| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Behavioral Ecology Vol. 13 No. 4: 467-471
© 2002 International Society for Behavioral Ecology
Previous agonistic experience determines both foraging behavior and territoriality in the limpet Lottia gigantea (Sowerby)
University of Oregon, Oregon Institute of Marine Biology, PO Box 5389, Charleston, OR 97420, USA
Address correspondence to A.L. Shanks. E-mail: ashanks{at}oimb.uoregon.edu .
Received 12 March 2001; revised 21 August 2001; accepted 6 September 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Lottia gigantea, the owl limpet, is an algal gardener. Territorial individuals actively defend gardens, which consist of cleared areas in the intertidal zone upon which a thick algal film develops and upon which the territory holder feeds. Smaller, nonterritory holders raid these gardens and graze the algal film. Territorial individuals must obtain an adequate ration without compromising the productivity of the garden. In contrast, a nonterritory holder grazing on another limpet's territory must obtain an adequate ration before it contacts the territory holder and is driven off. In the laboratory, replicate sets of 10 limpets were trained to behave territorially and nonterritorially. Training mimicked natural encounters between territorial and nonterritorial L. gigantea. Limpets given territorial training left significantly (t = -4.92, df = 9, p =.00041) more algal cover behind when grazing (on average 71%) than did limpets trained to be nonterritorial (on average 50%). Territorial limpets seldom grazed over the same area more than once (4% of the grazed area). In contrast, nonterritorial limpets frequently foraged over an area more than once; of the area grazed, 20% had been visited more than once. Previous agonistic experience determines both territorial behavior and foraging strategies, two of the critical behaviors necessary for successful gardening behavior. Nonterritorial limpets maximize consumption per unit area, whereas territorial limpets appear to forage prudently, leaving a significantly greater proportion of the plant biomass behind.
Key words: foraging, gardening, grazing, limpets, Lottia gigantea, territoriality.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
In a feeding (nonbreeding) territory, the resident controls or restricts access to food resources within some area (Wolf, 1970
). The energetic
economics of feeding territories has been actively studied in a variety of
species, particularly in birds (e.g.,
Carpenter and MacMillen, 1976;
Carpenter et al., 1983;
Davies, 1976;
Gill and Wolf, 1975). In these
territorial systems the utilization of the resource does not affect the
subsequent production of the resource; a hummingbird's consumption of the
nectar in a flower does not decrease the future production of nectar.
In some territorial systems the resident not only controls the food
resource within its territory, but also modifies the plant assemblage,
selectively enhancing particular plant species and increasing the food value
of the plants for the grazer (Branch et
al., 1992). This form of territoriality is known as
"gardening." There are a number of examples of gardening species
within aquatic habitats (reviewed in Branch
et al., 1992), but gardening is apparently rare in terrestrial
habitats. In a gardening system, the utilization of the resource by the
territory holder does alter the subsequent production of the resource. There
appear to be two keys to successful gardening behavior. First, the gardener
must modify its grazing to "prudent" levels. At intermediate
grazing pressure, many plants display maximal productivity, and the new growth
tends to be more nutritious (reviewed in
Branch et al., 1992). In the
long term, prudent resource utilization may lead to a greater energy gain for
the territory holder than aggressive, short-term resource exploitation. For
the gardener to benefit from this prudent behavior, it must defend the garden
from its competitors, and the gardener must also be territorial. In contrast,
an intruder on a gardener's territory gains nothing from prudent foraging, and
one would predict that intruders would forage to maximize their short-term
gain. In this study I compared the resource utilization of territorial
individuals of a gardening species, the intertidal limpet Lottia
gigantea, to the resource utilization by nonterritorial individuals.
