Bulletin

re) TRY CV OP eat Sy ee eee

q Le SF3S ISSN 0038-3872

SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

BULLETIN

Volume 85 Number 3

BCAS-A85(3) 129-184 (1986) DECEMBER 1986

Southern California Academy of Sciences

Founded 6 November 1891, incorporated 17 May 1907
© Southern California Academy of Sciences, 1986

OFFICERS

Peter L. Haaker, President
Robert G. Zahary, Vice-President
Camm C. Swift, Secretary
Takashi Hoshizaki, Treasurer
Jon E. Keeley, Technical Editor
Gretchen Sibley, Managing Editor

BOARD OF DIRECTORS

1984-1986 1985-1987 1986-1988
Larry G. Allen Hans M. Bozler Daniel M. Cohen
Peter L. Haaker Jules M. Crane Takashi Hoshizaki

Harlan Lewis Susanne Lawrenz-Miller Edward J. Kormondy
Charles E. Oxnard John D. Soule Camm C. Swift
June Lindstedt-Siva Gloria J. Takahashi Robert G. Zahary

Membership is open to scholars in the fields of natural and social sciences, and to any person interested
in the advancement of science. Dues for membership, changes of address, and requests for missing
numbers lost in shipment should be addressed to: Southern California Academy of Sciences, the Natural -
History Museum of Los Angeles County, Exposition Park, Los Angeles, California 90007.

AnnualuMembersi: x cecutaiet 6 2. Neer ees Re ee aan a ae ie gag ee a nr $ 15.00
Student Members)... s0s2 6 0) ces, we pene oes em) me ee) 10.00
Lite,;Members: 2). 30's anes cee BR nd eee ee i ee 300.00 —

Fellows: Elected by the Board of Directors for meritorious services.

The Bulletin is published three times each year by the Academy. Manuscripts for publication should

be sent to the appropriate editor as explained in “Instructions for Authors” on the inside back cover
of each number. All other communications should be addressed to the Southern California Academy

of Sciences in care of the Natural History Museum of Los Angeles County, Exposition Park, Los
Angeles, California 90007.

Date of this issue 19 December 1986

Bull. Southern California Acad. Sci.
85(3), 1986, pp. 129-138
© Southern California Academy of Sciences, 1986

The Mollusk Assemblage Associated with Fronds of Giant Kelp
(Macrocystis pyrifera) off Santa Catalina Island, California

James A. Coyer

Catalina Marine Science Center (University of Southern California)
Ayalon, California

Abstract.—The mollusk assemblage associated with giant kelp (Macrocystis
pyrifera) fronds off Santa Catalina Island, California was examined monthly from
June 1975 through December 1976. Mollusks comprised 1.0, 1.6, and 2.9% (by
number) of all invertebrates associated with kelp fronds in the canopy, middle,
and bottom zones of the kelp forest, respectively. Forty one species (36 gastropods,
5 bivalves) were collected, ranging from 14-28 for any given month. The mean
number of species present was greater in the bottom (16.9) and middle zones
(15.6) than in the canopy (9.3). Most species (29) were rare in occurrence (<1/
kg kelp) and small (<4 mm) in size. Throughout the study, more mollusks were
found in the bottom zone than in the middle; fewest numbers were present in the
canopy. Most species reflected this general pattern; only the nudibranch Polycera
tricolor was more abundant in the canopy than in the lower zones. As a group,
mollusks were most abundant on Macrocystis fronds during summer-fall.

Mollusks are an important group of invertebrates within giant kelp forests
(Macrocystis pyrifera) along the coast of California. Several species are herbivores
and utilize the kelp as food (Jones 1971; Leighton, 1971; Rosenthal et al. 1974),
whereas others serve as a food resource for invertebrates and fishes residing within
the kelp forests (Quast 1968; Feder et al. 1974; Hobson and Chess 1976; Schmitt
1982; Schmitt et al. 1983).

Little is known about the life history and distributions of most mollusks as-
sociated with the fronds of Macrocystis, and no studies have examined the frond-
associated assemblage over extended periods of time. North (1971) and Pearse
and Lowry (1974) listed the mollusks found within kelp forests off southern and
northern California, respectively. Wing and Clendenning (1971) examined the
motile invertebrates associated with clean and encrusted Macrocystis blades at
three different depths during spring and summer, but did not identify individual
species of mollusks. Rosenthal et al. (1974) identified invertebrate species asso-
ciated with southern California kelp forests, but only two species of mollusks were
chosen for intensive quantitative analysis during the seven-year study. During a
six-month study, Miller and Geibel (1973) collected nine species of canopy-as-
sociated mollusks, but the sampling method collected only those individuals larger
than 10 mm in length. Several studies (Lowry et al. 1974; Riedman et al. 1981;
Watanabe 1984) have examined the spatial segregation of Tegula spp. and/or

Present Address: Division of Science and Mathematics, Marymount Palos Verdes College, Rancho
Palos Verdes, California 90274.

129

130 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

@ Canopy
30 O Middle

© Bottom
4 Combined

NUMBER OF SPECIES
fo)

JJAS ON DJ FMA MJ Jv AS ON OD
975 1976

Fig. 1. The number of mollusk species present in each of the vertical zones. Many species are
present in more than one zone. Grand means (X + 95% C.I.): 9.3 + 1.5 (Canopy), 15.6 + 2.0 (Middle),
16.9 + 1.9 (Bottom).

Calliostoma spp. within central California kelp forests, but other mollusks were
not studied.

The present report examines the spatial distribution, abundance, and seasonal
dynamics of all mollusks associated with the fronds of giant kelp at Santa Catalina
Island (California) over a 19-month period. A general overview of the entire
assemblage and a detailed examination of the most abundant species are presented.

Materials and Methods

The study was conducted at Habitat Reef, located in Big Fisherman Cove, Santa
Catalina Island, California, USA (33°28’N, 118°20'W). Habitat Reef is a 100-m
underwater extension of the lava-breccia cliffs which form the south headland of
the cove. Beginning at the cliff base, the reef slopes gently to a depth of 18 m,
then becomes steeper and terminates at a depth of 25 m. The substrate of the
shallow portion (<3 m depth) was covered by a rich algal-invertebrate turf, where-
as the deeper portion (>3 m depth) was dominated by giant kelp and had very
little understory algae.

Monthly samples were collected from giant kelp plants in the central portion
of the Habitat Reef kelp forest (7-9 m depth) during tidal heights from +1.0 to
+1.3m MLLW. I divided the kelp forest at Habitat Reef into three vertical zones:
the canopy (water surface to a depth of 1 m), bottom (just above the holdfasts to
2 m above the substrate), and middle (area between the canopy and bottom).
Holdfasts were not examined. Three replicate samples were collected from each
zone from June through September 1975; five replicates were collected from
October 1975 through December 1976. Only one sample was collected from any

MOLLUSKS OF THE GIANT KELP 131

Table 1. Mean abundance (#/kg kelp) of all species of mollusks within each depth zone. All values
are averaged over the entire 19 month study. Parenthetical values are size ranges in mm; all specimens
were preserved before measuring. Abundances <0O.1 indicate that 1-3 specimens were collected during
the entire study.

Species Canopy Middle Bottom
Gastropoda
Prosobranchia
Archeogastropoda
Tricolia pulloides (Carpenter, 1865) (0.7—1.6) 0.2 3.5 26.1
Tricolia rubrilineata (Strong, 1928) (0.7—1.4) <0.1 <0.1 0.1
Norrisia norrisi (Sowerby, 1838) (4.0—49.0) <0.1 0.2 0.2
Tegula aureotincta (Forbes, 1852) (15.0-—26.0) <0.1 0.1 <0.1
Haliotis sp. (1.5—1.6) 0 0 <0.1
Mesogastropoda
Crepidula sp. (0.7—3.2) 7.1 28.0 79.6
Barleeia californica Bartsch, 1920 (0.6—2.0) 0.2 3.7 10.1
Amphithalamus inclusus Carpenter, 1864 (0.5—1.2) <0.1 1.0 6.0
Amphithalamus tenuis Bartsch, 1911 (1.0-1.2) <0.1 0.9 4.4
Lacuna unifasciata Carpenter, 1857 (0.8—5.4) 0.5 0.6 0.2
Merelina aequisculpta (Keep, 1887) (1.0—1.6) <0.1 0.1 0.1
Rissoina sp. (1.0-1.1) <0.1 <0.1 <0.1
Caecum californicum Dall, 1885 (1.4-2.6) 0 <0.1 <0.1

Fartulum occidentale (Dall, 1885) (1.5—1.6)

0
Cerithiopsis sp. (0.6—3.2) 0
Balcis thersites (Carpenter, 1864) (1.4—1.5) 0 0 <0.1
Bittium sp. (6.6) 0 <0.1 0)
Neogastropoda
Granulina margaritula (Carpenter, 1857) (0.6—-2.1) 4.6 52.6 32.9
Mitrella tuberosa (Carpenter, 1864) (1.4-7.1) 1.1 2.3 1.1
Amphissa sp. (1.0—2.4) 0 0.5 0.8
Mitrella carinata (Hinds, 1844) (6.7—7.2) 0.1 <0.1 0.2
Conus californicus Reeve, 1844 (1.5-1.8) 0 <0.1 <0.1
Opistobranchia
Cephalaspidea
Aglaja inermis (Cooper, 1862) (2.0-5.0) 0.2 0.9 0.3
Nudibranchia
Polycera tricolor Robilliard, 1971, (2.7—8.0) 8.5 5.6 1.1
Melibe leonina (Gould, 1852) (3.4—9.4) 1.1 2.1 DD,
Hermissenda crassicornis (Eschscholtz, 1831)

(2.4—6.4) <0.1 1.3 3.0
Eubranchus rustyus (Marcus, 1961) (1.5—3.4) 0.1 3.4 1.1
Corambe pacifica MacFarland and O’Donoghue,

1929 (1.0-2.2) <0.1 0.3 <0.1
Antiopella barbarensis (Cooper, 1863) (3.6—11.4) <0.1 0.1 0.2
Dendronotus frondosus (Ascanius, 1774) (1.9-9.6) <0.1 0.1 0.2
Flabellinopsis iodinea (Cooper, 1862) (2.8-6.2) 0 0.1 <0.1
Coryphella pricei MacFarland, 1966 (5.0-6.0) 0 <0.1 0

Pyramidella
Odostomia sp. A. (0.5—1.8) 0.1 1.7 1.8
O. navisa Dall & Bartsch, 1909 (1.7—1.8) 0 <0.1 <0.1
O. virginalis Dall & Bartsch, 1903 (0.6—-1.8) 0 0 <0.1

O. sp. B. (2.3) <0.1 0 0

132; SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

Table 1. Continued.

Species Canopy Middle Bottom
Bivalvia
Pteriomorpha
Arcoida
Philobrya setosa (Carpenter, 1864) (1.3—2.3) 0 <0.1 0.1
Pterioida
Lima sp. (1.1-1.2) 0 0 <0.1
Leptopecten sp. (2.0-9.2) 0 0 <0.1
Heterodonta
Myoida
Hiatella artica (Linnaeus, 1767) (0.8-2.3) <0.1 0.8 1.0
Veneroida
Halodakra brunnea (Dall, 1916)( (0.6—1.8) <0.1 0.6 3.0

plant. Fronds were severed at zone junctions; the middle and bottom zones were
collected by carefully severing the upper zones and allowing them to drift away.
Disturbance to the desired zones during this procedure was negligible. Similar
amounts of kelp were collected from each zone throughout the study (kg kelp =
2.5, Canopy; 2.1, Middle; 2.3, Bottom; all n = 19).

Scuba divers collected the kelp-associated mollusks (and other invertebrates)
by maneuvering a small plankton net (1 m diam, 3 m length, 0.33 mm mesh) over
the desired portion of the plant, severing the fronds, and closing the net. The
enclosed kelp was then placed into a large container of warm fresh water (providing
a thermal and salinity shock), vigorously agitated, and removed piece by piece.
The remaining water was filtered through a 0.25 mm sieve and the residue pre-
served. A repeat of the agitation-freshwater method four hours later indicated
that 94.1, 98.9, and 98.8% (by number) of all mollusks in the canopy, middle,
and bottom, respectively, were removed by the initial treatment.

Mollusks were identified (using McLean 1969) and measured using a dissecting
microscope with an ocular micrometer. The wet-weight of kelp from each sample
was measured, and abundances of all mollusks were expressed as the number of
individuals per kg (wet-weight) of kelp. All samples are deposited at the Catalina
Marine Science Center Museum.

Results

When averaged over the entire 19 month study, mollusks accounted for 1.0,
1.6, and 2.9% (by number) of all invertebrates in the canopy, middle, and bottom,
respectively (# mollusks/kg kelp + 95% C.I.: Canopy, 25.7 + 14.3; Middle, 111.3 +
21.7; Bottom, 176.5 + 47.0). Arthropods were dominant in each of the three
zones, accounting for 97% of all invertebrates in the canopy, 94% in the middle,
and 95% in the bottom (see Coyer, 1984). Forty-one species of mollusks (36
gastropods, 5 bivalves) were collected from all zones during the study, ranging
from 14—28 for any given month (Fig. 1, Table 1). When ranked by mean monthly
abundance (mean of all zones), only Crepidula sp. (unidentified juveniles) and
Granulina margaritula were common (10—100/kg kelp), whereas 10 species were
uncommon (1—10/kg kelp), and 29 were rare (<1/kg kelp).

MOLLUSKS OF THE GIANT KELP 133

Throughout the study, a greater number of species and individuals were present
in each of the lower zones than in the canopy. The mean number of species
(averaged over the 19 month study) in the bottom (16.9) and middle (15.6) were
similar, but greater than the mean number present in the canopy (9.3; Fig. 1).
The nudibranch, Polycera tricolor, was the most abundant species in the canopy,
largely because of high numbers in October 1975, and it was the only species
more abundant in the canopy than in the lower zones (Fig. 2, Table 1). For the
remaining species of gastropods, 11 were more abundant in the middle, 18 more
abundant in the bottom, and 6 were present equally in both zones (Table 1). Each
of the five bivalves collected were absent or rare in the upper zones; three were
rare and two were uncommon in the bottom.

With respect to numbers of individuals, more mollusks (x + 95% C.I.) were
present in the bottom zone (176.2 + 47.0) than in the middle (111.3 + 21.7; Fig.
3). Fewest numbers were found in the canopy (25.7 + 14.3). The general pattern
occurred throughout the study except for five months when the number of mol-
lusks present in the middle was slightly higher than in the bottom.

Crepidula sp. and G. margaritula were the most abundant species during the
study, collectively accounting for 45.5% (by number) of the mollusks in the canopy,
72.4% of those in the middle, and 63.7% of all mollusks in the bottom (Table 1).
Five species comprised 87, 84, and 88% (by number) of all mollusks in the canopy,
middle, and bottom, respectively (Table 2). With regard to overall occurrence,
each of the five canopy species were uncommon (1-10/kg kelp); two and four
species were common (10—100/kg kelp) in the middle and bottom, respectively.

As a group, mollusks were more abundant on Macrocystis fronds during sum-
mer-fall than during winter-spring (Fig. 3). Annual summer-fall peaks of abun-
dance were suggested for Barleeia californica and Tricolia pulloides, although the
latter was much more abundant in 1975 than in 1976 (Fig. 2). Crepidula sp. was
most abundant during spring-summer 1975 then declined in numbers, whereas
G. margaritula was present in low numbers during spring-summer 1975 then
increased to maximum abundance during summer-fall 1976. No temporal pat-
terns were evident for P. tricolor and abundances of all other mollusks were too
low to detect any seasonal trends.

Seventy-seven percent of the shelled species (Table 1) and 78-88% of shelled
individuals in each zone (Table 2) were less than 4 mm in total length. The two
most abundant species, Crepidula sp. and G. margaritula, never exceeded 3.2 and
2.1 mm, respectively. Tegula aureotincta (15-26 mm) and Norrisia norrisi (4-49
mm) were by far the largest species, but both rarely were encountered.

Discussion

Several invertebrate species display patterns of horizontal and/or vertical dis-
tribution within central and southern California kelp forests, including bryozoans
(Bernstein and Jung 1979), crustaceans (Hines 1982; Coyer 1984), and mollusks
(Lowry et al. 1974; Riedman et al. 1981; Watanabe 1984). Frond-associated
mollusks at Habitat Reef also revealed clear and consistent patterns of vertical
zonation, with far more species and individuals found in the lower zones than in
the canopy. Only the nudibranch, P. tricolor, was more abundant in the upper
zones than in the bottom.

Several factors may account for the patterns of vertical zonation observed at

134 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

J

Grepidula sp Canopy
S005 1 Middle
FE] Bottom
200;
100
re wy Mis g afl WAZ mts > ~ = =
a 200
uJ
ne A
@ 100 i
= (3 pe : | ;
SS ee toh Ft I LE IE nee i Bt Bu Gusts
ry
200
= r
[OOF P A
1 ae
Six =e Sz _
100; B casfornica
ee aS ee es = = me =
ek P tricolor
L
100 y |
Pie, Vth. — AL ro reo
J J A S O N D) J F M A J J S 0) N D
1975 1976

Fig. 2. Monthly abundances (¥ = 95% C.I.) of mollusks (all species) in each vertical zone. Each
value represents the mean of 3 (July-September 1975) or 5 (October 1975—December 1976) replicate
samples.

Habitat Reef. The most important factor may be water motion generated by storms
and/or tidal currents, which undoubtedly dislodges many mollusks from the kelp
fronds. As the degree of motion and removal probably is greatest in the canopy
and as all dislodged mollusks must crawl back onto the kelp via the bottom zone,
the observed patterns of vertical zonation could be explained most often by
differential water motion/dislodgement with subsequent concentration in the low-
er zones.

Other factors may be important, however, especially during extended periods
of relatively calm conditions. For example, mollusks may avoid the canopy during
calm summer months at Habitat Reef because of warmer surface temperatures
and/or greater light intensity.

