INTEG.
AND
COMP. BIOL., 42:872–880 (2002)
Biogeography, Competition, and Microclimate: The Barnacle Chthamalus fragilis
in New England1
DAVID S. WETHEY2
Department of Biological Sciences, University of South Carolina, Columbia, South Carolina 29208
SYNOPSIS. Geographic limits of species are commonly associated with climatic or physical boundaries, but
the mechanisms of exclusion at the limits of distribution are poorly understood. In some intertidal populations, the strengths of interactions with natural enemies are mediated by microclimate, and determine geographic limits. The northern limit of the barnacle Chthamalus fragilis in New England is the south side of
Cape Cod, Massachusetts. South of the cape, Chthamalus has a refuge from competition in the high intertidal, which is too hot for survival of its superior competitor Semibalanus balanoides. North of the cape, the
high intertidal is cooler, and Semibalanus survives, so Chthamalus has no refuge. Thus, geographic variation
in the strength of competition may determine the geographic limit of Chthamalus. Intolerance of cold by
Chthamalus cannot account for the geographic limit: transplants of Chthamalus 80 km beyond its northern
limit survived up to 8 yr in the absence of competition with Semibalanus. At the geographic limit of Chthamalus in the Cape Cod Canal there are two bridges, 5 km apart. On the southern bridge, Chthamalus is
abundant and occupies a refuge above Semibalanus. On the northern bridge in 2001, only 7 individual
Chthamalus were present. Despite the proximity of the bridges, their microclimates are very different. The
southern bridge, where Chthamalus is abundant, is up to 88C hotter than the northern bridge. This higher
temperature creates a refuge in the high intertidal for Chthamalus. On the cooler northern bridge, there is
no refuge for Chthamalus. Because of the difference in temperatures of the water masses that meet in the
canal, heat storage in the rock of the bridge piers causes the temperatures to differ between the bridges.
Thus, geographic change in microclimate alters the strength of competition, and determines the geographic
limit.
‘‘When we travel from south to north, or from a damp region to a dry, we invariably see some species
gradually getting rarer and rarer, and finally disappearing; and the change in climate being conspicuous,
we are tempted to attribute the whole effect to its direct action. But this is a very false view: we forget that
each species, even where it most abounds, is constantly suffering enormous destruction at some period of
its life, from enemies or from competitors for the same place and food; and if these enemies or competitors
be in the least degree favoured by any slight change of climate, they will increase in numbers, and as each
area is already fully stocked with inhabitants, the other species will decrease.’’—Charles Darwin, On the
Origin of Species, 1859, p. 69.
INTRODUCTION
way ecologists view systems, where interactions
among species are commonly the foci of our studies.
I will take a community ecologist’s view of biogeography and turn several of these ideas upside down.
In the end, I will demonstrate that the northern limit
of a tropical species is set not by intolerance of cold,
but rather in a strange way, by intolerance of heat. I
also hope to demonstrate that interaction strength in
the community is the key factor that explains the geographic pattern (e.g., Darwin, 1859, p. 69). I will use
a combination of field experiments, purely descriptive
ecology, biophysical ecology and models to dissect out
the underlying mechanisms.
On the basis of field manipulations of microclimate,
I have argued that the same mechanisms responsible
for creating local vertical zonation patterns, also determine the northern geographic limit of the barnacle
Chthamalus fragilis in New England (Wethey, 1983,
1984). Chthamalus has a geographic range from the
Caribbean to the south side of Cape Cod (Dando and
Southward, 1980). Near its northern limit, Chthamalus
lives in a narrow band in the high intertidal zone, below which lives the boreo-arctic barnacle Semibalanus
balanoides. This zonation is the result of competition
In this paper I pose the question ‘‘what mechanisms
set the geographic limits of species?’’ When considering the northern and southern limits of species, we
tend to put our minds into an autecology framework,
in which we think of species in isolation and assume
that they are limited by intolerance of cold at their pole
ward limits and by intolerance of heat at their equator
ward limits. This approach was formalized very elegantly by Hutchins (1947). He made the case that geographic limits are set by thermal tolerances of the most
sensitive life stages. He showed that one could take
advantage of the differences in temperature on opposite sides of the Atlantic and Pacific to determine
which kind of limitation was occurring at the northern
and southern extent of species ranges. This approach,
although appealing, is in opposition to the rest of the
1 From the Symposium Physiological Ecology of Rocky Intertidal
Organisms: From Molecules to Ecosystems presented at the Annual
Meeting of the Society for Comparative and Integrative Biology, 2–
7 January 2002, at Anaheim, California.
2 E-mail: [email protected]
872
THE BARNACLE CHTHAMALUS FRAGILIS
FIG. 1. The zonation of Chthamalus and Semibalanus as a function
of slope and aspect on Horse Island, Connecticut. On sloping surfaces, the Chthamalus zone (between black lines) is wider than on
vertical surfaces. The upper distribution limit of Semibalanus is the
lower of the two black lines.
between the species. When I removed the competitively dominant Semibalanus from the low intertidal,
Chthamalus invaded and persisted (Wethey, 1983).
