appeared subequal. It was
estimated to be more than 7 m long. On the basis of the attack behavior and
the nature of the wounds, it is most likely a Carcharadon.

c-i.slager

FIGURE 2. Illustration of cadaver of Jose Larenas Miranda, attacked at Punta Negra.

Observations by witnesses and the character of the wounds sustained suggest
that the victim saw the shark coming and tried to fend it off with his left arm.
Secondary wounds inflicted on the cadaver as it floated belly down show that
the shark continued to attack the wounded and bleeding upper torso. The

WHITE SHARK ATTACKS IN CHILE 177

displacement of the dorsal and ventral wounds on the body indicate that the
weight belt may have interfered with the shark's lower jaw. Thoracic wounds
were probably produced by lateral teeth.

Autopsy at the Los Vilos Hospital on 9 January showed decapitation at the 7th
cervical vertebra, including amputation of the shoulder and left arm. The clavicle
was lost, as was part of the scapula and a mass of corresponding dorsal muscula-
ture. The wounds were surrounded by jaggedly cut skin. The midsection was
wounded due to secondary bites by the shark, producing both dorsal and ventral
lacerations, vertebral and rib fracture, lung penetration, and penetration of the
abdominal cavity. The right arm showed a 6 cm laceration inside the elbow. No
shark tooth fragments were found.

ATTACK 3. BAHIA TOTORALILLO

Carlos Vergara M. was attacked at Bahia Totoralillo, ca. 80 km N of Coquimbo,
on 4 March 1981. He was diving ca. 700 m from a sea lion haulout area, along
a rocky shoreline which gently sloped to depth. Sea surface was calm, surface
temperature was 16°-18° C, and the turbid water had 2-3 m visibility. He and
two other sport divers wore full black neoprene wet suits, black fins and masks,
lead weight belts, and carried arbalette spearguns. They had been spearing fishes
in depths of 3-5 m for ca. 45 minutes prior to the attack at about 1030 h. None
carried speared fishes. About three hours earlier, fishermen in a small boat had
been dynamiting schools of mackerel near the site. Sea lions had been seen in
the area scavenging discarded fishes. The men had entered the bay at its south-
ern point and had advanced in a line parallel to the coast, separated from" one
another by a distance of about 10-15 m. The victim was trailing and located a
flounder which he pursued with difficulty due to turbidity. He reached a point
about 50 m offshore at about 5 m depth, and dived vertically about 1 .5 m to view
the bottom. As he did so, he was seized firmly by the right foot and dragged
toward the bottom. Looking down, he saw the head of a shark which he
estimated to be 30-40 cm in breadth. The shark was roughly estimated to be 2
m long and dark in color. He was able to see, but not in detail, the eye of the
shark and its dorsal fin, which he described as being broadly triangular. Unable
to free his foot, he violently stamped on the shark's head with the heel of his
left foot, at which the shark freed him, turned and swam away. The victim swam
rapidly to shore, where his friends helped him from the water. The boot of his
wet suit had been slashed, and he had suffered two deep lacerations of 6 and
8 cm length which bled profusely (Figure 3). There was an extensive zone of
erosion and contusion around the wounds. First aid was administered at the site
and he was taken to the hospital for suturing. The shark was not seen again. No
teeth fragments were found in the victim.

The identity of the attacking shark is not clear. The behavior would suggest
that it was a young white shark, however the puncture wounds ( rather than deep
lacerations) might indicate a more pointed dentition, such as that of a mako,
Isurus oxyrinchus. Randall and Levy's (1976) description of a mako attack in the
Red Sea suggests a similar attack scenario. The dynamiting of fish was probably
attractive to the shark, as well as to the sea lions, which would be a further
attractant to a lamnid shark. White sharks about 2 m usually feed on fishes. They
do not attack pinnipeds until they are about 3 m in length. The teeth of 2-3 m

178

CALIFORNIA FISH AND GAME

Carcharodon are considerably more pointed than those of an adult (Tricas and
McCosker 1984), and would tend to leave puncture wounds as well as a slash
mark. This fact, as well as the proximity of the shark to shallow water (unlikely
for a mako), would favor the identity of the attacker as a white shark rather than
a mako, which prefers warmer offshore waters.

contusion

contusion

contusion

slager

FIGURE 3. Diagram showing various aspects of right foot of Carlos Vergara M., attacked at Bahia
Totoralillo.

COMPARISON TO ATTACKS IN CALIFORNIA
The attack scenarios described above are similar to those off southern Aus-
tralia (Coppelson 1963) and off California and Oregon (Miller and Collier 1981 ).
Noteworthy similarities include: offshore attacks beyond algal beds (although in
shallow water); the wearing of black neoprene wet suits by divers; and surface
diving after hyperventilation and diving in the immediate vicinity of pinniped
rookeries. These conditions suggest that, as happens elsewhere in temperate
waters, white sharks mistake the appearance and behavior of divers for pin-
nipeds, their normal prey.

It is significant that two of the three Chilean divers died as a result of the
attacks, whereas only five of ca. 40 attack victims have died in California and
Oregon. It is also significant that deaths in North America resulted from massive
blood loss, rather than significant consumption by the sharks. We account this

WHITE SHARK ATTACKS IN CHILE 179

to a difference in human rather than shark behavior. In North America, it is
customary for divers and swimmers to enter the water in pairs or groups, such
that once attacked, a victim is nearly immediately taken from the water. It is
typical of white sharks to attack with one massive bite, then wait for five to ten
minutes until its victim (either human or pinniped) has bled to death before
attempting to consume it, thereby avoiding injury from a wounded but struggling
prey. This supposed "bite and spit" paradox was previously interpreted to be
related to a distasteful quality of neoprene and/or human flesh, but is now best
explained in the above manner (see Tricas and McCosker 1984, and McCosker
MS).

In that scuba diving is relatively new in Chilean waters we feel it imperative
that divers should be cognizant of the risk of shark attack. As intertidal and
shallow water resources become depleted by human behavior, it is likely that
more attacks will occur in Chilean waters upon divers. We therefore advise that
divers should enter the water in pairs or in groups, particularly in the vicinity of
pinniped rookeries, where a chance encounter with a white shark is most likely.

