Content uploaded by Ekaterina Kozuharova
Author content
All content in this area was uploaded by Ekaterina Kozuharova
Content may be subject to copyright.
Доклади на Българската академия на науките
Comptes rendus de l’Acad´emie bulgare des Sciences
Tome 66, No 10, 2013
BIOLOGIE
Botanique
POLLINATION ECOLOGY OF HABERLEA RHODOPENSIS
FRIV. (GESNERIACEAE), A TERTIARY RELICT ENDEMIC
TO THE BALKAN PENINSULA
Katerina Bogacheva-Milkoteva, Ekaterina Kozuharova,
Regine Claßen-Bockhoff∗, Andrej Gogala∗∗
(Submitted by Academician V. Golemansky on May 28, 2013)
Abstract
This study presents preliminary results on the pollination strategy and re-
productive success of the tertiary relict and endemic species Haberlea rhodopen-
sis Friv., Gesneriaceae in a highly fragmented habitat in Bulgaria and Greece.
The flowers of the species receive very scarce insect visits, mainly from
non-specialized insects. The species is obviously not attractive to honeybees
and bumblebees and we could not find a specific pollinator. Flowers offer pollen
and scarce nectar as a reward. Pollination success is characterized by high seed
production and rate of pollination.
Key words: Haberlea rhodopensis, pollination, nectar, seed set
The study was conducted with the financial support of a project of Deutsche Bundesstiftung
Umwelt, Medical University of Sofia and Johannes Guthenberg University of Mainz, 2011.
1427
Introduction. Haberlea rhodopensis is an endemic species of the Balkans.
It is a resurrection plant, belonging to the tropical family Gesneriaceae. In Eu-
rope there are only three genera of the family: Haberlea (in Bulgaria and Greece),
Ramonda (Balkan Peninsula and Pyrenees) and Jankaea (in Greece) [1]. All of
them are tertiary relicts, having evolved under climatic conditions, which differ
strongly from the recent ones. The flora of Southeastern Europe, the Early to
Middle Miocene flora, was rich, diverse in thermophilous elements, consisting of
polydominant mesophytic to hygromesophytic forests. Climatic changes after the
Miocene climatic optimum caused changes in the floristic composition and veg-
etation structure, characterized by a decrease in the abundance of palaeotropic
and thermophilous elements, a reduction of macrothermic elements, and the dis-
appearance of evergreen laurel forests. Together with these changes, the role of
arctotertiary species in plant communities in the mesophytic forests increased.
Major vegetation changes occurred in the late Miocene, under more diverse cli-
matic conditions [2].
The flowers of H. rhodopensis have a prominent entomophilous syndrome,
namely of a “gullet” type with pollen transferred nototribically [3–5] (Figs 1, 2)
and our previous studies show that they are incapable of self-pollination. Obser-
vations on actual pollination in the Gesneriaceae are scanty [3]. With regard to
the European representatives, there are studies on the pollination biology of the
related species Jankaea heldreichii [6], some observations on honeybee influence
on Haberlea rhodopensis [7] and on Ramonda myconi [8]. Therefore, it is of great
interest for the development of conservation strategies for the species to study its
pollination characteristics in the context of its relict nature, fragmented habitat
and the challenges of the changing environment.
Molecular analysis with nuclear and plastid markers for molecular analysis
of five actinomorphic genera in the Old World Gesneriaceae shows that floral
actinomorphy in the family has evolved convergently in different alliances. The
authors hypothesize that the evolution from zygomorphy to actinomorphy, with
novel combinations of characters, is possibly due to shifts in pollination strate-
gies, such as a switch from nectar to pollen rewards and a switch to generalist
pollinators. It is suggested that Haberlea rhodopensis might be a relict of the
stock from which the actinomorphic flowers of Ramonda evolved [9].
