Southeastern Naturalist
475
C.F. Noss and D.J. Levey
22001144 SOUTHEASTERN NATURALIST 1V3o(3l.) :1437,5 N–4o8. 33
Does Gut Passage Affect Post-dispersal Seed Fate in a Wild
Chili, Capsicum annuum?
Clay F. Noss1,2,* and Douglas J. Levey3,4
Abstract - Seeds of Capsicum spp. (wild chilies ) are coated with capsaicin, which deters
mammalian seed predators. During gut passage through frugivorous birds, its presence on
seeds likely is greatly reduced, presumably increasing the seeds’ susceptibility to postdispersal
seed predation by mammals. We tested whether gut passage influences the rate at
which dispersed seeds are removed from dispersal sites by different types of seed consumers.
We predicted that seeds passed through birds (passed seeds) would be removed at higher
rates than seeds taken directly from fruits (non-passed seeds). Removal rates of passed seeds
were either lower or no different than removal rates of non-passed seeds, contrary to our
prediction. In a second set of trials, we placed caged and exposed (control) seeds in pairs on
the ground to determine whether vertebrates or invertebrates were primarily responsible for
post-dispersal seed removal. We found an inconsistent effect of caging on frequency of seed
removal, indicating that both invertebrates and vertebrates harvest chili seeds at our site.
These results suggest that capsaicin’s role in mediating interactions with vertebrate seed
dispersers and predators is largely restricted to the wild chilies’ fruiting stage.
Introduction
Seed dispersal by animals is a critically important process in the ecology and
evolution of fruiting plants (Jordano 2000, Tiffney 2004). Despite many potential
advantages of fruits being ingested, carried, and deposited away from parent
plants, most plant-oriented studies of seed dispersal have narrowly focused on
consequences of where seeds are dispersed (Howe and Miriti 2004, Nathan and
Muller-Landau 2000). Relatively few studies have examined how gut treatment
may impact condition and subsequent fate of seeds, and of those studies, the vast
majority have tested solely for the effect of gut treatment on germination (Barnea et
al. 1991, Samuels and Levey 2005, Traveset et al. 2007). Any advantages associated
with germination, however, would be rendered irrelevant if seeds are consumed
soon after dispersal. In this context, it is becoming increasingly apparent that gut
treatment can change the chemical and physical structure of seeds in ways that impact
post-dispersal seed predation (Andresen and Levey 2004, Fricke et al. 2013,
Manzano et al. 2010, Mártinez-Mota et al. 2004).
Capsicum spp. (wild chilies ) provide an unusually good study system for examining
effects of gut treatment on the post-dispersal fate of animal-dispersed seeds.
1Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, FL
32611-8525. 2Current address - Department of Enviornmental Science, Policy, and Management,
University of California - Berkeley, Berkeley, CA. 3Department of Biology, University
of Florida, Gainesville, FL 32611-8525. 4Current address - National Science Foundation, 4201
Wilson Boulevard, Arlington, VA 22230. *Corresponding author - claynoss@gmail.com.
Manuscript Editor: Frank Moore
Southeastern Naturalist
C.F. Noss and D.J. Levey
2014 Vol. 13, No. 3
476
First, because chili fruits contain capsaicinoids, which elicit pain in mammals but
not birds (Jordt and Julius 2002, Mason and Clark 1995, Tewksbury and Nabhan
2001), practically all chili seeds are dispersed by birds (Levey et al. 2006). Given
that taxonomically dissimilar frugivorous birds process fruits in a similar manner
(Karasov and Levey 1990), there is presumably much less variation in the condition
of defecated chili seeds than in the seeds of species consumed and defecated
by both birds and mammals. Second, mechanisms by which gut passage of chili
seeds may affect seed condition are already established (Fricke et al. 2013). Digestive
processing likely reduces the presence of capsaicinoids that coat the outside of
chili seeds, thus leading to the prediction that chili seeds passed through a bird’s gut
will be more susceptible to post-dispersal seed predation by mammals than seeds
not passed through a bird’s gut. If this capsaicin-stripping prediction is supported,
it will provide a rare example of how vertebrate seed dispersal may be detrimental
to a fruiting plant.