ACKNOWLEDGMENTS
We wish to thank the witnesses and many individuals that provided us with
information and assistance. We also thank D. Miller and E. Reyes F. for advice
and information, L. H. DiSalvo for assistance with the preparation of this manu-
script, S. Smith for translation, and C. J. Slager for assistance with the figures. We
also thank L. J. Dempster, W. I. Follett, R. N. Lea, and J. E. Randall for constructive
comments regarding this manuscript.

LITERATURE CITED

Balbontin, F., and E. Reyes. 1981. Ataque de tiburon registrado en la costa de Chile central. Not. Mensual del Mus.
Nac. de Hist. Nat., Santiago, No. 298: 3-8.

Garcia, J. M., and R. Inzunza. 1980. No esta solo el tiburon asesino. La Estrella, Valparaiso, 8 January 1980. p. 12.

McCosker, ). E. 1981. Great white shark. Science, 81 (2): 42-51.

Miller, D. J., and R. S. Collier. 1981. Shark attacks in California and Oregon, 1926-1979. Calif. Fish Came,
67: 76-104.

Randall, J. E., and M. F. Levy. 1976. A near-fatal attack by a mako in the northern Red Sea. Israeli. Zool., 25: 61-70.
Tricas, T. C, and ). E. McCosker. 1984. Predatory behavior of the white shark, Carcharodon carcharias, with notes
on its biology. Proc. Calif. Acad. Sci. (In press).

180 CALIFORNIA FISH AND CAME

Calif. Fish and Came 70{i) : ^&0-^')2 1984

NOTES

ECOLOGICAL STATUS OF STRIPED BASS,

MORON E SAX A TILIS,
IN UPPER NEWPORT BAY, CALIFORNIA

INTRODUCTION

Each April from 1974 through 1977, the California Department of Fish and
Game stocked 10,000-14,000 juvenile striped bass (100-150 mm sl) into the
upper Newport Bay estuary in southern California.

In 1978 and early 1979, we (Horn and Allen 1981 ) completed a comprehen-
sive, multiple-gear survey of the fish populations of upper Newport Bay. This
study offered the first opportunity to assess the status of the introduced striped
bass in the upper bay environment. The specific objectives of the present paper
are, therefore, to provide information on the (i) seasonal abundance and distri-
bution, (ii) size and age composition, (iii) food habits, and (iv) ectoparasite
load of the striped bass population.

MATERIALS AND METHODS
Striped bass were collected by otter trawl and gill net during the bimonthly
study (Horn and Allen 1981) carried out from January 1978 through January
1979 at four stations in upper Newport Bay (Lat 33° 37'N, Long 117° 54'W).
Standard lengths (mm sl) and weights (g) of fish were recorded in the field.
Specimens were fixed in 10% buffered formalin after their guts had been inject-
ed with formalin via the anus. We recorded the age of each of 25 specimens
of striped bass by counting the annuli on scales taken from midbody just above
the lateral line. Stomach contents of each of 26 specimens encompassing the size
range of all fish caught were removed, enumerated, identified to the lowest
possible taxonomic level and dried to a constant weight in a 60° C oven. Further
details of the field and laboratory procedures are described in Horn and Allen
(1981).

RESULTS

We caught 36 striped bass during the study. Fifteen of these, from 201 to 265
mm SL, were collected in otter trawls while 21, mostly larger individuals (240-
460 mm SL), were captured by gill net. The smallest fish (201 mm sl) weighed
150 g while the largest fish (460 mm sl) weighed 1814 g. Of the fish in which
sex was determined, 11 were males and eight were females. Fish ranged in age
from two to four years with the majority two years old (Table 1 ). Most (86%)
of the specimens were collected during the first three sampling periods (January,
March and May, 1978). The fish may spend the rest of the year in the lower bay
and/or adjacent coastal waters, or may be subject to high mortality.

NOTES 181

TABLE 1. Length And Age Relationships For Striped Bass Collected In Upper Newport Bay
In 1978-1979.

Length interval Total No. No. fish in each age class

(mm SL) offish II III IV

201-272 19 19 - -

275-356 4 - 4 -

345^60 2 - - 2

The relationship of length and weight was described by the equation logio
weight = 2.8463 logio length - 4.3957.

Stomach contents of the striped bass consisted primarily of fishes, decapod
crustaceans, and polychaetes. Fishes were the most frequently occurring items
(65% of stomachs). Gobiids comprised the largest proportion (38%) of total
dietary items, whereas the topsmelt, Atherinops affinis, was the principal com-
ponent (68%) in terms of total dry weight. The decapods in the stomachs
included yellow shore crab, Hemigrapsus oregonensis; snapping shrimp, Al-
pheus californiensis, and bay ghost shrimp, Callianassa californiensis.

One or more of the fins of approximately 75% of the striped bass was either
infested with the cymothoid isopod, Nerocila californiensis, or had scars of
infestation assumed to be from this ectoparasite. All fins except the dorsal fin
were infested.

DISCUSSION

The principal findings of this study are that ( i ) some striped bass are surviving
and functioning in their expected ecological role several years after being
stocked in upper Newport Bay, but ( ii ) appear to be in relatively poor condition
possibly because the upper bay is a suboptimal habitat. The following discussion
provides the basis for these conclusions.

In their ecological classification of the upper Newport Bay fish community,
Horn and Allen (1981) described striped bass as a member of a periodic,
channel-inhabiting species group that during the winter also included striped
mullet, Mugil cepiialus, and deepbody anchovy, Anchoa compressa. Striped
bass joins California halibut, Paralichtliys californicus; spotted sand bass, Parala-
brax macu/atofasciatus; and yellowfin croaker, Umbrina roncador, as the princi-
pal top carnivores in upper Newport Bay (Allen 1980).

Age-size and length-weight relationships of striped bass in upper Newport Bay
were generally similar to those determined in other studies in California (Scofield
1931, Robinson 1960). The slope of our length-weight equation is about 5%
lower than that (logio weight = 3.0038 log,o length — 2.1393) derived by
Robinson (1960) for striped bass from the San Francisco Bay area and lower
Sacramento-San Joaquin Delta. The age classes we identified appear to corre-
spond to fish planted in the upper bay in the four years prior to our study. That
is, our age group II fish probably represented the 1976 and/or 1977 introduc-
tions, age class III the 1975 and/or 1976 introductions, and age class IV the 1974
and/or 1975 introductions.