It has been argued that many of the ways in which humans disturb natural
habitats are likely to disrupt plant-pollinator interactions, with consequent out-
cross pollen limitation, and that the ecological context of anthropogenic pollen
Fig. 1. Flower visitors of Haberlea rhodopensis. Left to right, first row: Pollinating
Lasioglossum morio.L. morio with pollen load. Left to right, second row: Syrphids
were frequent flower visitors. Evidence for predation on the flowers
→
1428 K. Bogacheva-Milkoteva, E. Kozuharova , R. Claßen-Bockhoff et al.
limitation can generate diverse selective forces on plant mating systems. In
many cases, we expect an evolutionary shift towards higher frequencies of self-
fertilization, although the selective forces and traits involved are likely to vary
with the nature of the disturbance. It will be particularly important to deter-
mine whether human disturbance is qualitatively different from natural factors
that cause PL and whether plant populations have the demographic and genetic
capacity for evolutionary responses adequate to ensure persistence in changing
environments [10 ].
Pollination systems are often more generalized and dynamic than the con-
cept of “pollination syndromes” suggests. Moderate to substantial generalization
often occurs. Plant generalization is predicted by a simple model: as long as tem-
poral and spatial variance in pollinator quality is appreciable, different pollinator
species do not fluctuate in unison and they are similar in their pollinator effec-
tiveness. In conservation biology, generalized pollination systems imply resilience
to linked extinctions, but also the possibility that introduced generalists might
displace natives with a net loss of diversity [11].
Flowers in six communities from three continents were scored for expression
of floral traits used in published descriptions of the pollination syndromes, and
simultaneously the pollinators of as many species as possible were characterized.
Ordination of flowers in a multivariate “phenotype space”, defined by the syn-
dromes, showed that almost no plant species fell within the discrete syndrome
clusters. Furthermore, in approximately two-thirds of the plant species, the most
common pollinator could not be successfully predicted by assuming that each
plant species belongs to the syndrome closest to it in the phenotype space. A
fresh look at how traits of flowers and pollinators relate to visitation and pollen
transfer is recommended [12].
In our study, we compared native populations and plants grown ex situ in
the Botanical Garden of Mainz University. We tested: 1) the identity of the pol-
linator agent; 2) the nature of the reward; 3) the extent to which the abundance
of the fruit and seed set could be used as estimators of pollination success.
Materials and methods. The study sites. The population at Ustovo
district is situated in the Central Rhodope Mts, at 800 m. a.s.l., along Cherna
River, on limestone rocks in broadleaf forest with meadows nearby. Co-flowering
genera during the observation time in the range of 30 m were: Taraxacum offici-
nale L., Geranium sp., Alliaria sp., Lamium sp., Veronica sp., Crataegus monog-
yna Jacq., Syringa vulgaris L., Apium sp., Rhinantus sp., Sambucus nigra L. The
weather during the observation period was mostly cloudy, with rain and storms
←Fig. 2. Flower visitors of Haberlea rhodopensis – small size generalist insects
Compt. rend. Acad. bulg. Sci., 66, No 10, 2013 1429
in the afternoon with temperatures at noon between 11.5 ◦C and 22 ◦C. In the
period between 8.06. and 10.06.2012, the weather was sunny with temperatures
22–30 ◦C.
The population in the village of Nenkovo is situated in the Eastern Rhodope
Mts, at 400 m a.s.l., along Borovitsa River, on limestone rocks in an open area
near meadows. Co-flowering genera during the observation time in the range of
30 m were: Malus sp., Prunus sp., Salix sp., Alnus sp., Muscari sp., Veronica
sp., Viola sp., Geranium sp. The weather was sunny, with temperatures of about
18.5 ◦C.
The cultivated population in the Botanical Garden of Mainz University is
in the alpinum of the garden at about 90 m a.s.l. A bumblebee nest was near
the stand of Haberlea rhodopensis. The weather was rainy, with temperatures of
about 14 ◦C between 26 and 29.04.2012, and in the period 29.04.2012–06.05.2012
there was warm (23 ◦C) and sunny weather.
