Convergent evolution of sunbird pollination systems of Impatiens species in tropical Africa and hummingbird systems of the New World: Bird pollination of Impatiens species

Abstract

The bird pollination systems of the New and Old Worlds evolved independently, and differ in many aspects. New World plants are often presented as those adapted to hovering birds while Old World plants to perching birds. Most Neotropical studies also demonstrate that in hummingbird species rich assemblages, only a small number of highly specialized birds exploits the most specialized plants with long corollas. Nevertheless, recent research on bird–plant pollination interactions suggest that sunbird pollination systems in the Old World have converged more with the highly specialized hummingbird pollination systems than previously thought. In this study we focus on the pollination systems of the bird pollination syndrome Impatiens species on Mt. Cameroon, West Africa. We show that despite the high diversity of sunbirds on Mt. Cameroon, only Cyanomitra oritis appear to be important pollinator of all Impatiens species. This asymmetry indicates the absence of pair wise co-evolution and points to a diffuse co-evolutionary process resulting in guilds of highly specialized plants and birds; a situation well known from hummingbirds and specialized plant communities of the New World. Additionally, the herbaceous habits of Impatiens species, the frequent adaptations to pollination by hovering birds, and the habitat preference for understory in tropical forests or epiphytic growth, resemble the highly specialized Neotropical plants. © 2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, ●●, ●●–●●.

Full text

Convergent evolution of sunbird pollination systems of
Impatiens species in tropical Africa and hummingbird
systems of the New World
ŠTE
ˇPÁN JANEC
ˇEK1,2*, MICHAEL BARTOŠ1and KEVIN YANA NJABO3
1Institute of Botany, Academy of Sciences of the Czech Republic, Dukelská 135, CZ-379 82 Trˇebonˇ,
Czech Republic
2Department of Ecology, Faculty of Science, Charles University in Prague, Vinicˇná 7, CZ-128 44
Praha 2, Czech Republic
3Center for Tropical Research, UCLA Institute of the Environment and Sustainability, Los Angeles,
CA 90095, USA
Received 14 October 2014; accepted for publication 28 November 2014
The bird pollination systems of the New and Old Worlds evolved independently, and differ in many aspects. New
World plants are often presented as those adapted to hovering birds while Old World plants to perching birds. Most
Neotropical studies also demonstrate that in hummingbird species rich assemblages, only a small number of highly
specialized birds exploits the most specialized plants with long corollas. Nevertheless, recent research on
bird–plant pollination interactions suggest that sunbird pollination systems in the Old World have converged more
with the highly specialized hummingbird pollination systems than previously thought. In this study we focus on
the pollination systems of the bird pollination syndrome Impatiens species on Mt. Cameroon, West Africa. We show
that despite the high diversity of sunbirds on Mt. Cameroon, only Cyanomitra oritis appear to be important
pollinator of all Impatiens species. This asymmetry indicates the absence of pair wise co-evolution and points to
a diffuse co-evolutionary process resulting in guilds of highly specialized plants and birds; a situation well known
from hummingbirds and specialized plant communities of the New World. Additionally, the herbaceous habits of
Impatiens species, the frequent adaptations to pollination by hovering birds, and the habitat preference for
understory in tropical forests or epiphytic growth, resemble the highly specialized Neotropical plants. © 2015 The
Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 127–133.
ADDITIONAL KEYWORDS: Cameroon – Cinnyris reichenowi – co-evolution – Cyanomitra oritis – guilds –
hovering – nectar robbing – rainy season.