Lottia gigantea, the owl limpet, is found on exposed rocky shores
from Neah Bay, Washington, USA, to Bahia de Tortuga, Baja California, Mexico
(Morris et al., 1980). L.
gigantea, a large limpet growing up to 10 cm in length, can be an
important space holder in the intertidal zone (Stimson,
1970,
1973;
Wright, 1985). The ecology of
L. gigantea is complex. Individuals can be either nonterritorial or
territorial (Stimson, 1970,
1973;
Wright, 1985). Territorial
individuals defend a clearing on the rock surface upon which a dense algal mat
grows; they are gardeners (Branch et al.,
1992). Smaller individuals are generally male and nonterritory
holders. With increased size and, especially, after the acquisition of a
territory, a male limpet changes sex to female
(Lindberg and Wright, 1985;
Wright, 1985,
1989). Most territory
holders/gardeners are, hence, large females
(Wright, 1985).
Territorial behavior in L. gigantea is ritualized
(Wright, 1982). Upon
contacting an intruding limpet, a territory holder bunches up its foot, crawls
rapidly forward (1.4 mm/s compared to 0.05 mm/s average speed), and rams its
shell under the shell of the intruder. Shell ramming can dislodge an intruder
from the rocks, causing the intruder to be washed away by the waves. The
latency of the territorial response is on average 57 s. Upon contacting a
territory holder, the intruding limpet generally draws back from the point of
contact, turns, and crawls quickly away (average speed 1.1 mm/s). The response
time for the evasive behavior is only 22 s. Thus, in most territorial
encounters, the intruding limpet has begun its escape and is some distance
away from the territory holder by the time the shell-ramming territorial
behavior commences (Wright,
1982). This set of behaviors results in less physical contact and,
generally, less danger for the intruder (few are actually knocked off the
rock); this is the essence of ritualization
(Wilson, 1975).
Stimson (1970,
1973) compared the amount of
algal film removed during foraging by different-sized L. gigantea and
interspecific competitors in the genus Lottia. The amount of algae
removed was inversely proportional to size; the larger the limpet, the less
algal film removed. Larger L. gigantea, however, displayed a
surprisingly large range in the amount of algae removed. A 6-cm L.
gigantea removed as little algae as a 2-cm Lottia spp. or
several times as much.
What might cause this large range in the amount of algae removed? Could it
be related to the territorial and gardening behaviors of L. gigantea?
Foraging territorial and nonterritorial L. gigantea are faced with
different sets of problems. A territory holder defends a 250-1000
cm2 territory (Stimson,
1970,
1973). The resident feeds on
the dense algal film characteristic of the territory. If the territory holder
is removed, the algal film is quickly consumed by other algal film grazers
(Stimson, 1970). The algal
film can be viewed as a garden with the resident limpet acting as the gardener
and defender (Branch et al.,
1992). A territory holder must obtain an adequate ration, but
without compromising the productivity of its garden. In contrast, a
nonterritory holder intruder must obtain an adequate ration before it contacts
the territory holder and is driven off. The goal of the foraging
nonterritorial limpet may be to obtain the needed ration in as short a
distance as possible and thereby minimize its chances of contacting the
territory holder.
Because of the different problems facing territorial and nonterritorial
L. gigantea, I hypothesized that a grazing territorial limpet would
leave behind more algal film than a non-territorial limpet. This hypothesis
was tested in the laboratory using the techniques of Wright and Shanks
(1993) to train limpets to use
territorial or nonterritorial behaviors. Trained limpets were then presented
with a glass plate with an undisturbed algal film. The first set of grazing
marks they made on this plate was analyzed for the amount of algal film left
behind.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Limpets between 50 and 75 mm in length were collected from the intertidal zone near Monterey and Santa Barbara, California, USA. These were shipped to the Oregon Institute of Marine Biology in Charleston, Oregon. Wright (1985
) found that nearly all
L. gigantea <30 to 40 mm shell length were males, whereas those
>50 mm shell length were predominantly female. Though I did not determine
the sex of the experimental animals, the larger size of the experimental
subjects suggests that they were females. In addition, Wright
(1985) found that limpets in
this size range were predominantly territorial as well. In fact, one trigger
for the ontogenetic change in sex from male to female is the acquisition of a
territory (Wright, 1985). Limpets were maintained outdoors on glass tiles in a running seawater system. The tiles were set on an Astroturf floor mat to prevent the limpets from wandering off their tiles. I randomly assigned each limpet a pair of tiles, a 10 x 10 cm central tile and an adjacent 5 x 5 cm peripheral tile. The central tiles had been placed outdoors in a running seawater table several weeks before the beginning of the experiment. The algal film that grew on these tiles provided food for the limpets. Daily at 0800 and 1700 h, the tiles and attached limpets were sprayed with water from the running seawater system for 3 h. The limpets were allowed to acclimate to their new surrounding for 24 h before the experiment began.