We experimentally tested effects of gut passage by birds on seed removal of
a wild chili, Capsicum annum L. (Bird Pepper). Our study had two goals: (1) to
test the capsaicin-stripping prediction, and (2) to determine whether vertebrates or
invertebrates are the most common post-dispersal seed predators at our study site.
The second goal was important in interpreting results of the experiment we conducted
to test the prediction. In particular, if ants removed most chili seeds then we
would not have expected a difference in seed removal of gut-passed and non-gutpassed
seeds because ants are not deterred by capsaicin (D.J. Levey, pers. observ.).
Methods
Field site description
This study took place on Seahorse Key, a 64-ha island in the Big Bend region
of Florida’s west coast, approximately 8 km from mainland Florida (29°07'30''N,
83°2'W; Lillywhite et al. 2002, Spears 1987). As part of the Lower Suwannee
National Wildlife Refuge, the island is undeveloped and retains much of its
original flora and fauna. In temperate woodlands (like Seahorse Key), rodents
are generally the most important granivores (Hulme 1998), but ants are usually
better at locating small seeds at low densities (like those used in this study; Kaspari
1993, Manzano et al. 2010), and in Bolivia, ants are post-dispersal predators
of chilies (Fricke et. al. 2013). Common mammalian seed predators present on
Seahorse Key include various rodents (Peromyscus spp. [deer mice], other Muridae,
Sciurus carolinensis Gmelin [Carolina Squirrel]), and common invertebrate
seed predators include several species of ants (Pogonomyrmex badius (Latrelle)
[Florida Harvester Ant], Solenopsis invicta Buren [Red Imported Fire Ant], and
Pheidole spp. [big-headed ants], D.J. Levey, pers. observ.). Avian seed predators
are uncommon—the only one we have detected is Cardinalis cardinalis L.
(Northern Cardinal). We worked on the interior of the island within Quercus virginiana
Mill. (Southern Live Oak) woodlands (Spears 1987) where chili plants
are common in the understory.
Southeastern Naturalist
477
C.F. Noss and D.J. Levey
2014 Vol. 13, No. 3
Field experiments
We collected all fruits and seeds used in this study on the island and we handled
them with latex gloves and tweezers to reduce the presence of human scent, which
is known to affect seed removal by ants and rodents (Duncan et al. 2002). Fruits
were fed to captive Corvus ossifragus Wilson (Fish Crows) and Acridotheres tristis
L. (Common Mynas). Fish Crows are abundant on Seahorse Key and consume a
wide variety of fruits (McGowan 2001), almost certainly including those of Bird
Peppers, although we have not observed them doing so. Common Mynas also
consume fruits (Pell and Tidemann 1997) and occur in southern Florida, although
not on Seahorse Key. We included them because we could not obtain a sufficient
number of Fish Crows, and because both species were readily available at a nearby
National Wildlife Research Center facility. Gut treatment of seeds by the two species
is likely to be similar, given their similar body size and omnivorous diet. We
fed chili fruits to birds, collected defecated (passed) seeds, and placed the seeds in
a refrigerator the next day. We removed a similar number of seeds from fruits by
hand (non-passed) and placed them in a refrigerator. We lightly blotted all seeds
with damp paper towels before storage to simulate removal of fecal material by rain
and to standardize the amount of pulp and feces remaining on seeds.
In 2009, we tested the capsaicin-stripping prediction by placing 122 pairs of
passed and non-passed seeds within 1 cm of each other, making it likely that seed
predators would encounter both seeds of a pair and make a choice between them
(Hulme 1994). We placed a Magnolia grandiflora L. (Southern Magnolia) leaf supported
by a toothpick over each pair of seeds (station), providing protection from
rain while allowing access by all granivores. Stations were placed every 5 m along
a transect, and presence or absence of seeds was recorded every 2–6 days (daily at
first) over the next 16 days, at which point very few seeds were being removed. We
assumed removal of seeds indicated seed predation, although this is not necessarily
the case (Levey and Byrne 1993).
