Reprinted{rom
2 l Nov em ber1975,V o l u me1 9 0 ,p p ' 8 0 4 -8 0 6
SCIEI\UCE
Host Tree LocationBehaviorof a TropicalVine
( M onstersgigontea) by S kototropism
Donald R. Strong, Jr. and Thomas S. Ray, Jr.
Copyrighta fi75
by the American Association lor the Aduancement of Science
Host Tree Location Behaviorof a Tropical Vine
( M onstera gigantea)by Skototropism
Abstract. Seedlings of the arboreal, ground-germinating, tropical v,r?eMonstera gigantea (Engler) are shown to grow directly toward potential host trees; they do not find
hosts by haphazard growth or random searching. Our experimentsshow that these vines
are attracted to the darkest sector of the horizon. In nature treesprovide thesedark sectors. We term this response skototropism (growth toward darkness). Skototropism is
probably produced by a modifcation of the molecular and cellular mechanisms that produce negativephototopism. Vle intoduce the new term to emphasize the adaptive nature of the response; whereas the term "negative phototropism" can imply either growth
away from light or growth toward darkness, only growth toward darkness can lead the
vine directly to a host. This is because,in nature, hosts will not be aligned l8U from the
lightest sector of the horizon relative to the vine.
Arboreal vines are a mainly tropical
phenomenon; species diversity of this
group is much higher at low latitudes (1)
where perennially warm and moist conditions allow the high surface-to-volumeratio of their elongate form. This form allows vines the unique advantage of lateral
movement acrossthe canopy, to occupy diverse areas not available to trees anchored
F ig . l. Se e d lin g s o f
gigantea
Monstera
converging upon a willbe host tree, at Finca
La Selva, Costa Rica.
Notice erect shoot tipg
which serve to constantly apprise of host
location relative to the
minor diversions of
forest floor debris.
at a single site. The capacity for lateral
canopy movement, however, costs vines
their self-supportive ability and presents
arboreal specieswith the problem of finding a host tree to climb. Many biologists
have been concernedwith the mechanism
of vine host location. Darwin interpreted
his experiments with the vine Bignonia
capreolata to show that hosts are found by
"apheliotropism" (growth away from
light) (2). Other mechanismshavebeenhypothesized through the years: negative
geotropism (J), random searching(2, 4),
and positive phototropism (5). In this report we give the first description of actual
host location in a natural vine population
and demonstratethe location mechanism
to be skototropism (growth toward darkness); seedlings of Monstera gigantea locate trees by growing toward the darkness
of a tree silhouette.
Monstera gigantea (Araceae) is a vine of
Caribbean lowland forests. lt germinates
terrestrially, immediately after dispersal
from inflorescencesset partway up trunks
of canopy trees.Seedlingshave singleslender prostrate stems with tiny bract-like
leaves at nodes. There are three distinct
ontogeneticmorphologiesduring the vine's
progress from ground to mature height a
seedling stage of horizontal terrestrial
growth, a "saucer-ieafed" stage maintained during host tree ascent,and an adult
stage.In the adult morphology the species
flowers repeatedly in direct sunlight and
continues to climb. We observed seed set
to be simultaneous,with the entire contents of an inflorescence falling to the
ground en masse within several meters.
This means that seedlingstend to climb the
parent tree, or very nearby ones.
Host finding behavior of Monstera gigantea seedlingscan be conspicuousin nature. Thousandsof bright green seedlings
are seenin contrast against the brown leaf
litter, creeping inward from all sidesof the
tree. The seedlingsappear as short spokes
of a wheel with the tree as the hub (Fig. I ).
In this behavior the tip of the elongating
stem simply stands up, bends,and grows
toward the tree, independently of compass
direction betweenthe two. The mean angle
of growth relative to the host tree is calculated as d
180
derlain by cellular and molecular mechanisms similar to thosecausingthe classical
negativephototropism of laboratory plant
physiology ( 6, Z).
_3
There is no apparenttrend for individual
Monslera seedlings more distant from a
potentialhost tree to havegrown more toro
o
oa
tuously than close seedlings. Seedling
U
ta o
o
!
stems at all distancesfrom potential host
trees are contorted. This contortion results
from growth around and over fallen leaves
c
and other forest floor objects.