Additionally, predators may influence the vertical distribution of mollusks at
Habitat Reef, although their effect is dificult to assess. Dense alga cover has been
demonstrated to decrease the effectiveness of sea star predation (Watanabe 1984)
as well as predation by fishes (Vince et al. 1976; Heck and Thoman 1981; Peterson
1982: Stoner 1982). Consequently, one could predict that most mollusks would
be present in the bottom and canopy regions at Habitat Reef, because kelp biomass
consists of a dense and “bushy” cluster of sporophytes in the bottom, a dense
mass of interdigitating fronds and blades in the canopy, and a relatively sparse
mass of fronds and blades in the middle. Furthermore, mollusks can avoid benthic
predators by climbing up the kelp fronds where fewer predators are present (Har-
rold 1982; Schmitt et al. 1983; Watanabe, 1984). Ifreduction of predation pressure

MOLLUSKS OF THE GIANT KELP 135

Canopy
( Middle

NUMBER / Kg KELP

Fig. 3. Monthly abundances (x + 95% C.I.) of the mollusks, Crepidula sp., Granulina margaritula,
Tricolia pulloides, Barleeia californica, and Polycera tricolor in each vertical zone. Each value rep-
resents the mean of 3 (July-September 1975) or 5 (October 1975—December 1976) replicate samples.

is an important factor in determining spatial abundance of mollusks at Habitat
Reef, the number of mollusks in the canopy should be similar to numbers found
in the bottom. The consistently low numbers of mollusks in the canopy, however,
suggests that decreased predation pressure probably does not influence the number
of mollusks in the canopy. The importance of predation in the bottom zone
remains to be determined.

The majority of shelled mollusks (species and individuals) associated with
fronds of giant kelp at Habitat Reef are less than 4 mm in length with only two
species exceeding 10 mm. In contrast, several species of shelled gastropods as-
sociated with fronds of giant kelp in central California (Monterey) are larger than
10 mm in length (Miller and Geibel 1973; Lowry et al. 1974; Riedman et al.
1981; Watanabe 1984) and the number of individual gastropods greater than 10
mm can be substantial, especially in summer (Miller and Geibel 1973; Watanabe
1984). Although none of the larger species found at Monterey were present at
Habitat Reef and all are rarely found off Santa Catalina Island (Engle, unpub.
data), the difference in mollusk sizes between Monterey and Habitat Reef is
intriguing and merits further investigation.

The total amount of kelp biomass is a potentially important mechanism gov-
erning the numbers of mollusks within a kelp forest. More kelp-associated mol-
lusks (species and individuals) might be expected in a dense, rather than a sparse
forest, and this pattern was observed at Habitat Reef. Kelp density was high and
relatively constant (Coyer 1984) during most of the study (June 1975 through
August 1976) when mollusk density also was highest. Beginning in June 1976, an
anomalous warming trend associated with the El Nino of 1976-77 (for discussion
of the El Nino, see Smith 1983; Rasmusson and Wallace 1983) resulted in a
dramatic reduction of kelp density by September 1976 and the near disappearance
of the canopy by December 1976 (Coyer 1984). Fewer mollusks were present in
the much reduced canopy during the latter portion of 1976.

Without experimental data, however, it is difficult to attribute the decline in
mollusk abundance solely to the decline in kelp biomass, as oceanographic con-
ditions undoubtedly are a confounding factor. For example, the onset of warm
water in mid-1976 may have reduced local larval settlement in those species
exhibiting lower abundances in summer-fall 1976 than in 1975 (Crepidula, T.
pulloides, B. californica), thereby influencing population sizes before the amount
of kelp became critical. The two most abundant mollusks during the study, Cre-

136 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

Table 2. The five most abundant mollusks in each zone. Abundances are mean monthly values
for the 19 month study.

2 kg/ 2 kg/ # kg/

Canopy kelp Middle kelp Bottom kelp
Polycera tricolor 8.5 Granulina margaritula 52.6 Crepidulasp. 79.6
Crepidula sp. fel Crepidula sp. 28.0 Granulina margaritula 32.9
Granulina margaritula 4.6 Polycera tricolor 5.6 Tricolia pulloides 26.1
Mitrella tuberosa 1.1 Barleeia californica 3.7 Barleeia californica 10.1
Melibe leonina ei Tricolia pulloides 3.5 Amphithalamus inclusus 6.0

pidula sp. and G. margaritula, displayed non-overlapping peaks of abundance
with the former most abundant during spring—summer 1975 and the latter during
summer-fall 1976. The non-overlapping patterns may represent differential re-
cruitment as a result of changing oceanographic conditions.

Little is known about the feeding biology of the most abundant mollusks as-
sociated with fronds of giant kelp at Habitat Reef, except for Melibe leonina which
feeds on micro-crustaceans (Ajeska and Nybakken 1976) and T. pulloides which
grazes diatoms from algal surfaces (Mooers 1982). Information from studies on
congenerics, however, suggests that feeding is diverse and includes bryozoan pred-
ators (Polycera), micro-carnivore/detrital feeders (Mitrella), and filter feeders
(Crepidula) (McDonald and Nybakken 1978; Abbott and Haderlie 1980).

Most of the mollusks associated with kelp fronds at Habitat Reef are at least
present, and many are more abundant, in intertidal and subtidal non-kelp habitats
along the California coast and all of the shelled gastropods are found on gravel
bottoms, as well (McLean 1969: Carlton and Roth 1975: Engle 1979: Abbott and
Haderlie 1980; Haderlie and Abbott 1980; Stewart and Myers 1980). Some genera
such as Haliotis, Crepidula, Amphissa, and Lima are found on kelp only as
juveniles (McLean, pers. comm.). In addition, several species reported to be
occasionally or seasonally abundant in other giant kelp forests along the southern
and central coast of California such as Lacuna unifasciata, Mitrella carinata, N.
norrisi, T. aureocinta, Leptopecten sp. and M. leonina (Limbaugh 1955; Jones
1971; Pearse and Lowry 1974; Ajeska and Nybakken 1976; Bernstein and Jung
1979) were present in low numbers at Habitat Reef throughout the 19-month
study. Such patterns suggest that local events and conditions strongly influence
composition and dynamics of the mollusk assemblage associated with fronds of
giant kelp. Additional non-kelp and giant kelp habitats need to be examined for
extended periods of time before general predictions can be proposed.

Acknowledgments

The manuscript was adopted from a portion ofa doctoral dissertation completed
at the University of Southern California. I thank my committee, chaired by J. N.
Kremer, and am grateful to R. L. Zimmer and R. R. Given for their support and
cooperation at the Catalina Marine Science Center. Special thanks to J. R. Chess,
J. F. Pilger, C. S. Shoemaker, and T. E. Audesirk for substantial field assistance
and to D. Cadien, J. H. McLean, and J. R. Chess for assistence with species
identification. I also thank R. R. Ambrose, J. H. McLean, and R. J. Schmitt for
comments on an earlier draft.

MOLLUSKS OF THE GIANT KELP 137

The research was supported in part by the NOAA Office of Sea Grant under
Grant No. USDC 04-158-44881 to the University of Southern California and by
Sea Grant Traineeships. This is Contribution No. 83 from the Catalina Marine
Science Center.

Literature Cited

Abbott, D. P., and E. C. Haderlie. 1980. Prosobranchia: Marine snails. Pp. 230-307 in Intertidal
invertebrates of California. (R. H. Morris, D. P. Abbott, and E. C. Haderlie, eds.), Stanford
Univ. Press, xiv + 690 pp.

Ajeska, R. A., and J. Nybakken. 1976. Contributions to the biology of Melibe leonina (Gould, 1852)
(Molluska: Opistobranchia). The Veliger, 19:19-26.

Bernstein, B. B., and N. Jung. 1979. Selective pressures and coevolution in a kelp canopy community
in southern California. Ecol. Monogr., 49:335-355.

Carlton, J. T., and B. Roth. 1975. Phylum Molluska: shelled gastropods. Pp. 467-514 in Lights
manual, intertidal invertebrates of the central California coast, 3rd Ed. (R. I. Smith and J. T.
Carlton, eds.), Univ. California Press, xvii + 716 pp.

Coyer, J. A. 1984. The invertebrate assemblage associated with the giant kelp Macrocystis pyrifera,
at Santa Catalina Island, California: a general description with emphasis on amphipods, co-
pepods, mysids, and shrimps. Fish. Bull., U.S., 82:55-66.

Engle, J. M. 1979. Ecology and growth of juvenile California spiny lobster, Panulirus interruptus
(Randall). Ph.D. Thesis, Univ. Southern Calif., 289 pp.

Feder, H. M., C. H. Turner, and C. Limbaugh. 1974. Observations of fishes associated with kelp
beds in southern California. Calif. Dept. Fish and Game, Fish Bull., 160:1-144.

Haderlie, E. C., and D. P. Abbott. 1980. Bivalvia: the clams and allies. Pp. 355-411 in Intertidal
invertebrates of California. (R. H. Morris, D. P. Abbott, and E. C. Haderlie, eds.), Stanford
Univ. Press, xiv + 690 pp.

Harrold, C. 1982. Escape responses and prey availability in a kelp forest predator-prey system. Amer.
Nat., 119:132-135.

Heck, K. L., Jr.. and T. A. Thoman. 1981. Experiments on predator-prey interactions in vegetated
aquatic habitats. J. Exp. Mar. Biol. Ecol., 53:125-134.

Hines, A.H. 1982. Coexistence in a kelp forest: size, population dynamics, and resource partitioning
in a guild of spider crabs (Brachyura, Majidae). Ecol. Monogr., 52:179-198.

Hobson, E. S., and J. R. Chess. 1976. Trophic interactions among fishes and zooplankters near shore
at Santa Catalina Island, California. Fish. Bull., U.S., 74:567-598.

Jones, L.G. 1971. Studies on selected small herbivorous invertebrates inhabiting Macrocystis can-
opies and holdfasts in southern California kelp beds. Pp. 343-367 in The biology of giant kelp
beds (Macrocystis) in California. (W. J. North, ed.), Beih. Nova Hedwigia 32, xv + 600 pp.

Leighton, D. L. 1971. Grazing activities of benthic invertebrates in southern California kelp beds.
Pp. 421-453 in The biology of giant kelp beds (Macrocystis) in California. (W. J. North, ed.),
Beih. Nova Hedwigia 32, xv + 600 pp.

Limbaugh, C. 1955. Fish life in the kelp beds and the effects of harvesting. Univ. Calif. Inst. Mar.
Res., IMR Ref. 55-9.

Lowry, L. F., A. J. McElroy, and J. S. Pearse. 1974. The distribution of six species of gastropod
mollusks in a California kelp forest. Biol. Bull., 147:386-396.

McDonald, G. W., and J. W. Nybakken. 1978. Additional notes on the food of some California
nudibranchs with a summary of known food habits of California species. The Veliger 21:110-
119.

McLean, J. H. 1969. Marine shells of southern California. Los Angeles County Mus. Nat. Hist.
Science Series, 24:1-104.

Miller, D. J., and J. J. Geibel. 1973. Summary of blue rockfish and lingcod life histories: a reef
ecology study; and giant kelp, Macrocystis pyrifera, experiments in Monterey Bay, California.
Calif. Dep. Fish Game, Fish Bull., 158:1—137.

Mooers, M. G. 1982. Diet and reproductive biology of the rocky intertidal prosobranch gastropod
Tricolia pulloides. The Veliger 24:103-108.

North, W. J. 1971. Introduction and background. Pp. 1-96 in The biology of giant kelp beds
(Macrocystis) in California. (W. J. North, ed.), Beih. Nova Hedwigia 32, xv + 600 pp.

138 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

Pearse, J. S., and L. F. Lowry. 1974. An annotated species list of the benthic algae and invertebrates
in the kelp forest community at Point Cabrillo, Pacific Grove, California. Coast. Mar. Lab.,
Univ. Calif., Santa Cruz, Tech. Reprt., 1:1-73.

Peterson, C. H. 1982. Clam predation by whelks (Busycon sp.): experimental tests of the importance
of the prey size density, and seagrass cover. Mar. Biol., 66:159-170.

Quast, J. C. 1968. Observations on the food of the kelp-bed fishes. Pp. 109-143 in Utilization
of kelp-bed resources in southern California. (W. J. North and C. L. Hubbs, eds.), Calif. Dept.
Fish and Game, Fish Bull., 139.

Rasmusson, E. M., and J. M. Wallace. 1983. Meteorological aspects of the El Nino/Southern Os-
cillation. Science, 222:1195-—1202.

Riedman, M., A. H. Hines, and J.S. Pearse. 1981. Spatial segregation of four species of turban snails
(Gastropoda: Tegula) in Central California. The Veliger, 24:97—102.
Rosenthal, R. J., W. D. Clarke, and P. K. Dayton. 1974. Ecology and natural history of a stand of
giant kelp, Macrocystis pyrifera off Del Mar, California. Fish. Bull., U.S., 72:670-684.
Schmitt, R. J. 1982. Consequences of dissimilar defenses against predation in a subtidal marine
community. Ecology, 63:1588-1601.

———,, S. W. Osenberg, and M. G. Bercovitch. 1983. Mechanisms and consequences of shell fouling
in the kelp snail, Norrisia norrisi (Sowerby) (Trochidae): indirect effects of octopus drilling. J.
Exp. Mar. Biol. Ecol., 69:267—282.

Smith, R. L. 1983. Peru coastal currents during El Nino: 1976 and 1982. Science, 221:1397-1399.

Stewart, J. G., and B. Myers. 1980. Assemblages of algae and invertebrates in southern California
Phyllospadix-dominated intertidal habitats. Aquatic Botany, 9:73—94.

Stoner, A. 1982. The influence of benthic macrophytes on the foraging behavior of pinfish, Lagodon
rhomboides (Linnaeus). J. Exp. Mar. Biol. Ecol., 58:271-284.

Vince, S., I. Valiela, N. Backus, and J. M. Teal. 1976. Predation by the salt marsh killifish Fundulus
heteroclitus (L.) in relation to prey size and habitat structure: consequences for prey distribution
and abundance. J. Exp. Mar. Biol. Ecol., 23:255-266.

Watanabe, J. M. 1984. The influence of recruitment, competition, and benthic predators on spatial
distributions of three species of kelp forest gastropods (Trochidae:Tegula). Ecology, 65:920-
936.

Wing, B. L., and K. A. Clendenning. 1971. Kelp surfaces and associated invertebrates. Pp. 319-339
in The biology of giant kelp beds (Macrocystis) in California. (W. J. North, ed.), Beih. Nova
Hedwigia 32, xv + 600 pp.

Accepted for publication 27 May 1985.

Bull. Southern California Acad. Sci.
85(3), 1986, pp. 139-151
© Southern California Academy of Sciences, 1986

Two New Tiger Beetles of the Genus Cicindela from Western
United States (Cicindelidae: Coleoptera)

Norman L. Rumpp
P.O. Box 60178, Las Vegas, Nevada 87160

Abstract.—Two new subspecies of Cicindela are described: C. scutellaris yam-
pae of northwestern Colorado, and Cicindela formosa rutilovirescens of eastcentral
New Mexico. Both subspecies are found at the periphery of the ranges of their
respective species. This study traces their migration into their present locations
in an attempt to define the forces that cause changes leading to subspeciation. C.
scutellaris is retained as a species group because of the singularity of parts within
the internal sac of the aedeagus in the male genitalia. C. formosa is significantly
different in both male and female genitalia from other members of Rivalier’s
Group VII so that it constitutes a separate group of its own. Rivalier’s Group VII
should be renamed the Purpurea Group.

The two subspecies described in this paper belong to the tiger beetle species
Cicindela scutellaris and C. formosa. They represent populations located at the
limit of the known ranges of these species where contact with other populations
has not happened for a long time. Thomas Say described C. formosa (1817:19)
and C. scutellaris (1823:40) from midwestern United States populations that were
found sympatrically in sand blowouts, or, as Say called these, “‘sand alluvions.”’
Both species have extensive ranges sufficiently varied to have resulted in subspe-
ciation. It can be generalized that wherever C. formosa is found C. scutellaris will
occur, but the reverse is not always so because C. scutellaris has extended its
range into southeastern United States by invading sandy woodlands.

Key to the Subspecies of Cicindela scutellaris

1. Bicolored: Head and pronotum bright green, blue-green or blue: elytra

from a bright red (sometimes alutaceous), to a darkerred............. 2
— Unicolored, where head, pronotum and elytra are the same background

Colomysometimes+ withivariations sii). eel One aan meee Ge 3
Pane lytrar immaculateys vem eel) tnd ee ae iehiot la actin Ueaa te ea scutellaris

— Elytra with a continuous white maculation at outer edge, with a confused
area at middle band position; rarely reduced to broad spots at outer edge

wip Buh oS sicyh tate Geb: 2 leh GAT aE aI PPP RR es ta leg eRe eon Per yampae
3. Color bright to dark red, sometimes suffused with green .............. 4
-— Color bright green to deep violet-blue. Variable in size depending on
SUS PSCIE SMA, NERO ee eee ed BS rate NOL ae oy eee AHMAR Pel cae un Sie acti wine 6
4. Maculation consisting of white dots on outer edge of elytra ....... lecontei
maculation connected at outer edge of elytra ........................ 5)
5. Broad maculation at outer edge of elytra, with a non-confused broadened
MATCH ATA Pec cs Hae ie gee a nee seem ah hPa ce Se ge ae inte ae criddlei

140 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

— Maculation reduced to outer edge dots, infrequently narrowly connected

II ee te bok Soir. ea hes SoS ya rugifrons
6. Smaller (10-11 mm long). Dark blue, blue-green or green, sometimes with

ADIGA OTS MERA Py ee ke eR cs os Bands dane Rs se ee unicolor
= Larger (ales = 12 symm Blog) i iss Ge thew nes (base. eee 7
7; Color purplishvomolucsenreen)..,2.. 04.5. /.. 052 2) a eee rugata
= ColoryelloweoneemQarewwents ot cle e 6. 7 salt. ats Be ee flavoviridis

Cicindela scutellaris yampae, new subspecies
Figure | (1)

Holotype. — Male, in the collection of the American Museum of Natural History
(AMNH), New York, NY: collected by Mont A. Cazier, September 1953.

Locality. —Maybell Sand Hills which lie between two and 6 km east of Maybell,
Moffat County, Colorado. All specimens of the type series were captured in this
region. The first 51 specimens by M. A. Cazier in 1953. The remaining specimens
of this series were collected between June 1959 and April 1983.

Type series.—There are 370 specimens in the series, distributed as follows:
Holotype and 26 paratypes in the collection of the AMNH, New York, NY;
allotype female and 80 paratypes (41 6, 39 2) in my collection designated the
NLRC by Arnett et al. (1969:28). The remainder are in the following collections:
two at the National Museum of Natural History (NMNH), Washington, DC; two
at the California Academy of Sciences (CAS), San Francisco, CA; Walter Johnson,
Minneapolis, MN, 13 specimens; David W. Brzoska, Lawrence, KS, 149 speci-
mens (88 6, 61 2); Howard P. Boyd, Vincentown, NJ, 37 specimens (21 4, 16 9).