I randomly assigned 20 limpets to two training groups (i.e., there were no
significant differences in average shell length or the sexual composition of
the groups). One group received training mimicking consecutive agonistic
victories; the second group experienced training mimicking consecutive
agonistic defeats. This training procedure (see
Wright and Shanks, 1993)
attempts to mimic the natural interaction between territorial and
nonterritorial limpets (Stimson,
1970; Wright,
1982). Training progressed as follows: Starting approximately 20
min after the water was turned on, the tiles were inspected for moving
limpets. To all limpets that had moved onto their central tile, a stimulus
limpet (a randomly selected L. gigantea not part of the study) was
presented in a staged encounter. The cephalic tentacles and shell of the
stimulus limpet were held in gentle contact with the cephalic tentacles of the
moving subject limpet. This contact was maintained for 90 s (the maximum
response latency observed in the field
Wright, 1982) or until the
subject limpet displayed a territorial or evasive response. I defined a
territorial response as forward movement of greater than one shell length
during the 90-s interval (Wright,
1982; Wright and Shanks,
1993). I defined evasive behavior as a >90° turn by the
subject from the point of contact. At the end of the 90-s stimulus period, or
after a subject limpet completed a response, the subject limpet was given one
of two opposing agonistic experiences. Simply removing the stimulus limpet
simulated the experience of agonistic victory; the opponent retreated. This
produced a limpet with territorial behavior. Limpets trained to be
nonterritory holders experienced agonistic defeat, the experience of being
pushed by an aggressive limpet (Stimson,
1970; Wright,
1982). Using the stimulus limpet's shell, the subject limpet was
manually pushed slowly to the peripheral tile, or was pushed until it clamped
so tightly to the central tile that it could not be safely moved. Limpets were
trained each time they moved onto the central tile. Depending on how active
the individual was, it could have received training once, twice, or not at all
during a 24-h period.
Training ran from 8 to 24 November. At the end of this period, I removed the central tile and replaced it with a fresh tile with a uniform algal film. These tiles were preconditioned by being placed outdoors in a running seawater table for several weeks before their use in the experiment. After the first foraging event, the fresh tile was assigned a random number, wrapped in aluminum foil, and stored in a freezer until it was analyzed.
I hypothesized that grazing territorial limpets would leave more of the algal cover behind than would nonterritorial limpets. Territorial limpets could remove less of the algal film during each grazing event, or they could avoid grazing in areas that had already been visited. Both possibilities were tested.
To test if territorial limpets removed less of the algal film during a
grazing event, I measured the percent algal cover left behind. All
measurements were made blind; the tiles were coded with random numbers, and
the data were not assigned to a treatment until the analysis had been
completed. The tiles were viewed under a dissecting microscope at a
magnification of 40x with dark field illumination. A 10 x 10
ocular grid was centered over a section of the grazing path. I noted the
presence or absence of algal cover at each grid intersection. On each tile,
five areas of grazing trails were haphazardly selected and the percent cover
determined. I excluded sections of trails if they had been visited more than
once (see below) or if the grazed area was smaller than the occular grid.
Using these data, I calculated the average percent cover left behind after a
grazing event. Data were arcsine transformed
(Sokal and Rohlf, 1995) and,
using a t test, I tested the hypothesis that the percent cover of
algal film was higher after foraging by a territorial than by a nonterritorial
limpet.