To meet our second objective, we conducted a similar experiment in which two
passed seeds were placed within 2 cm of each other. We covered 1 seed in each
pair with an 8-cm3 cage of 1-cm-wire mesh (hardware cloth) to exclude vertebrate
granivores and left the other seed fully accessible to all granivores (control). One
Southern Magnolia leaf protected each seed pair from rain. We placed 40 of these
stations 5 m apart along eight 25-m transects, and recorded the presence or absence
of seeds for 15 days.
We repeated both experiments in 2010 with 64 replicates that we checked for 14
days. In both years, approximately 10% of stations were disrupted by wind or rain;
such cases were excluded from analysis.
Data analysis
Data were analyzed using Pearson’s chi-squared test. We used data only from
stations where, on the last day of the experiment, one of the paired seeds was missing
and the other remained.
Southeastern Naturalist
C.F. Noss and D.J. Levey
2014 Vol. 13, No. 3
478
Results
Removal rates of passed and non-passed seeds differed dramatically in 2009
(Fig. 1a). In 2009, over 50% of seeds disappeared in the first 4 days, and the removal
rate of non-passed seeds was higher than that for passed seeds. After the
initial divergence, the removal rate was similar for both seed types. At the end of
the trial, significantly fewer non-passed than passed seeds remained (c2 = 12.73,
Figure 1. Percentage of chili seeds defecated by birds (passed) or taken directly out of ripe
fruit (non-passed) remaining at stations on successive days in 2009 (a) and 2010 (b). The
difference in number of passed and non-passed seeds remaining at the end of the trial was
significant in 2009 but not in 2010.
Southeastern Naturalist
479
C.F. Noss and D.J. Levey
2014 Vol. 13, No. 3
df = 1, P < 0.001), contrary to our prediction. In 2010 the pattern of seed removal
was obviously different; removal rates were lower, more constant, and closely similar
between passed and non-passed seeds (c2 = 0.06, df = 1, P < 0.81; Fig. 1b), also
contrary to our prediction.
Figure 2. Percentage of chili seeds protected from vertebrate granivores (caged) or unprotected
(exposed) remaining at stations on successive days in 2009 (a) and 2010 (b). The
difference in number of caged and exposed seeds remaining at the end of the trial was significant
in 2010 but not in 2009.
Southeastern Naturalist
C.F. Noss and D.J. Levey
2014 Vol. 13, No. 3
480
There was an inconsistent effect of caging on seed removal between years. In
2009, most seed removal occurred in the first 4 days and then co ntinued at a lower
rate for the remainder of the experiment, resulting in an insignificant difference between
exposed and caged seeds by the end of the trial (c2 = 1.92, df = 1, P < 0.17;
Fig. 2a). In 2010, there was a relatively steady decline in caged seeds remaining
during the experiment but the removal rate for exposed seeds was variable between
checks. At the end of this experiment, significantly more caged than exposed seeds
remained (c2 = 4.76, df = 1, P = 0.0291; Fig. 2b), indicating that mammalian granivores
were present and had been prevented from removing caged seeds.
We did not observe moved seeds, seed fragments, or other indicators of consumption
by granivores near the experimental trials.
Discussion
We found no support for the prediction that treatment of chili seeds in bird
guts increases the seeds’ risk of post-dispersal seed predation. To the contrary,
gut passage apparently decreased the overall risk of seed removal in both years
of our study, but the difference between passed and non-passed seeds was significant
only in 2009. Results of the caging experiment in 2010 (but not 2009)
showed that vertebrate (as well as invertebrate) granivores were present at our
study site and that they removed chili seeds passed through bird guts, a finding
that suggests that vertebrate granivores may be significant post-disperal seed
predators. Seed removal by vertebrate granivores was important for us to document
at our site because the capsaicin-stripping prediction assumes presence of
mammalian seed predators.
Previous studies have documented many benefits to plants of seed dispersal by
vertebrates, including escape from high levels of seed and seedling mortality near
the parent plant, reduction in microbial infection, directed dispersal to especially
suitable microhabitats, and colonization of disturbed or novel habitats (Fricke et al.