Description. —Medium size, robust; head and pronotum bluish-green, elytra red
with golden birefringence, and broad band of maculation along outer edge which
broadens out into confused pattern where middle band would enter normally.

Head: Color mostly bright bluish-green, more green at vertex and near eyes,
interocular space nearly blue; only slightly wrinkled at vertex, more so on frons;
interocular striae not deep; frons punctured and bearing many decumbent setae;
setae between eyes nearly erect but spaced in growth of medium density; clypeus
green, blue toward the outer edges, bare; labrum white, outer edges dark, six long
white setae near outer edge, tridentate and extended forward but less than half as
long as wide; penultimate labial segment longest, heavily setose and metallic blue
in color, terminal segment similarly pigmented but bare; segments of labral palpi
metallic blue, second segment longest and heavily setose, outer two bare except
for a few setae near anterior surface of penultimate segment; first four segments
of antennae green, first large and covered with many punctures bearing long white
erect setae on anterior surface only, length of fourth segment seven-eights that of
third.

Thorax: pronotum narrower than head across eyes, wider than long by more
than one-fifth; color green, disc golden, margins bluer, wrinkled; median sulcus
blue and shallow, crossed with transverse impressions; anterior sulcus blue, deep,
micro-reticulated anteriorly only, not wrinkled; posterior sulci also blue and deep;
longitudinal row of white, erect setae two-thirds distance from centerline, another
row of setae at epipleural edge, recumbent toward center, but distinct from very
long sub-epipleurual setae.

TIGER BEETLES 141

10 mm

Fig. 1. Dorsal views. 1. Cicindela scutellaris yampae, 2. C. formosa rutilovirescens.

Elytra: Broad, widest posterior to middle; micro-reticulate and covered with
very small impressions, surface transparent in appearance with birefringence of
golden green on disc against basic red color; disc with row of very small setigerous
punctures near suture; another short row of punctures located in humeral impres-
sions that delimit disc; maculation continuous at outer edge of elytra but lunules
and middle band traceable, pattern as in Figure 1 (1); pigment at discal edge of
maculation deeper red, almost purple in small impressions.

Venter: Bright greenish-blue throughout; genae with longitudinal stria, glabrous;
epimera and coxae with numerous long setae; femora and tibia greenish blue, with
long white setae, tibiae more densely covered with setae, tarsal claws nearly as
long as last tarsal segment; penial notch present as a slight declivity on penultimate
sternite.

Comment.—Females are similar to males, both in size and proportion, but
differ in that the frons is nearly glabrous.

Dimensions. —Holotype, male—length 11.8 mm, width 4.9 mm. Allotype, fe-
male—length 12.0 mm, width 5.1 mm.

Etymology. —Named for the Yampa River Valley of Utah.

Population variables. —In a population sample of 56 specimens studied in more
detail, the most striking variations are in color and maculation.

Color: The head and pronotum are green in 51 specimens, blue in four, and
golden green in one. Elytra are red, varying between dark red and golden red,
superimposed with varying amounts of green suffusion—the darker the red, the
less the green suffusion.

Maculation: In 46 specimens the maculation is connected along the entire
margin of the elytra. In four specimens the humeral lunule, or dot, is disconnected
from the marginal band. In seven specimens the maculation is reduced to marginal
dots, and in two cases, the humeral dots are lacking. In all samples the middle
band consists of an expansion from the outer edge toward the center of the elytra,
with a confused area at the basal edge in the majority of specimens. In 27 specimens
this confused area expands considerably anteriorly toward the base of the elytra

142 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

such that these specimens are more heavily maculated than the rest. Of the more
heavily maculated specimens 23 have epicentral spots which, in some cases,
consist of a narrow vertical line bordered by a purplish-red pigmentation; in two
cases the white disappears, but the area is marked by the deeper red color.

Conclusions Based on Faunal Distribution

1. During the glacial periods of the Pleistocene, a broad region of southcentral
Wyoming became a refugium for a number of Canadian species of Cicindela. The
species most sensitive to change developed noticeable characteristics in response
to climate change, and possibly through some hybridization. This seems to have
been the situation with C. scutellaris.

2. A region of northern Montana, centered around Great Falls, was an isolated
haven for species from Alberta as they migrated southward ahead of glacial ad-
vances. Here, the environmental impact on C. scutellaris was slight, thereby
inducing only slight color change. Thus, the climate in this region was hardly
more severe than today’s climate in central Alberta.

3. Isolated populations of C. scutellaris, C. formosa and C. limbata may still
exist in the Bighorn and Powder River Basins. The discovery of such populations
would aid in tracing the migratory routes of these species.

Key to the Subspecies of Cicindela formosa

L:. Elytra. broadly maculated@. 2222s. ee ee 2
= Elytra immaculate or narrowly, maculated’.” >... 5. 2.5.) ee eee =
2. Broad white maculation of elytra consisting of humeral and apical lunules,

and a middle band; all narrowly connected at edge ................... 3
— Maculation coalesced into total coverage of elytra ............... gibsoni
3: Bright red in color. Midwestcmniorm) 90. 2a. eee formosa
= _ Dark brown in color; Eastemaormi -.. 5,-2.4... eee generosa
4. Elytra with apical lunule white, often with other lunules pigmented, or

lacking. Color dark red with upper half of elytra a deep purple ........

ge ae ta ene ener AS LAr ckie ht Bete ed Nl aos a 2 pigmentosignata
— Elytra with either narrow apical lunule only or none at all. Color bright

red ‘with ereen’ birciningence: 2 ee eee rutilovirescens

Cicindela formosa rutilovirescens, new subspecies
Figure 1 (2)

Holotype. —Male, no. 12984 in the collection of the CAS, San Francisco, CA,
captured by Mont A. Cazier, October 1953.

Locality.—The type location is in the Mescalero Sands, 55 to 65 km due east
of Roswell, New Mexico, in the eastern part of Chaves County, at a mean elevation
of 1300 m. The sands trend from southwest to northeast for nearly 50 km, their
width varying from 6 to 14 km. The area is made up of low hills, 2 to 3 m high,
covered with a short scrub oak (Quercus harvardi) which partially stabilizes the
sand, although in a few places bare dunes rise to heights not exceeding 8 m. The
ground sand is light in color, somehat darker where vegetation growth contami-
nates it with debris. The innumerable blowouts in the troughs of the oak-covered
area are bare.

TIGER BEETLES 143

Type series. —There are 185 specimens from the type area, distributed as fol-
lows: Holotype male and three paratypes at the CAS; allotype female and 28
paratypes in the NLRC, collected in October 1953 by M. A. Cazier, in September
1957 by E. Tinkham, and at various times by me from 1960 to 1980. Paratypes
are in the private collections of the following individuals: W. D. Sumlin, 12 (3 4,
9 2); E. V. Gage, 3 (2 6, 1 2); G. Gaumer, 30 captured in May 1971; D. Brzoska
who collected in May and September from 1978 to 1982, 56 (35 6, 21 9); J.
Stamatov, 27, and H. Boyd, 23 specimens which they collected in September
1976.

There are 24 paratypes from locations other than the type area which are
distributed as follows: One female from Terry County, TX collected by D. Kelly
deposited in the collection of Walter Johnson; eight specimens collected east of
Plains, Yoakum County, TX, two of which are in the NLRC; one from SSE of
Loco Hills, Eddy County, NM by G. Gaumer in September 1971; three from
Portales, Roosevelt County. NM in the East New Mexico University collection;
one male in the NLRC collected by J. Sheppard in May 1974 south of Milnesand,
Roosevelt County, NM.

Description. —Large in size, form robust, upper surface red, birefringent with
green, with only slight indication of maculation at apex of elytra.

Head: Bright golden red vertex finely wrinkled, interocular striae brightly col-
ored red with golden tinges; frons wrinkled and punctate, greenish at edges, bright
deep blue toward antennae, punctures with long white setae, nearly recumbent
anteriorly, anteriormost setae lying toward vertex, middle setae lying toward sides,
and sparse setae between eyes lying generally toward middle; clypeus bright blue,
violaceous toward sides; first four segments of antennae green, first segment with
three or four setae from middle to apex on outer edge, four or five setae around
apex and near base of second segment, with third segment nearly one-third longer
than fourth; labrum tridentate, extended forward, one half as long as wide, whitish
with outer edge dark, six setigerous punctures near outer edge; long heavily setose
penultimate labial segment nearly 1.5 times as long as glabrous last segment, both
metallic blue; second labral palpal segment unpigmented with white setae on basal
half, 2.1 times longer than penultimate segment and 1.3 times longer than last
segment, both of the latter metallic green.

Thorax: Pronotum nearly as broad as head, ratio of width to length = 1.15 and
broadest just basal to anterior transverse impression, tapered to rounded posterior
angle; surface finely rugose; median sulcus of disc shallow, anterior and posterior
sulci deep; color red with golden tinge, more so laterally and at bottom of impres-
sions; two rows of recumbent setae, one near lateral edge, other just above epi-
pleurae; venter bright violaceous blue; all three thoracic sterna and coxae with
medium growth of long white setae.

Elytra: Wider at shoulder than head, tapered slightly wider for two-thirds, then
broadly rounded to apex; entire elytral surface micro-reticulate, impressed
throughout with minute shallow pits, each with slightly raised anterior edge; one
row of very small setigerous punctures near suture, not readily seen except under
magnification, entire surface with birefringent transparency such that surface ap-
pears red when viewed from above, but when viewed from oblique angles it
appears greenish; outer edge narrowly bluish green at epipleural ridge; no mac-

144 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

ulation except for barely visible narrow band of white representing vestigial apical
lunule.

Venter: Deep metallic violaceous blue; genae wrinkled, with a few recumbent
setae ventrally and remote from the eyes; all sterna and coxae covered with
medium growth of long white decumbent setae; trochanters galbrous except for
one long setigerous puncture at apex; femora and tibiae blue, with long white
setae; tarsi dull bluish green, each segment with a few pits near apex from which
rise long setae; tarsal claws long, those of hind tarsus nearly as long as last tarsal
segment; all segments of abdomen sparsely covered with white setae; penial notch
symmetrical but shallow.

Dimensions. — Female slightly longer and proportionately broader than the male.
Holotype male—Length 12.0 mm, width 5.8 mm. Allotype female—length 16.2
mm, width 6.8 mm.

Etymology. —Named for its primary color red (rutilis), and its secondary green
(viridis) birefringence.

Population variables. —Of 85 specimens studied in detail 13 were completely
immaculate. Nearly all the rest appeared to be immaculate also, but actually bore
either a minute apical spot, or a very thin white line that paralleled the apical
portion of the elytra for a short distance, but not recurved. Only one specimen
had markings consisting of a narrow uninterrupted humeral lunule, a thin middle
band without recurved tip, the latter connected along the edge to a thin apical
lunule.

The degree of color transparency is varied. Nearly one third of the specimens
studied had a deeper color. When viewed from above they appeared less bire-
fringent than the lighter colored specimens.

Hybrid populations.—The slight variations in color and maculation reported
above suggest that intergradation of this form with C. f formosa has occurred
where the Llano Estacado blends into the Great Plains to form the hybrid pop-
ulation of rutilovirescens X formosa. Such hybrids were taken by G. Gaumer in
Texas, both at Muleshoe, Bailey County and north of Sudan, Lamb County in
October 1971. D. Brzoska located another hybrid group north of San Jon, Quay
County, New Mexico. Of six specimens from San Jon, two are similar to ssp.
rutilovirescens, one with a high degree of birefringence, the other a darker red;
another is similar to ssp. formosa, whereas the other three have narrower mac-
ulation with short middle band, one with some birefringence.

On the Status of Cicindela formosa manitoba

In the northwest region of the Great Plains, in a broad area encompassing parts
of the United States and Canada, the maculation of Cicindela formosa broadens
until the limit is reached where the elytra are nearly white. C. f, manitoba was
described by C. W. Leng (1902:137) from broadly marked Manitoba populations,
many years before still more broadly marked populations were discovered farther
west, in Saskatchewan, and subsequently described by W. J. Brown (1940:181)
as Cicindela formosa gibsoni. This last subspecies represents the broadest mac-
ulation form, a condition mentioned earlier as having been associated with mi-
gration to an area where such a type of maculation was common to several related
species of Cicindela. C. f. gibsoni developed through isolation, whereas manitoba
is a hybrid created wherever ssp. gibsoni and ssp. formosa met upon returning

TIGER BEETLES 145

north during Pleistocene interglacials. Therefore, the name manitoba has no status
in the nomenclature, and as such becomes a junior synonym of C. formosa.

Systematics of Cicindela scutellaris and Cicindela formosa

An early attempt to classify the Cicindelidae of the United States was made by
J. L. LeConte (1856:33-40). He placed Cicindela scutellaris in a Group VI with
C. pulchra, C. lecontei, and C. nigrocoerulea. Cicindela formosa was placed in
Group VII with C. ancocisconensis, C. venusta (=C. lengi), C. vulgaris (=C. tran-
quebarica) and C. fulgida. The LeConte classification was based largely on sim-
ilarities of markings, and to a lesser extent on punctation and pilosity. Leng (1902:
124-125), in his revised classification, altered the LeConte scheme by placing C.
scutellaris in a group by itself. Leng (1902:135) placed C. formosa with C. late-
signata and C. tenuicincta. W. Horn (1926:262) took a broader view by combining
some Nearctic Cicindelidae into a Formosa—purpurea—oregona Group. Here, he
included both C. scutellaris and C. formosa. In his Catalogue, Leng (1920:40)
reflects the Horn viewpoint.

In 1954, Rivalier revised the Nearctic genus Cicindela under an earlier stated
intention (Rivalier 1950:15) to dismember the classical genus Cicindela into a
number of new genera. The primary method to accomplish this was based on a
comparative study of the structures within the aedeagus. The intention of my
paper is to go a step further and use the genitalia of both sexes. The aedeagi of
the males, and the female external genitalia (those parts used to dig holes in which
to deposit eggs), are shown in Figures 2 through 5. The definition of terms to
describe the aedeagus are the same used earlier (Rumpp 1967:134). Those used
to describe the female parts are the same as used by V. M. Tanner (1927).

Genitalia of Cicindela scutellaris
Figures 2 and 3

Male. —Nineteen aedeagi were examined, some in the resting position, some
everted. Six were mounted on slides, the rest preserved in alcohol. These aedeagi
were from specimens taken in Colorado, Nebraska, Oklahoma, and Texas.

The penis is arched, and the central bulge is long to accomodate an unusually
long internal sac. The apex is nearly concentric when viewed from the inner side.
The apical ridges are long but not prominent. The bulges on the mid-dorsal and
mid-ventral sides are equal in size, but the ventral one is lower.

The internal sac is long, as noted by Rivalier (1954:252). The structures that
form a central plate are feebly developed, so much so that the tip, or tooth that
caps the central plate, is reduced to a very small scutum (Fig. 2, c and d). Other
structures are well developed, especially so in the great length of the large stiffening
rib, and long filiform flagellum with its loose basal whorl. The elongated sac
resembles the general design found in species of the subgenus Habroscelimorpha,
also mentioned by Rivalier (1954:252), and depicted, in part, by Rumpp (1957:
146-147), but it differs internally where the flagellum of C. scutellaris does not
sustain a membrane.

In the copulatory evaginated position, the internal sac of C. scutellaris everts
at an acute angle of about 25° from the centerline of the penis. This is similar to
the process in most of the species of Cicindela (sensu Rivalier). The everted

146 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

Imm

Cicindela scutellaris

Fig. 2. Aedeagus of Cicindela scutellaris. View a and b of the dorsal and inner sides of the penis,
ssp. scutellaris (Lakeside, NE); views c and d of the dorsal and ventral sides of the internal sac, ssp.
rugata (Nacogdoches, TX); e, f and g are views of the everted sac, ssp. nov. (Maybell, CO—type
location); h is a through view of the everted sac showing location of the sclerites, ssp. scutellaris (Boise
City, OK).

convolutions are uncomplicated, and the external fields are not pronounced. There
is no apical fold near the ventral side. This departure from other species may
have induced advanced development of the flagellum and large stiffening rib,
while concurrently inducing atrophy of the tooth structure.

Female. —The external genitalia are shown in Fig. 3. The cleft in the hypopygi-
um is V-shaped, much as it is in closely related species of the Purpurea Group.
The styli are long, and the tip of the proctiger is wider at the top than in either
C. formosa or the species of the Purpurea Group.

Conclusion. —The placement of C. scutellaris in Rivalier’s phylogeny is correct.
However, the slight morphological departure from Cicindela (sensu Rivalier) does
not warrant his erecting the separate subgenus Pachydela for this one species. The

TIGER BEETLES 147

Fig. 3. Cicindela scutellaris yampae, external female genitalia. 1. Dorsal, or inside view of sternum
eight (hypopygium); 2. Ventral, or outer view of sternum eight—upper part only; 3. Ventral view of
second gonocoxa (coxite) and second gonopophysis (stylus); 4. Dorsal view of syntergum 9 and 10
(proctiger and paraprocts).

name Pachydela remains a junior synonym of Cicindela as recorded in the Check-
list of Cicindelidae by H.P. Boyd et al. (1982:6).

Genitalia of Cicindela formosa

Figures 4 and 5

Male. —Twenty-seven aedeagi were examined in both the resting and everted
positions. Of these, 12 were mounted on slides, and 15 were preserved in alcohol.
These organs were extracted from specimens collected in Colorado, Oklahoma,
Texas, and New Mexico.

148 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

Cicindela formosa

Fig. 4. Aedeagus of Cicindela formosa. View a of the dorsal side of the penis; 5 of the inner side:
cand d the dorsal and ventral sides of the internal sac, ssp. nov. (Mescalero Sands, NM—type location):
é, f, g, and h are views of the everted sac, ssp. gibsoni (Maybell, CO).

The penis of C. formosa is proportionately small for a tiger beetle of its size
(Fig. 4, a and b). It is smooth on the exterior, and only slightly arched. The lateral
ridges are not pronounced. The overall shape is nearly concentric, with no bulges
on either the dorsal or ventral side.

Rivalier (1954:253) erected Group VII, designating C. formosa as the type
species. Included in this group were C. purpurea, C. limbalis, and C. sexguttata,
all because the penis was not arched, the internal sac was provided with a thin
flagellum, and a median tooth was claimed to be absent. Such a description seems
to fit C. formosa to some degree, as shown in Fig. 4. However, it does differ
radically from the other species placed in Group VII because is does not possess
a median tooth, whereas the others do.