Connell (1961) had similar results near the northern
limit of Chthamalus in Scotland.
Local microclimate strongly determines the local
zonation of the two species. On north-facing vertical
surfaces, Chthamalus occupies a narrow zone; on
south-facing verticals it occupies a wider zone, and on
nearly horizontal surfaces, its zone is even broader
(Wethey, 1983). Figure 1 illustrates this pattern very
clearly. The area between the black lines is the
Chthamalus zone, and the region below the lower
black line is the Semibalanus zone. On sloping surfaces of the left hand boulder, Chthamalus occupies a
wider zone than on the vertical surfaces (Fig. 1). The
mechanism responsible for this pattern involves the
mediation of interaction strength by microclimate. The
low shore competitive dominant Semibalanus is intolerant of heat and desiccation, and its upper shore limit
of distribution is strongly influenced by direct solar
radiation (which contributes the largest component of
heating in the system). Chthamalus is extremely resistant to heat and desiccation (Southward, 1958; Foster,
1969, 1971a, b), able to withstand rock temperatures
up to 458C (Aveni-Deforge and Wethey, 2001). In microhabitats where Semibalanus survives, it competitively excludes Chthamalus. I demonstrated this experimentally by placing opaque shades over small areas of shore, using clear plastic roofs as controls. In
shade treatments, Semibalanus recruits survived on the
upper shore and were responsible for 94% of the mortality of Chthamalus by crushing, undercutting and
overgrowing. In the sun treatments, more than 90% of
the Semibalanus recruits died, and Chthamalus suffered little mortality (Wethey, 1984).
North of Cape Cod, Semibalanus is only slightly
affected by local microclimate (Wethey, 1983, 1984).
Semibalanus survives over the entire intertidal zone in
873
FIG. 2. The distribution of Chthamalus in the region of its northern
limit on Cape Cod, Massachusetts. Solid symbols—dense Chthamalus zone in high intertidal. Half colored symbols—sparse Chthamalus in high intertidal. Open symbols—Chthamalus rare to absent
in high intertidal: density ,1 m22.
northern New England, even when it dies back further
south. In consequence, I hypothesized that north of
Cape Cod, Chthamalus has no refuge from competition on the high shore, and it is therefore excluded
from northern New England (Wethey, 1983, 1984; see
also Southward and Crisp, 1956; Lewis, 1964, p. 252).
Distribution of Chthamalus near its northern limit
The geographic distribution of Chthamalus around
its northern limit (Fig. 2) has a very sharp boundary
at Cape Cod, Massachusetts. On the south side of Cape
Cod, from Woods Hole to the western half of the Cape
Cod Canal (completed in 1914), Chthamalus forms a
dense band in the high intertidal. In these areas,
Chthamalus can occupy up to 100% of the available
substratum in the refuge zone. Two highway bridges,
built in the 1930s cross the canal, approximately 5 km
apart. The Bourne Bridge (418449500N, 708359250W)
is nearer the western end of the canal and its barnacle
populations are more representative of southern New
England distributions, with a dense (10 per 0.01 m2)
Chthamalus band in the high intertidal zone (Fig. 2).
The Sagamore Bridge (418469330N, 708329370W), is
near the eastern end of the canal and its barnacle populations are more characteristic of northern New England, with very few Chthamalus. In 1983, there were
between 0.1 and 0.5 Chthamalus per 0.01 m2 at the
top of the barnacle zone on the Sagamore Bridge. In
2000–2001, only 7 Chthamalus individuals were alive
on the north piers of the bridge, representing a population density of 0.09 per 0.01 m2 at the top of the
barnacle zone. The sharp change in population density
between the two bridges on the Cape Cod Canal is
remarkable because it takes place over a distance of 5
km.
In this paper I examine two hypotheses that might
explain the geographic distribution of Chthamalus near
874
DAVID S. WETHEY
its northern limit. The strict microclimate hypothesis
is that Chthamalus is limited to the region south of
Cape Cod by intolerance of cold. I tested this experimentally with a transplant. The combined microclimate-competition hypothesis is that the refuge from
competition disappears as one crosses the geographic
limit. To test this hypothesis, first I quantified the microclimatic conditions associated with the refuge zone.
I hypothesized that the difference in water temperature
between the colder Massachusetts Bay to the north and
the warmer Buzzards Bay to the south caused the intertidal shore temperatures to differ at low tide. Shore
temperatures could differ if there was heat storage in
the rock, and if the rate of heat conduction to and from
the rock was high enough to affect the surface temperature of the rock. I tested this hypothesis by quantifying the temperatures experienced by Chthamalus
on the two bridges, and by developing a numerical
model of rock temperature with which I could examine
the effect of the heat storage terms on the temperature
of the rock.
TESTS OF HYPOTHESES
A test of the intolerance of cold hypothesis
To test the hypothesis that the northern limit of
Chthamalus is set by intolerance of cold conditions to
the north, I transplanted Chthamalus beyond their
northern limit, and measured their survival. Stones and
plastic disks with Chthamalus from the Yale Field Station in Connecticut (418169N, 728449W), 180 km south
of the northern limit, were transported by car at night
to avoid overheating, and glued with epoxy to the intertidal zone in Nahant Massachusetts (428259020N,
708549250W), 80 km beyond the northern limit. Water
temperatures are lower in Massachusetts than in Connecticut (128C lower in August 1990, 48C lower in
June 2001).