2013, Herrera 2002). Our study, based on an examination of gut treatment, provides
yet another potential benefit of vertebrate seed dispersal. We note, however, that the
few studies comparing removal of passed and non-passed seeds, or seeds embedded
or non-embedded in feces, do not yield a clear or consistent picture of any impact
of gut passage on seed predation (Andresen 1999, Bermejo et al. 1998, Cochrane
2003, Fricke et al. 2013, Mártinez-Mota et al. 2004, Roberts and Heithaus 1986).
Furthermore, at least some observed impacts from these studies are inconsistent
between seasons (Mártinez-Mota et al. 2004) or within a season (Andresen 1999).
For small-seeded temperate species like Bird Peppers, post-dispersal predation of
defecated seeds is poorly understood; the vast majority of previous studies concern
large-seeded tropical species (Manzano et al. 2010). An exception is a recent study
by Fricke et al (2013), which found that gut passage increased Capsicum chacoense
Hunz. (Tova) seed survival 370% by reducing ant predation and pathogenic fungal
infection. Ant predation was reduced through gut removal of volatile chemicals that
attract granivorous ants, a mechanism that may also explain the reduction in seed
predation of gut-passed seeds that we observed in 2009.
Southeastern Naturalist
481
C.F. Noss and D.J. Levey
2014 Vol. 13, No. 3
We provide two explanations for why our results differed between years. First,
such variation is not surprising—practically all studies on seed predation that have
looked for temporal variation have found it (e.g., Andresen 1999, LoGiudice and
Ostdelf 2002, Mártinez-Mota et al. 2004, Schupp 1990, Willson and Whelan 1990).
In our study, the rate of removal was appreciably higher in both sets of experiments
in 2009 than in 2010, suggesting a biological difference in the community of seed
predators. For example, activity of granivorous ants may have been higher in 2009
if daytime temperatures were higher than in 2010 (Traniello et al. 1984, Vickery
and Bider 1981). However, average daytime temperature during our study periods
did not differ (National Climatic Data Center 2010). A non-mutually exclusive
explanation for the annual differences concerns statistical power. In both experiments,
significant differences between treatments and controls were found in the
year with greater replication—2009 for the comparison of passed and non-passed
seeds, and 2010 for the comparison of caged and exposed seeds. This is a likely
explanation in the experiments comparing removal of caged and exposed seeds
because the sample size in 2010 was more than 50% greater than in 2009 (n = 64
and 40 pairs, respectively) and the overall patterns of removal were similar between
years (compare the two panels of Fig. 2). On the other hand, in the experiments
comparing removal of passed and non-passed seeds, there was an immediate, large,
and lasting difference in removal in 2009 that was not apparent in 2010.
Previous studies on dispersal of wild chilies have documented multiple benefits
of capsaicinoids, including protection from fungal pathogens, longer gut-retention
time in avian seed dispersers (which presumably increases dispersal distance), and
deterrence of mammalian seed predators (Chichewicz and Thorpe 1996; Fricke
et al. 2013; Levey et al. 2006, 2007; Tewksbury et al. 2008a, b). Although we
expected the last of these benefits to be reduced or eliminated by gut treatment in
frugivorous birds, we conclude that gut passage can increase any post-dispersal
benefits already present due to capsaicin. More generally, we suggest that the traditional
emphasis on dispersal benefits related to escape from high mortality near
parent plants should be broadened to examine potential advantages associated with
gut treatment (Fricke et al. 2013).
Acknowledgments
We thank Henry Coulter and Bronko Gukanvich for transportation to and from Seahorse
Key; Dr. Michael Avery and Kandy Keacher, of the National Wildlife Research Center,
Gainesville, FL, for help collecting defecated seeds from captive Fish Crows and Common
Mynas; Keilani Jacquot and Dylan Lee for assistance in the field; and the Howard Hughes
Medical Institute for funding.
Literature Cited
Andresen, E. 1999. Seed dispersal by monkeys and the fate of dispersed seeds in a Peruvian
rainforest. Biotropica 31:145–158.
Andresen, E., and D.J. Levey. 2004. Effects of dung and seed size on secondary dispersal,
seed predation, and seedling establishment of rain forest trees. Oecologia 139:45–54.