Even though the tile was removed after the first foraging event, areas could have been grazed more than once. In areas that have been grazed once, the radula marks are parallel to each other (Figure 1A,B). In areas that have been grazed more than once, there are additional sets of radula marks that are not parallel but cross at an angle (Figure 1C,D). I used the following procedure to determine the percentage of the foraging area on the tile that was grazed over more than once (note that this analysis was also run blind). The tile was viewed with dark field illumination at a magnification of 6.5x. A piece of clear acetate was placed over the tile, and a map was made of the entire area over which the limpet had foraged. The map was cut into pieces representing the sections grazed once and those grazed more than once. The pieces were weighted to the nearest 0.001 g. I used these data to calculate the percentage of the grazing paths that had been grazed over more than once. To test if territorial limpets visited areas grazed earlier less frequently than nonterritorial limpets, the data were arcsine transformed and tested with a t test.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Out of the 20 limpets that started this experiment, three died during the experiment, and five were unresponsive, died shortly after the completion of the experiment, and were excluded from the analysis. Twelve limpets, seven territorially and five nonterritorially trained, were healthy and responsive at the end of the training. The behavioral responses of these limpets were identical to those found by Wright and Shanks (1993
). Their behavior was
typical of territorial L. gigantea in the field (Wright,
1982,
1985;
Wright and Shanks, 1993).
After contact with the stimulus limpet, they moved rapidly toward the stimulus
limpet and attempted to push it. All of the limpets trained to be
nonterritorial displayed the typical behavior of nonterritorial individuals
(Wright, 1982,
1985;
Wright and Shanks, 1993).
After contact with the stimulus limpet, they turned and moved rapidly away
from the point of contact, or they clamped down hard to the substrate. When a limpet is actively grazing, it swings its head through an arc, rasping the surface at regular interval with its radula. At the end of the arc, the limpet crawls a short distance forward and swings its head through another arc. The spacing between rasping events, the size of the arc through which it swings its head, and the distance the animal crawls forward before beginning a new head swing govern the amount of algal cover removed. In nonterritorial limpets, sets of radula marks, rasping events, were frequently so closely spaced that I could not separate them (Figure 1B). Further, the radula marks in the separate swings frequently overlapped (Figure 1B). In contrast, sets of radula marks made by territorial limpets were often distinct and separated by an undisturbed band of algal film (Figure 1A). An undisturbed algal film usually separated the radula marks in separate head swings of the feeding limpet as well (Figure 1A).
There was not a significant difference in the total area grazed by limpets that received territorial and nonterritorial training (average ± SE area grazed: territorially trained, 48% ± 6.2%; nonterritorially trained, 44% ± 12.2%), suggesting that the motivation to forage (hunger) was comparable in the two groups. Limpets given territorial training, however, left significantly (t = -4.92, df = 9, p =.00041) more algal film behind than did limpets trained to be nonterritorial (Figure 2). On average, territorial limpets left 71% of the algal film behind, compared to only 50% of the film left behind after grazing by nonterritorial limpets. The small standard error around the mean algal cover left behind by individual limpets indicates that individuals were highly consistent in their pattern of grazing (Figure 2).