Transplants were divided into several treatments:
FIG. 3. Survival of Chthamalus in transplants beyond its northern
limit. The fraction alive is plotted as a function of date from 1982
to 1991. Top panels: comparison of unmanipulated areas (dotted
lines) and areas where Semibalanus was removed (solid lines) on
the same stones within the Semibalanus zone. Left panel (a) is a
horizontal surface, right panel (b) is a vertical surface. Bottom panel
(c) is a set of transplants above the upper zonation limit of Semibalanus.
a. Vertical surface in Semibalanus zone, exposed to
competition.
b. Vertical surface in Semibalanus zone, competitors
removed in spring 1984.
c. Horizontal surface in Semibalanus zone, exposed to
competition.
d. Horizontal surface in Semibalanus zone, competitors removed in spring 1984.
e. Vertical surface above Semibalanus zone, where
there is no natural barnacle settlement, and therefore no competition.
Semibalanus (Fig. 3c). Since the upper limit of settlement of larvae is approximately the same in both
Chthamalus and Semibalanus (Wethey, 1983), the
zone of treatment (e) is an area where recruitment
cannot or does not occur. However, if transplanted
to this zone, Chthamalus can survive for very long
periods, far longer than Semibalanus. The maximum
lifespan of Semibalanus in the high intertidal at Nahant, Massachusetts is approximately 3 yr (Wethey,
1985), compared to the 8 yr lifespan of the Chthamalus transplants (Fig. 3c). Therefore Chthamalus
survives very well in areas beyond its northern limit,
but only at shore levels where it cannot settle and
recruit naturally, or where its competitor is experimentally removed. These results are consistent with
the hypothesis (Southward and Crisp, 1956; Lewis,
1964; Wethey, 1983, 1984) that Chthamalus is not
limited by cold at its northern limit, but rather by
competition with Semibalanus.
A total of 89 individuals were in treatments (a–d),
and treatment (e) included 443 individuals.
Chthamalus that were transplanted to the Semibalanus zone at Nahant died rapidly if exposed to
competition from Semibalanus, but survived well in
treatments where competitors had been removed
(Fig. 3a). This result was similar both on vertical and
horizontal surfaces (Fig. 3a, b). Most remarkably,
Chthamalus survived as long as 8 yr in the zone
above the upper limit of settlement and survival of
Tests of the combined microclimate—competition
hypothesis
Quantification of the conditions in the Chthamalus
refuge zone. To quantify the refuge of Chthamalus
from competition, I measured the upper zonation limits
of both Chthamalus and Semibalanus on rock surfaces
with a variety of slopes and orientations to the sun in
June and July of 1983 and 1984 at sites in Long Island
Sound on Horse Island (418159N, 728459W), and the
THE BARNACLE CHTHAMALUS FRAGILIS
875
FIG. 4. The effect of rock temperature on the size of the Chthamalus refuge zone. Open symbols represent the upper distribution limit of
Chthamalus and solid symbols represent the upper distribution limit of Semibalanus at Horse Island, Connecticut, on surfaces with a variety
of slopes and orientations. The vertical axis represents the elevation in meters above mean low water. The horizontal axis represents the
maximum temperature in June 1983 and 1984, predicted by a numerical heat model for each site. The set of 3 solid lines represent the
regression (wide line) and upper and lower 95 percent confidence values (thin lines) for the individual observations of the elevation of the
upper shore limit versus maximum rock temperature for Semibalanus. The set of 3 dotted lines represent the regression (wide line) and upper
and lower 95 percent confidence values (thin lines) for the individual observations of the elevation of the upper shore limit of Chthamalus.
The triangular region in the upper right corner is the Chthamalus refuge. Semibalanus does not occur in sites where temperatures exceeded
428C.
Yale Peabody Museum Field Station (418169N,
728449W), Connecticut. Elevations of all sites were
measured with a telescopic contractor’s level and leveling rod, and referenced to a benchmark of known
elevation at the Yale Field Station. Slopes and orientations of the rock surfaces at the zonation limits were
measured with an inclinometer and sighting compass
(Silva Ranger). Magnetic orientations were corrected
for local magnetic declination using published values
on U.S. Geological Survey topographic maps of the
region, for the period 1983–1984. A total of 140 elevations were measured.
If body temperature affects Semibalanus survival as
suggested by Figs. 1 and 3, then the upper shore zonation limit should correlate more strongly with temperature for Semibalanus than for Chthamalus. The
thermal conditions at each of the zonation limits were
estimated using a numerical model of rock temperature. The model was derived from methods described
in Denny and Wethey (2001) (see Fig. 7 for model
validation). Since barnacles are so small, and since
they have a large surface area in contact with the rock,
their temperatures were assumed to be the same as the
rock (Thomas, 1987; Bertness, 1989). The model uses
air temperature, water temperature, height of the tide,
and solar radiation at 10-min intervals, to predict the
surface temperature of the rock. Solar radiation was
calculated from the position of the sun in the sky, assuming a maximum solar irradiance of 1,000 W m22.