Southeastern Naturalist
C.F. Noss and D.J. Levey
2014 Vol. 13, No. 3
482
Barnea, A., Y. Yom-Tov, and J. Friedman. 1991. Does ingestion by birds affect seed germination?
Functional Ecology 5:394–402.
Bermejo, T., A. Traveset, and M.F. Willson. 1998. Post-dispersal seed predation in the temperate
rainforest of southeast Alaska. Canadian Field-Naturalist 112:510–512.
Chichewicz, R.H., and P.A. Thorpe. 1996. The antimicrobial properties of chile peppers
(Capsicum species) and their uses in Mayan medicine. Journal of Ethnopharmacology
2:61–70.
Cochrane, E.P. 2003. The need to be eaten: Balanites wilsoniana with and without elephant
seed-dispersal. Journal of Tropical Ecology 19:579–589.
Duncan, R.S., D.G. Wenny, M.D. Spritzer, and C.J. Whelan. 2002. Does human scent bias
seed-removal studies? Ecology 83:2630–2636.
Fricke, E.C., M.J. Simon, K.M. Reagan, D.J. Levey, J.A. Riffell, T.A. Carlo, and J.T.
Tewksbury. 2013. When condition trumps location: Seed consumption by fruit-eating
birds removes pathogens and predator attractants. Ecology Letters 16:1031–1036.
Herrera, C.M. 2002. Seed dispersal by vertebrates. Pp. 185–208, In C.M. Herrera and O.
Pellmyr (Eds.). Plant–Animal Interactions: An Evolutionary Approach. Blackwell Science,
Oxford, UK. 328 pp.
Howe, H.F., and M.N. Miriti. 2004. When seed dispersal matters. Bioscience 54:651–660.
Hulme, P.E. 1994. Post-dispersal seed predation in grassland: Its magnitude and sources of
variation. Journal of Ecology 82:645–652.
Hulme, P.E. 1998. Post-dispersal seed predation: Consequences for plant demography and
evolution. Perspectives in Plant Ecology, Evolution, and Systematics 1:32–46.
Jordano, P. 2000. Chapter 6: Fruits and frugivory. Pp. 125–166, In M. Fenner (Ed.). Seeds:
The Ecology of Regeneration in Natural Plant Communities, 2nd Edition. Commonwealth
Agricultural Bureau International, Wallingford, UK. 410 pp.
Jordt, S.E., and D. Julius. 2002. Molecular basis for species-specific sensitivity to hot chili
peppers. Cell 108:421–430.
Karasov, W.H., and D.J. Levey. 1990. Digestive system trade-offs and adaptations of frugivorous
passerine birds. Physiological Ecology 63:1248–1270.
Kaspari, M. 1993. Removal of seeds from Neotropical frugivore droppings: Ant responses
to seed number. Oecologia 95:81–88.
Levey, D.J., and M.M. Byrne. 1993. Complex ant–plant interactions: Rainforest ants as
secondary dispersers and post-dispersal seed predators. Ecology 74:1802–1812.
Levey, D.J., J.J. Tewksbury, M.L. Cipollini, and T.A. Carlo. 2006. A field test of the directed-
deterrence hypothesis in two species of wild chili. Oecologia 150:61–68.
Levey, D.J., J.J. Tewksbury, I. Izhaki, E. Tsahar, and D.C. Haak. 2007. Evolutionary ecology
of secondary compounds in ripe fruit: Case studies with capsaicin and emodin. Pp.
3–58, In A.J. Dennis, E.W. Schupp, R.A. Green, and D.A. Westcott (Eds.). Seed Dispersal:
Theory and Its Applications in a Changing World. CABI Publishing Cambridge,
MA. 702 pp.
Lillywhite, H.B., C.M. Sheehy, and M.D. McCue. 2002. Scavenging behaviors of Cottonmouth
Snakes at island bird rookeries. Herpetological Review 33:259–261.
LoGiudice, K., and R.S. Ostfeld. 2002. Interactions between mammals and trees: Predation
on mammal-dispersed seeds and the effect of ambient food. Oecologia 130:420–425.