|
During their initial foraging foray onto the fresh central tile, limpets that received territorial training seldom grazed over the same area more than once (average 4% of grazed area; Figure 3), suggesting that they recognize and avoid areas upon which they had earlier grazed (Figure 1C). In contrast, limpets that received nonterritorial training frequently grazed over an area more than once; of the area grazed, 20% had been visited more than once (Figure 3). The grazing paths of nonterritorial limpets were frequently observed to continue through areas upon which they had earlier grazed (Figure 1D). Nonterritorial limpets grazed over the same area more than once significantly more often than territorial limpets (t test, t = -3.60, df = 10, p =.0024). There was no significant difference in the total area grazed by territorial and nonterritorial limpets (see above). Hence, this difference in grazing behavior was not due to nonterritorial limpets being more reluctant to explore the new tile and thus concentrating their foraging in a smaller area where there would be a greater probability of crossing previous grazed areas. Rather, the data suggest that territorial limpets avoided crossing previous grazing trails, but nonterritorial limpets did not.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The limpets in this foraging experiment were first trained (Wright and Shanks, 1993
) to
display territorial or nonterritorial behavior. As found by Wright and Shanks,
limpets that experienced consecutive agonistic victories responded with
species-typical territorial behavior, while those that experienced consecutive
agonistic defeats displayed species typical nonterritorial behavior. Not only did previous agonistic experience determine an individual's response after contact with another limpet, but it also determined the individual's foraging pattern. Individuals trained to be territorial removed less algal film during foraging and grazed previously visited areas significantly less often than did individuals trained to display nonterritorial behavior. Territorial individuals appeared to forage prudently. Prudent foraging by territory holders is not a novel observation (see below), but it is interesting that a relatively simple animal such as a limpet displays such a complex set behaviors.
Gardening is only possible if the garden is productive enough to support
the gardener. An individual L. gigantea can occupy the same territory
for at least 4 years (Wright,
1985). For such sustained residence to be possible, the territory
holder may need to avoid overgrazing so that the productivity of the garden is
maintained. This can be accomplished by prudent grazing at some intermediate
level of consumption. A prudently grazing gardener may experience lower
immediate food intake but higher average intake over the long term.
A gardener only reaps the benefits of prudent grazing if competitors are
kept out of the garden; the gardener must also be territorial. Because of the
defense of their garden, few other limpets (either L. gigantea or
other Lottia spp.) are found on the territories of L.
gigantea (Stimson, 1970,
1973). If the territory holder
is removed, the algal film characteristic of the L. gigantea
territory is quickly consumed by intruding limpets
(Stimson, 1970). In L.
gigantea, two of the crucial gardening behaviors, prudent grazing and
territoriality, are determined by previous agonistic experience; individuals
experiencing repeated agonistic victories graze prudently and defend a
territory.
In addition to territorial behavior and prudent grazing, gardening L.
gigantea also rotate foraging forays to different portions of their
territory (Stimson, 1970),
actively remove sessile animals within their territory
(Stimson, 1970), and lay down
a mucous trail during foraging that both captures algal cells from the water
column and enhances their growth (Connor
and Quinn, 1984; Davies and
Hawkins, 1998). L. gigantea may also vary foraging
intensity in response to the local algal density. During my microscopic
inspection of the foraging trails, territorial limpets appeared to vary the
density of radula scraps depending on the local algal density; radula marks
appeared to be more tightly packed in areas with higher algal density than
where algae was more thinly distributed. Future experiments are planned to
explore this question. Like prudent foraging and territoriality, these diverse
gardening behaviors may also be tied to previous agonistic experience;
repeated agonistic victories may trigger a complex set of gardening behaviors
as well as the observed sex change from male to female
(Lindberg and Wright, 1985;
Wright, 1985,
1989).
In nature, most nonterritory-holding L. gigantea are small males
living on the periphery of territory held by a larger, usually female, limpet
(Wright, 1985,
1989). These small males
regularly make foraging intrusions onto the defended territory where they have
a high probability of encountering the territory holder. In Wright's
(1982) observations, 43% of
the intruders contacted the territory holder. Usually, the intruder quickly
withdrew from the territory ending its foraging; but in one instance (4% of
observed intrusions); the territory holder knocked the intruder off the rock,
undoubtedly leading to the intruder's death. The longer an intruder can forage
on a territory without contacting the territory holder, the larger the ration
it should be able to obtain. A simple way to minimize the chances of
encountering the territory holder is to obtain a ration while foraging over as
little area as possible. The foraging behavior of the limpets in this study
that were trained to be nonterritorial appears to follow this rule. Their
grazing paths were characterized by close packing of the individual rasp marks
(they often overlapped), they advanced only a short distance forward before
initiating a new grazing sweep, and they often foraged in areas they had
visited earlier. To obtain the same sized ration, a nonterritorial individual
foraged over half the area that a territorial individual would cover.