Pollinators/visitors. The tests were performed according to standard
methods [13]. Most observations were conducted in 2010–2012, during the flow-
ering period from April to July. For in situ observations we chose two natural lo-
calities in Central and Eastern Rhodope Mts Bulgaria, and a population from the
Botanical Garden at Johannes–Gutenberg University in Mainz, Germany. The
total observation time of 52 h was adjusted according to the peak of flowering
of the populations and distributed during the periods 16.05.2010–20.05.2010 and
01.06.2012–10.06.2012 at Ustovo district, 24.04.2010–26.04.2010 in the village of
Nenkovo and 26.04.2012–06.05.2012 in the Botanical garden of Mainz from 08.00
till 22.00 h. We used the transect method [9]. Previous sporadic observations in
four more localities in Rhodope Mts in 2008 and 2009 (villages of Beden, Trigrad
and Yagodina) supplied additional information for the study.
Reward – nectar presence and concentration. We obtained preliminary
results on nectar presence in the flowers of H. rhodopensis, using glucose test
strips. In the second phase of the study, nectar concentration was measured at
different times of the day on bagged and free flowers with hand refractometer
Eclipse 0–50% Sucrose. For estimating nectar quantity, we used glass capillaries
of 1 µl, 2 µl, 5 µl and 10µl.
Pollination success. Pollination success was evaluated by: a) free pollina-
tion fruit set of a random sample; we collected randomly 36 inflorescences with
177 flowers from four sampling sites at the village of Yagodina and the village
of Trigrad in Rhodope Mts; b) seed set per fruit of a random sample on plants
of different sampling sites; we counted the seeds, non-developed ovules and dam-
aged seeds from a total of 27 capsules from four different sampling sites. For each
sampling site, we calculated the average for each category.
Results and discussion. Pollinating agents. Our observation showed
very scarce insect visits, even at the peak of flowering of the population and at
a time of high insect activity. Honeybees and bumblebees avoid the flowers of
1430 K. Bogacheva-Milkoteva, E. Kozuharova , R. Claßen-Bockhoff et al.
Fig. 3. Pollination rate of a random sample of H. rhodopensis
Haberlea rhodopensis and visit other co-flowering species. They rarely approach
the flower and do not try to reach pollen or nectar. We compared the populations
of H. rhodopensis and Veronica sp. in Ustovo district, a distance of 1 meter apart,
undertaking 40 min observation in conditions of high insect activity. Veronica sp.
received 18 visits by Apis meliffera, while H. rhodopensis received 3 visits with
duration of no more than two seconds. Many insects approach the flowers without
trying to get inside. In the Botanical Garden of Mainz, a bumblebee nest was
near the site, but the bumblebees did not visit the flowers of H. rhodopensis.
In both natural habitats, we found many small-size visitors, probably us-
ing the flowers as shelter or feeding on pollen – ants, caterpillars, beetles, flies
(Formicidae, Coleoptera, Diptera, larvae, Fig. 2). We found evidence for preda-
tion, pollen or nectar robbing. In the Botanical Garden in warm and dry weather
(temperatures above 23 ◦C), in addition to the above-mentioned, we noted reg-
ular pollination visits by Lasioglossum morio (Fabricius 1793), collecting pollen,
storing it and moving from flower to flower. A group of five H. rhodopensis indi-
vidual flowers received 20 visits in 20 min. Similarly high activity was observed
during four days of warm weather between 10 a.m. and 12 at noon. Additionally,
syrphids were active flower visitors (Fig. 1).
Compt. rend. Acad. bulg. Sci., 66, No 10, 2013 1431
Fig. 4. Seed set per fruit of H. rhodopensis
Flowers do not close at night, but by observations after sunset between 6
p.m. and 10 p.m. we could not find any insects at any of the experimental sites.