Our study suggests that the striped bass in upper Newport Bay constitute a
nonbreeding population. We base this conclusion on the lack of mature or
maturing gonads in individuals old enough to be in breeding condition (2-3 yrs

182 CALIFORNIA FISH AND CAME

for males, 4-5 yrs for females — see Setzler et al. 1980) and the absence of both
eggs and larvae of striped bass in the ichthyoplankton samples taken during the
1978-79 study. Furthermore, striped bass are probably unable to spawn in the
upper bay because of the lack of suitable spawning conditions. Spawning gener-
ally occurs in fresh or nearly fresh waters in the late spring and summer ( March-
July, peak in M^y and June) as temperatures reach 15-20°C (Westinand Rogers
1978; Setzler et al. 1980). Although upper bay temperatures are appropriate for
spawning during these months, the small volume of fresh water entering the
upper bay during this period of the year may be inadequate to stimulate or
sustain reproductive activity.

The high infestation frequency of ectoparasites may have a detrimental effect
on the striped bass population. Several of the largest fish we caught had lost
virtually all of their fins to the isopod parasites. Such severe fin damage must
reduce swimming and presumably feeding capabilities. Setzler et al. (1980) in
their summary of striped bass parasites listed two isopods (in genera other than
Nerocila) known to occur on the fish, but reported a much lower infestation
frequency ( < 1 to 4%) than we found for N. californiensis on the upper bay
striped bass population. The heavy ectoparasite load of striped bass in the upper
bay may be consistent with a pattern we have generally observed, namely that
bay-estuarine fish populations have a higher frequency of ectoparasite infesta-
tions than conspecifics in open coastal waters.

ACKNOWLEDGMENTS

We thank all those people who assisted in the field work, especially K. L.
Heath, W. S. White, J. A. Jones, S. M. Harvey, J. M. Fancher and R. J. Bellmer.
D. Ganssle provided the information on the striped bass stockings while T. F.
Morefield and M. D. Wilson kindly verified our age determinations of striped
bass. This study was supported by a contract from the California Department of
Fish and Game and by a grant from the Orange County Fish and Game Commis-
sion.

LITERATURE CITED

Allen, L. C. 1980. Structure and productivity of the littoral fish assemblage of upper Newport Bay, California.
Dissertation, Univ. Southern Calif., Los Angeles, Calif. 175 p.

Horn, M. H., and L. C. Allen. 1981. Ecology of fishes in upper Newport Bay, California: seasonal dynamics and
community structure. Calif. Dept. Fish Game, Mar. Resources Tech. Rep. 45, 102 p.

Robinson, ). B. 1960. The age and growth of striped bass (Roccus saxatilis) in California. Calif. Fish Came,
46:279-290.

Scofield, E. C. 1931. The striped bass of California (Roccus lineatus). Calif. Div. Fish Game, Fish Bull. (29):1-82.

Setzler, E. M., W. R. Boynton, K. V. Wood, H. H. Zion, L. Lubbers, N. K. Mountford, P. Frere, L. Tucker and ).
A. Mihursky. 1980. Synopsis of biological data on striped bass, Morone saxatilis (Walbaum). NOAA Tech.
Rep. NMFS Circ 433. 69 p.

Westin, D. F., and B. A. Rogers. 1978. Synopsis of biological data on the striped bass, Morone saxatilis (Walbaum)
1792. Grad. Sch. Oceanogr., Univ. Rhode Island, Mar. Tech. Rep. 67. 154 p.

— Michael H. Horn, Larry C. Allen, and F. Dennis Hagner, Department of
Biological Science, California State University, Fullerton, California 92634 and
Southern California Ocean Studies Consortium, Long Beach, California 90840.
Dr. Allen's current address is: Department of Biological Science, California State
University, North ridge, California 91324. Contribution No. 35 from the Southern
California Ocean Studies Consortium. Accepted for publication October 1982.

NOTES 183

FOOD HABITS OF BLACK-TAILED DEER, ODOCOILEUS

HEM ION US COLUMBIANUS, IN TRINITY

COUNTY, CALIFORNIA

Rumen contents from 40 black-tailed deer in Trinity County, California, were
analyzed to determine their food habits. Between 16 December 1979 and 5
February 1980, we shot 10 deer from the Weaverville herd and 9 from the
Hayfork herd. Between 21 March and 22 April 1980, 11 Weaverville and 10
Hayfork deer were collected. Herd boundaries, habitat descriptions, methods of
collection and analysis, and information on the physical condition of individual
deer are presented by Kie, Burton, and Menke (1984).

During early winter, Weaverville deer diets, by volume, consisted of browse
(91%),forbs (3%), and grasses and grasslike plants (6%) (Table 1 ).Ceanothus
species made up 71% of the diet, with wedgeleaf ceanothus accounting for
39%. Hayfork deer ate less browse (63%), and more forbs (11%) and grasses
(26%) early in winter than did Weaverville deer (Table 1 ). White fir (10%) and
Douglas-fir (13%) contributed substantially to Hayfork deer diets in early win-
ter, but less to diets of Weaverville deer. Hayfork deer also ate more grasses in
early winter. However, availability of different forage classes to each herd was
not measured.

TABLE 1. Percent by Volume of Forage Species in Rumens of Deer from the Weaverville and
Hayfork Herds Collected During Early Winter (16 December 1979 to 5 February
1980) and Late Winter-spring (21 March to 22 April 1980) in Trinity County,
California. Number of Deer Collected in Parentheses, Tr = Trace Amourtts.