E
The relativedarknessas well as distance
of objects also influencesthe propensity to
attract thesevine seedlings.We set pairs of
seedlingsparallel to both sides of shade
cloth panels at 5-cm intervals from I to
100 cm away from the panels.We usedfive
o
o
panels with one to five thicknesses
different
50
@
of cloth. A singlethicknesspasses60 percent ofthe light. The panelswere placedon
0
391521273339
the forest floor, and faced east-west.SeedDecimeters
from tree
lings respondeddifferently only to the first
Fig. 2. Averageangleofgrowth (Z)andstandard two panels, to 40 percent as opposed to 64
deviationof growth angle(s) amongvine seed- percent shade. Responsesto the four panlings found along a 4-m transectout from the els of shade of 64, 78, 81, and 89 percent
baseof a potentialhost tree.When the average were identical. At these panels all of the
directionof growth is directly toward the tree
E : 0, and whenall seedlings
aregrowingin the closer seedlingsturned 90" and grew tosamedirections = 0. The formulasaregivenin ward the panel within 6 days;the most disthe text. Note that the vinepopulation'saccura- tant seedlingshad turned toward the pancy (a) and precision(s) of host location de- els within 30 days. Closer seedlingsat the
creaseswith increasingdistancefrom the tree.
panel of 40 percent shade began to turn
Thesepatternsobtainin all directionsfrom potential host trees,indicatingthat the seedlings only after l2 days, and only about halfof
those within 30 cm had turned by the end
areattracteddirectlyto the trees.
of the 34-day experiment; there was no responseat greater than 30 cm. Over 90 perIn Fig. 2 we show how the accuracyand cent of seedlingsat the higher shadeswere
precision of seedling growth angle de- growing toward the panelsat the end of the
creases with distance from the tree; the experiment,whereasonly 7 of l4 surviving
mean growth angle wanders from 0', and on the eastsideanci9 of l4 survivingon the
the variance among seedlings in growth west side of the 40 percent darkness panel
angle increases with increasing distance were doing so at termination. These exfrom the potential host tree. The same periments were done beneath dense canorelationship was observedto hold on all py, where the seedlingsoccur naturally.
sidesof the tree (seeFig. 1). Seedlingsfall- Hence, the seedlingsare discriminating40
en naturally around 19 treesand in several percent light intensity differenies at the
n
experimental situations produced con- very low intensity(10 to 100lux) ofthe for) sin a;
vergencefrom all quarters.
est floor.
l- l
a =a rcta n
That the seedlingsconvergefrom all diThe attractivenessof a tree trunk is also
) c os a,
rections
indicates
that
a
function
they
are
finding
the
of its diameter; trees with large
i= I
tree by skototropism, by growth toward diameters will orient seedlingsat greater
wheren is the number of seedlingsand a; is the darkest sector of the horizon indepen- distancethan those with small trunks. The
the angle of growth of individual seedlings dently of the angle between this dark sec- seedlings whose growth angles are repre(/). When a : 0, average growth among tor and the lightest sector of the horizon. sented in Fig. 2 were
orienting toward a
seedlings is directly toward the tree. The We introduce this new term to emphasize tree 75 cm in diameter. Although the oristandard deviation in growth angle among the adaptive nature of the response.On the entation decreasesgradually with distance,
seedlingsrelativeto the host tree is s (1)
forest floor potential host trees are dark a distinct drop-off of orientation appears
sectorson a haphazardly chiaroscuro hori- between1.9 and 2.I m. At this distancethe
zon, and direction toward a potential host silhouette comprised roughly 22" to 20" of
is almost never 180" from the lightest sec- the horizon. Our experimentswith the attor, relative to the seedling.Hence, this re- tractive power of experimentally placed
sponse is not growth away from the light, vertical boards of various widths (l to 30
but an adaptation that serves to lead the cm) confirms that the silhouette must comvine directly to a host tree. There is no rea- prise more than a few degreesof the horison to suspectthat skototropismis not un- zon in order to attract seedlines.The nar@
[ +{-,,"[e#).
I,i,"o"'\'l'l."
\-, /l I
rowest boards attracted from only a few
centimeters,whereas the widest ones attracted seedlingsat distancesup to 70 cm.
Seedling mortality resulting from burial
under falling debris,leaves,and branchesis
very high, and the growth of seedlingsis
very slow (lessthan l0 cm/month); therefore only those seedlingswithin a few decimeters are able to reach the host before
being buried under dead leaves.
As skototropism might be dangerousfor
plant, we set up a dark
a photosynthesizing
cul-de-sacto see how far vine seedlings
would grow into the dark. We placeda box
with three sides and a top on the ground
next to severalseedlings.The open side of
the box, I m wide and l5 cm high, was perpendicular to the growth direction of the
vines.After 2 months the vines had grown
into the box and in a sinusoidal pattern
parallel to the open end.They werecaught,
oscillatingjust insidethe mouth of the box.
The vines apparently switch to positive
phototropism when the light grows very
dim, but switch again to skototropism
when the light grows brighter.This switching could be homologousto the first positive and first negative phototropic responsesfound in taxa used in laboratory
plant physiology. However, Mo,nstera gi'
gantea shows no second positive phototropic response.At all but minimal light
levels these vines are skototropic, hence
this responseis distinct from those describedby classicallaboratory plant physiology (6,7).