The position of the sun in the sky was calculated at
10-min intervals for June 1983 and June 1984 using
the NASA-JPL long ephemeris (Giorgini et al., 1996).
Air temperatures for East Wareham Massachusetts, in
June 1983 and June 1984 were used as proxies for air
temperatures in Connecticut. East Wareham is 180 km
east of the study sites in Long Island Sound, and has
very similar climate. Water temperatures at New London Connecticut for June 1997 were used as a proxy
for water temperature in Long Island Sound in June
1983 and 1984. The air temperature conditions in
1983–84 and 1997 were similar. New London is 55
km east of the sites where barnacle zonation was quantified. Tidal elevations were calculated at 10-min intervals for June 1983 and 1984 with the computer program WXTide (Flater, 2000), which uses the U.S. National Oceanic and Atmospheric Administration
(NOAA) tidal prediction algorithms (Schureman,
1958). The maximum predicted rock temperature at
each of the barnacle survey locations was used as a
summary of the thermal conditions at that site.
As expected, the rock temperature model results indicate that the upper zonation limit of Semibalanus
moves down the shore as maximum rock temperature
increases (Fig. 4). In the figure, the lines represent the
linear regression of the height of the empirically measured zonation limit versus the predicted maximum
rock temperature, and the 95 percent confidence band
around the individual estimates. The solid lines are for
876
DAVID S. WETHEY
FIG. 5. Water temperature (8C) at the western end of the Cape Cod
Canal as a function of tidal height (meters above mean low water).
Values were recorded during August 1990 at the US Army Corps
of Engineers District Field Office on the canal. Tidal heights are
predictions from the NOAA tide model.
Semibalanus, dotted lines are for Chthamalus. The upper zonation limit of Chthamalus is less sensitive to
maximum rock temperature as seen by the smaller
slope of its regression line (Fig. 4). The refuge for
Chthamalus from competition is the triangular region
in the upper right hand corner of the graph, where
Chthamalus survives above the upper zonation limit
of Semibalanus. The refuge is the region where
Chthamalus lives above the 95% confidence band of
Semibalanus distribution limit. The refuge does not exist when the maximum rock temperature is below 268C
(Fig. 4). The refuge increases in size as the maximum
rock temperature increases. As maximum rock temperatures increase above 268C, the refuge expands
from a small region at 12.2 m above MLW to a zone
spanning 11.5 m to 11.8 m above MLW when the
maximum temperature are 408C. Semibalanus was not
found in sites whose maximum predicted temperatures
exceeded 428C (Fig. 4).
Hypotheses that might explain the sharp boundary
within the Cape Cod Canal. The distribution of
Chthamalus (Fig. 2) corresponds to a large water temperature difference between the two ends of the Cape
Cod Canal, with cold water on the Massachusetts Bay
(north) side and warmer water on the Buzzards Bay
(south) side. The temperature difference is as much as
128C. The tidal range is 1.22 m in Buzzards Bay and
2.86 m in Massachusetts Bay, where the mean tide
level is 0.12 m lower. High tide in Massachusetts Bay
is approximately 3 hr later than Buzzards Bay. As a
result, there is a current that reverses every 6 hr and
reaches a mean maximum velocity of 2 m sec21. Although the water temperature in the canal shifts
throughout the tidal cycle, at high tide the Bourne
Bridge is immersed in warmer Buzzards Bay water
(Fig. 5) and the Sagamore Bridge is immersed in colder Massachusetts Bay water. As a result of convective
heat exchange during immersion, and heat storage in
their granite blocks, the piers of the Bourne Bridge are
likely to begin each low tide at a higher temperature
than those of the Sagamore Bridge. Heat conduction
within the piers is then likely to be an important contributor to temperature excursions of the surfaces of
the piers at low tide. These factors are likely to cause
the temperatures of barnacles on the piers to be higher
on the Bourne Bridge than on the Sagamore Bridge,
at equivalent tidal heights. These temperature differences may be responsible for the presence of a
Chthamalus refuge zone on the Bourne Bridge and the
absence of one on the Sagamore Bridge (Fig. 2).
To test the hypothesis that the southern bridge is
consistently warmer than the northern bridge at low
tide, temperature recordings were made at the upper
distribution limits of Chthamalus and Semibalanus on
both bridges. Small self contained temperature data
loggers (Thermochron model 1921 iButton, DallasMaxim Semiconductor Corporation) were glued to the
piers of the Bourne and Sagamore Bridges with underwater epoxy putty (Splash-Zone Compound, Z Spar
Corporation). The loggers are similar in shape to a
watch battery (stainless steel case, 17 mm diameter, 5
mm thick), and were programmed to record observations every 30 min for 45 days, at a resolution of
0.58C. The loggers were programmed and data were
downloaded with a Palm-Pilot hand held computer interface (IConnection model 3301A, Scanning Devices
Corporation).