Weaverville Hayfork

Forage species early late early late

(W) (II) (9) (10)

Percent

Browse

White fir [Abies concolor) 4 - 10 -

Chamise (Adenostoma fasciculatum) 2 3 7 2

Q,\ee'n\e3.\ mAnzAri\\aL (Arctostaphylos patula) 5 16 3 3

Whiteleaf manzanita (/I. k/5c/b'j) Tr 6 2 Tr

Ceanothus ((r(?<?/7Ci/Aw5 spp.) 9 - - -

Wedgeleaf ceanothus (C Ci//7e<3/iy5) 39 1 10 13

Lemon ceanothus (C /e/770/7//) 22 5 3 5

Deerbrush (C. integerhmus) 1 - - -

Mountain whitethorn (C. cordulatus) - - 5 6

Squaw carpet (C. prostratus) - - 1 -

Moun\.3i\n ma\\o%3,ny [Cercocarpus betuloides) Tr 6 Tr 6

Y erba sanla {£riodictyon californicum) 3 - 1 -

S\\k-[asse\ iCarrya fremonti/) - Tr - -

Joyon (Heteromeles arbutifolia) Tr - - -

\ncer\se-cedar {Libocedrus decurrens) 2-51

Honeysuckle (Zc/7/c6>A<3 spp.) - 2 - -

Pine (Pinus spp.) Tr

Stonefruits iPrunus spp.)

Douglas-fir (Pseudotsuga menziesii) Tr

Interior liveoak (Quercus wislizenii)

Hollyleaf redberry (Rhamnus crocea) Tr

Sierra gooseberry (Ribes roezlii)

Snowberry (Symphoricarpos mollis)

Other 4

Subtotal 91

-

-

Tr

7

13

14

1

1

-

2

_

Tr

5

-

4

1

2

Tr

55

63

55

184 CALIFORNIA FISH AND CAME

TABLE 1. Continued

Forbs, fungi, lichen

Mountain dandelion (/l^o^ew spp.) - - - Tr

Sunflower family (Cci/77^<95//<3e) - 2 - Tr

F\\aree (Frod/um spp.) - Tr - 2

Redstem filaree (£ cicutarium) - - 2

Bedstraw { C<3//i//7j spp. ) - - _ jr

Geranium (CeA<3/7/y/77 spp.) - Tr - Tr

W a^erleai (Hydrophy//um spp.) _ . _ _ . Tr

Mint family (Labiatae) - Tr - -

Pea (Lathyrus spp.) - 1 _ Tr

Peppetgrasi (Lep/d/um nit/'dum) - Tr - -

Lotus (Zo/i/5 spp.) - 1 1 Tr

Lupine (Zy/7//7y5 spp.) - Tr - Tr

Miner's lettuce (A^(3/7/«/7eM9/«/'<3) - - 1 3

Popcorn flower (/%^/ci6o/A/-K5 spp.) - Tr 1 -

Crowfoot family (Ranunculaceae) - _ _ ji-

Madder family (Rubiaceae) - Tr - Tr

Thimbleberry [Rubus parviflorus) - 3 _ _

Clover ( 7>-//'o//iy/77 spp. ) - - Tr Tr

Vetch ( i//c« spp.) - _ _ Jr

Other forbs - Tr - -

Fungi (Basidiomycetes) 2 Tr 2 -

Crustose lichen (Lecideaceae) - Tr 1 1

Foliose lichen (Parmeliaceae) Tr Tr 1 Tr

Fruticose lichen (Usneaceae) 1 2 2 Tr

M\s{\e\.oe (Phoradendron flavescens) - - Tr -

Subtotal 3 9 11 6

Grasses, grasslike

Grasses (Gramineae) 6 36 26 39

Rushes (Juncaceae) - Tr - -

Subtotal , 6 36 26 39

Total 100 100 100 100

During late winter-early spring, consumption of different forage classes by
Weaverville and Hayfork deer was nearly identical. Weaverville deer consumed
55% browse, 9% forbs, and 36% grasses; and Hayfork deer, 55% browse, 6%
forbs, and 39% grasses (Table 1 ). However, species of browse and forbs eaten
by each herd differed. Furthermore, differences in grass species eaten may have
also occurred, although individual species were not identified.

Consumption of all ceanothus species by Weaverville deer dropped from 71 %
to 6% between collection periods; consumption of wedgeleaf ceanothus
dropped from 39% to 1%. The percentage of wedgeleaf ceanothus in Hayfork
deer diets remained nearly constant at 10-13%. These results may be explained
in several ways: in early winter, Weaverville deer heavily used ceanothus growth
from the previous growing season; or Hayfork deer had greater access to forbs
and grasses than did Weaverville deer. Hayfork deer ate mountain whitethorn
during both collection periods; Weaverville deer did not appear to eat it at all.
Consumption of the deciduous deerbrush was low during both collection peri-
ods, even though it grew where many of the deer were collected and new leaves
were present by late winter-early spring.

In most cases, food items were related to habitat types from which the deer
were collected. Consumption of chamise was relatively low by all deer, and was
related to local availability. Three Weaverville deer that had eaten chamise

NOTES 185

(17% of their individual diets) were all collected where chamise was abundant.
All four of the Hayfork deer that had eaten chamise (22% of their diets) were
also collected where it was abundant. Higher proportions of white fir, Douglas-
fir, mountain whitethorn, and incense-cedar in the diets of Hayfork deer than
in those of Weaverville deer reflected the availability of those forage species at
the higher elevation, mixed-conifer sites where many of the Hayfork deer were
collected. We suggest that deer selected some habitats with desirable forage
(third order selection) and then chose forage species within those habitats
(fourth order selection) (Johnson 1980).

We believe that the relatively poor physical condition in Weaverville deer
reported by Kie, Burton, and Menke (1984) was correlated with diet. Consump-
tion of large amounts of wedgeleaf ceanothus and small amounts of grasses by
Weaverville deer in early winter suggests that they should have been in better
condition than Hayfork deer. Most ceanothus species are taken readily by deer
in California (Dixon 1934, Leach and Hiehle 1958). Mature grasses, low in
digestible nutrients, are not readily used by deer (Nagy, Hakonson, and Knox
1969). However, the concept that deer are browsers rather than grazers by
choice has been questioned (Wilson 1969, Gill 1976), and Evans (1976) has
suggested that high consumption of wedgeleaf ceanothus by deer may be related
to lack of more palatable forages. Furthermore, the volume of vigorously grow-
ing, green grasses in the diets of black-tailed deer in the spring has been estimat-
ed to be as as high as 55% in Tehama County (Leach and Hiehle 1958:171 ) and
90% in Mendocino County, California (Longhurst et al. 1979:221).