INTRODUCTION
Ornithophilous plants evolved independently in many
phylogenetic lineages of angiosperms both in the New
and Old Worlds (Cronk & Ojeda, 2008). Although
these plants share many similar traits, other features
of their pollination systems seem to be determined by
specific evolutionary history and/or environmental
factors in the different geographical regions. Conse-
quently we find plants specialized to hovering hum-
mingbirds in America, and many plants specialized to
perching sunbirds with blossoms providing a well
defined perch in South Africa and Australia
(Siegfried, Rebelo & Jones, 1985; Westerkamp, 1990;
Anderson, Cole & Barrett, 2005; Fleming &
Muchhala, 2007). Flowers visited by hummingbirds
are typically more phenotypically specialized, with a
less open flower mouth, than those pollinated by
perching birds (Willmer, 2011). Most ornithophilous
plants of the New World grow epiphytically, whereas
those pollinated by sunbirds are often trees and
shrubs (Fleming & Muchhala, 2007). Many studies
have shown that hummingbirds and ornithophilous
plant communities in the New World are structured
into individual guilds by different degrees of
*Corresponding author. E-mail: stepan.janecek@ibot.cas.cz
bs_bs_banner
Biological Journal of the Linnean Society, 2015, 115, 127–133. With 2 figures
© 2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 127–133 127

specialization, indicating a rather diffuse than pair
wise co-evolutionary processes (Feinsinger, 1976,
1978; Wolf, Stiles & Hainsworth, 1976; Stiles, 1981;
Arizmendi & Ornelas, 1990; Dalsgaard et al., 2008).
Recent studies on sunbird-pollinated plants from
Africa now show that they share more similar traits
with plants pollinated by hummingbirds from
America than previously thought. For example, they
have similar nectar volume, similar concentration
and sucrose content (Johnson & Nicolson, 2008;
Bartoš et al., 2012). Highly specialized plants in
South Africa are only pollinated by one specialized
sunbird Nectarinia famosa (Geerts & Pauw, 2009a),
which resembles the special guild of plants (e.g.
Heliconia species) and related specialized hermits in
America (Stiles, 1975). The first example of a native
African plant adapted to bird hovering, (similar to
many plants in America), was reported by Janecˇek
et al. (2011). Unfortunately, there is still very limited
knowledge on specialized sunbird-pollinated plants
in tropical Africa (e.g. Evans, 1996; Ley &
Claβen-Bockhoff, 2009) where paradoxically, the
highest diversity of sunbirds occurs (Cheke, Mann &
Allen, 2001).
In tropical Africa the genus Impatiens, has been
suggested to be highly specialized to sunbird pollina-
tion and to play an important role in sunbird-plant
co-evolution (Grey-Wilson, 1980; Janecˇek et al., 2011).
This hypothesis is so far, mainly supported by indirect
evidence. The early Pliocene and Pleistocene radia-
tions of the genus Impatiens (Janssens et al., 2009)
seem to be similar to those of some groups of sunbirds
(Fjeldså & Lovett, 1997). These Impatiens and
sunbird species are mostly found on the mountains
(Fjeldså & Lovett, 1997) where their distribution is
often restricted to small areas (Grey-Wilson, 1980;
Cheke et al., 2001). However, only the pollination
system of Impatiens sakeriana, a species endemic to
the Cameroonian mountains and Bioko, has been
studied in the field (Janecˇek et al., 2011, 2012). That
study showed a close relationship between
I. sakeriana and the Cameroon Blue-headed sunbird
Cyanomitra oritis, a bird species sharing the same
geographical area as I. sakeriana.
In this study, we focus on the pollination interac-
tions between sunbirds and Impatiens species of Mt.
Cameroon in West Africa and attempt to show to
which extent these interactions resemble those found
in hummingbird pollination systems. There are six
described Impatiens species bearing bird pollination
syndrome in this area (Grey-Wilson, 1980; Cable &
Cheek, 1998; Cheek & Fischer, 1999). Most of the
Impatiens species flower during the rainy season,
when access to the region is very challenging, thus,
very little information is so far known on their polli-
nation systems. Here we attempt to answer two main
questions I) Do we have guilds of highly specialized
sunbirds pollinating Impatiens species similar to
guilds of specialized hermits which pollinate special-
ized groups of plants in America? II) Are Impatiens
species pollinated by hovering birds as many hum-
mingbird plant species and are sunbird and/or plant
traits decisive for hovering?
MATERIAL AND METHODS
STUDY SITE
Mt Cameroon is one of the richest biodiversity centres
in Africa (Myers, 1988; Küper et al., 2004) with a
unique vegetation zonation along an altitudinal gra-
dient (Hall, 1973; Cable & Cheek, 1998; Proctor et al.,
2007). The montane forest of this region is found at
relatively low elevations, generally above 800 m. On
the seaward southern slopes, montane species of trees
appear as low as 500 m, presumably because the
extensive cloud cover and frequent mists caused by
the proximity of the sea moderate temperatures
(White, 1981), and result in high rainfall (Graham,
Smith & Languy, 2005).
