WHY STRANGLER FIGS DONT STRANGLE STRANGLER FIGS
Jonathan H+Tttustrangler figs (Ficus spp.) are trees that germi-
nate in soil cavities on other trees, grow down
to the ground, and subsequently "strangle" the
tree upon which they grew by completely encircling
it (see Fig. 3.16). A question has long puzzled observ-
ers: Why are strangler figs almost never observed
growing on another strangler fig (Titus et al. 1990)?
The rarity of strangler figs on strangler figs may be a
case of a plant avoiding competition from its own
progeny in a unique way. Although a few observations
of fig seedlings and saplings have been described (W.
Haber, pers. obs.), their occurrence is far more rare
than expected, Mature fig trees produce numerous
fleshy fruits that are relished by monkeys, bats, and
birds. One would therefore expect an abundant rain
of fig propagules on the parent tree. The convoluted
growth form of strangler fig trees creates cavities that
collect organic debris and seem to provide optimal es-
tablishment sites for strangler figs (Putz and Holbrook
1989).
Four hypotheses, not mutually exclusive, have
been posed to explain the apparent lack of epiphytic
fig seedlings on mature fig trees: (1) strangler figs are
autotoxic, that is, produce substances inimical to the
establishment of their own seedlings (an example of
autopathy, the chemical inhibition of one's own spe-
cies; Smith 1979, Rice 1984); (2) fig seedlings do not
survive on reproductively mature strangler fig trees
because soil cavities on strangler figs are not of suffi-
cient volume or depth to support fig seedlings; (3) fig
seedlings suffer mortality by pathogens and predators
that frequent adult fig trees; and (4) fig seedlings fuse
with conspecific hosts and are thus undetectable.
In May—August 1985, my colleagues and I investi-
gated the first two hypotheses. Two strangler fig spe-
cies, Ficus pertusa and F. tuerckheimii, common on
trees in the pastures and forests of Monteverde, were
selected for this study (Titus et al. 1990). To examine
potential autotoxic effects at the time of germination,
seeds from both fig species were collected and planted
in pots and petri plates and subjected to the follow-
ing six treatments (treatments 4 and 5 were conducted
in pots only): (1) watered with conspecific stemflow
(water flowing down the trunks of trees was col-
lected); (2) watered with mixed samples of stemflow
collected from tree species that frequently host stran-
gler figs (Inga brenesii, Acnistus arborescens, Cedrela
tonduzii, and Ocotea tonduzii); (3) watered with un-
treated spring water; (4) mulched by shredded fresh
leaves of F. pertusa and F. tuerckheimii mixed into
the top 2.5 cm of soil and watered with spring water;
(5) mulched with shredded fresh leaves of host spe-
cies mixed into the top 2.5 cm of soil and watered with
spring water; or (6) placed into a dark box and wa-
tered with spring water.
In another set of trials, seeds were planted into
soil-filled branch cavities on 10 F. pertusa and 10 F.
tuerckheimii trees located in pastures and forest frag-
ments. Two large cavities were selected on each tree
and 50 seeds were planted in each; one cavity of each
pair was watered with conspecific stem flow and the
other cavity with spring water. Seeds were also
planted in soil-filled branch cavities found on known
host trees. To determine if the quantity of soil in fig
tree soil cavities was less than that in host tree soil
cavities, the volume of soil in cavities was compared
between strangler figs and host trees. To determine if
F. pertusa and F. tuerckheimii seedlings can grow on
the ground, fig seeds were planted 1 cm deep in
areas that receive full sun and in areas under complete
(non-Ficus) canopy cover. Germination was moni-
tored in all experiments at 2-3-day intervals for 9-
12 weeks.
Results showed that neither stemflow nor extracts
of macerated Ficus leaves affected fig germination;
germination levels were similar in all treatments,
thereby ruling out autotoxic effects on germination,
Germination of fig seeds planted in soil cavities was
not correlated with either host species (conspecific
Ficus or other tree species) or watering; fig seeds ger-
minated in all the soil cavity treatments. No seeds
germinated from the ground plots. Seeds have often
been observed growing directly out of fruits; however,
these seeds quickly succumb to fungal infection. Soil
cavity size does not appear to determine germination
success since the volume of soil in strangler fig soil
cavities was not different from the soil volume of cavi-
ties on host trees where epiphytic figs were growing.
We observed marginally higher germination in
the petri plates than in the soil cavities, suggesting
that pathogens on parent or host trees may inhibit
F. pertusa and F. tuerckheimii seed germination and/
or survival. Germination was highest in the constantly
moist petri plates and lowest on the ground where
surface soil dried despite daily watering. This suggests
that high humidity levels may improve germination.71 Plants and VegetationS