It appeared that Weaverville deer collected on winter ranges were in poorer
condition than Hayfork deer (Kie, Burton, and Menke 1984). We found that
over one-third of the early winter diet of Hayfork deer consisted of grasses and
forbs. Therefore, we suggest that in addition to prescribed burning (Kie and
Menke 1980), seeding of grass-forb mixtures to provide abundant herbaceous
forage in early winter should be investigated.

ACKNOWLEDGMENTS
We thank the Trinity River Basin Fish and Wildlife Task Force and the USDA
Forest Service for funding and administering this study. We also thank D. Jessup,
P. McLaughlin, G. Monroe, B. Pyshora, T. Ramsey, D. Smith, T. Stone, S. Thomp-
son, and D. Yount of the California Department of Fish and Game; and H. Black,
Jr., P. Brouha, T. Brumley, J. Okula, R. Pedersen, and H. Salwasser of the USDA
Forest Service for field assistance and advice. C. Evans and J. Verner also of the
USDA Forest Service offered critical comments on this manuscript.

LITERATURE CITED

Dixon, ). S. 1934. A study of the life history and food habits of mule deer in California. Part II, Food habits. Calif.
Fish Came, 20:315-354.

Evans, C. ). 1976. Winter and spring food preferences of mule deer of the North Kings deer herd. Unpubl. M.A.
thesis. Calif. State Univ., Fresno. 102 pp.

Gill, R. B. 1976. Mule deer management myths and the mule deer population decline. Pages 99-106 in C. W.
Workman and ). B. Low, eds. Mule deer decline in the west, a symposium. Utah State Univ. College of Nat.
Res. and Utah Agric. Exp. Sta., Logan, 134 pp.

lohnson, D. H. 1980. The comparison of usage and availability measurements for evaluating resource preference.
Ecology, 61:65-71.

Kie, J. C, and J. W. Menke. 1980. Deer response to Trinity River habitat manipulation, a cooperative study. Univ.
of Calif. Davis, USDA Forest Service, Calif. Dept. Fish and Came, Trinity River Basin Fish and Wildlife Task
Force. 158 pp.

186 CALIFORNIA FISH AND GAME

Kie, J. C., t. S. Burton, and ). W. Menke. 1984, Comparative condition of black-tailed deer, Odocoileus hemionus
columbianus, in two herds in Trinity County, California. Calif. Fish Game, 70(2);78-88.

Leach, H. R., and J. L. Hiehle. 1958. Food habits of the Tehama deer herd. Calif. Fish Came, 43:161-178.

Longhurst, W. M., G. E. Connolly, B. M. Browning, and E. O. Carton. 1979. Food interrelationships of deer and
sheep in parts of Mendocino and Lake Counties, California. Hilgardia, 47:191-247.

Nagy, ). C, T. Hakonson, and K.L. Knox. 1969. Effects of quality on food intake in deer. Trans. N. Amer. Wildl.
Conf., 34:146-154.

Wilson, A. D. 1969. A review of browsing in the nutrition of grazing animals. J. Range Manage., 22:23-28.

—John C. Kie, US DA Forest Service, Pacific Southwest Forest and Range Experi-
ment Station, 2081 £ Sierra Ave., Fresno, CA 93710; Timothy S. Burton, Califor-
nia Department of Fish and Came, Redding, CA 96001; John W. Menke,
Department of Agronomy and Range Science, University of California, Davis,
CA 95616; and William £ Grenfell, jr., California Department of Fish and Game,
Wildlife Investigations Laboratory, Sacramento, CA 95819. This research was
supported by the Pacific Southwest Region and Shasta-Trinity National Forest,
USDA Forest Service; the Trinity River Basin Fish and Wildlife Task Force; and
Federal Aid to Wildlife W-52-R-24, Study IV, Job 1.0. Accepted for publication
December 1982.

ADDITIONAL NOTES ON THE REPEAT SPAWNING BY

PACIFIC LAMPREY

Prior to Michael's (1980) report of repeat spawning for Pacific lamprey,
Lampetra tridentatus, it was assumed adults died after spawning (Wydoski and
Whitney 1979). The report of repeat spawning drew several responses and
comments. This paper will attempt to answer questions raised in the letters, and
further examine the data from Snow and Salmon creeks.

The Washington Department of Game operated permanent fish traps near the
mouths of the Snow and Salmon creeks December 1 977-June 1 981 . The streams
are located on the northeastern corner of the Olympic Peninsula and drain into
the Strait of Juan de Fuca. The traps operated year-round and trapped upstream
and downstream migrants. Lampreys captured in the traps were measured to the
nearest millimetre total length (tl) . During September to mid-February the traps
were selective and did not capture all lampreys migrating past them. This al-
lowed an unknown number of lampreys to pass upstream undetected. When the
smolt trapping screens were installed in mid-February, the traps captured all
lampreys longer than about 200 mm tl.

Upstream migrating adult lampreys were captured from September through
May. Most were trapped during periods of high streamflow. Downstream mi-
grating lampreys were trapped in the spring, mostly during May.

Both upstream and downstream migrant adult anadromous Pacific lampreys
were observed in Snow and Salmon creeks (Figure 1 ). Upstream migrants were
generally longer than 440 mm tl, while downstream migrants were generally less
than 430 mm tl. The absence of fish less than 420 mm tl captured as upstream
migrants was due to the smaller fish passing through the traps undetected. Fish
as large as 600 mm tl had been observed to squeeze through the trap pickets.
The absence of downstream migrants greater than 440 mm tl is probably due
to post-spawning mortality, as the smolt trapping screens in operation at the time
would stop all downstream migrating adults. Some of the adult downstream
migrants ("kelts") had eggs which were easily extruded, suggesting the fish

NOTES

187

were mature and capable of spawning. Operational considerations precluded
trapping all upstream migrants at Snow and Salmon creeks, so it is not possible
to determine the magnitude of the run or the relative abundance of the two
groups.