Our research was done along the trail from Mann’s
Spring (4.144°N, 9.121°E, 2280 m a.s.l.), where the
forest-grassland transition occurs, to Bakingili village
on the seacoast.
TARGET IMPATIENS SPECIES
We targeted Impatiens species bearing bird pollina-
tion syndrome i.e. plant species with red flowers and
bucciniform lower sepals (Grey-Wilson, 1980). Six of
these species have been recorded on Mt. Cameroon
(Grey-Wilson, 1980; Cable & Cheek, 1998), Impatiens
sakeriana Hook.f. is a terrestrial herb; Impatiens
hians Hook.f and Impatiens niamniamensis Gilg., can
both grow on the ground and epiphytically; Impatiens
grandisepala Grey-Wilson, Impatiens etindensis
Cheek & Eb. Fisch., and Impatiens frithii Cheek only
grow epiphytically.
Six camps were established along the transect at
different elevations (approximately 2200, 1800, 1500,
1100, 650 and 300 m a.s.l) from Mann’s Spring to the
Bakingili trail, and short trips were conducted at
these camp sites to find Impatiens species. At each
elevation, we selected five or six plants of each species
if present.
SUNBIRD OBSERVATION
Sunbirds were observed using remote video systems
during the rainy season (24 July to 2 September
2013) when all the Impatiens species were in flower.
The objective was to observe each plant for 2 days
(from 6:00 am to 6.15 pm). However, errors in video-
128 Š. JANEC
ˇEK ET AL.
© 2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 127–133

taping sometimes occurred or we were not able to
take the video systems away after 2 days because of
logistic hurdles resulting in some video recordings
differing in duration (full duration of video records
from individual plants is given in Supporting Infor-
mation, Table S1). We systematically recorded the
changes in flowers during each minute of the obser-
vational period (i.e. changes as the flowers fall down
or as the plants start to flower); individual visits of
sunbirds, (if the visit was legitimate, i.e. sunbird
makes contact with reproductive organs; or illegiti-
mate, i.e. sunbird steals nectar by piercing the flower
spur), and if sunbird visits flower by perching or
hovering. We then calculated the frequency of visits,
proportion of legitimate/illegitimate visits and propor-
tion of hovering/perching behaviour for each plant.
STATISTICAL ANALYSES
We used non-parametric permutation analysis of vari-
ance (ANOVA) in the program PERMANOVA+ for
PRIMER (Anderson, Gorley & Clarke, 2008) in the
analyses as the data on visitation frequency from
individual sunbird species on individual plants as
well as data on proportions of legitimate visits and
proportions of hovering behaviour did not fulfil the
assumption of a normal distribution (there were
many zeros in the data set).
RESULTS
Apart from I. grandisepala, we found all the Impa-
tiens species bearing bird pollination syndrome pre-
viously known from the region. In total, we recorded
1230 h of videos and 630 visits of sunbirds on 53
plants (the observed altitudinal ranges of Impatiens
and sunbirds are given in Supporting Information,
Fig. S2). All of the observed plant species bearing
bird pollination syndrome were shown to be
pollinated by sunbirds. Small bees or ants were
occasionally observed sitting on the flowers or
consuming nectar from holes in the spur which were
probably pierced by nectar robbing sunbirds, but
without contact with the reproductive organs. Of the
total sunbird visits, seven sunbird species visited
just one plant species, while only three species
visited more than one plant species: Cinnyris
reichenowi (16 plants of three species), Cyanomitra
oritis (35 plants of five species) and Cyanomitra
obscura (nine plants of three species). Five of the
seven sunbird species were young birds and/or adult
females of the genus Cinnyris and were unidentifi-
able at the species level. One plant of I. hians was
only visited by Cinnyris ursulae and one plant of
I. frithii by Cinnyris johannae.
Total visitation frequency and frequency of legiti-
mate visits were affected by both plant and sunbird
species (Table 1). Cyanomitra oritis was the most
frequently observed sunbird dominating visits on
I. niamniamensis,I. frithii and I. hians. Both
Cyanomitra oritis and Cinnyris reichenowi
co-dominated visits on I. sakeriana and I. etindensis
(Fig. 1).