The southern bridge is consistently warmer than the
northern bridge. On north-facing surfaces of the northern (Sagamore) bridge, the temperature exceeded 268C
on only 1 low tide in 2001, yet on surfaces of similar
aspect on the southern (Bourne) bridge, the temperature commonly exceeded 268C during low tides
throughout the summer of 2001 (Fig. 6). On south facing surfaces of the two bridges the temperatures are
higher and there is a consistent difference between the
northern and southern bridges. On the south facing
surfaces of the Sagamore bridge the temperature exceeded 328C on only 1 low tide in 2001, whereas on
south facing surfaces at the same tidal height on the
Bourne bridge, the temperature exceeded 328C on 15
low tides (Fig. 6). Chthamalus should have a refuge
from competition only on surfaces whose maximum
temperature exceeds 268C (Fig. 4). The refuge from
competition becomes large when the maximum surface
temperature exceeds 328C (Fig. 4). These results indicate that the southern Bourne Bridge has thermal
conditions consistent with the presence of a refuge
from competition for Chthamalus, and the northern
Sagamore Bridge does not.
In order to determine whether heat storage in the
piers of the bridges could account for the difference
in low tide conditions on the two bridges, the numerical model of rock temperature (see above) was used
to simulate rock temperatures during the period June
1 to June 20, 2001. Water levels at the two bridges
were obtained from 6-min records for June 2001 from
tide gauges deployed by NOAA, adjacent to each
bridge. The Bourne tide gauge (Canal Station 320) was
THE BARNACLE CHTHAMALUS FRAGILIS
877
FIG. 7. Model predictions and rock temperature observations from
the Cape Cod Canal, July 1–8, 2001. The dotted line is the model
prediction and the solid line is the record from a Thermochron iButton data logger.
FIG. 6. Rock temperatures (July 2000–November 2001) at the top
of the Semibalanus zone on the Sagamore (left column) and Bourne
(right column) bridges. The upper pair are records from the north
facing piers on the north side of the canal, and the lower pair are
records from the south facing piers on the south side of the canal.
Horizontal lines are plotted at 268C, the temperature below which
Chthamalus does not have a refuge from Semibalanus.
304 m west of the bridge, and the Sagamore tide gauge
(Canal Station 115) was 730 m east of the bridge. Water temperatures for Massachusetts Bay were from the
NOAA tide station in Boston, and water temperatures
for Buzzards Bay were from the NOAA tide station in
Woods Hole, for the same period. Air temperatures for
the period were from the NOAA cooperative weather
station at East Wareham (418469N, 708409W), 6.5 km
west of the Bourne Bridge. Solar radiation was estimated using calculated azimuths and elevations of the
sun for the same dates and times of day, calculated
with the NASA-JPL long ephemeris (Giorgini et al.,
1996). Two model scenarios were used. One scenario
used the normal pattern of colder water on the Massachusetts Bay end of the canal and warmer water on
the Buzzards Bay end of the canal. In order to test the
hypothesis that differences in water temperature during
immersion could change the rock temperatures during
low tide, in the second model scenario the northern
bridge was exposed to simulated southern water temperatures and the southern bridge was exposed to simulated northern water temperatures, while maintaining
the normal tidal oscillations at each site.
The model predicts very well the timing and magnitude of temperature fluctuations during low tide (Fig.
7). Since the model was based on idealized solar radiation values, much of the difference between the
model and the observed temperatures are likely the
result of clouds and rainy conditions that occurred on
some of the days during the model run (July 2 and 4).
If heat storage in the piers of the bridge were important, then differences in water temperature at high
tide should be reflected in temperature differences at
low tide. The results of the simulation indicate that
heat storage in the piers of the bridges accounts for at
least 28C of the temperature difference between the
bridges. The mean difference in water temperature between the north and south sides of the canal during
the period of the model runs was 4.48C. In the simulations where northern bridge was exposed to southern
water at high tide, the rock temperatures at low tide
were on average 28C hotter (maximum of 5.48C hotter)
in the simulations where it was exposed to northern
water at high tide. In the simulations where the southern bridge was exposed to southern water at high tide,
rock temperatures were on average 28C hotter (maximum of 6.18C hotter) than in the simulations where it
was exposed to northern water at high tide. If the
northern bridge were exposed at high tide to warmer
water from the south, it would create a small refuge
for Chthamalus in the highest sites on the piers.
DISCUSSION
This paper examines the mechanisms responsible
for setting the geographic limits of species. It specifically tests two general classes of hypotheses regarding
polar limits: 1) that species are strictly limited by intolerance of physical conditions beyond their polar
limit (the intolerance of cold hypothesis), and 2) that
species are limited by interactions with natural enemies, whose strength is a function of physical conditions (the combined microclimate-competition hypothesis).