TIGER BEETLES 149

es

tan)

$
;
‘
uv:
;
aS

=—S=

Fig. 5. Cicindela formosa rutilovirescens, external female genitalia. 1. Dorsal, or inside view of
sternum eight (hypopygium) h; 2. Ventral or outer view of sternum eight— upper party only; 3. Ventral
view of second gonocoxa (coxite) c and second gonopophysis (stylus) s; 4. Dorsal view of syntergum
9 and 10 (proctiger and paraprocts) Pr&p.

Another significant departure from other species in Rivalier’s Group VII is the
unique development of the everted sac shown in Fig. 4 (e, f, g, and fA). In C.
formosa the internal sac everts apically, nearly in line with the centerline of the
penis, and not at the angle so typical of the other species in Rivalier’s Group VII.
The eversion of the C. formosa sac is at an angle similar to the eversion of the
sac in species of the subgenus Habroscelimorpha, but otherwise differs from these

150 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

77)

A %@
Denver

MEXICO

@C. formosa rutilovirescens mC. scutellaris yampae
@ hybrid: FOLMOSE XX hUVIlOViTfescens

Fig. 6. Map ofa portion of southwestern United States showing location of some new tiger beetles
of the Genus Cicindela.

by having a median tooth and a stubby-based flagellum that does not sustain a
large membrane.

Female.—The external female genitalia are shown in Fig. 5. The outstanding
feature is the curved bottom of the hypopygium notch. This is a radical departure

TIGER BEETLES 151

from all other species in Rivalier’s Group VII; these other species have an acute
angle to their hypopygium notches.

Conclusion. —From an evolutionary standpoint, C. formosa fits into a singular
niche of the subgenus Cicindela. It is a species with adults equal in size to, or
larger than, those of the Longilabris, Pulchra, or Tranquebarica Groups. Ecolog-
ically it is adapted to the various climates and environments of the upper North
American continent where it remains fixed in a sand-dwelling niche. It has an
extensive distribution across the United States and Canada, from the Rocky
Mountains to the eastern seaboard. The generic features of this species have few
attributes to fix its ancestral type with other extant species. Therefore, it must be
assumed that its derivation stems from great antiquity. It is a singular development
of the Nearctic regions.

Phylogenetically Cicindela formosa fits between the Scutellaris Group (sand
blowout habitat; long internal sac of the aedeagus, with long unsustained flagellum
and atrophied tooth), and the Purpurea Group (coarse sand habitat; internal sac
of the aedeagus with a very small tooth and a short sinuous flagellum).

Referenced Locations
Figure 6

Locations are shown on this map with reference to C. scutellaris yampae and
C. formosa rutilovirescens.

Literature Cited

Arnett, R. H. and Samuelson, G. A. 1969. Directory of Coleoptera Collections of North America
(Canada through Panama), 123 pp.
Boyd, H. P. and Associates 1982. Checklist of the Cicindelidae. Plexus Publishing, Inc., 31 pp.
Brown, W. J. 1940. Some new and Poorly Known Species of Coleoptera. Canadian Entomologist,
72:183-184.
Horn, W. 1926. S. Schenkling on Carabidae—Cicindelidae, 85, W. Junk Coleopterum Catalogus,
ed. pp. 258-302.
LeConte, J. L. 1856. Revision of the Cicindelidae of the United States. Trans. Am. Phil. Soc., II
(New Series):27-62.
Leng, C. W. 1902. Revision of the Cicindelidae of Boreal America. Trans. Am. Entom. Soc., 28:
93-190.
1920. Catalogue of the Coleoptera of America North of Mexico, compiled by John D.
Sherman Jr., Mount Vernon, NY. 470 pp.
Rivalier, E. 1950. Démembrement du Genre Cicindela Linné—Faune Pal€éarctique. Revue Frang.
d’Entom., 17:217-244.
1954. Démembrement du Genre Cicindela Linné—Faune Nearctique. Revue Frang. d’En-
tom., 21:248—268.
Rumpp, N. L. 1957. Notes on the Cicindela praetextata—californica Complex; Description of New
Subspecies from Death Valley, California. Bull. So. Calif. Acad. Sci., 58:144-154.
1961. Three New Tiger Beetles of the Genus Cicindela from Southwestern United States.
Bull. So. Acad. Sci., 60:165-187.
1967. A New Species of Cicindela from Idaho. Proc. Calif. Acad. Sci., 35:129-140.
Say, T. 1817. American Entomology. Journ. Acad. Nat. Sci. of Phila., 1:19-23.
1823. Descr. of New Species of Coleoptera from an Expedition to the Rocky Mnts. Journ.
Acad. Nat. Sci. of Phila., 3:139-216.
Tanner, V.M. 1927. A Prelim. Study of the Genitalia of Female Coleop. Trans. Amer. Entom. Soc.,
53:5-50.
Wallis, J. B. 1961. The Cicindelidae of Canada. Univ. of Toronto Press, 74 pp.

Accepted for publication 25 May 1985.

Bull. Southern California Acad. Sci.
85(3), 1986, pp. 152-157
© Souther California Academy of Sciences. 1986

Coyote Diets, Five Years Later, at Cuyamaca Rancho State Park

Joel D. Weintraub

Department of Biological Science, California State University,
Fullerton, California 92634

Abstract.— Food habits of coyote (Canis latrans) on East Mesa, Cuyamaca Ran-
cho State Park, San Diego County, California were studied by analysis of scats.
Remains of pocket gopher (Thomomys bottae), mule deer (Odocoileus hemionus),
vole (Microtus californicus), ground squirrel (Spermophilus beecheyi), and white-
footed mouse (Peromyscus spp.) were predominate in scats collected from No-
vember 1983 to July 1984. Five years earlier coyote on East Mesa were reported
to feed primarily on beaver (Castor canadensis), white-footed mouse, opossum
(Didelphis virginiana), pocket gopher, mule deer, and badger (Taxidea taxus).
Explanations are offered for the differences noted.

The coyote (Canis latrans) has been described as an opportunistic predator
(Bekoff 1977). In California, Ferrel et al. (1953) found rodent in 49%, rabbit in
29%, deer in 18%, and carnivore in 4% of 2222 coyote stomachs. MacCracken
(1981) noted that when leporids were poorly represented in western coyote diets,
the proportion of vole, pocket gopher, or ground squirrel was usually high. Bowyer
et al. (1983) reported on coyote diets over a twenty month period (June 1977-
January 1979) from the East Mesa area of Cuyamaca Rancho State Park, Cali-
fornia. Leporids were found in only 1% of their sample while beaver (Castor
canadensis) was present in 44% of 223 scats. Carnivores were present in 33%,
opossum (Didelphis virginiana) in 24%, pocket gopher (Thomomys bottae) in 22%,
and mule deer (Odocoileus hemionus) in 21% of the coyote droppings analyzed
by Bowyer et al. (1983).

A review of 57 coyote diet studies (list available from the author) indicates that
the earlier East Mesa coyote study (Bowyer et al. 1983; Bowyer, pers. comm.)
recorded the highest frequency of opossum, beaver, gray fox (Urocyon cinereoar-
enteus), bobcat (Lynx rufus), long-tailed weasel (Mustela frenata), and badger
(Taxidea taxus) in any North American study, as well as the only records of
mountain lion (Felis concolor) and man (Homo sapiens). The present study in-
vestigated whether this unusual coyote diet persists at Cuyamaca Rancho State
Park.

Methods

East Mesa encompasses about 800 hectares of upland meadows and oak-pine
woodlands at an elevation of 1525 m:; the site is in the southern part of Cuyamaca
Rancho State Park, San Diego County, California and is surrounded by lower
elevation chaparral (Bowyer and Bleich 1980). The Laguna Mountain area of
Cleveland National Forest is to the east and the towns of Guatay and Descanso
are located about 4.5 km south and 5.5 km southwest, respectively. Weather data

152

CUYAMACA RANCHO STATE PARK COYOTE DIETS 153

are available for over 60 years from Descanso Ranger Station (elevation 1066
m). A rainfall gauge is also at park headquarters, 1.5 km west of East Mesa at an
elevation of about 1280 m.

Scats were collected along the East Mesa Fire Road on East Mesa on 19 No-
vember 1983 (N = 26), 27 January 1984 (N = 25), 9 April 1984 (N = 28), and
7 July 1984 (N = 35). All scats collected were greater than 20 mm in diameter
which should exclude gray fox droppings, but also excludes about half of the
coyote scats produced in the area (Danner and Dodd 1982). Scats that might be
from bobcat (regular annular constrictions) were not collected.

Scats were dried at 80°C for 48 hrs to destroy tapeworm stages and then washed
in a flour sifter (mesh opening about 1.7 mm) under running water. Most of the
bones, teeth, and claws were separated from the hair by placing the washed
material in water in an enamel tray and floating off the hairs into the sifter. Hair
and heavier material were separately dried and analyzed. The apparatus was
carefully cleaned between scat preparations. Claw, bone, and tooth material were
identified to species using reference collections at California State University
Fullerton (CSUF). A key to rodent incisors also aided the process (Weintraub and
Shockley 1980). Mayer (1952), Moore et al. (1974), and Stains (1958) were used
for hair identification as well as a hair slide collection from CSUF’s specimens.
Intact tufts of hair from unidentified prey were stored on two-sided sticky tape
placed on blue paper for color viewing and measurements of the hairs. If further
microscope viewing was needed, they were cleaned in carbon tetrachloride and
scale pattern molds of the hair made using techniques of Hilton and Kutscha
(1978). Hairs were then immersed in xylene for '2 hr before permanent mounting
on slides. Hair slides were viewed under polarized-light microscopy to accentuate
the medullary features. Voucher materials were kept for future reference.

Differences between seasons or between studies in the proportion of scats with
a particular prey item were analyzed by an exact test for 2 x 2 contingency tables
(Wells and King 1980). Significance was set at the 5% probability level. An ex-
panded species list was available from Bowyer for the 1977-1979 East Mesa study
and accounts for most of the ‘““Bowyer, pers. comm.” citations in this paper.

Results

Table 1 compares coyote prey items found in the present study with those found
in the 1977-1979 study. I recorded 17 mammal prey species including a tentative
identification of house cat (Felis catus) and Bowyer et al. (1983) recorded 26
mammal prey species (includes both Neotoma fuscipes and N. lepida). 1 had 6
unidentified rodent items. Seasonal differences occurred in the proportion of scats
containing a given prey in 1983-1984. Between the November and January sam-
ples the proportion of seed showed a significant reduction and the proportion of
vole (Microtus californicus) significantly increased in the droppings. Between the
April and July samples, both vole and white-footed mouse (Peromyscus spp.)
showed significant reductions in scat occurrences while the proportion of ground
squirrel (Spermophilus beecheyi) and seed material significantly increased. Neither
pocket gopher nor deer showed significant changes in their proportions in coyote
droppings between sampling periods, but deer showed a higher proportion for the
fall and summer samples.

Three of the six most commonly recorded mammal prey species in the 1977-

154 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

Table 1. Coyote food items (% occurrence in scats; tr = <0.5%) based on analysis of 223 droppings
from June 1977 to January 1979 (~BOW: Bowyer et al. 1983; Bowyer, pers. comm.) and on analysis
of 114 scats from November 1983 to July 1984 (=WEI: current study) from the East Mesa area of
Cuyamaca Rancho State Park, San Diego Co., California.

Study Study
Prey item BOW WEI Prey item BOW WEI
Mammals
Didelphis virginiana 24 — Procyon lotor Dy, 1
Scapanus latimanus — 3 Mustela frenata 2 —
Eumops perotis tr _ Taxidea taxus 11 —
Lepus californicus tr — Spilogale gracilis 1 _
Sylvilagus auduboni 1 _— Mephitis mephitis 2 1
Eutamias merriami 1 1 Felis concolor 3 —
Spermophilus beecheyi 3 24 Lynx rufus 3 1
Sciurus griseus 2 1 Felis catus — 1 (?)
Thomomys bottae DD 64 Odocoileus hemionus 21 41
Perognathus californicus 1 — Homo sapiens tr —
Dipodomys agilis tr 1
Castor canadensis 44 _
Reithrodontomys megalotis 4 Birds 3 6

Peromyscus spp. 26
Neotoma spp. 1 4 Reptiles 4 12
Microtus californicus —
Canis latrans 5
Urocyon cinereoargenteus 3
Bassariscus astutus 1

1 Arthropods 34 47

1 Vegetation 84 33

1979 study were not recorded from coyote droppings in 1983-1984 (i.e. beaver,
opossum, and badger); five of these six (Peromyscus spp. the exception) showed
significantly different frequencies of occurrence between the two studies. Of the
five most common mammal prey species encountered in coyote droppings in
1983-1984, one species was not recorded in 1977-1979 (i.e. vole); four of these
five species (Peromyscus spp. the exception) showed significantly different pro-
portions in coyote scats between the two studies.

Discussion

Coyote on East Mesa fed largely on gopher, deer, vole, and ground squirrel in
1983-1984, a pattern similar to other diets from western areas with low leporid
intake (e.g., Hawthorne 1972; Murie 1940; Weaver 1980). This study documents
the highest intake of pocket gopher of any of 57 previous coyote diets. High
populations of deer (approximately 200) and ground squirrel were noted on the
East Mesa area in the late 1970s (Bowyer and Bleich 1980), and were still prevalent
in and around this area during the present study. Overgrazing by cattle in the
1950s and a major chaparral fire during the period may have favored these species
as well as pocket gopher.

Seasonal! dietary shifts by southern California coyote were reported by Ferrel
et al. (1953). Vegetation was a major fall food item while animal consumption
was highest in spring and deer intake highest in the summer. Both East Mesa
studies found large amounts of vegetation taken most frequently during the fall.

CUYAMACA RANCHO STATE PARK COYOTE DIETS 155

Deer intake by coyote appeared to increase during the summer months, presum-
ably corresponding to fawn production.

The two East Mesa coyote studies differ significantly in the proportion of many
of the prey items in the scats. Coyote from East Mesa appear to have fed on rare
or restricted prey during 1977-1979 (e.g., beaver, badger, and mountain lion).
San Diego County is outside the historic range of the beaver (Tappe 1942). How-
ever, four beaver were released near Cuyamaca Rancho State Park in 1945 (Calif.
Fish and Game, no date) as part of a statewide transplant program. The nearest
beaver sites in the late 1970s were located on the Sweetwater River, about 3 km
west of East Mesa. Only one colony was known to be in the park (at the Green
Valley campground) with two colonies located south of the park boundary in 1978
(Treanor, pers. comm.). Beaver were last seen in the park in 1979 (Treanor, pers.
comm.; Bowyer, pers. comm.) and last seen below the park in 1981 (Treanor,
pers. comm.).

The California Department of Fish and Game carried out a large-mammal
trapping program in the late 1970s to the east of East Mesa. Gould (pers. comm.)
estimated, based on trapping data from that program, that badger were present
here in densities of 1.3 animals per sq km. Badger are rarely recorded as coyote
prey. Rathbun et al. (1980) observed coyote killing a young badger by cooperative
predation under low prey availability conditions. Gould (pers. comm.) also es-
timated that mountain lion density was 0.1 to 0.3 per sq km in the trapping area.
They were recorded in coyote scats from three separate seasons by the 1977-1979
study (Bowyer, pers. comm.). A mountain lion can kill a coyote (Gianini, 1935).

Opossum were common on East Mesa in the late 1970s (Curto, pers. comm.;
Bowyer, pers. comm.). Their present abundance is unknown. Although opossum
were reported taken by coyote in about one sixth of the other reviewed diet studies,
their frequency in coyote scats or stomachs never exceeded 3%. Whiteman (1940)
found that captive coyote rejected opossum, insectivora, and weasel flesh. This
discussion suggests that coyote may have been under food stress during 1977-
1979 and were utilizing uncommon, perhaps less appetizing, and potentially dan-
gerous prey species in spite of the apparently high deer, ground squirrel, and
probably pocket gopher populations in the park.

The 1977-1979 study occurred at the end of a period of below average rainfall
and during the very heavy 1978 rainfall year. Rainfall (% of normal) at Descanso
Ranger Station was as follows: 1973 = 96%; 1974 = 86%; 1975 = 91%; 1976 =
103%; 1977 = 86%; 1978 = 199%; 1979 = 126%; 1980 = 193%; 1981 = 75%;
1982 = 168%; 1983 = 185% (U.S. Dept. Commerce 1973-1983). The present
study occurred during an unusually low spring rainfall period, as only 7 cm of
rain was recorded at park headquarters from January to June 1984. Vole were
not recorded as food items in the earlier East Mesa study but they were a major
prey item in the current study. Recent rainfall patterns may have favored high
vole populations in the park. Vole are also known to exhibit cyclic population
fluctuations (Pearson 1966). Vole were considered to be the most important rodent
prey for California coyote (Ferrel et al. 1953) and changes in the population of
this species might influence food selection by coyote. Blankenship (1982) found
no vole in 5430 trap nights between December 1979 and December 1980 on a
burned pine forest site on East Mesa. Vole made up a small part (1%) of the diet
of spotted owl (Strix occidentalis) in the park in 1978 and 1979 while most of

156 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

the owl diet was dusky-footed woodrat (Neotoma fuscipes, 33%), white-footed
mouse (31%), and gopher (8%) (Barrows 1980).

Another environmental factor that may have influenced prey populations is
controlled burns on and around the East Mesa area. This program started with
test plots in 1978 (Lathrop and Martin 1982) and then was expanded in coverage.
Lawrence (1966) and Lubina (1978) found that grassland rodent species increased
in number after fire in California chaparral. Blankenship (1982), working in pine
forest on East Mesa, did not find this pattern.

Possible experimental errors should also be considered in evaluating the two
East Mesa coyote diets. Four such factors relating to the proportion of a prey item
in scats are presented. First, Bowyer et al. (1983) recorded a high frequency of
vegetation in scats. Andelt and Andelt (1984) found that the proportion of in-
digestible material in a meal influenced the defecation rate of the coyote; when
coyote eat vegetation the number of scats produced is higher per biomass intake.
Second, remains of a single mammal prey (depending on size) may end up in
more than one coyote scat (Weaver and Hoffman 1979). As a result, frequency
data of mammal prey in scats cannot be directly equated with number of prey
individuals. I did not collect droppings under 20 mm in diameter which should
bias my sample towards larger prey since they lead to larger diameter coyote scat
(Danner and Dodd 1982). Third, the estimated density of coyote in the Laguna
Mountain area (Gould, pers. comm.) would predict about 6 individuals on East
Mesa; coyote may produce 6 scats a day (Fichter et al. 1955). Based on these two
values, both East Mesa coyote studies sampled less than 2% of the available coyote
scats. These small samples may not be representative of the prey taken by coyote
in the region. Finally, sample sizes of droppings per given time period was dis-
similar for the two East Mesa studies and would influence the yearly proportion
of prey items in the scats.