Gardening in L. gigantea is a form of central-place foraging
(Hamilton and Watt, 1970); the
territory holder returns to its home scar (central resting place within the
territory) at low tide and disperses out to forage within the garden during
high tide. As with any territorial species, the energy input from the
exploitation of the defended resources must be at least equal to the energy
needed to maintain the basic needs of the individual plus the added energy
expense of defending the territory
(Carpenter and MacMillen,
1976). In addition, not only must gardening be at least a
break-even strategy, it must be more profitable then other foraging strategies
(Hamilton and Watt, 1970).
There are several species of birds that are both territorial and prudently
exploit their defended resources. Territorial hummingbirds, golden-winged
sunbirds (Nectarina reichenowi), and Hawaiian honeycreepers
(Vestiaria coccinea) all protect a renewing resource, nectar
accumulating in flowers (Carpenter and
MacMillen, 1976; Carpenter et
al., 1983; Gill and Wolf,
1975). In addition, sunbirds may delay their return to flowers so
as maximize the accumulation of nectar
(Gill and Wolf, 1975). Davies
and co-workers (Davies, 1976;
Davies and Houston, 1981;
Houston et al., 1985)
described a system in which wintering pied wagtails (Motacilla alba)
defend sections of river shoreline. The territorial wagtails follow a regular
circuit around their territories, feeding on insects that wash ashore. During
the territory holder's absence, insects accumulate so that when the bird
returns to feed, its feeding rate is high. In each case, the territory holder
defends a resource upon which it forages and, between foraging events, the
resource increases in value. Like L. gigantea, the foraging behavior
of these bird species appears to be prudent. There is, however, an important
difference in resource utilization that is related to the fact that L.
gigantea is a gardener and the bird species are not.
In these bird examples, all of the resource is consumed during each foraging pass or event. The wagtail consumes all of the insects that have washed ashore in its absence, and nectar-feeding birds consume all of the nectar that has accumulated in a flower. In these systems, consumption of the resource has no affect on its subsequent replenishment. In contrast, foraging by L. gigantea does affect resource replenishment. Consumed algal cells cannot grow and produce new cells. Territorial L. gigantea forage prudently perhaps to avoid compromising future productivity.
From the perspective of the intruder, the bird and L. gigantea
systems are not so different. In the bird examples, resource density within
the territory may on average be higher than outside the territory
(Davies, 1976;
Gill and Wolf, 1975), making an
intrusion onto the territory relatively profitable; successful foraging
intrusions decrease the value of the defended resource for the territory
holder (Carpenter and MacMillen,
1976; Davies, 1976;
Gill and Wolf, 1975), and the
intruder gains more the longer it can remain on the territory without being
perceived and chased off by the territory holder. In L. gigantea, the
territory holder must apparently contact the intruder before the intruder is
perceived. The apparent lack of distance perception by the territory holder
may contribute to the intruder's foraging intensively. In this way an adequate
ration may be obtained within a small area, decreasing the probability of
contacting the territory holder.
As is typical of most research, this study has generated far more questions
than it has answered. The data suggest that territorial L. gigantea
forage prudently. I suggest or hypothesize that this behavior leads to higher
sustained productivity in the territory and that the territory holder, by
somewhat limiting its immediate food intake, experiences a higher energy
intake over the long term. I speculate that gardening individuals may adjust
their grazing to match the local algal density. These suggestions seem quite
reasonable, but they have not been tested. In nature, most males are small and
nonterritorial, whereas most females are large and territorial gardeners. How
much of this difference in behavior is driven by differences in the energy
requirements of reproduction in the sexes? Gardening is a stable strategy as
long as the gardener is able to reap the benefits. With increasing pressure
from intruders, does gardening eventually become unprofitable? Are female
territory holders more tolerant of male intruders? When a territory holder
intrudes on an adjacent territory (Wright,
1985) does it shift its foraging from prudent to nonprudent? In
the laboratory, we should be able to experimentally test these questions as
well as many others.