Manzano, P., F.M. Azcárate, P. Begoña, and J.E. Malo. 2010. Are ecologists blind to small
things? The missed stories on non-tropical seed predation on feces. Oikos 119:1537–1545.
Martínez-Mota R., J.C. Serio-Silva, and V. Rico-Gray. 2004. The role of canopy ants in removing
Ficus perforata seeds from Howler Monkey (Alouattapalliata mexicana) feces
at Los Tuxtlas, México. Biotropica 36:429–432.
Southeastern Naturalist
483
C.F. Noss and D.J. Levey
2014 Vol. 13, No. 3
Mason, J.R., and L. Clark. 1995. Capsaicin detection in trained European starlings: The importance
of olfaction and trigeminal chemoreception. The Wilson Bulletin 107:165–169.
McGowan, K.J. 2001. Fish Crows (Corvus ossifragus). In A. Poole (Ed.).The Birds of
North America Online, Cornell Lab of Ornithology, Ithaca, NY. Available online at
http://bna.birds.cornell.edu.bnaproxy.birds.cornell.edu/bna/species/589. Accessed 6
February 2012.
Nathan, R., and H.C. Muller-Landau. 2000. Spatial patterns of seed dispersal, their determinants,
and consequences for recruitment. Trends in Ecology and Evolution 15:278–285.
National Climatic Data Center (NCDC). 2010. Levy County daily high summary. Available
online at http://www.ncdc.noaa.gov/cdo-web/datasets. Accessed 6 February 2012.
Pell, A.S., and C.R. Tidemann. 1997. The ecology of the Common Myna in urban nature
reserves in the Australian capital territory. Emu 91:141–149.
Roberts, T., and E.R. Heithaus. 1986. Ants rearrange the vertebrate-generated seed shadow
of a Neotropical fig tree. Ecology 67:1046–1051.
Samuels, I.A., and D.J. Levey. 2005. Effects of gut passage on seed germination: Do experiments
answer the questions they ask? Functional Ecology 19:365–368.
Schupp, E.W. 1990. Annual variation in seedfall, post-dispersal predation, and recruitment
of a Neotropical tree. Ecology 71:504–515.
Spears, E.E. 1987. Island and mainland pollination ecology of Centrosema virginianum and
Opuntia stricta. Journal of Ecology 75:351–362.
Tewksbury, J.J., and G.P. Nabhan. 2001. Directed deterrence by capsaicin in chillies. Nature
412:403–404.
Tewksbury, J.J., D.J. Levey, M. Huizinga, and D.C. Haak. 2008a. Costs and benefits of
capsaicin-mediated control of gut retention in dispersers of wild chilies. Ecology
89:107–117.
Tewksbury, J.J., K.M. Reagan, N.J. Machnicki, T.A. Carlo, D.C. Haak, A.L.C. Peñaloza,
and D.J. Levey. 2008b. Evolutionary ecology of pungency in wild chilies. Proceedings of
the National Academy of Sciences of the United States of America 105:11,808–11,811.
Tiffney, B.H. 2004. Vertebrate dispersal of seed plants through time. Annual Review of
Ecology, Evolution, and Systematics 35:1–29.
Traniello, J.F.A., M.S. Fujita, and R.V. Bowen. 1984. Ant foraging behavior: Ambient
temperature influences prey selection. Behavioral Ecology and Sociobiology 15:65–68.
Traveset, A., A. Robertson, and J. Rodriguez-Perez. 2007. A review on the role of endozoochory
on seed germination. Pp. 78-103, In A.J. Dennis, R.J. Green, E.W. Schupp, and
D.A. Wescott (Eds.). Seed Dispersal: Theory and Its Application in a changing world.
CABI Publishing, Wallingford ,UK. 684 pp.
Vickery, W.L., and J.R. Bider. 1981. The influence of weather on rodent activity. Journal of
Mammalogy 62:140–145.
Willson, M.F., and C.J. Whelan. 1990. Variation in postdispersal survival of vertebratedispersed
seeds: Effects of density, habitat, location, season, and species. Oikos
57:191–198.