Acknowledgments

I thank Chief Ranger Steven Treanor of Cuyamaca Rancho State Park for his
help and Wildlife Biologist Gordon Gould of the California State Department of
Fish and Game for information about mammal densities in the area. Others who -
contributed to this study are V. Bleich, T. Bowyer, M. Curto, G. Jones, E. Lathrop,
M. Lembeck, G. Stewart, and G. Weintraub. I would like to thank Jody Weintraub,
Rema Weston, and Axhel Munoz for help in the field work. Thanks are also due
the reviewers of this paper. This study was made possible by a collecting permit
from the California Department of Parks and Recreation.

Literature Cited

Andelt, W. F., and S. H. Andelt. 1984. Diet bias in scat deposition-rate surveys of coyote density.
Wildl. Soc. Bull. 12(1):74-77.

Barrows, C. 1980. Feeding ecology of the spotted owl in California. Raptor Research 14(3):73-78.

Bekoff, M. 1977. Canis latrans. Mammalian Species 79:1-9, Amer. Soc. Mammal.

Blankenship, D. J. 1982. Influence of prescribed burning on small mammals in Cuyamaca Rancho
State Park, California. P. 587 in C. E. Conrad and W. C. Oechel (Tech. Coords.). Proceeds.
Symp. Dynamics and Manage. Mediterranean-Type Ecosys. U.S. For. Serv. Gen. Tech. Rept.
PSW-58. 637 pp.

Bowyer, R. T. and V. C. Bleich. 1980. Ecological relationships between southern mule deer and
California black oak. Pp. 292-296 in T. R. Plumb (Tech. Coord.). Symp. Ecol., Manage. and
Util. of Calif. Oaks. U.S. For. Serv. Gen. Tech. Rep. PSW-44. 368 pp.

CUYAMACA RANCHO STATE PARK COYOTE DIETS 113)7/

—,, S. A. McKenna, and M. E. Shea. 1983. Seasonal changes in coyote food habits as determined
by fecal analysis. Amer. Midl. Natur. 109(2):266-273.

Calif. Dept. Fish and Game. No date. Beaver Transplanting. Final report, Project California 18-D-
2. Mimeo.

Danner, D. A. and N. Dodd. 1982. Comparison of coyote and gray fox scat diameters. J. Wildl.
Manage. 46(1):240-241.

Ferrel, C. M., H. L. Leach, and D. F. Tillotson. 1953. Food habits of the coyote in California. Calif.
Fish and Game 39(3):301-341.

Fichter, E., G. Schildman, and J. H. Sather. 1955. Some feeding patterns of coyotes in Nebraska.
Ecol. Monogr. 25(1):1-37.

Gianini, C. A. 1935. Cougar and coyote. J. Mammal. 16(3):229.

Hawthorne, V.M. 1972. Coyote food habits in Sagehen Creek Basin, northeastern California. Calif.
Fish and Game, 58(1):4—-12.

Hilton, H. and N. P. Kutscha. 1978. Distinguishing characteristics of the hairs of eastern coyote,
domestic dog, red fox and bobcat in Maine. Amer. Midl. Natur. 100(2):223-227.
Lathrop, E. W. and B. D. Martin. 1982. Response of understory vegetation to prescribed burning
in yellow pine forests of Cuyamaca Rancho State Park, California. Aliso 10(2):329-343.
Lawrence, G.E. 1966. Ecology of vertebrate animals in relation to chaparral fire in the Sierra Nevada
foothills. Ecol. 47(2):278-291.

Lubina, J. A. 1978. The effects of fire on rodent populations in the chaparral of southern California:
a comparative approach. M.S. Thesis, Calif. State Univ., Long Beach.

MacCracken, J. G. 1981. Coyote foods in southwestern Colorado. Southw. Nat. 26(3):317-318.

Mayer, W. V. 1952. The hair of California mammals with keys to the dorsal guard hairs of California
mammals. Amer. Midl. Natur. 48(2):480-—512.

Moore, T. D., L. E. Spence, and C. E. Dugnolle. 1974. Identification of the Dorsal Guard Hairs of
Some Mammals of Wyoming. Wyo. Game and Fish Dept. Bull. 14, Cheyenne.

Murie, A. 1940. Ecology of the coyote in the Yellowstone. Fauna of the Natl. Parks of the U.S.
Cons. Bull. No. 4.

Pearson, O. P. 1966. The prey of carnivores during one cycle of mouse abundance. J. Anim. Ecol.
35(2):217-233.

Rathbun, A. P., M. C. Wells, and M. Bekoff. 1980. Cooperative predation by coyotes on badgers.
J. Mammal. 61(2):375-376.

Stains, H. J. 1958. Field key to guard hair of middle western furbearers. J. Wildl. Manage. 22(1):
95-97.

Tappe, D. T. 1942. The status of beavers in California. Calif. Dept. Fish and Game Bull. 3.

U.S. Dept. Commerce. 1973-1983. Climatological Data, Annual Summary(ies), California. N.O.A.A.
VOLS. 77-87, Nos. 13.

Weaver, J. L. 1980. Influence of food supply on coyote populations in Jackson Hole, Wyo. Pp. 185-

226 in Proceeds. Second Conf. Sci. Res. Natl. Parks, Vol. 12: Terrestrial Biology: Zoology. U.

S. Dept. Interior Rpt. NPS/ST-80/02-12. 411 pp.

, and S. W. Hoffman. 1979. Differential detectability of rodents in coyote scats. J. Wildl.

Manage. 43(3):783-786.

Weintraub, J. D. and G. Shockley. 1980. Use of incisors to identify genera in owl pellets. Bull. S.
Calif. Acad. Sci. 79(3):127-129.

Wells, H. and J. L. King. 1980. A general ‘exact test’ for N x M contingency tables. Bull. S. Calif.
Acad. Sci. 79(2):65-77.

Whiteman, E. E. 1940. Habits and pelage changes in captive coyotes. J. Mammal. 21(4):435—438.

Accepted for publication 18 February 1985.

Bull. Southern California Acad. Sci.
85(3), 1986, pp. 158-172
© Southern California Academy of Sciences, 1986

The Influence of Mima Mounds on Vegetation Patterns in the
Tijuana Estuary Salt Marsh, San Diego County, California

George W. Cox and Joy B. Zedler

Department of Biology, San Diego State University, San Diego,
California 92182-0057

Abstract.—Mima-type earth mounds, up to 18.6 m in diameter, occur in marsh,
transition, and upland areas of the Tijuana Estuary, but are best developed at
intermediate tidal elevations. Plant dominants on mounds grade from low marsh
to upland species. Upland and high marsh species show steeper species-area re-
lationships than low marsh species, reflecting insular patch dynamics. Mounds
are activity centers for several herbivorous mammals that probably influence
mound and intermound vegetation. The mounds create a highly patchy transition,
and indicate that landward limits of marsh species are more reliable marsh bound-
ary indicators than seaward limits of upland species.

Mima-type earth mounds, attributed to the long-term soil translocation activ-
ities of geomyid pocket gophers, occur in many locations in western North Amer-
ica (Cox 1984a). These mounds, which reach 2 m in height, 40 or more m in
diameter, and 50 or more ha“! in density, seem to be restricted to areas of original
grassland habitat, including lower alpine tundra and areas transitional to desert
scrub and salt marsh (Cox in press).

A survey of Mima mound distribution in San Diego County, in the southwestern
corner of California, revealed mounds at several locations bordering the Tijuana
Estuary salt marsh (Cox 1984b). This marsh has been studied extensively by
Zedler (1977, 1982), Zedler et al. (1980), and Winfield (1980), but the marsh-
upland transition has not been described. The salt marsh is relatively undisturbed
and species-rich, while the transition zone from marsh to upland is heavily dis-
turbed. The mounds, concentrated in the marsh-upland transition, are conspic-
uously different in their vegetation from intermound areas, and show extensive
tunneling and digging by mammals. Elsewhere in California, Brown (1951) noted
Mima mounds in the transition zone and upper salt marsh of San Francisco Bay
near Port Chicago and Pittsburgh, Contra Costa County. Along the Gulf Coast
of Louisiana and Texas, Mima-type mounds border salt marshes along the inner
side of coastal sounds (Dietz 1945, Price 1949).

The presence of these mounds at the Tijuana Estuary provided an opportunity
to examine questions relating to the influence of topographic heterogeneity on
salt marsh vegetation, and to the mechanism of Mima mound formation. Pre-
viously, the transition from salt marsh to upland has been assumed to be a
vegetational continuum along a simple environmental gradient (Harvey et al.
1978, Eilers et al. 1983). Mounds within the marsh-upland transition modify
salinity and moisture relationships and create islands of habitat suitable for upland

158

MIMA MARSH MOUNDS 159

IMPERIAL
BEACH

IMPERIAL
BEACH
NAVAL
AIR
UPLAND STATION

09%
00%,
59)
%,
%
O =
=
Soovsoooenn

*ounnoe

Ln
RT2%

UPLAND

PACIFIC
OCEAN

Tijuana River

0.5 km
Vertical datum = MLLW

Fig. 1. Map of the Tijuana Estuary, San Diego County, California, showing Mima mounds fields
examined in this study. The approximate 2 m MLLW contour is shown within the estuary. Study
plots are coded as follows: NE—North End, W— Wheels, RB—Reed Bed, C—Central, RT— Ridge
Top. The symbol UT designates the marsh-to-upland transition area that was examined but not
sampled quantitatively.

plants and animals. Invasion of such areas of habitat by upland plants and by
herbivorous mammals that utilize both upland and marsh plants may greatly
complicate wetland boundary determination. Defining ecologically valid criteria
for salt marsh boundaries is a matter of current major concern (Zedler and Cox
1985).

The origin of Mima mounds in Southern California has been the object of
several physical and biotic hypotheses, but Cox (1984b) has strongly supported
the hypothesis of their origin by the long-term process of soil translocation by
geomyid pocket gophers. Under this hypothesis, maximum mound development
should occur on soils of intermediate depth above a basement layer that causes
frequent soil waterlogging and gives an advantage to animals living in spots of

160 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

elevated, well drained soils created by such translocation (Gakahu and Cox 1984).
Thus, we hypothesized that the Tijuana Estuary mounds should show evidence
of pocket gopher occupation, and should exhibit maximum size at an intermediate
point in the marsh-upland transition, where their influence on vegetational fea-
tures might be greatest.

Our studies were therefore designed to answer the following specific questions:

1. What is the extent of the mounded zone, and the pattern of heights, di-
ameters, and spacing of the mounds?

2. How do soil salinity and other substrate conditions important to plant es-
tablishment and growth vary with mound size and location?

3. How are distribution and diversity of marsh and non-marsh plants related
to mound size and location?

4. Are the floras of non-marsh species on mounds influenced by insular col-
onization and extinction processes?

Procedure
Study Area

The Tijuana Estuary (32°34'N, 117°7’W) lies just north of the Mexican Border
at the mouth of the Tijuana River (an intermittent stream). The estuary (432 ha)
includes about 166 ha of intertidal marshland and 164 ha of transition zone
habitat (MclIllwee 1970). Although the estuary has a long record of good tidal
flushing (Zedler 1982), the ocean entrance became blocked during our study. The
estuary proper then became hyposaline due to freshwater inflow from the Tijuana
River, and later hypersaline due to evaporation during the dry summer and early
autumn. The transition zone and upland habitat have been disturbed heavily by
excavation and filling, construction of dikes and channels, and vehicle activity.

The Tijuana Estuary salt marsh has the richest flora of vascular halophytes of
the 23 major southern California coastal wetlands (Zedler 1982). In common with
other marshes that have experienced regular tidal flushing, the lowest marsh levels
are dominated by cordgrass (Spartina foliosa). At higher levels, the marsh dom-
inants are succulents, primarily pickleweed and glasswort (Salicornia virginica
and S. subterminalis, respectively), and low, perennial, halophytic grasses (salt-
grass, Distichlis spicata; shoregrass, Monanthochloe littoralis). In the transition
zone, halophytes gradually give way to a variety of herbaceous plants (many of
which are annual Mediterranean exotics), cacti, semi-woody subshrubs, and ev-
ergreen woody shrubs typical of California annual grassland and coastal sage scrub
communities.

Methods

Mounds occur in five distinct locations in the northern and northeastern portion
of the estuary (Fig. 1). In each location a plot spanning the available elevational
range was selected and the mounds present were marked with numbered stakes.
These mound groups were the main source of data for this study, but general
observations were also made at several other locations. Mounds were present, for
example, along most of the western side of the central ridge which separated the
inner and outer marsh areas, and on which the Central plot was located (Fig. 1).

The physical and biotic characteristics of 86 mounds were assessed between 12

MIMA MARSH MOUNDS 161

and 23 April 1984. Mound top and edge elevations were surveyed with a Wild
Automatic Level; elevations were referenced to U.S. Army Corps of Engineers
benchmark TJE-35, located at 2.03 m (6.58 ft) above Mean Lower Low Water
(MLLW). Mound heights were obtained as the difference between top and edge
elevations. Distance and direction from a single survey point in each area were
also recorded to map each mound field. The maximum and minimum diameters
of each mound and the distance to its nearest neighbor (center to center) were
measured. Vegetational characteristics and the area of surface disturbed by animal
digging were evaluated on a six-point cover scale: <1%, 1-5%, 5—25%, 25-50%,
50-75%, 75-100%. The percent coverage for each vacular plant species with live
tissue or tissue produced during the 1983-84 winter growing season was estimated
for each mound and for an equal-area intermound circle adjacent to, and at the
same elevation as, the mound base. The areas of bare soil surface and of flotsam
coverage were estimated in the same way. The probable animal agent responsible
for any digging or tunneling was also noted.

Soil samples were collected from the top and base of 53 mounds in three
locations (North End, Wheels, Reed Bed) on 17 June 1984. At this time the
estuary entrance was closed, and river inflow had become impounded. On this
date mounds at the lowest elevations (below about 1.95 m MLLW) were com-
pletely above water level, although they had been surrounded by shallow water
(<10 cm deep) with a salinity of 15 parts per thousand during the April sampling
period. Soil samples for 29 mounds in the Central area were collected on 14
September 1984, after the impounded water had receded. At this later date the
estuary entrance was also closed, but re-entry of sea water had occurred and
evaporation had caused water level to drop well below (about 54 cm) the basal
elevation of the lowest mounds.

Soil salinities were measured as the electrical conductivity of uniform soil pastes,
using a Lab-line Mark IV Mho-meter and the procedure outlined by Richards
(1954).

Techniques of statistical analysis follow Zar (1984). Most correlation and regres-
sion analyses were performed only on mound data from North End, Wheels, Reed
Bed, and Central areas, which overlapped in elevation. The four Ridge Top
mounds were more than a meter disjunct in elevation. Plant nomenclature follows
Munz (1974).

Results
Extent and Internal Geometry of Mounded Areas

The four mound areas in the upper marsh and transition zone (Fig. 1) ranged
in size from 0.79 to 1.59 ha, and contained 15-29 mounds each (Table 1). Basal
elevations of the mounds in these areas ranged from 1.87 to 2.88 m MLLW (Table
1). Mounds with basal elevations below 2.31 m MLLW are surrounded by water
at regular high tides; those below 2.93 m MLLW are occasionally surrounded by
storm tides. The lowest basal elevations in the Wheels and Reed Bed areas were
15-21 cm above those for the North End and Central areas. Of these four areas,
Central showed the least evidence of disturbance by vehicle and foot traffic.

The Ridge Top area (Fig. 1) was less than 0.1 ha in area and lay at an elevation
of 4.29-4.51 m MLLW (Table 1). The vegetation and soils of this small site

SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

162

"SOO < d 1 (SU) JUROYIUSIS 1OU PoJOpIsSUOD | y

CS oe

100 >dLit pO + SET ¢9°81-09'S LS'0-80°0 IS p-L8'1 98 Isp Te10.L

SU 87 pic + SS'6 Ov’ 1 1-00°8 L70-¥1 0 IS p-6C v 60°0 doy ospra

100 >dL7el I0€ + ¥9TI Sp et—09'S ¢v'0-80°0 6t C-16 1 6C STI [eua)

su 60'1 I8'€ + 86 v1 Cr sI-O1'9 yy 0-010 £9°C-80°C IZ 6S T ped Pood

su oT I 90°C + SE ET ¢9°81-00°9 LS'0-60°0 88°C-90'C LI 680 STOOUMA

#SU CCT c67 + CO VI CL vI-09'S tv0-v10 9¢°C-L8 I cl 6L°0 pug YON

Y ‘JUSTOYJI09 uOTsIod (aS = Ww) (WI) sIN}9WIIp (uw) s}Yy3I0Y (MTIW 2A0ge WwW) = spunoul (8H) dnoi3 punoyy

-SIP SUPAQ-YIL[D IOqysiIoU jsoreOU jo osuey jo oduey suoneAgo jo Joquinny eoly
0} 90URISIP ULI] jeseq jo odury

Ea

“BIUIOJIED ‘AyUNOD OdsIq] ues ‘ArenjoOURS oULIeN\sq [eUONeN JoaTY vuenfiy oy 18 spjoypunow pue spunoul odA}-eUI] JO SONSLIQ}ORIEYD “| 9IGRL

MIMA MARSH MOUNDS 163

BASAL ELEVATION MOUND PLANT
(m above MLLW) MOUND HEIGHT, RADIUS, VEGETATION COVER (%)
100 110 120
30) 429-451 U

100 110 120

100 110 120

HEIGHT (cm)

100 110 120

100 110 120

6 5 4 3 2 1 O 1 2 3 4 5 6
MOUND RADIUS (m)

Fig. 2. Mound size and form, mound and intermound vegetation composition, and percent plant
cover of Mima mounds in different elevational belts at the Tijuana Estuary, San Diego County,
California. Symbols designate species categories: L—low marsh, H—high marsh, U—upland; categories
are shown in decreasing order of total cover from left to right for mound and adjacent intermound
areas.

suggested, however, that it was a relatively undisturbed remnant of the original
upland habitat.