1100

1000

900

800

B 700

X

O 600h

z

UJ

^ 500h

oc
O

(/>

400
300
200
100

Cumulative Growth

-0.1982(Ago - 0.2207)

,° /' FL=95.47{1 - e )

^ (From Collins 1982)

Annual Increment

160
150
140
130
120
110
100
90
80
70
60
50
40
30
-20
10

z
c
>
r-

O
3}

o

z
o

3]

m

3
rti

z

3
3

65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81
0 1 2 3 4 5 6 7 8 9 10 1 1 12 13 14 15 16

YEAR/ANNULUS

FIGURE 1 . Length frequency distribution for adult lampreys captured in traps on Snow and Salmon
Creeks.

Downstream migration of adult lampreys has also been documented by
Washington Department of Game personnel operating downstream migrant
traps on the Kalama River system in southwest Washington. In 1979, Chilcote,
Leider, and Crawford ( 1 980) estimated over 7,000 adult lampreys moved down-
stream. The 1981 estimate for Kalama River was greater than 13,000 (S. Leider,
Wash. Dept. of Came biologist, pers. commun.). Leider indicated that most of
these fish were in good condition when captured in the Kalama River. Washing-
ton Department of Fisheries biologist D. Seiler (pers. commun.) has also trapped
downstream migrating "adult" Pacific lampreys in some Puget Sound streams.
The general appearnace and size of fish were similar to that observed in Snow
and Salmon creeks.

Several questions have been put forth by other individuals concerning the
report of repeat spawning by Pacific lamprey.

John Heinrich (U.S. Dept. of Interior biologist, Marquette, Michigan) suggest-
ed that the marks observed on the returning adults were natural. In his experi-
ence with sea lamprey, Pteromyzon marinus, fin marks occurred naturally. It is
possible that the initiation of a marking program resulted in closer examination

188 CALIFORNIA FISH AND CAME

of the lampreys. Still, the marks were only noted after marking was initiated and
not seen after marking ceased.

Dr. R. J. Beamish ( Director, Resource Services Branch of the Pacific Biological
Station, Nanaimo, Canada) suggested that the smaller lampreys, the "kelts",
were fish that would remain in freshwater for 12 months prior to spawning, as
had been observed in Babine Lake. However, extensive electrofishing surveys
in the Snow and Salmon creeks failed to find lampreys, other than ammocoetes,
during the summer months. The marked fish also showed substantial growth
between captures. It is unlikely that the small creeks could provide sufficient
food for adults to grow as much as was observed. Minimum growth observed
at Salmon Creek (largest marked "kelt" to smallest ingeminator) was 19 mm and
maximum (smallest "kelt" to largest ingeminator) was 284 mm.

The other comments simply stated that based on the writers' experience, it
was not possible for lampreys to spawn more than once. The observations at
Snow and Salmon creeks indicate that Pacific lampreys can survive two returns
to saltwater; the first follows smolting and the second appears to follow spawn-
ing. The adult fish captured on downstream migrations had all the external
appearances of spawning conditions: abrasions, pale body color, easily extruda-
ble eggs, etc. Marks were applied to some of these fish, and during the following
spawning season some marked upstream migrants were captured. These fish
were substantially larger than when marked, indicating they found a good food
source.

A number of respondents mentioned that their studies of lamprey (mainly the
sea lamprey) physiology proved that they all died after spawning. Similar re-
marks have been and are still being made for Pacific salmon, Oncorhyncus spp.,
even though there have been reports since the 1950's (Robertson 1957) that
some chinook, O. Tshawytscha, and perhaps kokanee, O. nerka, survived suc-
cessful spawning and became sexually mature the following year. The observa-
tions on chinook were from hatchery-reared fish; the kokanee were wild.

Circumstantial evidence suggests that the Pacific lamprey can spawn more
than once, since marked fish which appeared to be mature were captured on
succeeding spawning runs and showed substantial growth between captures.
The presence of downstream migrations of healthy adult Pacific lampreys in
several streams in Washington suggests that the observations from Snow and
Salmon creeks are not isolated. Further study is necessary to determine the life
history and taxonomic status of Washington's Pacific lampreys. It is possible that
a taxonomic investigation similar to Beamish's (1982) work in British Columbia
will prove illuminating.

LITERATURE CITED

Beamish, R. J. 1982. Lampetra macrostoma, a new species of freshwater parasitic lamprey from the west coast

of Canada. Can. J. Fish, and Aq. Sci., 39(5): 736-747.
Chilcote, M. W., S. A. Leider, and B, A. Crawford. 1980. Kalama River salmonid studies. Wash. Dept. Came, Res.

Rept. 80-10, Olympia. 37 p.
Michael, ). H., Jr. 1980. Repeat spawning of Pacific lamprey. Calif. Fish Came, 66(3): 186-187.
Robertson, O. H. 1957. Survival of precociously mature king salmon male parr (Oncorhynchus tshawytscha) after

spawning. Calif. Fish Game, 43(2): 119-130.
Wydoski, R. S., and R. R. Whitney. 1979. Inland fishes of Washington. Univ. of Wash. Press, Seattle. 270 p.

—John H. Michael, Jr. Washington Department of Fisheries, Room 115 General
Administration Building, Olympia, Washington 98504. Accepted for publication
August 1982.

NOTES 189

A NORTHERN FUR SEAL, CALLORHINUS URSINUS,
FOUND IN THE SACRAMENTO-SAN JOAQUIN DELTA.

Pinniped sightings in rivers and estuaries of the west coast of North America
are on the increase. Harbor seals, Phoca vitulina, California sea lions, Zaiophus
californianus, and Steller sea lions, Eumetopias jubata, have been reported
( Paulbitski 1 974; Sitts, Hayes, and Knight 1 978 ) . However, there are no recorded
sightings of northern fur seals, Callorhinus ursinus, in inland waters of the west
coast.

On 6 December 1981 the California Marine Mammal Center (CMMC) re-
ceived a rescue call from the Port of Stockton, San Joaquin County, California.
Mr. Bob Gaides had been salmon fishing with his family, an estimated 144 km
upstream from the Pacific Ocean on the Sacramento River in the canal across
from the Port of Stockton turning basin. They saw an animal they throught to
be a beaver swimming around the boat, but as it came closer, they realized it
was a small seal. They picked the animal up in their fishing net and called
CMMC.