|
ACKNOWLEDGEMENTS |
|---|
Laura Rice—limpet trainer extraordinary—handled much of the limpet training for this experiment. Tom Rippetoe constructed the experimental apparatus, the limpet circus. Bill Wright offered helpful comments and wise advice. Comments by S.J. Hawkins and an anonymous reviewer on an earlier version of the manuscript are much appreciated.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Branch GM, Harris JM, Parkins C, Bustamante RH, Eekhout S, 1992. Algal `gardening' by grazers: a comparison of the ecological effects of territorial fish and limpets. In: Plant—animal interactions in the marine benthos (John DM, Hawkins SJ, Price H, eds). Oxford: Clarendon Press; 405-423.
Carpenter FL, MacMillen RE, 1976. Threshold model of
feeding territoriality and test with a Hawaiian honeycreeper.
Science 194:
639-641.
Carpenter FL, Paton DC, Hixon MA, 1983. Weight gain
and adjustment of feeding territory size in migrant hummingbirds. Proc
Natl Acad Sci USA 80:
7259-7263.
Connor VM, Quinn JF, 1984. Stimulation of food species
growth by limpet mucus. Science 225:
843-844.
Davies MS, Hawkins SJ, 1998. Mucus from marine molluscs. Adv Mar Biol 34: 2-71.
Davies NB, 1976. Food, flocking and territorial behaviour of the pied wagtail (Motacilla alba yarrellii Gould) in winter. J Anim Ecol 45: 235-253.
Davies NB, Houston AI, 1981. Owners and satellites: the economics of territory defence in the pied wagtail, Motacilla alba. J Anim Ecol 50: 157-180.
Gill FB, Wolf LL, 1975. Economics of feeding territoriality in the golden-winged sunbird. Ecology 56: 333-345.[ISI]
Hamilton WJ III, Watt KEF, 1970. Refuging. Annu Rev Ecol Syst 1: 263-286.
Houston AI, McCleery RH, Davies NB, 1985. Territory size, prey renewal and feeding rates: interpretation of observations on the pied wagtail (Motacilla alba) by simulation. J Anim Ecol 54: 227-239.
Lindberg DR, Wright WG, 1985. Patterns of sex change of the protandric patellacean limpet Lottia gigantea (Mollusca:Gastropoda). Veliger 27: 261-265.
Morris RH, Abbott DP, Haderlie EC, 1980. Intertidal invertebrates of California. Stanford, California: Stanford University Press.
Sokal RR, Rohlf FJ, 1995. Biometry. San Francisco, California: WH Freeman.
Stimson J, 1970. Territorial behavior of the owl limpet Lottia gigantea. Ecology 51: 113-118.[ISI]
Stimson J, 1973. The role of the territory in the ecology of the intertidal limpet Lottia gigantea (Gray). Ecology 54: 1020-1030.
Wilson EO, 1975. Sociobiology. Cambridge: Belknap Press.
Wolf LL, 1970. The impact of seasonal flowering on the biology of some tropical hummingbirds. Condor 72: 1-14.
Wright WG, 1982. Ritualized behavior in a territorial limpet. J Exp Mar Biol Ecol 60: 245-251.
Wright WG, 1985. The behavioral ecology of the limpet Lottia gigantea: interaction between territoriality, demography, and protandric hermaphroditism (thesis). San Diego: University of California, Scripps Institute of Oceanography.
Wright WG, 1989. Intraspecific density mediates sex-change in the territorial patellacean limpet Lottia gigantea. Mar Biol 100: 353-364.
Wright WG, Shanks AL, 1993. Previous experience determines territorial behavior in an archaeogastropod limpet. J Exp Mar Biol Ecol 166: 217-229.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||