In the marsh and transition zone, mound heights and diameters varied with
elevation (Fig. 2). Mound height increased significantly with elevation between
1.87 and 2.35 m MLLW (r = 0.43, df = 50, t = 3.36, P < 0.001). Between 2.32
and 4.51 m MLLW, however, mound height declined significantly with increasing
elevation (r = —0.332, df = 39, t = 2.20, P < 0.05). The tallest mounds (n =
11), exceeding 0.4 m in height, all occurred between 2.28 and 2.62 m MLLW.
The broadest mounds, exceeding 13 m in diameter (n = 10), spanned a wider
elevational range, 2.18—2.84 m MLLW. Mound diameter was significantly cor-
related with elevation above MLLW (r = 0.44, df = 80, t = 4.37, P < 0.001).
Mound height and diameter showed a significant correlation, as well (r = 0.35,
df = 80, t = 3.36, P < 0.01). The regression equation for mound diameter (Y)
as a function of height (X) was:

Y = 6.76 + 9.29X. (1)

Analysis of mound spacing, using the technique of Clark and Evans (1954),
indicated that the pattern tended to deviate from random in the direction of
uniformity, the coefficient R exceeding 1.0 (Table 1). This deviation was signif-

164 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

North End
17 June 1984
N=15

ia.
aie Base
2 ENG P<0.01

Top

Wheels + Reed Bed
17 June 1984
N= 38

b= -2.43 P<0.001

Top
b=-1.74 P<0.001

=
==
==
==

Central
9 September 1984
N= 29

SOIL PASTE CONDUCTIVITY (mmhos cm’)

b=-5.56 P<0.001

18 2.2 2.5 2.8
ELEVATION ABOVE MLLW (m)

Fig. 3. Regression lines for mound top and mound edge soil salinities on elevation above Mean
Lower Low Water for mound fields at the Tijuana Estuary, San Diego County, California.

icant, however, only for the Central area, the least disturbed mound group, and
for all mound areas taken together. No significant relation existed between ele-
vation above MLLW and distance to nearest neighbor.

Soil Salinity and Soil Disturbance by Animals

Soil salinity, expressed as the electrical conductivity of the soil paste, was higher
for mound edges than for mound tops in all cases (n = 82). Salinity decreased
with increasing height above MLLW (Fig. 3), but patterns differed for different
mound areas. In areas sampled in June, 1984 (North End, Wheels, Reed Bed),
salinities at mound edges were significantly higher than those on mound tops
(North End: t = 6.78, df = 27, P < 0.001. Wheels and Reed Bed: t = 14.46, df =
73, P < 0.001). The patterns of change of mound top and mound edge salinity
with elevation also did not differ between these two areas. However, salinities for
the Wheels and Reed Bed areas were significantly higher than those for North
End (Tops: t = 5.35, df = 50, P < 0.001. Edges: t = 20.38, df = 50, P < 0.001).

MIMA MARSH MOUNDS 165

For the Central area, sampled later in the year, salinity showed a steeper rela-
tionship with elevation for mound tops than for mound edges (t = 2.11, df = 54,
P < 0.05). In addition, the relation for mound base salinity and elevation for the
Central area was significantly steeper than that for the North End area (t = 2.91,
df = 40, P < 0.01), which was similar in its general position within the estuary
system (Fig. 1). Mound height was also inversely correlated with mound top
salinity (r = —0.40, df = 80, t = 3.95, P < 0.001).

Digging activity or tunneling by various animals was noted on all but 19 of 82
mounds in the marsh and transition zone. Surface heaps of the valley pocket
gopher (Thomomys bottae) were found on 17 of 82 mounds, all with basal ele-
vations over 2.08 m MLLW. In all but one case, a mound in the Wheels area
with a basal elevation of 2.71 m MLLW, these heaps were many months old and
appeared to have been flooded. Burrow systems of the California ground squirrel
(Spermophilus beecheyi) were noted on 40 of 82 mounds, some as low as 1.90 m
MLLW. Of these, 27 showed signs of very recent tunneling. Digging by dogs was
noted on 12 mounds, usually in association with burrows of ground squirrels.
One mound contained a burrow occupied by a striped skunk (Mephitis mephitis).

The extent of surface disturbance by digging (mid-point of percent cover class)
by all animals together increased significantly with mound height (r = 0.31, df =
80, t = 2.88, P < 0.01). The area of bare soil, due both to digging and other
factors, also increased significantly with increasing mound height (r = 0.35, df =
80, t = 3.38, P < 0.001) and also with increasing elevation above MLLW (r =
Om2adte— (S05 t= 3:07. P= 0:01):

Distribution of Mound and Intermound Plants

In the marsh and transition zones, 34 species of vascular plants were recorded
on mounds (Fig. 4); 2 additional species, Spergularia villosa and Gasoul nodiflo-
rum, were found in the intermound areas sampled. Of the former, 11 were low
marsh species with distributions extending below the elevations sampled, and 23
were high marsh or upland species whose lower limits occurred within the ele-
vational range sampled. We sought prominent distributional breaks where several
(3-5) species reached their upper or lower limits, and noted these at 2.08-2.09,
2.27—2.28, and 2.42-2.43 m MLLW (Fig. 4). In the Ridge Top area, which was
over 4 m MLLW, 6 additional vascular plant species were encountered (Table
2). Only one high marsh species, of minor importance, occurred at this location.

Below 2.08 m MLLW, plant cover consisted mainly of low marsh species (Table
2, Fig. 2), most of which were non-succulents (Table 3). From 2.09 to 2.27 m
marsh succulents increased in cover and succulents of upland affinity became
prominent (Table 3). Between 2.28 and 2.42 m MLLW several other high marsh
and upland forbs and grasses appeared, along with three species of cacti and other
succulents, two coastal sage scrub subshrubs, and the perennial bunchgrass Spo-
robolus airoides (Table 2, Fig. 2). Over 77% coverage by succulent species was
recorded in this zone (Table 3). Several of the typical low marsh species disap-
peared, as well. Above 2.43 m MLLW in the transition zone, a number of ad-
ditional upland species appeared, notably the evergreen shrub Rhus integrifolia.
Plant cover at this elevation was still dominated by the glasswort, Salicornia
subterminalis, but the subshrub Eriogonum fasciculatum and the perennial bunch-
grass S. airoides reached their maximum abundance here. The mounds of the

166 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

SPECIES BASAL ELEVATION (m above MLLW)
1.90 2.00 2A0 > 2:20) 2:30! 240, 2:50) 260RR 20m oO
2 ee ee ee ee ee ress

nf
® ood 3

—™nHOs

ro)

5 w ‘<

nvOO >
rc Ay

o

scens

(ene)
30 3

nm So Ww
5

As]
>

Fig. 4. Elevational ranges above Mean Lower Low Water for plant species occurring on Mima
mounds at the Tijuana Estuary, San Diego County, California.

Ridge Top area, separated from other areas by an elevational gap of 1.41 m, were
dominated by the woody evergreen shrub Malosma laurina and by coastal sage
subshrubs (Tables 2, 3).

A small area spanning the elevational gap from transition zone to upland was
located at the extreme north end of the estuary, near the North End mound group
(Fig. 1, UT). Eleven mounds were present at this location, but they exhibited such
extensive disturbance that no quantitative sampling was attempted. The dominant
plants on these mounds were the woody evergreen shrubs Rhus integrifolia and
Simmondsia chinensis (5 mounds), Malosma laurina and S. chinensis (5 mounds)
and all three of these species together with Atriplex canescens (1 mound). Inter-
mound areas here contained a scattering of marsh species at the lowest levels and
a variety of coastal sage subshrubs and annual grasses and forbs at higher levels.
Active pocket gopher digging was evident at this location.

Mean total plant cover on mounds was greatest in the lowest marsh zone and
second greatest in the uppermost part of the transition zone (Table 2, Fig. 2).
Mean cover was lower in the intermediate parts of the transition zone, where
succulents reached their maximum abundance, and least on the upland site where
marsh species were absent. Total plant cover for individual mounds bore a sig-
nificant, but rather weak, direct correlation with mound area (r = 0.28, df = 80,
t= 2.56, F< 0105)

Similarity of mound vegetation to that of the adjacent intermound areas was
determined by computing the coefficient of community, C, using mid-range values
of percent cover classes for each species, by the equation

C = 2S/A + B, (2)

MIMA MARSH MOUNDS 167

Table 2. Mean percent cover of low marsh (LM), high marsh (HM), and upland (UP) plant species
on Mima mounds at the Tijuana River National Estuarine Sanctuary in relation to elevation in meters
above MLLW. Total cover is the sum of cover values for individual species, and includes overlap.

Elevation above MLLW (m)
1.87-2.08 2.09-2.27 2.28-2.42 2.43-2.88 4.29-4.51

Species N = 21 N= 14 N = 23 N = 24 N=4
Triglochin concinnum (LM) 0.02
Cuscuta salina (LM) 0.02
Salicornia virginica (LM) 0.83
Suaeda californica (LM) 0.93 0.04
Jaumea carnosa (LM) 0.31 0.02
Monanthochloe littoralis (LM) 60.36 45.75 6.30 3.96
Frankenia grandifolia (LM) Deval 5.25 5.15 2.42
Cressa truxillensis (LM) 1.26 0.28 0.17 0.81
Limonium californicum (LM) 0.64 1.46 0.41 0.98
Salicornia subterminalis (LM) 27.76 47.14 56.76 34.92
Distichlis spicata (LM) 1.17 0.82 10.30 11.73
Carpobrotus edulis (HM) 0.04
Lycium californicum (UP) 0.88 9.54 15.85 13.38
Atriplex watsonii (HM) 0.17 0.46 1.93 3.68
Opuntia prolifera (UP) 1.28 3.91 6.67
Opuntia littoralis (UP) 0.78
Suaeda fruticosa (HM) 0.02
Rumex crispus (UP) 0.20
Amblyopappus pusillus (HM) 0.04
Chrysanthemum coronarium (UP) 0.02
Bromus mollis (UP) 0.04 0.27
Bromus rubens (UP) 0.22 3.73
Dudleya edulis (UP) 0.02 0.33
Atriplex parishii (HM) 0.28 0.06
Ferocactus viridescens (UP) 0.04 0.04
Haplopappus venetus (UP) 3.30 7.44 0.88
Eriogonum fasciculatum (UP) 2.28 18.12 15.00
Sporobolus airoides (UP) 1.63 5.73 0.38
Rhus integrifolia (UP) 0.62
Sonchus asper (UP) 0.02
Solanum xantii (UP) 0.02
Avena sp. (UP) 0.02
Astragalus anemopsis (UP) 0.06 0.25
Atriplex semibaccata (UP) 0.44 0.12
Rhus laurina (UP) 56.25
Artemisia californica (UP) 26.20
Mirabilis laevis (UP) 3.00
Yucca schidigera (UP) 0.75
Lotus scoparius (UP) 0.12
Limonium sinuatum (HM) 0.12

Total 120.06 112.06 109.67 115.45 103.07

where S is the sum of shared percentages for species in common, and 4 and B
are the sums of percent:ges for all species of the mound and intermound areas,
respectively. Similarity values ranged from 0.06 to 0.98, and showed a significant
negative relationship with mound area (r = —0.31, df = 80, t = 2.92, P < 0.01).

168 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

Table 3. Mean percent cover of major life form groupings of vascular plants on Mima mounds at
the Tijuana River National Estuarine Sanctuary in relation to elevation in meters above MLLW.

Height above MLLW

Life form group 1.87-2.08 2.09-2.27 2.28-2.42 2.43-2.88 4.29-4.51
Low marsh species
Non-succulent monocots 61.55 46.57 16.60 15.69
Non-succulent dicots 27.63 6.99 S515 4.21
Succulents 29.83 47.18 56.78 34.92
High marsh/upland species
Succulents 0.88 10.86 20.62 20.42 0.75
Non-succulent forbs 0.17 0.46 2.47 4.26 0.49
Annual grasses 0.26 4.02
Perennial grasses 1.63 5.73 0.38
Coastal sage subshrubs 3.58 25.58 45.20
Evergreen woody shrubs 0.62 56.25

Total 120.60 112.06 109.67 115.45 103.07

Species-Area Relations of Mound and Intermound Plants

From 3 to 15 vascular plant species were recorded on mounds and 3 to 14
species in the adjacent intermound areas of same size. Within the marsh and
transition zones, number of mound species increased significantly with elevation
above MLLW (r = 0.38, df = 80, t = 9.69, P < 0.001) and with mound height
(r = 0.41, df = 80, t = 4.00, P < 0.001). Combined in the following multiple
regression equation, elevation in m above MLLW (X,) and mound height (X,)
gave a multiple correlation coefficient (R) of 0.76 with number of mound species
(S):

S=— —13.02) S662G) 1 ON Oxe (3)

Distinctive species-area relationships existed for numbers (S) of mound and
intermound species, as reflected in the coefficients of the power function,

S = bA’, (4)
in which A is mound area in m?. The slope, z, of the species-area relationship
was much steeper for mound than for intermound areas, differing little whether
all species or only perennial species were considered (Table 4). The species-area
slope was also much steeper for high marsh and upland species than for low marsh
species (Table 4). Slope values for marsh species were not significantly different
from 0. Because of the strong correlation of number of mound species with
elevation above MLLW,, the species-area relationship for high marsh and upland
species was examined by a two-factor regression that related number of species
to both area in m? (A) and elevation in m MLLW (X) by the equation,

Log S = log b + z,log A + z,log X. (5)
This analysis yielded the specific coefficients
Log S = —2.461 + 0.338 log A + 6.138X, (6)

which possessed a multiple correlation coefficient, R, of 0.78. In this equation,
the slope, z, of the species-area relation was reduced to 0.338, which was still

MIMA MARSH MOUNDS 169

Table 4. Coefficients of the species-area power function, S = bA?, in which S is number of species
and A area in m?, for mound and intermound areas at the Tijuana Estuary, San Diego County,
California.

Mounds Intermounds
b Z b Z
All vascular plant species 0.077 0.453 0.422 0.246
Vascular perennials only 0.098 0.429 0.451 0.218
High marsh/upland species —0.594 0.582 —0.327 0.391
Low marsh species 0.487 0.126 0.726 0.032

much steeper than for low marsh species on mounds. Efforts to incorporate other
factors, such as the area of soil distubance by animal digging, into this equation
gave no additional significant variables.

Discussion

Mima-type mounds at the Tijuana Estuary tend toward maximum development
at an intermediate point in the marsh—upland transition, roughly between 2.2 and
2.8 m MLLW. Their dispersion shows a significant tendency toward uniformity,
but spacing distance does not vary significantly with elevation above MLLW.
This geometry is consistent with the hypothesis that these mounds are formed by
geomyid pocket gophers and their spacing determined by the territorial behavior
of these animals. Their maximum development at an intermediate position in
the gradient matches the prediction of the pocket gopher hypothesis that mound
growth will be most active under conditions of intermediate limitation of use of
intermound soils by fossorial rodents (Gakahu and Cox 1984). At the lowest
levels, the dense silty-clay marsh soil (Zedler et al. 1980) and frequent flooding
severely limit the extent of tunneling into intermound areas when animals are
present; periods of unusually high water also eliminate the animals from these
mounds periodically. In the loamier, better drained soils of the upland areas, the
advantages of nest location in areas of elevated soil are weaker, and the tendency
for soil to be translocated consistently toward such spots much less. In the tran-
sition zone, however, flooding appears to be frequent enough to force animals to
locate their permanent nest chambers in elevated areas, yet infrequent enough to
allow extensive tunneling into surrounding lower areas, with a resultant strong
displacement of soil moundward.

The critical factor limiting occupancy of mounds by upland species is probably
soil salinity, which differed markedly between the tops and bases of even the
lowest mounds. Neuenschwander et al. (1979) also found salinity to be the limiting
factor for upland species at Bahia San Quintin, Baja California. The fact that
Mima mounds create low salinity conditions and permit upland species to extend
far beyond the point at which intermounds become dominated by marsh species
means that the maximum limits of penetration of upland species cannot be used
as a meaningful criterion for the boundary of this marsh area.

Several additional factors probably determine precisely which upland species
are present on mounds, however. The high species-area slope for high marsh and
upland species indicates that mound habitats are highly variable in the floras of
these species in both space and time. Extinctions of these species are probably

170 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

frequent on smaller mounds, where they are also slow to be offset by seedling
reestablishment. Low marsh species, on the other hand, show a very flat species-
area relationship both for mound and intermound areas. The fact that low marsh
species are mostly perennials that reproduce vegetatively (Neuenschwander et al.
1979, Zedler 1982) means that they are able to spread quickly and colonize suitable
habitat. Their presence on mounds is probably due mostly to vegetative expansion
from adjacent intermound areas.

The steep species-area relationships of high marsh and upland species (Table
4) probably reflect several factors. For intermounds, the steep relation (z = 0.391)
presumably reflects only the effects of environmental instability on the establish-
ment and survival of high marsh and upland species in a marginal habitat together
with factors affecting the dispersal of their seeds from source areas at higher
elevations. For mounds, the relationship probably reflects three additional factors:
1) differential elevational distribution of large and small mounds, 2) differential
environmental stability of large and small mounds, and 3) differential establish-
ment opportunities for plants on large and small mounds. The largest mounds
occur at an intermediate elevation where conditions near the mound base favor
low marsh species and those at the top high marsh and upland species; soil salinity
is probably the most important of these conditions. Secondly, variations in water
levels and salinity within the estuary, due to events such as winter rainfall, river
flooding, storm tidal surges, and occlusion of the estuary mouth, create highly
variable soil conditions, particularly on the smaller mounds that are most frequent
at lower elevations above MLLW. Alternation of hypo- and hypersaline conditions
may thus cause high turnover of species on small mounds, while the floras of
large mounds are less severely affected. Finally, as suggested by the fact that a
steep species-area relation was still shown for non-marsh mound species when
elevation was taken into account (Equation 5), establishment of species on larger
mounds is probably easier than on small mounds. The establishment of high
marsh and upland species that reproduce by seed is probably favored on large
mounds by the significantly greater percentages of bare soil and soil disturbed by
animal digging, and their persistance by the larger areas of suitable habitat.

The overwhelming dominance of succulents on mounds at intermediate ele-
vations (Table 3) suggests that these sites pose critical moisture challenges to both
marsh and upland species. Total plant cover was lower in this elevational range
than in other levels of the marsh-upland transition, perhaps as a reflection of this
challenge.