Proportions of legitimate visits differed among indi-
vidual sunbird species as well as plant species. Indi-
vidual sunbirds behaved differently on different
Impatiens species (see Plant × Sunbird species inter-
action in Table 1). Whereas all visits of Cyanomitra
oritis were legitimate, two sunbirds, Cyanomitra
obscura and Cinnyris reichenowi more or less often
visited flowers illegitimately by piercing the spur
without making contact with stigma or anthers
(Fig. 1.).
The proportion of hovering behaviour was also
different on individual Impatiens species (Table 1).
Hovering behaviour was observed on three of the
five Impatiens species (Fig. 2). The most frequent
was Cyanomitra oritis on I. frithii (86%). Cyanomitra
oritis and Cinnyris reichenowi also frequently
hovered on I. sakeriana (34 and 25% respectively).
Less frequent hovering by these two species
was also observed on I. etindensis. In contrast
Table 1. Differences in behaviour of three common sunbirds on Impatiens species. Permutation ANOVA with plant and
sunbird species as factors
Visit frequency
(all visits)
Visit frequency
(legitimate visits)
Proportion of
legitimate visits
Proportion of
hovering
df Fps pperm df Fps pperm df Fps pperm df Fps pperm
Plant species (Pl) 4 6.8 <0.001 4 7.4 <0.001 4 6.0 0.001 2 8.7 <0.001
Sunbird species (Su) 2 7.5 <0.001 2 10.1 <0.001 2 11.5 <0.001 4 3.0 0.083
Pl × Su* 8 2.5 0.017 8 2.4 0.017 4 5.4 0.003 2 3.3 0.038
*Term for Proportion of legitimate visits and Proportion of hovering has empty cells.
BIRD POLLINATION OF IMPATIENS SPECIES 129
© 2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 127–133

I. niamniamensis and I. hians were only visited by
perching (Fig. 2).
DISCUSSION
Our study supported the hypothesis that sunbird
assemblages in Africa evolved guilds of highly spe-
cialized birds pollinating highly adapted plants,
similar in hummingbird assemblages in America.
Despite the high sunbird diversity on Mt Cameroon
(about 22 species, Cheke et al., 2001; Bayly &
Motombe, 2004), very few are involved in this system
resulting in extremely high specialization on Impa-
tiens species. This was especially true for the three
long-spurred plant species found in middle and lower
elevations, I. niamniamensis,I. frithii and I. hians
which were only pollinated by C. oritis. This speciali-
zation is however highly asymmetrical. Only two
sunbirds visited Impatiens species frequently and
only C. oritis seems to be principal pollinator of all the
Impatiens species. As C. oritis as well as the other
sunbirds are long-billed bird species, it is highly prob-
able that Impatiens species may have coevolved with
the long-billed sunbirds.
Interestingly, a similar trend in altitudinal gradient
has been shown for hummingbird-pollinated flora,
where the highly specialized plants of genus
Heliconia are found in the lowlands and the less
specialized plants in higher elevations (Stiles, 1981).
In contrast with C. oritis, we observed more or less
frequent nectar robbing behaviour of Ci. reichenowi
and C. obscura. Similar nectar robbing behaviours
have been observed in bird pollination systems in
America (Araujo & Sazima, 2003), South Africa
(Geerts & Pauw, 2009a), and Australia (Paton & Ford,
1977). Similarly, (e.g. see studies by Lara & Ornelas,
2001; Geerts & Pauw, 2009a) the nectar robbing
behaviour in our systems seems to be the consequence
of a mismatch between plant and sunbird traits.
Cinnyris reichenowi mostly visited flowers of
I. sakeriana and I. etindensis whose spur lengths
overlap with its bill length. The birds just robbed
nectar on long-spurred I. niamniamensis. Not only
can spur length limit visitation by some sunbird
species, long pedicels and peduncles could do also (see
Supporting Information, Movie S3), which are also
typical for plants pollinated by hovering humming-
birds in the New World (Rocca & Sazima, 2010).
Figure 1. Frequency of legitimate and illegitimate visits of sunbirds on individual Impatiens species observed on more
than one plant. Cobs,Cyanomitra obscura;Corit,Cyanomitra oritis;Creich,Cinnyris reichenowi.