The intolerance of cold hypothesis was falsified by
the results of the transplant experiment. Chthamalus
survived up to 8 yr in transplants 80 km beyond its
northern limit, in areas where its competitor could not
settle and survive (Fig. 3). At the same geographic
878
DAVID S. WETHEY
location Semibalanus survives typically no more than
3 yr in the high intertidal zone (Wethey, 1985). The
evidence in Britain is similar. In Europe, the northern
and eastern limit of Chthamalus is near Aberdeen on
the east coast of Scotland, the Isle of Wight and the
Cap de la Hague west of Cherbourg on the English
Channel (Crisp et al., 1981). Crisp (1950) transplanted
Chthamalus stellatus beyond its northern limit to the
North Sea coast at Whitley Bay, Northumberland,
where it survived two winters and produced viable larvae. In the extraordinarily cold winter of 1962–63,
Chthamalus did not suffer increased mortality, although other subtropical species did (Crisp, 1964). At
its northern geographic limit both in New England and
in Britain, Chthamalus is most abundant at the highest
shore levels, where the effect of cold would be the
greatest. If Chthamalus were intolerant of cold, it
should die in winter at its northern limit, however this
does not appear to be the case (Crisp and Southward,
1958; Lewis, 1964, pp. 251–252; Wethey, 1984). Additionally, since larval stages tend to be more sensitive
to environmental conditions than adults (Southward,
1958; Crisp and Ritz, 1967; Foster, 1969, 1971a, b),
the fact that Chthamalus settles in autumn near its
northern limit (Hatton, 1938; Wethey, 1983, 1984;
Lewis, 1986; O’Riordan et al., 1995) suggests that all
life stages of Chthamalus are tolerant of cold.
The results of this study are consistent with the hypothesis that the strength of interactions with natural
enemies determines the geographic limit. Chthamalus
fragilis in southern New England has a refuge from
competition whose size is correlated with the maximum temperature reached by the rock surface in summer (Fig. 4). In sites with a maximum temperature
below 268C, there is no real refuge. The size of the
refuge increases with temperature above 268C (Fig. 4).
When maximum rock temperature is above 408C, the
refuge can be 20 cm or wider (Fig. 4). Above 428C,
Semibalanus appears to be unable to survive (Fig. 4).
This is consistent with the observation that Semibalanus balanoides enters heat coma between 358C and
378C, and that its LT50 is 44.38C (Southward, 1958;
Foster, 1969). Presumably the competitive ability of
Semibalanus is greatly impaired when it is in heat
coma. Selective death of some Semibalanus genotypes
in high temperature microhabitats (Schmidt and Rand,
1999, 2001) may also affect average competitive ability. Chthamalus stellatus reaches heat coma at 438C
and its LT50 is approximately 508C (Southward, 1958;
Foster, 1969). Chthamalus fragilis presumably has
similar tolerance for heat to C. stellatus so it is likely
to take advantage of the competition refuge where conditions are too hot for Semibalanus. In South Carolina,
Chthamalus fragilis survives on rock surfaces that
commonly reach temperatures above 408C, and withstands rock temperatures of 458C (Aveni-Deforge and
Wethey, 2001).
The geographic limit of Chthamalus appears to be
in the middle of the Cape Cod Canal, which connects
the warm waters of Buzzards Bay, south of Cape Cod,
to the cold waters of Massachusetts Bay, north of Cape
Cod (Fig. 2). There is a refuge from competition for
Chthamalus on the southern of two highway bridges
across the Cape Cod Canal, because the piers of the
bridge store heat from the warm water of Buzzards
Bay, and warm up to high enough temperatures that
Semibalanus dies in the high intertidal zone. There is
no refuge for Chthamalus on the northern of the two
highway bridges; the piers remain colder at low tide
because they are immersed at high tide in the colder
water of Massachusetts Bay. The importance of heat
storage in the substratum documented here is consistent with Bertness’ (1989) observations that large intertidal boulders and cobbles embedded in the substratum remain cooler than loose cobbles. This sort of fine
balance between the two species near their geographic
limits is consistent with the 40 yr of observations by
Southward (1991), who found that Semibalanus increased at the expense of Chthamalus after summers
with cooler sea surface temperatures in southern England.
The temperature differences observed here between
the bridges on the Cape Cod Canal have similar biological effects over a much larger geographic scale on
opposite sides of the cape. Leonard (2000) compared
Semibalanus balanoides zonation, recruitment, survival, and growth between sites north of Cape Cod (Damariscotta, Maine) and south of Cape Cod (Newport,
Rhode Island). He found that Semibalanus survived
higher on the shore in Maine than in Rhode Island,
consistent with my observations in Massachusetts and
Connecticut (Wethey, 1983, 1984, this paper). His
temperature measurements indicate that the Rhode Island sites were hotter than those in Maine (Leonard,
2000). In addition, he found that shade from algal canopies enhanced Semibalanus survival in Rhode Island,
but not in Maine. Bertness (1989), working in Rhode
Island, found that crowded Semibalanus on cobbles
had lower body temperatures and higher survival than
solitary individuals on cobbles of similar size. These
results are consistent with the observation that Semibalanus dies back from hot exposed sites on the high
shore south of Cape Cod (Fig. 3). Based on the results
of the present paper, the positive effect of density in
Semibalanus should disappear north of Cape Cod,
where thermal stress is reduced. This expectation is
consistent with my (Wethey, 1979, p. 69) observation
north of Cape Cod, that crowded Semibalanus in the
high intertidal had lower survival than solitary individuals, a pattern opposite to that observed south of
Cape Cod by Bertness (1989).