Mounds appeared to provide optimal habitat for a number of non-marsh species
that were largely absent from the nearby remnants of upland habitat. These in-
cluded the succulents Lycium californicum, Ferocactus viridescens, and Dudleya
edulis. Although all three of these species also occurred in intermound areas in
the uppermost transition zone, the most robust, and presumably oldest, individ-
uals were located on mounds. These species obviously possess a high degree of
tolerance of desiccation stress, but their strong association with marsh mounds
may also reflect the action of other factors that exclude them from nearby upland
areas. One such factor, perhaps more important in the past than at present, is
fire. All three species are absent from inland areas where fire is frequent, or are
confined to rocky sites and cliffs where fire effects are weak or absent.

Mima-type mounds at the Tijuana Estuary are heavily utilized by herbivorous

MIMA MARSH MOUNDS 171

mammals, including burrowing species such as the valley pocket gopher and
California ground squirrel and nonburrowing species such as the Audubon’s cot-
tontail (Sylvilagus auduboni) and black-tailed jackrabbit (Lepus californicus), both
of which maintain hiding places in the dense mound-top vegetation. Both the
pocket gopher and ground squirrel were apparently eliminated from mounds in
most of the transition zone and marsh proper by prolonged flooding during the
winter of 1982-1983. The ground squirrel is rapidly recolonizing mounds far into
the marsh, but recolonization by the pocket gopher will probably be slower because
of the limited above-ground exploratory behavior of this species. Both cottontails
and jackrabbits are abundant. The combined population of these herbivores is
considerable, and their above- and below-ground browsing activities may influ-
ence the composition of the marsh vegetation in significant fashion. Neuensch-
wander et al. (1979) also noted that the transition zone at Bahia San Quintin,
Baja California, received heavy use by upland animals.

The presence of Sporobolus airoides in mound and intermound sites at several
locations in the Tijuana Estuary raises the question of the vegetation type that
bordered the salt marsh in presettlement time. S. airoides forms several extensive
patches in the Reed Bed mound area, and occurs on and between mounds at
several locations on the ridge between eastern and western marsh areas. The
present vegetation of upland areas that do not appear to have experienced ex-
cavation or filling is now annual grassland with a mixture of coastal sage species.
The presence of this important perennial bunchgrass, a species typical of mod-
erately saline or alkaline prairie soils throughout the western United States, sug-
gests that the Tijuana Estuary may have been bordered by California Valley
Grassland prior to European settlement.

Acknowledgments

We wish to thank John Beezley, Christopher Gakahu, and Abby White for
assistance in mound field surveys and soil sampling, and Mark Dodero and Abby
White for analyses of soil salinity. We also thank Charles F. Cooper, Jochen
Kummerow, Richard J. Vogl, and an anonymous reviewer for comments and
criticism on earlier drafts of the manuscript. This study was partially supported
by Grant No. INT-8211521 from the U.S. National Science Foundation and Grant
No. R/NP-1-13B from the California Sea Grant College Program.

Literature Cited

Brown, H. C. 1951. Mound microrelief of the Columbia Plateau and adjacent areas. Unpublished
manuscript.

Clark, P. J. and F. C. Evans. 1954. Distance to nearest neighbor as a measure of spatial relationships
in populations. Ecology, 35:445-453.

Cox, G. W. 1984a. Mounds of mystery. Natural History, 93(6):26—45.

. 1984b. The distribution and origin of Mima mound grasslands in San Diego County, Cal-

ifornia. Ecology, 65:1397-1405.

. In press. Mima mounds as an indicator of the presettlement grassland-chaparral boundary

in San Diego County, California. American Midland Naturalist.

Dietz, R.S. 1945. The small mounds of the Gulf coastal plain. Science, 102:596—-597.

Eilers, H. P., A. Taylor, and W. Sanville. 1983. Vegetative delineation of coastal salt marsh bound-
aries. Environmental Management, 7:443-452.

Gakahu, C. G. and G. W. Cox. 1984. The occurrence and origin of Mima mound terrain in Kenya.
Afr. J. Ecol., 22:31-42.

172 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

Harvey, H. H., M. J. Kutilek, and K. M. DiVittorio. 1978. Determination of transition zone limits
in coastal California wetlands. Unpublished report to the U.S. Environmental Protection Agen-
cy, Washington, D.C.

Mclllwee, W. R. 1970. San Diego County coastal wetland inventory: Tijuana Slough. Unpublished
report, California Fish and Game Department, Sacramento, CA.

Munz, P. A. 1974. A flora of southern California. Univ. of California Press, Berkeley, CA.

Neuenschwander, L. F., T. H. Thorstad, Jr., and R. J. Vogl. 1979. The salt marsh and transitional
vegetation of Bahia San Quintin. Bull. S. Calif. Acad. Sci., 78:163-182.

Price, W. A. 1949. Pocket gophers as architects of Mima (pimple) mounds of the western United
States. Texas J. Sci., 1:1-17.

Richards, L. A. (Ed.) 1954. Diagnosis and improvement of saline and alkali soils. USDA Agricultural
Handbook No. 60, Washington, D.C.

Winfield, T. P. 1980. Dynamics of carbon and nitrogen in a southern California salt marsh. Ph.D.
Dissertation, Univ. of California, Riverside and San Diego State University, San Diego, CA.

Zar, J. H. 1984. Biostatistical analysis. Second Ed. Prentice-Hall, Inc., Englewood Cliffs, NJ.

Zedler, J. B. 1977. Salt marsh community structure in the Tijuana Estuary, California. Estuarine

and Coastal Marine Science, 5:39—-53.

1982. The ecology of southern California coastal salt marshes: a community profile. U.S.
Fish and Wildlife Service, Biological Services Program, Washington, D.C. FWS/OBS-81/54.
110 pp.

,and G. W. Cox. 1985. Characterizing wetland boundaries: a Pacific Coast example. Wetlands,
4:43-55.

, 1. Winfield, and P. Williams. 1980. Salt marsh productivity with natural and altered tidal
circulation. Oecologia (Berl.), 44:236-240.

Accepted for publication 18 February 1985.

Bull. Southern California Acad. Sci.
85(3), 1986, pp. 173-176
© Southern California Academy of Sciences, 1986

Research Notes

Inter- and Intralocular Distribution of Yucca Moth Larvae in
Yucca whip plei (Agavaceae)

Pollination of Yucca species is dependent upon the services of a moth in the
genus 7egeticula (Trelease 1893). In this mutualistic relationship the moth is in
turn dependent upon the flowers for oviposition sites. Previous studies have shown
that the number of moth larvae, and consequently seed loss, are highly variable,
among Yucca species (Keeley et al. 1984).

Yucca whipplei is a common species on shallow soils throughout the foothills
of southern California and the southern Sierra Nevada. This species is distinct
from other yuccas in that it is the only one with monocarpic reproduction, i.e.,
dying after reproduction, (at least in two of the five subspecies, see Haines 1941).
In addition, while the majority of Yucca species are pollinated by Tegeticula
yuccasella, Y. whipplei has a different species of moth (7. maculata) restricted to
it. A recent study of yucca moth seed predation on Yucca whipplei (Keeley et al.
1986 press) has shown that the distribution of moth larvae is highly variable, among
subspecies, among individuals within populations, and even among capsules with-
in an inflorescence.

This variation prompted an investigation into the distribution of moth larvae
within the fruits of Yucca whipplei. Two questions are addressed. 1) For fruits
with more than a single moth larva, are the larvae randomly distributed among
the six locules in the fruit? 2) Within a locule, are larvae randomly distributed
along the length of the capsule?

Methods

Mature capsules were collected from plants throughout the range of Yucca
whipplei (see Keeley et al. 1986 for localities). Capsules were opened and the
relative distribution of larvae within capsules was indicated by numbering each
of the six locules beginning with the first locule in which a larva was encountered
and noting this number for each subsequent larva. Distribution of larvae within
locules was described by measuring the distance from the base of the capsule to
the nearest seed destroyed.

Results

To determine if multiple larvae within a capsule were randomly distributed
between locules, the number of capsules with only two larvae were used to de-
termine the proportion of larvae in the adjacent locule vs. nonadjacent locules.
If larvae are randomly distributed in a capsule with six locules, the probability
that a second larva would be adjacent to an occupied locule would be 2/5, and
3/5 that it would occur in a non-adjacent locule. The Chi-square test for goodness
of fit showed no significant departure from expectation (P > 0.05, N = 180).
Thus, moth larvae were randomly distributed among locules within a capsule.

The distribution of larvae within a single locule, however, was highly skewed.

173

174 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

Y. whipplei

No. of larvae
Ol
(@)

Onl 2A Sioa ao onlOn . 15 20
Distance from capsule base (mm)

Fig. 1. Distribution of Tegeticula maculata larvae within locules of Yucca whipplei fruits (sample
size: N = 2912 larvae).

The Chi-square tests for goodness of fit showed a highly significant departure
from normality (P < 0.01). As seen in Fig. 1, the vast majority of moth larvae
in Yucca whipplei fruits were situated at the base of the capsule.

Due to this very skewed distribution the question arose as to whether this was
a characteristic of Yucca species in general. The intralocular distribution of larvae
was determined from data collected previously (Keeley et al. 1984) for nine other
Yucca species from the southwestern U.S. These included species with dehiscent
capsular fruits (Y. angustissima, Y. elata, Y. glauca, Y. reverchoni) and species
with indehiscent baccate fruits (Y. baccata, Y. brevifolia, Y. schidigera, Y. schottii,
Y. torreyi). Fig. 2 shows the intralocular distribution of larvae for one of these

2 Y. schidigera

No. of larvae
O

4812 20 28 36 44 48 56 62 70 78
Distance from fruit base (mm)

Fig.2. Distribution of Tegeticula yuccasella larvae within locules of Yucca schidigera fruits (sample
size: N = 82 larvae).

RESEARCH NOTES 175

Fig. 3. Open capsules of Yucca whipplei illustrating the typical site of oviposition (point of con-
striction of the capsule), larval chamber (arrow) consisting of destroyed seeds glued together and most
common position of the larval chamber within the capsule by Teresa Montygierd-Loyba.

species which closely resembles the pattern observed for the other eight species.
For all nine of these species the distribution of larvae along the length of the
capsule did not depart significantly from normality (P > 0.05).

Discussion

Particularly intriguing is the distribution of larvae within the locules of Yucca
whipplei (Fig. 1). The highly skewed distribution of moth larvae in Y. whipplei
capsules is in marked contrast to the intralocular distributions of the larvae in
nine other Yucca species from the southwestern U.S. These other nine species
represent most of the range of variation in the genus with respect to fruit char-
acteristics.

There are characteristics unique to Yucca whipplei which could explain the
highly skewed intralocular distribution of larvae in this species but not in others.
For example differences in floral structure could affect oviposition location; in Y.
whipplei flowers, the stamens tend to spread whereas in other yuccas they are
erect or appressed to the carpel. Behavioural differences in oviposition site and/
or orientation of larvae may be involved as Y. whipplei has a unique Tegeticula
species. One characteristic of T. maculata that could be involved is the fact that
the larvae of this species produce a silk that glues together the remains of the
seeds they eat and thus forms a chamber (Fig. 3). Experiments have shown that
if these chambers, rather than being situated at the base of the capsule, are arti-
ficially placed higher up, they act as a plug and block the dispersal of all seeds
situated beneath them (A. Meyers and J. Keeley, unpublished data). This is due

176 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

to the fact that the upright capsules do not open completely and, due to a rudi-
mentary false septum the seeds must escape upwards (McKelvey 1947). The
positioning of larval chambers at the base of the capsules is apparently due to the
orientation of the feeding larva since the female typically oviposits near the middle,
as evidenced by direct observation (Trelease 1893) or presence of apparent ovi-
position scars (J. Keeley personal observation).

In other Yucca species the moth larvae may not create a seed dispersal problem
for two reasons. In the baccate-fruited species the fruits are indehiscent and thus
the position of the larvae should not have any effect on ultimate seed dispersal.
In the other Yucca species with dehiscent capsules, the yucca moth (Tegeticula
yuccasella) larvae are larger and do not produce a silk which binds the seed remains
together into a chamber; thus less seed debris is left behind and since it is not
glued together it disperses as well as or better than good seeds.

It is concluded that non-random orientation of moth larvae within the locules
of Yucca whipplei fruits has been selected for as a means of enhancing the dispersal
of seeds not consumed by the yucca moth.

Literature Cited

Haines, L. 1941. Variation in Yucca whipplei. Madrono, 6:33-45.

Keeley, J. E., S. C. Keeley, and D. A. Ikeda. 1986. Seed predation by yucca moths on semelparous,
iteroparous, and vegetatively reproducing subspecies of Yucca whipplei (Agavaceae). Am. Midl.
Nat., 115:1-9.

Keeley, J. E., S. C. Keeley, C. C. Swift, and J. Lee. 1984. Seed predation due to the Yucca-moth
symbiosis. Am. Mid]. Nat., 112:187-191.

McKelvey, S. D. 1947. Yuccas of southwestern United States, Pt. 2. Arn. Arbor., Harvard Univ.,
Jamaica Plain, Mass.

Trelease, W. 1893. Further studies of yuccas and their pollination. Ann. Rep. Mo. Bot. Gard., 4:
181-226.

Accepted for publication 1 April 1985.

Jon E. Keeley, Department of Biology, Occidental College, Los Angeles, California
90041

Bull. Southern California Acad. Sci.
85(3), 1986, pp. 177-179
© Southern California Academy of Sciences, 1986

A Note on the Trigger Pollination Mechanism in the Camphor
Weed (Trichostema lanceolatum) as Related to Pollinator
Weight and Behavior

The camphor weed or vinegar weed (Trichostema lanceolatum Benth.) is one
of the North American members of the mint family (Lamiaceae). The upper
portion of its light blue corolla forms a thin tube which contains the anthers and
style (Fig. 1A). When a foraging insect alights on the lower lobes of the corolla,
and inserts its mouth parts into the nectar-containing lower section of the same
tube, the narrow corolla portion above is straightened and snaps rapidly downward
brushing pollen onto the back of the visiting insect (Fig. 1B). After the pollinating
insect leaves, the narrow tube flips back up into its original position. This action
has been previously described by T. Spira (1978) who attributed the trigger effect
to the weight of the pollinating insect on the corolla lobes. In this note an attempt
is made to determine if some insect behavior (in addition to its weight or even
instead of its weight) may be responsible for flexing the tube in certain circum-
stances.

Relationships where a flower dusts pollen onto the dorsal surface of a visiting
insect are called “‘nototribic”’ and are not uncommon in other species of the genus
Trichostema as noted by T. Spira (1978, 1980) or other genera in the mint family —
see Proctor and Yeo (1972) and L. W. Macior (1974). Although he described these
and other features of camphor weed, T. Spira reported no study of this flower’s
action in response to weight or behavior of its pollinating species. Such data may
be valuable in understanding the relationships between insects and pollination in
this species and may shed light on the pollination process in other nototribic taxa
as well.

In order to estimate the amount of pollinator weight alone required on the
lower corolla lobes to trip the mechanism, the central lower lobe of the corolla
was pierced with a staple, from which additional staples could then be hung. These
were Bates Standard staples and each had a weight of 33 mg. From tests performed
on 60 different Trichostema lanceolatum flowers in 1983 and 1984 the following
was recorded: 23% of the 60 flowers were triggered by only one staple (33 mg),
58% by two staples (66 mg), and the remaining 19% by three staples (99 mg) so
that all flowers tested required less than 99 mg to trigger their pollen apparati.

These data should be compared to the weights of the various bees and one of
the moth species which according to Spira (1980) and Howe (1985) routinely visit
the camphor weed flowers. Each insect was collected, stunned in an ethyl acetate
jar, and then immediately weighed on a Sartorius balance. The mean weights and
standard deviations for each species were as follows where n = 6: bumblebees
(Bombus sonorus Say) 223 + 21 mg, honey bees (Apis mellifera L.) 90 + 4.5 mg,
female mason bees (Anthophora urbana Cresson) 98.3 + 13 mg, and the woodland
skipper moth (Ochlodes sylvanoides sylvanoides Boisduval) 87.8 + 7.2 mg.

Of the bees which regularly pollinate 7. Janceolatum, Howe (1985), all Bombus
sonorus individuals are obviously heavy enough to trigger this mechanism by
weight alone. The honey bees, the female mason bees, and the woodland skipper

177

178 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

Fig. 1A. Trichostema lanceolatum (camphor weed) flower before the bumble bee (Bombus sonorus)
has alighted in quest of nectar. Drawing by Ross Marshall. Fig. 1B. T. Janceolatum flowers after the
bumble bee has landed. When the bee alights, the weight of its body and the action of its mouth parts
in the floral tube cause the anthers and the style to flex rapidly downward, dusting the pollen onto
the dorsal surface of the bee’s abdomen. Drawing by Ross Marshall.

moths would also be able to flex most camphor weed flowers by their weight on
the corolla lobes alone. The two mason bee males collected, however, had weights
of only 59 and 53 mg respectively and were thus too light to have tripped many
of the flowers tested. Since 23%—8 1% of the flowers were tripped with 1-2 staples,
perhaps some flowers were tripped by weight and other heavier flowers are un-
available to mason bee males.

An attempt was made to see if a light insect such as these 4. urbana males might
be triggering the camphor weed flower mechanism by simply thrusting its mouth
parts into the curved corolla—thereby mechanically straightening out the S-shaped
upper tube. Upon intromission into the corolla tube, each of the following objects
would routinely flex the flower tube without any weight on the lower corolla lobes:
bristle of “soft” toothbrush, polyolefin bristle of two-inch paint brush, the mouth
parts of honey bee heads held with tweezers. Since these objects can be used to
flex the flower mechanism, it is possible that the action of insect mouth parts
and/or tongues normally plays an important role in this process.

It has thus been shown that there is a close relationship between the force
required to flex the pollinating apparatus of camphor weed and the actual weight
ofits key pollinating species. The weight of most of its pollinating insects is slightly
greater than the minimum force required to spring the pollinating mechanism.
While other workers have assumed that it is simply the weight of the pollinator
that is solely responsible for flexing the corolla, it is shown here that some smaller
insects might trigger the flower by thrusting their mouth parts into the corolla

RESEARCH NOTES N72)

tube and thereby supplying some or all of the force required to bend the pollinating
tube of the flower.

Acknowledgments

I thank Roy Snelling and Julian Donahue of the Los Angeles County Museum
of Natural History for identifying the bees and moths, respectively. I am grateful
to Ross Marshall for preparing the figure.