130 Š. JANEC
ˇEK ET AL.
© 2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 127–133

Hovering behaviour during feeding by C. oritis and
Ci. reichenowi on I. sakeriana has already been
reported from the Bamenda Highlands, Northwest
Region of Cameroon by Janecˇek et al. (2011). In this
study, C. oritis was shown to hover frequently, a clear
indication of the special ability of this species to
hover. Although our results concur with others
(Geerts & Pauw, 2009b; Wester, 2013a, b), that the
hovering behaviour of sunbirds is largely affected by
plant traits, we also show that, possibility of hovering
on individual plant species can differ between indi-
vidual sunbird species. For example, among our
plants, I. frithii represented the most adapted to
sunbird hovering with a mean proportion of hovering
per plant by C. oritis greater than 85 percent (see
Supporting Information Movie S4). The perching
behaviour of C. oritis species was only detected on one
plant where the surrounding vegetation served as a
perch (see Supporting Information, Movie S5). This
high frequency of hovering is unprecedented and
could be related to higher precision of pollen place-
ment (Bartoš & Janecˇek, 2014). It has already been
shown that plants pollinated by hovering humming-
birds are often able to place pollens more precisely
than plants pollinated by perching birds (Rocca &
Sazima, 2010).
In this study we have demonstrated that the polli-
nation systems of Impatiens species are in many
respects, similar to those of hummingbird-pollinated
plants in America. New World ornithophilous plants
such as herbs growing near the forest floor (Snow &
Snow, 1972) or epiphytes (Buzato, Sazima & Sazima,
2000; Araújo, Sazima & Oliveira, 2013) are similar to
Impatiens species growing on Mt. Cameroon.
However, contrary to the results of Cotton (1998), who
studied co-evolution in an Amazonian hummingbird
plant community, we did not find any evidence of pair
wise co-evolution, but rather asymmetric relation-
ships where one sunbird was the main pollinator of
all Impatiens species on Mt. Cameroon. This indicates
a rather diffuse co-evolution resulting in guilds of
highly specialized plants and guilds of highly special-
ized birds. Such guild structure is also well known
from America (Stiles, 1975), and more recently from
South Africa (Geerts & Pauw, 2009a). Moreover we
provide evidence to further support the hypotheses
Figure 2. Proportions of perching and hovering behaviour of sunbirds on individual Impatiens species observed on more
than one plant. Cobs, Cyanomitra obscura;Corit, Cyanomitra oritis;Creich, Cinnyris reichenowi.
BIRD POLLINATION OF IMPATIENS SPECIES 131
© 2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 127–133

that plants adapted to hovering not only evolved in
America but also in the Old World (Janecˇek et al.,
2011). Although we still lack a complete understand-
ing of the general patterns of co-evolution between
plants and their bird pollinators, this preliminary
study expands our knowledge of the pollination
systems of ornithophilous plants in tropical Africa.
ACKNOWLEDGEMENT
We thank Luma Francis Ewome, Jakques Esembe
and Moki George Mbonde for their assistance in the
field and David Horˇák and Ondrˇej Sedlácˇek for logis-
tic support. The study was supported by the Czech
Science Foundation (project no. P505/11/1617), the
National Geographic Foundation (project no. 923012)
and the long-term research development project RVO
67985939.
REFERENCES
Anderson B, Cole WW, Barrett SCH. 2005. Specialized bird
perch aids cross-pollination. Nature 435: 41–42.
Anderson MJ, Gorley RN, Clarke KR. 2008. PER-
MANOVA+ for PRIMER: guide to software and statistical
methods. Plymouth: PRIMER-E.
Araujo AC, Sazima M. 2003. The assemblage of flowers
visited by hummingbirds in the ‘capões’ of Southern
pantanal, Mato Grosso do Sul, Brazil. Flora 198: 427–435.
Araújo FP, Sazima M, Oliveira PE. 2013. The assembly of
plants used as nectar sources by hummingbirds in a
Cerrado area of Central Brazil. Plant Systematics and Evo-
lution 299: 1119–1133.
Arizmendi M del C, Ornelas JF. 1990. Hummingbirds and
their floral resources in a tropical dry forest in Mexico.
Biotropica 22: 172–180.