Skototropism ceaseswhen the horizontally growing seedlingmeetsa tall, vertical
surface. In our experiments, seedling
growth changesto positively phototropic
within I month of meetingthe tree.During
this period leaf rnorphology changesinto
l0 . l'1,*v*,
7,
58 I
that of the second,saucer-leafedstage.The
positively phototropic responseof the second stage is demonstratedby the growth of
seedlingsthat encounterthe tree deep in a
growth
crotch betweena pair of buttresses;
in this case is upward but also distinctly
out war d i n t o a r e a s m o r e e x p o s e dt o s u n light. The first saucer leaves are small,
about I cm in diameter, and are Iight
green.During ascentthe leavesdarken;the
highestsaucerleavescan be 25 cm in diameter and are often dark green and covered
with epiphyllae.The translormation from
the saucer-leafto adult stageoccurs when
the vine emergesinto direct sunlight. We
infer this from observationsof the transition positionson the host trunk from saucer-leafed to adult Monstera gigantea.
When the species ascends a trunk in
unshaded sunlight it assumesadult morphology at only a few meters above the
ground. Also, we have not found the adult
morphology beneath understory canopy.
Skototropism is the only meansyet proposed that will lead a vine directly to a
host. Random searching (2, 4), negative
geotropism (3), positive phototropism (5),
or growth away from the light (2, 8) can
only lead the seedlinginto an area where
there may be a host, or into a light area
until it
where the vine can photosynthesize
finds a host by random movement. Becauseof its obvious adaptivevalue skototropism may be a general mechanismfor
host location in ground-germinatingarboreal vines (9). We can find only one previous author, though, with data that can be
interpretedto indicate a whole-plant skototropic response.In this case an unidentified vine speciesis describedas pursuing
a stake that was moved daily to a different
PS,loqY, A 6,tazs
position relative to the sun's diurnal
course.However the author offersthis, not
as evidence of skototropism, but as
of plants,"
"proof " of the "consciousness
"that they think," and that "plants belong
to the philosopherclass" (10).
D o N n r - o R . Sr n o N c, Jn .
T s o u n s S. R ,q v,Jn .
Biological Science Department,
Florida State U niversity,
Tallahassee 32306
Referencesand Notes
l. E. Batshelet.Statistical Methods for the Analvsis
of Problems in Animal Orientation and Certain
Biologicat Rhvthms (American lnstitute of Biol oqi c i l Sc i enc e,w as hi ngton. D .C .' 1 9 6 5 ).
2 C. R. Darwin. rlre Movements and Habits ol
Climbing Plqnts (Mvruy, London. I875).
J. E. A. Menninger. f lowering l/ines of the World:
An Encvclopedia o[ Climbing P/artt (Hcarthside.
N ew Yor k . 1970) .
4. A. C. Hottes, A Little Book oJ Climbing Plants
( D e La M ar e. N ew Yor k , l 9l l ) .
5. b. H . J anz en, " Vi ne popul ati on d y n a mi c s a n d
c om m unr l y s tr uc tur e. O r gani z ati o n f o r Tro p i e u l
Studies course report," in Advanced Eiolog.u
Ecolo7y in the Tropic.s(Organization for Tropical
Studi e;. U ni v er s i tv of C os ta R i c a. S a n J o s e . 1 9 6 9 )
6. A. W. Galston, iir Encyclopedia of Plant Ph,-si'
ol os y , W . R uhl and. Ed. ( Spr i nge r V e rl a g . B e rl i n .
f95C ) ,v ol . l ?. N o. l . pp.492 529.
7 H E Streer and H. Opif. The Phvsiologv ol
Flowering P/anr.r(American Elsevier, New York,
r970).
8. T . A. Kni eht, " Sec hs pfl anz enp h y s i o l o g i s c haeb handl uns en.' i n O s tw ai d's , Kl as si k e r d e r E ra k t e n
W i s s enc - haJ ten.
N . R .62 ( Lr epz i g. 1 8 9 5 t
9. Si nc e thi s r epor t w as s ubm i tted w e h a v e g a t h c rc d
s i m i l ar ev i denc e that another C o s t a R i c a n v i n e
species [Syzgoniaz, new species(Birdsey. nomen
in edit)l is skototropic. This second species,howev er . hi s a di fl er eni l i fe hi s tor y a n d mo rp h o J o g l
s tr r teg) . l t * i l l be des c r i bedr n s ub s e q u e npt u b l rc a ti ons ( T . S. R ay , J r ., i n pr epar atr o n ).
10. A. Bierce. "Moxon's master," in Ten Tales (Cur'
w en, London, 1925) .
I L W e i hank C . H ar r i s . M . E. T hor n t o n ' R D re s s l e r,
R . F ox , P. H om ann, M . Bi r ds ey ,a n d C . V e z a t -R o oi c al for hel p w i th v ar r ous por ti o n s o f t h i s w o rk '
T h" doto w er e eather ed at i he La S e l v a l i e l d u n rt
of the Organiz-ation for Tropical Studies' Saraoi oui . C os i a R i c a, dur i ng J une th ro u g h N o v c mb e r
19i 4. Suppor t w as fr om N Sh gra n t G B 4 2 5 1 5 ro
D .R .S.
23 May 1975