On admission to CMMC the animal was identified as a 9Kg, 0.6m long, six
month old, female northern fur seal. The northern fur seal is normally totally
pelagic during the winter months off the coast of California unless injured or
diseased ( Daugherty 1 979) . The animal had pink mucous membranes, complete
dentition, and was alert; medically it had a soft cough and was constipated.

Fecal samples were obtained for dietary analysis and parasitic examination.
No conclusions on the animal's feeding patterns could be determined and
parasitic examination was negative.

Following two and one-half months of rehabilitation involving laboratory tests,
radiography, respiratory therapy, antibiotics, and a weight gain of 2.25Kg, the
animal was released at Southeast Farallon Island, 43 km west of San Francisco
on 27 February 1982.

On 8 March 1982, the animal was again picked up, this time in Berkeley,
California, having been hit by a car.

Why then did this animal, a species which has the most pelagic distribution
of all north Pacific pinnipeds, on two occasions swim inland? This individual's
birth, migration, and feeding habit histories are unknown. Human-animal in-
teraction at an early age might be suspected, since the animal was noted as being
"friendly" when initially netted by Mr. Gaides in the Sacramento River. At
CMMC the animal was aggressive and at no time exhibited disorientation.

It has been suggested that pinnipeds swim upstream and inland in search of
food or territory (Sitts, Hayes, and Knight 1978); in this individual's case, the
problem may have been medical impairment. The animal has been placed in a
display facility.

LITERATURE CITED

Paulbitski, P. A. 1974. Pinnipeds observed in rivers of northern California. Calif. Fish Came, 60(1): 48-49.

Sitts, R. M., Hayes, P., Knight, A. W. 1978. Sighting of a California sea lion, Zaiophus californianus, in the
Sacramento-San Joaquin Estuary. Calif. Fish Came, 64(4): 307-308.

Daugherty, A. E. 1979. Marine Mammals of California. Third Revision. University of California Sea Grant Program
(information developed by the California Department of Fish and Game). 61 p.

— L. A. Dierauf, V. M. D. California Marine Mammal Center, Fort Cronkhite,
California, 94965. Accepted for publication March 1983.

190 ■ CALIFORNIA FISH AND GAME

EXTREME MERCURY CONCENTRATIONS OF A STRIPED

BASS, MORONE SAXATILIS, WITH A KNOWN
RESIDENCE TIME IN LAHONTAN RESERVOIR, NEVADA

In September 1981, a male striped bass, Morone saxatilis, introduced into
Lahontan Reservoir, Nevada in 1965 was found freshly dead, apparently from
natural causes. It weighed 15.2 kg and was 104 cm fork length. This bass is
unusual for two reasons: it's known residence time and it's tissues contained
extremely high concentrations of mercury. Since the species is a voracious
piscivore at the highest trophic level and the fish was in the reservoir for 1 6 years,
the bioaccumulation of mercury in this fish probably represents maximum val-
ues.

This bass is known to have been stocked in 1965 because about 4,000 striped
bass were seined from the Sacramento-San Joaquin Delta, California, and
stocked into Lahontan Reservoir, during that year. There is no evidence of
subsequent reproduction, stocking was discontinued (Nevada Department of
Wildlife, unpublished data), and scale analysis by the California Department of
Fish and Game's Bay-Delta project indicated it was 16 years old. Because of the
small population size, striped bass are not generally caught by anglers in Lahon-
tan Reservoir.

Lahontan Reservoir, 72 km southeast of Reno, was created in 1915 when the
Carson River was impounded for the Newlands Irrigation Project. The reservoir
has a surface area of 4,410 ha. It receives the entire flow of the lower Carson
River and about 27.5% of the total flow of the Truckee River. In addition to
storage of irrigation water, the reservoir is one of the most heavily used recrea-
tion areas in northern Nevada.

Richins and Risser ( 1 975 ) previously documented that fish in the Carson River
system, including Lahontan Reservoir, contained levels of mercury in muscle
tissue which exceeded concentrations considered safe by the U.S. Food and
Drug Administration (FDA). Mercury was introduced into the Carson River
during the mining of gold and silver which was discovered in 1859 in the
Comstock Lode, near Virginia City. During the 30-yr peak of the Comstock
(1865-1895) it is estimated that 200,000 flasks of mercury or 6.75 X 10^ kg were
lost in the milling process; only 0.5% of the amount was later recovered ( Bailey
and Phoenix 1944).

The objectives of this note are: ( i ) to document the concentrations of mercury
found in the tissues of this striped bass, and ( ii ) to compare its growth to growth
of striped bass from other areas.

Cold vapor atomic absorption was used to determine mercury concentration
in several tissues. Procedures followed Stewart (1977). Approximately lOg of
muscle tissue was taken below the dorsal fin and above the lateral line and
macerated in a tissue grinder. Whole liver and heart tissues were prepared
similarly. The wet tissue was weighed, lyophilized and reweighed to determine
percent moisture. The dried sample was then ground to a powder with a mortar
and pestle and weighed in a teflon parr bomb cup with 2.5 ml concentrated
HNO3, 1.9 ml of 30% H2O2 and 25 mg K2S2O8. The sample was then digested
in a parr bomb for 8 h at 100-110°C.

NOTES 191

Recovery of mercury from fish samples spiked prior to digestion averaged 92.1
percent. Results are presented as the concentration of total mercury per wet
weight of tissue. Cumulative length and incremental growth at each annulus and
corresponding year was estimated by measuring the distance between annuli on
a scale of the striped bass and back-calculating the fish length at each annulus
by direct proportion.

Total mercury concentration in the striped bass muscle tissue was 9.52 mg/kg;
i.e., over nine times the 1 mg/kg tolerance level established by the FDA for
edible fish tissue. Likewise the concentrations in heart and liver tissue were
excessive, 5.58 and 23.65 mg/kg, respectively.