Despite its typical bee flower characteristics – corolla blue (rarely white) with
yellow spots in the mouth, tube widening above, limb bilabiate, nectary annu-
lar (Weber) – Haberlea rhodopensis is obviously not attractive to bumblebees or
honeybees, as our observations in situ revealed. Our previous studies on the pol-
linators of this plant species show that honeybees do not visit its flowers. The
species receives no insect visits except single nectar-collecting Bombus terrestris
Linnaeus queen and we suggested that pollination is accomplished by nectar-
collecting bumblebee queens with long proboscises [7]. This corresponds with
studies on the closely related species Jankaea heldreichii. The plant is an insect-
pollinated generalist without any particular relationship with its insect partners.
J. heldreichii receives scarce visits only from bumblebee queens – Bombus lu-
corum, Pyrobombus soroeensis and P. pratorum [6]. Ramonda myconi, another
sister species to Haberlea rhodopensis, is mainly visited by bumblebees and syr-
phids (M. Riba and F.X. Pic´o, unpublished result). Insects involved in the pol-
lination of the Gesneriacese include bees (Apiidae, Euglossini in the neotropics,
Anthophoridae, Xylocopinae, etc.), rare butterflies and very rare moths. Wasps,
dipterans, gnats, etc. are not so far definitely known as pollinators; they probably
1432 K. Bogacheva-Milkoteva, E. Kozuharova , R. Claßen-Bockhoff et al.
play an insignificant role, if any at all [3]. In the recent climatic conditions the
plants exist as clone-populations. That is why geitonogamy is probable.
Reward. Haberlea rhodopensis has a strikingly orange-coloured nectar disc,
but we noted a very low quantity of nectar <1µl per flower. Nectar concentration
varied between 10 and 42% sucrose, but concentrations above 15% were noted only
in sunny weather with temperatures of above 23 ◦C. We suppose that nectar is
produced in the morning, since the highest concentrations are measured between
9 a.m. and 12 at noon. There was no difference between bagged and free flowers.
The flowers had a very weak scent.
Flowers of Gesneriaceae are primarily nectar flowers. Lack of nectar (rather
rare) is a phylogenetically secondary condition. There are taxa in which the
nectary is non-functional or completely reduced – deceptive nectar flowers and
pollen flowers. Attraction of pollinators is primarily by visual cues. Most of
the flowers are devoid of any floral scent [3]. The genera Ramonda and Jankaea
do not have nectaries [3, 6]. Possibly Haberlea rhodopensis is on the way to lose
its nectar, due to changes in the pollinator complex. Haberlea rhodopensis flow-
ers are self-compatible but do not self-pollinate spontaneously. This plant pos-
sesses the ability for vegetative propagation by sending thin, horizontal rhizomes
(Bogacheva-Milkoteva, Kozuharova, in press).
Pollination success. From the 177 flowers tested for pollination rate, 141
(79.6%) had fruits, 29 (16.4%) were not pollinated and 7 (4%) were predated.
We recorded an average of 768 (Ncapsules = 27) ovules per fruit. An average
of 458 (59.8%) of them matured to seeds. An average of 232 (30.3%) unset ovules
per fruit were counted. An average of 76 (9.9%) predated or damaged ovules
per fruit were counted. Generally in Gesneriaceae seeds are produced in great
numbers, are small and have little nutrition storage [3]. R. myconi fruits contain
many small seeds (on average 500 seedsper capsule; M. Riba, unpublished result).
In Jankaea heldreichii, the number of seeds per capsule varies between 134 and
613. Just 50% of the flowers of J. heldreichii survive to fruiting stage [6].
Conclusion. The hypothesis for the pollination syndromes [4, 5] predicts
that flowers with a definite complex of flower traits are linked to definite type of
pollinators. According to this hypothesis, Haberlea rhodopensis flowers should be
pollinated by bees. However, we could not observe such a relationship. In situ
practically no visitors were observed, except a sporadic one by a nectar-collecting
queen Bombus terrestris Linnaeus [6]. However, the pollination is effective, if we
take into account the results from reproductive success. Ex situ we found only
non-specialized visitors and the only bee species we observed was Lasioglossum
morio, Fig. 1. This social bee is a polylectic [14], widely distributed species [15].