Bartoš M, Janecˇ ek Š. 2014. Pollinator-induced twisting of
flowers sidesteps floral architecture constraints. Current
Biology 24: R792–R795.
Bartoš M, Janecˇ ek Š, Padyšáková E, Patácˇ ová E,
Altman J, Pešata M, Kantorová J, Tropek R. 2012.
Nectar properties of the sunbird-pollinated plant Impatiens
sakeriana: a comparison with six other co-flowering species.
South African Journal of Botany 78: 63–74.
Bayly N, Motombe FN. 2004. Final report of an ornithologi-
cal survey of Mount Cameroon. Wildlife Conservation
Society, Cameroon/Nigeria trans-Boundary Project.
Buzato S, Sazima M, Sazima I. 2000. Hummingbird-
pollinated floras at three Atlantic forest sites. Biotropica 32:
824–841.
Cable S, Cheek M. 1998. The plants of Mount Cameroon a
conservation checklist. Kew: Royal Botanic Gardens.
Cheek M, Fischer E. 1999. A tuberous and epiphytic new
species of Impatiens (Balsaminaceae) from southwest Cam-
eroon. Kew Bulletin 54: 471–475.
Cheke RA, Mann CF, Allen R. 2001. Sunbirds: a guide to
the sunbirds, flowerpeckers, spiderhunters, and sugarbirds
of the world. New Haven: Yale University Press.
Cotton PA. 1998. Coevolution in an Amazonian
hummingbird-plant community. Ibis 140: 639–646.
Cronk Q, Ojeda I. 2008. Bird-pollinated flowers in an evo-
lutionary and molecular context. Journal of Experimental
Botany 59: 715–727.
Dalsgaard B, González MM, Olesen JM, Timmermann A,
Andersen LH, Ollerton J. 2008. Pollination networks and
functional specialization: a test using Lesser Antillean
plant-hummingbird assemblages. Oikos 117: 789–793.
Evans MR. 1996. Nectar and flower production of Lobelia
telekii inflorescences, and their influence on territorial
behaviour of the scarlet-tufted malachite sunbird
(Nectarinia johnstonii). Biological Journal of the Linnean
Society 57: 89–105.
Feinsinger P. 1976. Organization of a tropical guild of
nectarivorous birds. Ecological Monographs 46: 257–291.
Feinsinger P. 1978. Community organization among
Neotropical nectar-feeding birds. American Zoologist 18:
779–795.
Fjeldså J, Lovett JC. 1997. Geographical patterns of old and
young species in African forest biota: the significance of
specific montane areas as evolutionary centres. Biodiversity
& Conservation 6: 325–346.
Fleming TH, Muchhala N. 2007. Nectar-feeding bird and
bat niches in two worlds: pantropical comparisons of verte-
brate pollination systems. Journal of Biogeography 35: 764–
780.
Geerts S, Pauw A. 2009a. Hyper specialization for long-
billed bird pollination in a guild of South African plants: the
Malachite Sunbird pollination syndrome. South African
Journal of Botany 75: 699–706.
Geerts S, Pauw A. 2009b. African sunbirds hover to polli-
nate an invasive hummingbird-pollinated plant. Oikos 118:
573–579.
Graham CH, Smith TB, Languy M. 2005. Investigating the
current and historical factors on patterns of species richness
and turnover of birds in the Gulf of Guinea Highlands.
Journal of Biogeography 32: 1371–1384.
Grey-Wilson C. 1980. Impatiens of Africa. Rotterdam: A.A.
Balkema.
Hall JN. 1973. Vegetational zones on the southern slopes of
Mount Cameroon. Vegetatio 27: 49–69.
Janecˇ ek Š, Patácˇ ová E, Bartoš M, Padyšáková E,
Spitzer L, Tropek R. 2011. Hovering sunbirds in the Old
World: occasional behaviour or evolutionary trend? Oikos
120: 178–183.
Janecˇ ek Š, Riegert J, Bartoš M, Horˇ ák D, Reif J,
Padyšáková E, Fainová D, Antczak M, Pešata M,
Mikeš V, Patácˇ ová E, Altman J, Kantorová J, Hrázský
Z, Bróm J, Doležal J. 2012. Food selection by avian floral
visitors: an important aspect of plant–flower interactions in
West Africa. Biological Journal of the Linnean Society 107:
355–367.