The results of the heat model indicate that the thermal conditions during high tide have a potentially important influence on temperatures experienced by barnacles at low tide, because of heat storage in the rock,
and the conduction term in the heat equation. As we
have moved to more mechanistic models of intertidal
thermal conditions, we have tended to assume that sub
aerial conditions were rather independent of submerged conditions. However for organisms whose
THE BARNACLE CHTHAMALUS FRAGILIS
temperatures are largely influenced by their substratum, like limpets (Hayworth and Quinn, 1990) and
barnacles (Thomas, 1987; Bertness, 1989), sea surface
temperature has an influence on body temperatures at
low tide via heat storage in the rock. This is in contrast
to organisms that sit above the substratum, like mussels, whose body temperature at low tide is more
weakly influenced by sea surface temperature (Helmuth and Hoffman, 2001).
Failure of recruitment could also determine the location of a geographic limit. Lewis et al. (1982) hypothesized that recruitment failure may set the northern limit of Chthamalus in Scotland. However, in New
England, especially in the Cape Cod Canal, this is very
unlikely. There is a very fast current that moves back
and forth with the tides in the canal, and there are
dense populations (10 per 0.01 m2) of Chthamalus on
the western half of the canal. It is hard to imagine how
Chthamalus larvae could fail to be transported to the
eastern half of the canal in this hydrodynamic environment.
The community ecology approach to thermal tolerances of organisms is very useful in developing predictive models of biogeographic distribution. Since the
strengths of interactions in communities help determine the distribution and abundance of species, it
should not be terribly surprising that such interactions
may influence the geographic distribution of species
(e.g., Darwin, 1859). In the case of Chthamalus fragilis in New England, it is the interplay between the
thermal environment and the strength of competition
with Semibalanus balanoides, which varies geographically and determines the geographic limit of distribution (confirming the conjectures by Southward and
Crisp [1956], Lewis [1964], and Wethey [1983]).
ACKNOWLEDGMENTS
This work was supported by the National Science
Foundation (field work: OCE 82-08176, OCE 8600531; data analysis: DEB-0108412). Sarah Woodin
helped in all phases of this project, and improved the
manuscript. Richard Grosberg, Matthew Reed, and
Ron Etter, assisted in the field work during the 1980s.
Leo Buss provided access to the Yale Peabody Museum Field Station and to Horse Island, and provided
housing and transportation in the 1980s. Frank Ciccone, Engineer in Charge, US Army Corps of Engineers Cape Cod Canal Field Office permitted access
to the Cape Cod Canal bridges, and Frank Morris, Assistant Engineer in Charge, provided engineering
drawings of the elevations of the bridge piers, and water temperature data. Nathan Riser, Robert Shepard,
Henry Werntz, Kenneth Sebens, and Joseph Ayers, directors of the Northeastern University Marine Science
Center at Nahant, provided access to study areas. M.
Patricia Morse provided housing and transportation in
Massachusetts in the 1980s. Brian Helmuth introduced
me to Ibuttons. Dave Chanoux of Scanning Devices
Inc. provided software support for the Ibutton/Palm
Pilot interface. This paper is dedicated to Jack Lewis
879
who started my interest in geographic limits of species,
and for whom this is merely confirmation of what he
knew in 1964.
REFERENCES
Aveni-Deforge, K. and D. S. Wethey. 2001. Physical constraints on
zonation patterns of the barnacle Chthamalus fragilis. Amer.
Zool. 41:1383.
Bertness, M. D. 1989. Intraspecific competition and facilitation in a
northern acorn barnacle population. Ecology 70:257–268.
Connell, J. H. 1961. The influence of interspecific competition and
other factors on the distribution of the barnacle Chthamalus
stellatus. Ecology 41:710–723.
Crisp, D. J. 1950. Breeding and distribution of Chthamalus stellatus.
Nature 166:311–312.
Crisp, D. J. (ed.) 1964. The effects of the severe winter of 1962–63
on marine life in Britain. J. Anim. Ecol. 33:165–210.
Crisp, D. J. and D. A. Ritz. 1967. Changes in the temperature tolerance of Balanus balanoides during its life cycle. Helgoländer
Wiss. Meeresunters. 15:98–115.
Crisp, D. J. and A. J. Southward. 1958. The distribution of intertidal
organisms along the coasts of the English Channel. J. Mar. Biol.
Ass. U. K. 37:157–208.
Crisp, D. J., A. J. Southward and E. C. Southward. 1981. On the
distribution of the intertidal barnacles Chthamalus stellatus,
Chthamalus montagui and Euraphia depressa. J. Mar. Biol. Ass.
U. K. 61:359–380.
Dando, P. R. and A. J. Southward. 1980. A new species of Chthamalus (Crustacea Cirripedia) characterized by enzyme electrophoresis and shell morphology: With a revision of the other
species of Chthamalus from the western shores of the Atlantic
Ocean. J. Mar. Biol. Ass. U. K. 60:787–831.