It was a frequent visitor to all flowers in the Botanic Garden.
Our hypothesis. Haberlea rhodopensis is a Tertiary relict which, due to its
desiccation tolerance and abilities for vegetative propagation, survived dramatic
climate changes. During these climatic changes, the local flora changed. This
Compt. rend. Acad. bulg. Sci., 66, No 10, 2013 1433
is probably connected with shifts in the pollinator complexes. In many cases,
when there was a specialization to a particular pollinator, because of the climatic
changes, the flowers had to switch to more generalized pollination system. Rarely
has the path of adaptation to mainly pollen-feeding insects been taken. In the
Balkans, Gesneriaceae relicts are such an example of adaptation traits. Haberlea
rhodopensis has small nectar quantity and lacks specialized visitors in spite of
the zygomorphic flower. Jankaea could be regarded as the “missing link” in the
evolution of the European Gesneriaceae (zygomorphic but without nectary) and
Ramonda is the next stage – a typical pollen flower.
A similar evolutionary trend is observed in Dalechampia (Euphorbiaceae)
vines. There is molecular phylogenetic evidence for an evolutionary shift in
Dalechampia from a highly specialized relationship with resin-collecting bees to
generalized pollination by a variety of pollen-feeding insects. This shift was associ-
ated with dispersal from Africa to Madagascar, where the specific resin-collecting
pollinators are absent. These results show that plants dispersing beyond the
range of their specific pollinators may succeed by evolving more generalized pol-
lination systems. Recent fieldwork in Madagascar has shown that the species
of Dalechampia found there offer only pollen as a reward for pollinators, and
that most are pollinated by a variety of pollen-feeding insects, including beetles
(Cerambycidae, Scarabidae), muscoid flies (Diptera) and several bees (Halictidae,
Anthophoridae, Apidae). It seems that resin-collecting megachilid bees, which are
the only pollinators of Dalechampia in Africa, failed to colonize Madagascar. The
ancestral Dalechampia colonists of Madagascar were probably pollinated inciden-
tally by other pollen-feeding insects. They subsequently adapted to the absence
of their specific pollinators by losing the gland that secretes the resin reward and
by effectively using diverse pollen-feeding insects as pollinators. These changes
were sufficiently successful to allow secondary diversification on the isolated is-
land of Madagascar [16]. Historical climate change influences the modularity
and nestedness of pollination networks, although this possibility remains poorly
tested. This lack of research is in sharp contrast to the considerable efforts to
disentangle the role of historical climate change and contemporary climate on
species distributions, richness and community composition patterns. On the ba-
sis of a global database of pollination networks, it is shown that historical climate
change is at least as important as contemporary climate in shaping modularity
and nestedness of pollination networks. Specifically, on the mainland is found
a relatively strong negative association between Quaternary climate change and
modularity, whereas nestedness is most prominent in areas having experienced
high Quaternary climate change [17].
What we have to take into account is how viable the seedlings are. Our
observations show that in nature very few seedlings survive. Possibly the seedlings
cannot compete effectively with the other vegetation. Alternatively, perhaps, they
have specialized to mycorrhiza, like the orchids, which are well known for their
1434 K. Bogacheva-Milkoteva, E. Kozuharova , R. Claßen-Bockhoff et al.
symbiosis requirements [18]. The seeds of Haberlea rhodopensis are also numerous
and tiny. These are implications for our further research.
Jankaea heldreichii occurs only in Mt. Olympus, Haberlea rhodopensis
reaches further, more to the north in the Rhodope and Stara Planina Mts and
Ramonda serbica, being least specialised with regard to pollinators, is found on
northernmost sites. One possible explanation for the distribution pattern of Ges-
neriaceae members on the Balkans could be their pollination adaptations, together
with other environmental requirements.