Literature Cited

Howe, G. 1985. A census of pollinator bees in large and small populations of the camphor weed
(Trichostema lanceolatum). Crossosoma 11(2):1-7.

Macior, L. W. 1974. Behavioral aspects of coadaptations between flowers and insect pollinators.
Ann. Mo. Bot. Gard. 61:760-769.

Proctor, M., and P. Yeo. 1972. The pollination of flowers. Taplinger Publ. Co., New York.

Spira, T. 1978. Floral parameters associated with breeding system and pollinator type in Trichostema

(Labiatae). Master of Arts Thesis, Cal. State Univ. Chico.

1980. Floral parameters, breeding system and pollinator type in Trichostema (Labiatae).
Amer. J. Bot. 67(3):278-284.

Accepted for publication 11 January 1985.

George F. Howe, Division of Natural Science and Mathematics, The Master’s
College, Newhall, California 91322

Bull. Southern California Acad. Sci.
85(3), 1986, pp. 180-181
© Souther California Academy of Sciences, 1986

Some Observations of the Alpheid Shrimp Betaeus setosus Hart
with Its Host, Pachycheles rudis Stimpson

Shrimps of the family Alpheidae are well known for their symbiotic relation-
ships with other species of crustaceans, molluscs, echinoderms, or fish (Lassig
1977; Karplus et al. 1972: Ache and Davenport 1972). Along the west coast of
North America, Betaeus harfordi is found in the mantle cavities of abalones
(Haliotus spp.), while B. macginitieae occurs in pairs under sea urchins of the
genus Strongylocentrotus and B. harrimani is associated with the burrows of the
ghost shrimps Callianassa and Upogebia (Hart 1964).

Betaeus setosus, a small species ranging from British Columbia to central Cal-
ifornia, has been considered an outer coast species as it has not previously been
recorded from sheltered waters in British Columbia or Washington (Butler 1980;
Kozloff 1974). Collections made at several sites within Puget Sound, however,
have shown that this species is fairly common in suitable habitats and is invariably
associated with the anomuran crab Pachycheles rudis. Hart (1964) noted that B.
setosus had been found in cavities under the holdfasts of kelp or eelgrass roots
with Pachycheles, but considered the species to be free-living.

In Puget Sound, Pachycheles rudis is found in relatively high current areas, with
male-female pairs frequently inhabiting empty giant barnacles (Balanus nubilis)
or clinging to the underside of large rocks. In each case only one shrimp occurred
with each pair of Pachycheles, where it was found lying on its side underneath
the abdomen of one of the crabs.

In order to study this relationship, the shrimps and their hosts were placed in
an aquarium in which sections of giant barnacle shells had been attached to the
glass. A “‘blind’’ was made to allow close-up observations without disturbing the
inhabitants. The Betaeus remained in constant contact with their hosts throughout
the day and left the shells at night, possibly to forage. Shrimps were sometimes
pushed away by their hosts upon returning to the shell, but would persist and
eventually get under the crab from the side or behind. No apparent signals were
visible in either the shrimp or the host crab, as have been reported for some
tropical species (Vannini 1985; Karplus et al. 1972); however the hurried return
of the shrimp (induced by artificial light) may differ from their return under natural
lighting conditions. Once in position, even a great deal of movement on the part
of the shrimp failed to elicit any visible response from the crab.

Betaeus setosus has been found in pairs (Hart 1964), however, the captive
specimens were quite territorial and would not tolerate another specimen in their
shell. The large chelipeds were used to push other shrimps away, but in no case
were they used to pinch the intruder. Unfortunately, the sexes of the specimens
involved in these encounters was not known. Betaeus was never observed feeding
inside the shells; it is suspected that the primary advantage of the relationship is
one of protection for the shrimp, and that feeding is done outside the shell at
night. Pachycheles is strictly a filter feeder and herbivore and represents no threat
to Betaeus, while its large, powerful chelae may effectively discourage predators.

1380

RESEARCH NOTES 181

Whether Pachycheles derives any benefit from this relationship or merely tolerates
it is not yet known.

Betaeus harfordi has been demonstrated to locate its host by chemosensory
means, while B. macginitieae uses both visual and chemosensory cues (Ache and
Case 1969; Ache and Davenport 1972). Visual cues seem unlikely in the case of
B. setosus due to the retiring habits of its host, but further investigation is needed
to determine how Pachycheles is located and whether a shrimp is always associated
with a specific host pair.

Literature Cited

Ache, B. W.,and J. Case. 1969. An analysis of antennular chemoreception in two commensal shrimps

of the genus Betaeus. Physiol. Zool., 42:361-371.

, and D. Davenport. 1972. The sensory basis of host recognition by symbiotic shrimps, genus

Betaeus. Biol. Bull., 143:94-111.

Butler, T. H. 1980. Shrimps of the Pacific coast of Canada. Canadian Bulletins of Fisheries and
Aquatic Sciences, 202:280 pp.

Hart, J. F. L. 1964. Shrimps of the genus Betaeus on the Pacific coast of North America with
descriptions of three new species. Proc. U.S. Nat. Mus., 115:431-466.

Karplus, I., Tsurnamal, M., and R. Szlep. 1972. Associative behavior of the fish Cryptocentrus
cryptocentrus (Gobiidae) and the pistol shrimp Alpheus djiboutensis (Alpheidae) in artificial
burrows. Mar. Biol., 15:95-104.

Kozloff, E. N. 1974. Keys to the marine invertebrates of Puget Sound, the San Juan Archipelago,
and adjacent regions. Univ. Washington Press, 226 pp.

Lassig, B. R. 1977. Communication and coexistance in a coral community. Mar. Biol. 42:85-92.

Vannini, M. 1985. A shrimp that speaks crab-ese. J. Crust. Biol., 5:160—-167.

Accepted for publication 20 September 1985.

Gregory C. Jensen, School of Fisheries, WH-10, University of Washington, Se-
attle, Washington 98195

INDEX TO VOLUME 85

Blake, James A.: A New Species of Boccardia (Polychaeta: Spionidae) from the
Galapagos Islands and a Redescription of Boccardia basilaria Hartman from
Southern California, 16

Bodkin, James I., see Glenn R. VanBlaricom

Cicindela formosa rutilovirescens, n. subsp., 139

Cicindela scutellaris yampae, n. subsp., 139

Coan, Eugene V., see Barry Roth

Cornett, James E., Jon Stewart, and Theo Glenn: Washingtoni robusta Naturalized
in Southeastern California, 56

Cox, George W., and Joy B. Zedler: The Influence of Mima Mounds on Vegeta-
tion Patterns in the Tijuana Estuary Salt Marsh, San Diego County, Cali-
fornia, 158

Coyer, James A.: The Mollusk Assemblage Associated with Fronds of Giant Kelp
(Macrocystis pyrifera) off Santa Catalina Island, California, 129

Dorsey, John H., see Andrew L. Lissner

Gilbertson, Larry, see Thomas P. O’Farrell

Glenn, Theo, see James E. Cornett

Goldberg, Stephen R., and Marie C. Pizzorno: Notes on the Spawning Cycles of
Labrisomus philippii (Labrisomidae) and Trachinotus paitensis (Carangidae)
from Peru, 126

Harrold, Christopher, see Glenn R. VanBlaricom

Heyning, John E.: First Record of the Dolphin Steno bredanensis from the Gulf
of California, 62

Howe, George F.: A Note on the Trigger Pollination Mechanism in the Camphor
Weed (Trichostema lanceolatum) as Related to Pollinator Weight and Be-
havior, 177

Jensen, Gregory C.: Some Observations of the Alpheid Shrimp Betaeua setosus
Hart with Its Host, Pachycheles rudis Stimpson, 180

Keeley, Jon E.: Inter- and Intraocular Distribution of Yucca Moth Larvae in
Yucca whipplei (Agavaceae), 173

Lees, Dennis C.: Marine Hydroid Assemblages in Soft-Bottom Habitats on the
Hueneme Shelf off Southern California, and Factors Influencing Hydroid
Distribution, 102

Lissner, Andrew L. and John H. Dorsey: Deep-Water Biological Assemblages of
a Hard-Bottom Bank-Ridge Complex of the Southern California Continental
Borderland, 87

O’Farrell, Thomas P., and Larry Gilbertson: Ecology of the Desert Kit Fox Vulpes
macrotis arsipus in the Mohave Desert of Southern California, 1

INDEX TO VOLUME 85 183
Pizzorno, Marie C., see Stephen R. Goldberg

Reed, Daniel C., see Glenn R. VanBlaricom

Roth, Barry: Notes on Three European Land Mollusks Introduced to Califor-
nia, 22

Roth, Barry and Eugene V. Coan: Rediscovery and Identity of the Holotype of
Helminthoglypta diabloensis (Cooper) (Gastropoda: Pulmonata), 65

Rumpp, Norman L.: Two New Tiger Beetles of the Genus Cicindela from Western
United States (Cicindelidae: Coleoptera), 139

Single, Jeffrey R.: A Priori Estimation of Sample Size and Number of Variables
for Principal Components Analyses, 123

Snelling, Roy R.: The Taxonomic Status of Two North American Lithurge (Hy-
menoptera: Megachilicae), 29

Stebbins, Timothy D.: Density, Distribution, and Feeding of the Marine Snail
Norrisia norrisi (Mollusca: Gastropoda) on the Kelp Macrocystis pyrifera
(Phaeophyta: Laminariales), 60

Stewart, Jon, see James E. Cornett

VanBlaricom, Glenn R., Daniel C. Reed, Christopher Harrold and James I. Bod-
kin: A Sublittoral Population of pleurophycus gardneri Stechell and Saunders
1900 (Phaeophyceae: Laminariaceae) in Central California, 102

Weintraub, Joel D.: Coyote Diets Five Years Later, at Cuyamaca State Park, 152
Wicksten, Mary K.: A New Species of Heptacarpus from California, with a Re-
description of Heptacarpus palpator (Owen) (Caridea: Hippolytidae), 35

Zedler, Joy B.: Catastrophic Flooding and Distributional Pattern of Pacific Cord-
grass (Spartina foliosa Trin.), 74
Zedler, Joy B., see George W. Cox

Announcing the
THIRD CALIFORNIA ISLANDS SYMPOSIUM
2-6 March, 1987 @ Santa Barbara, California

Hosted by: Santa Barbara Museum of Natural History, Santa Barbara Botanic Garden, Southern
California Academy of Sciences.

Old Timers’ Brunch opens the program on Monday, March 2; contributed paper sessions during the
five-day event include sections on History and Resource Management; Birds, Fishes, and Marine
Botany; Anthropology and Oceanography; Marine Invertebrates and Terrestrial Botany; Terrestrial
Invertebrates and Marine Mammals; Terrestrial Vertebrates; Geology and Geography. Between ses-
sions, you may want to examine the Channel Islands Archive at the Museum of Natural History and
the Channel Islands Collections at both the Museum and the Santa Barbara Botanic Garden. These
two institutions are currently the principal archival facilities in California for voucher collections and
specimens which document scientific research projects and environmental studies on the islands.

Early registration is advised. In Santa Barbara, Dr. F. G. Hochberg is Program Coordinator for the
event and all inquiries should be addressed to him at the Santa Barbara Museum of Natural History,
2559 Puesta del Sol Road, Santa Barbara, CA 93105.

BIOLOGY OF THE WHITE SHARK
Memoir #9

Papers from a symposium held by the Southern California Academy of Sciences.

Contents include material on shark distribution, ecology, age and growth, visual system,
hematology, cardiac morphology, feeding, temperature, heat production and exchange,
and attack behavior.

Seni 5 2) ORDER FORM . 2) =) eee

Make check or money order payable to Southern California Academy of Sciences, and
mail to:
SOUTHERN CALIFORNIA ACADEMY OF SCIENCES

900 Exposition Blvd.
Los Angeles, CA 90007

I would like to order
price of $22.50 per copy.

copies of the “Biology of the White Shark” papers at the

Enclosed is my check for $_________. Please ship my order to me as follows:
Name

Address

Cifyaemitaiev as “erdnies be ss DEAS ee] ee Sinica eee

g ‘nT

INSTRUCTIONS FOR AUTHORS

The BULLETIN is published three times each year (April, August, and November) and includes articles in English
in any field of science with an emphasis on the southern California area. Manuscripts submitted for publication
should contain results of original research, embrace sound principles of scientific investigation, and present data in
a clear and concise manner. The current AIBS Style Manual for Biological Journals is recommended as a guide for
contributors. Consult also recent issues of the BULLETIN.

MANUSCRIPT PREPARATION

The author should submit at /east two additional copies with the original, on 82 x 11 opaque, nonerasable paper,
double spacing the entire manuscript. Do not break words at right-hand margin anywhere in the manuscript. Foot-
notes should be avoided. Manuscripts which do not conform to the style of the BULLETIN will be returned to the
author.

An abstract summarizing in concise terms the methods, findings, and implications discussed in the paper must
accompany a feature article. Abstract should not exceed 100 words.

A feature article comprises approximately five to thirty typewritten pages. Papers should usually be divided into
the following sections: abstract, introduction, methods, results, discussion and conclusions, acknowledgments,
literature cited, tables, figure legend page, and figures. Avoid using more than two levels of subheadings.

A research note is usually one to six typewritten pages and rarely utilizes subheadings. Consult a recent issue of
the BULLETIN for the format of notes. Abstracts are not used for notes.

Abbreviations: Use of abbreviations and symbols can be determined by inspection of a recent issue of the
BULLETIN. Omit periods after standard abbreviations: 1.2 mm, 2 km, 30 cm, but Figs. 1-2. Use numerals before
units of measurements: 5 ml, but nine spines (10 or numbers above, such as 13 spines). The metric system of
weights and measurements should be used wherever possible.

Taxonomic procedures: Authors are advised to adhere to the taxonomic procedures as outlined in the International
Code of Botanical Nomenclature (Lawjouw et al. 1956), the International Code of Nomenclature of Bacteria and
Viruses (Buchanan et al. 1958), and the International Code of Zoological Nomenclature (Stoll et al. 1961). Special
attention should be given to the description of new taxa, designation of holotype, etc. Reference to new taxa in
titles and abstracts should be avoided.

The literature cited: Entries for books and articles should take these forms.

MeWilliams, K. L. 1970. Insect mimicry. Academic Press, vii + 326 pp.

Holmes, T. Jr., and S. Speak. 1971. Reproductive biology of Myotis lucifugus. J. Mamm., 54:452-458.

Brattstrom, B. H. 1969. The Condor in California. Pp. 369-382 in Vertebrates of California. (S. E. Payne, ed.),

Univ. California Press, xii + 635 pp.

Tables should not repeat data in figures (line drawings, graphs, or black and white photographs) or contained in
the text. The author must provide numbers and short legends for tables and figures and place reference to each of
them in the text. Each table with legend must be on a separate sheet of paper. All figure legends should be placed
together on a separate sheet. Illustrations and lettering thereon should be of sufficient size and clarity to permit
reduction to standard page size; ordinarily they should not exceed 81% by 11 inches in size and after final reduction
lettering must equal or exceed the size of the typeset. All half-tone illustrations will have light screen (grey)
backgrounds. Special handling such as dropout half-tones, special screens, etc., must be requested by and will be
charged to authors. As changes may be required after review, the authors should retain the original figures in their
files until acceptance of the manuscript for publication.

Assemble the manuscript as follows: cover page (with title, authors’ names and addresses), abstract, introduction,
methods, results, discussion, acknowledgements, literature cited, appendices, tables, figure legends, and figures.

A cover illustration pertaining to an article in the issue or one of general scientific interest will be printed on the
cover of each issue. Such illustrations along with a brief caption should be sent to the Editor for review.

PROCEDURE

All manuscripts should be submitted to the Technical Editor, Jon E. Keeley, Biology Department, Occidental
College, 1600 Campus Road, Los Angeles, California 90041. Evaluation of a paper submitted to the BULLETIN
begins with a critical reading by the Editor; several referees also check the paper for scientific content, originality,
and clarity of presentation. Judgments as to the acceptability of the paper and suggestions for enhancing it are sent
to the author at which time he or she may be requested to rework portions of the paper considering these recom-
mendations. The paper then is resubmitted and may be re-evaluated before final acceptance. Authors are requested
to submit the names, addresses and specialities of three persons who are capable of reviewing the manuscript.

Proof: The galley proof and manuscript, as well as reprint order blanks, will be sent to the author. He or she
should promptly and carefully read the proof sheets for errors and omissions in text, tables, illustrations, legends,
and bibliographical references. He or she marks corrections on the galley (copy editing and proof procedures in
Style Manual) and promptly returns both galley and manuscript to the Editor. Manuscripts and original illustrations
will not be returned unless requested at this time. All changes in galley proof attributable to the author (misspellings,
inconsistent abbreviations, deviations from style, etc.) will be charged to the author. Reprint orders are placed with

’ the printer, not the Editor.

CONTENTS

The Mollusk Assemblage Associated with Fronds of Giant Kelp (Macro-
cystis pyrifera) off Santa Catalina Island, California. By James A.
COVEN SE RERISE TER DR H A AN5 7 SS NRT ante (le

Two New Tiger Beetles of the eae iS ndela pee Wesieun United States
(Cicindelidae: Coleoptera). By Norman 1. Rumpp ass

Coyote Diets, Five Years Later, at Cuyamaca Rancho State Park. By Joel
D. Weintratih 220 05. AEE a af

The Influence of Mima Mounds on Vegetation Patterns in the Tijuana
Estuary Salt Marsh, San Diego County, California. By George W. Cox
and Joy B. Zedler 22.20 00.005 SN

Research Notes

Inter- and Intralocular Distribution of Yucca Moth Larvae in Yucca whipplei (Agavaceae).
By Jon: EB. Keeley: 2 0 SE

A Note on the Trigger Pollination Mechanism in the Camphor Weed (Trichostema lanceolatum)
as Related to Pollinator Weight and Behavior. By George F. Howe

Some Observations of the Alpheid Shrimp Betaeus setosus Hart with Its Host, Pachycheles
rudis Stimpson: | By ‘Gregory C. Jensen: 222 8 ee

Dp a a a

LIBRARY

APR ~ 2 1996

NEW YoR
BOTANICAL Ga GARDEN

COVER: The kelp forest at Santa Catalina Island, California (USA) where 41 species of mollusks

fy

139

152

158

173

177

180
182

were associated with the kelp fronds from June 1975 through December 1976. James A.

Coyer

Se ee

IES SE ee a ee