Janssens SB, Knox EB, Huysmans S, Smets EF, Merckx
VSFT. 2009. Rapid radiation of Impatiens (Balsaminaceae)
during Pliocene and Pleistocene: results of a global climate
change. Molecular Phylogenetics and Evolution 52: 806–
824.
132 Š. JANEC
ˇEK ET AL.
© 2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 127–133

Johnson SD, Nicolson SW. 2008. Evolutionary associations
between nectar properties and specificity in bird pollination
systems. Biology Letters 4: 49–52.
Küper W, Sommer JH, Lovett JC, Beentje HJ, van
Rompaey RSAR, Chatelain C, Sosef M, Barthlott W.
2004. Africa’s hotspots of biodiversity redefined. Annals of
the Missouri Botanical Garden 91: 525–535.
Lara C, Ornelas JF. 2001. Preferential nectar robbing of
flowers with long corollas: experimental studies of two hum-
mingbird species visiting three plant species. Oecologia 128:
263–273.
Ley AC, Claβen-Bockhoff R. 2009. Pollination syndromes
in African Marantaceae. Annals of Botany 104: 41–
56.
Myers N. 1988. Threatened biotas: ‘hot spots’ in tropical
forests. The Environmentalist 8: 187–208.
Paton DC, Ford HA. 1977. Pollination by birds of native
plants in South Australia. Emu 77: 73–85.
Proctor J, Edwards ID, Payton RW, Nagy L. 2007. Zona-
tion of forest vegetation and soils of Mount Cameroon, West
Africa. Plant Ecology 192: 251–269.
Rocca MA, Sazima M. 2010. Beyond hummingbird-flowers:
the other side of ornithophily in the neotropics. Oecologia
Australis 14: 67–99.
Siegfried WR, Rebelo AG, Jones RPP. 1985. Stem
thickness of Erica plants in relation to avian pollination.
Oikos 45: 153–155.
Snow BK, Snow DW. 1972. Feeding niches of hummingbirds
in Trinidad valley. Journal of Animal Ecology 41: 471–485.
Stiles FG. 1975. Ecology, flowering phenology, and humming-
bird pollination of some Costa Rican Heliconia species.
Ecology 41: 285–301.
Stiles FG. 1981. Geographical aspects of bird-flower
co-evolution, with particular reference to central America.
Annals of the Missouri Botanical Garden 68: 323–351.
Wester P. 2013a. Feeding on the wing: hovering in nectar-
drinking Old World birds – more common than expected.
Emu 114: 171–183.
Wester P. 2013b. Sunbirds hover at flowers of Salvia and
Lycium.Ostrich 84: 27–32.
Westerkamp C. 1990. Bird-flowers: hovering versus perching
exploitation. Botanica Acta 103: 366–371.
White F. 1981. The history of the Afromontane archipelago
and the scientific need for its conservation. African Journal
of Ecology 19: 33–54.
Willmer P. 2011. Pollination and Floral Ecology. Princeton,
NJ: Princeton University Press.
Wolf LL, Stiles FG, Hainsworth FR. 1976. Ecological
organization of a tropical, highland hummingbird commu-
nity. Journal of Animal Ecology 45: 349–379.
SUPPORTING INFORMATION
Additional Supporting Information may be found in the online version of this article at the publisher’s web-site:
Table S1. Observation periods of individual plants.
Fig. S2. Altitudinal distribution of Impatiens species and sunbirds observed on Mt. Cameroon. Vegetation belts
(1–5) following Cable and Cheek 1998 with small modifications according to our field experiences. 1, montane
grassland; 2, montane forest; 3, open canopy forest disturbed by forest elephants; 4, evergreen transitional to
littoral Atlantic forest; 5, Oil Palm plantations. Photo by Šteˇpán Janecˇek (1–5; A–G) and Ondrˇej Sedlácˇek (H).
Movie S3 Cinnyris reichenowi robbing nectar in Impatiens sakeriana.
Movie S4 Cyanomitra oritis hovering in front of Impatiens frithii flowers.
Movie S5 Cyanomitra oritis perching on surrounding vegetation when feeding on Impatiens frithii.
BIRD POLLINATION OF IMPATIENS SPECIES 133
© 2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 127–133