The concentrations of mercury in the Carson River drainage appear to be
extreme in relation to other drainages in the United States. Muscle mercury
concentrations ranged from 0.1 1 to 3.95 mg/kg in 56 fish of 10 species collected
from the reservoir in 1981 (Cooper 1983). Elsewhere in the Carson River system,
concentrations as high as 11.5 mg/kg (carp, Cyprinus carpio) in muscle have
been observed (Ekechukwu 1976). Striped bass from the Sacramento-San Joa-
quin River system, California, have been found with up to 2.0 mg/kg mercury
(California Department of Fish and Game, unpublished data). The highest re-
corded mercury levels that we have been able to find for freshwater fish in the
Western Hemisphere were reported for fish taken from the highly contaminated
Clay Lake on the Wabigoon-English-Winnipeg river system, Canada ( Bishop and
Neary 1976). Walleye, Stizostedion v/treum; p\ke, Esox /ucius; and burbot. Lota
lota, had muscle concentrations of 24.0, 27.8 and 24.8 mg/kg, respectively. A
largemouth bass, Micropterus salmoides, from Calero Reservoir, California had
5.1 mg/kg (Calif. Dept. Public FHealth 1971 ); and fish from Lake St. Clair and St.
Clair River, Michigan had levels ranging from 5.0-6.0 mg/kg (Turney 1971).

The age of 16 years is relatively old, but not uncommon, for striped bass.
Collins (1982) reports the oldest individual from 125,000 scale determinations
to be 19 years, with a tag return indicating that striped bass can live to at least
age 23.

The growth of our bass was slightly greater than the growth which Collins
(1982) estimated for males in the Sacramento-San Joaquin Estuary (Figure 1 ).
This result agrees with reports of many landlocked freshwater populations (Wil-
son and Christenson 1965, Goodson 1966, Allan and Roden 1978) having faster
growth than their marine counterparts (Robinson 1960, Talbot 1966). Therefore,
it does not appear that the high mercury concentration interfered with the
specimen's growth rate.

ACKNOWLEDGMENTS
Nevada Department of Wildlife provided data from stocking records. T. More-
field, California Department of Fish and Game, Stockton, analyzed a sample
scale and measured the annuli. J. Heidker, Desert Research Institute Water
Laboratory, conducted the mercury determinations. The study was funded by
the Office of Water Research and Technology (Contract 14-34-0001-0271).

192

CALIFORNIA FISH AND CAME

UPSTTSAH MIGRANTS

D

DOWNSTREAM MIGRANTS

ID 20 304O50 60 70 80908l0 20X)4O5O60 7D 809o8lO20 304O 50 50 70 809o8lD20 3O4O5O60 70 809O

TOTAL LENGTH (im)

FICURE 1. Back calculated growth of a male striped bass from Lahontan Reservoir, Nevada
( 1 965-1 981 ) compared to mean growth of male striped bass from the Sacramento-San
Joaquin Estuary, California (From Collins 1982).

LITERATURE CITED

Allan, R. C. and D. L. Roden. 1978. Fish of Lake Mead and Lake Mohave. Nevada Department of Wildlife. Biological
Bulletin No. 7. 105 pp.

Bailey, E. H. and D. A. Phoenix. 1944. Quicksilver deposits in Nevada. Univ. of Nev. Bull., 38(5): 12-46.

Bishop, J. N. and B. P. Neary. 1976. Mercury levels in fish from Northwestern Ontario, 1970-1975. Inorganic trace
Contaminants Section, Ministry of the Environment, Rexdale, Ontario: Lab. Services Branch.

California Department of Public Health. 1 971 . Mercury in the California environment. Compiled by the Interagency
Committee on Environmental Mercury. 15 pp -I- Appendices.

Cooper, J. ). 1983. Total mercury in fishes and selected biota in Lahontan Reservoir, Nevada-1981. Bull. Environ.
Contam. Toxicol., 31 (1).

Collins, B. W. 1982. Growth of adult striped bass in the Sacramento-San Joaquin estuary. Calif. Fish Game,
68(3):146-159.

Ekechukwu, G. C. A. 1976. Pharmacokinetics of methyl mercury bioaccumulation in carp, Cyprinus carpio Linna-
eus. Unpublished PhD Dissertation No. 1049. Univ. of Nev., Reno. 156 pp.

Coodson, L. F. 1966. Landlocked striped bass. Pages 407-412. In A. Calhoun (ed.) Inland Fisheries Management.
Calif. Dept. of Fish and Game.

Richins, R. T. and A. C. Risser, 1975. Total mercury in water, sediment, and selected aquatic organisms, Carson
River, Nevada-1972. Pestic. Monit. J., 9(1):44-54.

Robinson, J. B. 1960. The age and growth of striped bass, Roccus saxatilis, in California. Calif. Fish. Game,

46(3):279-290.
Stewart, R. M. 1977. Laboratories section procedures for the characterization of water and wastes. County

Sanitation Districts of Los Angeles County. Pg. 207-215.
Talbot, C. B. 1966. Estuarine environmental requirements and limiting factors for striped bass. A symposium on

estuarine fisheries. Amer. Fish. Soc. Spec. Pub. No. 3. pp. 37-49.
Turney, W. G. 1971. Mercury pollution: Michigan's action program. Jour. Wat. Pol. Cont. Fed. pp. 1427-1438.
Wilson, M. E. and D. P. Christenson. 1965. The Millerton Lake, Fresno/Madera counties striped bass fishery. Calif.

Dept. Fish and Game, Inland FTsh. Admin. Rept. No. 65-3. 15 pp.

—James J. Cooper and Steven Vigg, Biological Sciences, Center, Desert Research
Institute, P.O. Box 60220, Reno, Nevada 89506. Accepted for publication April
1983.

Photoelectronic composition by

CAUFORhOA OFFICE OF STATE PRINTING

78391—800 3-84 3,600 LDA

800007

m

z

e

"~ HI

o>

0)

2 o.

-4

z

z
-1
I

"1

S

a 3»

U)

m r-

-H

•o I—

a>

> "n

m

*

So

"-t

(fl

:^3B

e

>

O

m — -■

so

z

n

z %>

•fl

>

-•■T,

o

z
PI

z

2*.

en

1*1

0
0

>

(D

3=0

n

c

*•
z

CO

e» m

•H

x>

ftl