Darwin, C. 1859. On the origin of species by natural selection. John
Murray, London.
Denny, M. W. and D. S. Wethey. 2001. Physical processes that generate pattern in marine communities. In M. D. Bertness, S. D.
Gaines, and M. E. Hay (eds.), Marine community ecology, pp.
3–37. Sinauer Associates, Sunderland, Massachusetts.
Flater, D. 2001. WXTide. Flaterco. http://www.flaterco.com/xtide.
(see also http://tbone.geol.sc.edu).
Foster, B. A. 1969. Tolerance of high temperature by some intertidal
barnacles. Mar. Biol. 4:326–332.
Foster, B. A. 1971a. Desiccation as a factor in the intertidal zonation
of barnacles. Mar. Biol. 8:12–29.
Foster, B. A. 1971b. On the determinants of the upper limit of intertidal distribution of barnacles (Crustacea: Cirripedia). J.
Anim. Ecol. 40:33–48.
Giorgini, J. D., D. K. Yeomans, A. B. Chamberlin, P. W. Chodas,
R. A. Jacobson, M. S. Keensey, J. H. Lieske, S. J. Ostro, E. M.
Standish, and R. N. Wimberly. 1996. JPL’s on-line solar system
data service. Bull. Am. Astr. Soc. 28:1158.
Hatton, H. 1938. Essais de bionomie explicative sur quelques espèces intercotidales d’algues et d’animaux. Annls. Inst. Océanogr. Monaco 17:241–348.
Hayworth, A. M. and J. F. Quinn. 1990. Temperature of limpets in
the rocky intertidal zone—effects of caging and substratum.
Limnol. Oceanogr. 35:967–970.
Helmuth, B. S. T. and G. Hoffman. 2001. Microhabitats, thermal
heterogeneity, and patterns of physiological stress in the rocky
intertidal zone. Biol. Bull. 201:374–384.
Hutchins, L. W. 1947. The bases for temperature zonation in geographical distribution. Ecol. Monogr. 17:325–335.
Leonard, G. H. 2000. Latitudinal variation in species interactions:
A test in the New England rocky intertidal zone. Ecology 81:
1015–1030.
Lewis, J. R. 1964. The ecology of rocky shores. Hodder and Stoughton, London.
Lewis, J. R. 1986. Latitudinal trends in reproduction, recruitment
and population characteristics of some rocky littoral molluscs
and cirripedes. Hydrobiologia 142:1–13.
Lewis, J. R., R. S. Bowman, M. A. Kendall, and P. Williamson.
1982. Some geographical components in population dynamics:
880
DAVID S. WETHEY
Possibilities and realities in some littoral species. Neth. J. Sea
Res. 16:18–28.
O’Riordan, R. M., A. A. Myers and T. F. Cross. 1995. The reproductive cycles of Chthamalus stellatus (Poli) and C. montagui
Southward in south-western Ireland. J. Exp. Mar. Biol. Ecol.
190:17–38.
Schmidt, P. S. and D. M. Rand. 1999. Intertidal microhabitat and
selection at MPI: Interlocus contrasts in the northern acorn barnacle, Semibalanus balanoides. Evolution 53:135–146.
Schmidt, P. S. and D. M. Rand. 2001. Adaptive maintenance of
genetic polymorphism in an intertidal barnacle habitat: Habitatand life-stage-specific survivorship of Mpi genotypes. Evolution
55:1336–1344.
Schureman, P. 1958. Manual of harmonic analysis and prediction of
tides. Special Publication 98. U.S. Dept. of Commerce, Coast
and Geodetic Survey, U.S. Government Printing Office, Washington, DC.
Southward, A. J. 1958. Note on the temperature tolerances of some
intertidal animals in relation to environmental temperatures and
geographic distribution. J. Mar. Biol. Ass. U. K. 37:49–66.
Southward, A. J. 1991. Forty years of changes in species composition and population density of barnacles on a rocky shore near
Plymouth. J. Mar. Biol. Ass. U. K. 71:495–513.
Southward, A. J. and D. J. Crisp. 1956. Fluctuations in the distribution and abundance of barnacles. J. Mar. Biol. Ass. U. K. 35:
211–229.
Thomas, F. I. M. 1987. The hot and cold of life on rocks: Determinants of body temperature in the northern rock barnacle, Semibalanus balanoides. Department of Ecology and Evolutionary
Biology, Brown University, Providence, Rhode Island.
Wethey, D. S. 1979. Demographic variation in intertidal barnacles.
Department of Biology, University of Michigan, Ann Arbor.
Wethey, D. S. 1983. Geographic limits and local zonation: The barnacles Semibalanus (Balanus) and Chthamalus in New England.
Biol. Bull. 165:330–341.
Wethey, D. S. 1984. Sun and shade mediate competition in the barnacles Chthamalus and Semibalanus: A field experiment. Biol.
Bull. 167:176–185.
Wethey, D. S. 1985. Catastrophe, extinction and species diversity:
A rocky intertidal example. Ecology 66:445–456.