Acknowledgements. We express our gratitude to Frank O’Reilly, B.A.,
Ph.D (Uni. London) for the correction of the English language.
REFERENCES
[1]Ganchev I. Rep. Inst. Bot. Bulg. Acad. Sci., 1, 1950, No 1, 191–214.
[2]Ivanov D., T. Utescher, V. Mosbrugger, S. Syabryaj, D. Djordjevi´
c-
Milutinovi´
c, S. Molchanoff. Palaeogeography, Palaeoclimatology, Palaeoecol-
ogy, 304, 2011, Nos 3–4, 262–275.
[3]Weber A. Gesneriaceae. In: The families and genera of vascular plants, Springer,
No VII, 2004, 63–158.
[4]Faegri K., L. van der Pijl. The principles of pollination ecology, 3rd rev. edn.
Oxford, Pergamon Press, 1979.
[5]Richards A. J. Plant breeding systems, 2 ed., London, Chapman & Hall. 1997,
529 pp.
[6]Vokou D., Th. Petanidou, D. Bellos. Biol. Conserv., 52, 1990, No 2, 125–133.
[7]Kozuharova E. Phytol. Balcan., 14, 2008, No 3, 410–412.
[8]Xavier Pico F., M. Riba. Plant Ecology, 161, 2002, No 1, 1–13.
[9]Daphni A. Pollination ecology: a practical approach. Oxford Univ. Press, 1992.
[10]Westrich P. Die Wildbienen Baden-W¨urttembergs, 2. verb. Aufl., Ulmer,
Stuttgart, 1990, 891.
[11]Ebmer A. W. Linzer biol. Beitr., 20, 1988, No 1, 527–711.
[12]Wang Y. Z., R. H. Liang, B. H. Wang, J. M. Li, Z. J. Qui, Z. Y. Li, A.
Weber. Taxon, 59 , 2010, No 4, 1044–1052.
[13]Eckert C., S. Kalisz, M. Geber, R. Sargent, E. Elle, P. O. Cheptou, C.
Goodwillie, M. Johnston, J. Kelly, D. Moeller, E. Porcher, R. Ree, M.
V. Marıon, A. Winn. Trends in Ecology and Evolution, 25, 2010, No 1, 35–43.
[14]Waser N., L. Chittka, M. Price, N. Williams, J. Ollerton. Ecology, 77,
1996, No 4, 1043–1060.
[15]Ollerton J., R. Alarcon, N. Waser, M. Price, S. Watts, L. Cranmer,
A. Hingston, C. Peter, J. Rotenberry. Annals of Botany, 103, 2009, No 9,
1471–1480.
[16]Armbruster W., B. Baldwin. Nature 394, 1998, 632.
Compt. rend. Acad. bulg. Sci., 66, No 10, 2013 1435
[17]Dalsgaard B., K. Trøjelsgaard, A. M. Mart
´
ın Gonz´
alez, D. Nogu´
es-
Bravo, J. Ollerton, T. Petanidou, B. Sandel, M. Schleuning, Z. Wang,
C. Rahbek, W. J. Sutherland, J.-C. Svenning, J. M. Olesen. Ecography,
2013, Early View (Online version of record published before inclusion in an issue).
[18]Baskin C. C., J. M. Baskin. Seeds. Ecology, Biogeography and Evolution of
Dormancy and Germination, San Diego, Academic Press, 2001, 890.
Department of Pharmacognosy
Faculty of Pharmacy
Medical University of Sofia
2, Dunav Str.
1000 Sofia, Bulgaria
∗Johannes Gutenberg-University
Institut f¨ur Spezielle Botanik
55099 Mainz, Germany
∗∗Slovenian Museum of Natural History
Preˇsernova 20, p.p. 290
SI-1001 Ljubljana, Slovenia
1436 K. Bogacheva-Milkoteva, E. Kozuharova , R. Claßen-Bockhoff et al.
