Color profile: Disabled
Composite 150 lpi at 45 degrees
Acta Bot. Croat. 67 (2), 201–208, 2008 CODEN: ABCRA 25
ISSN 0365–0588
Reproductive ecology of mangrove trees
Ceriops decandra (Griff.) Ding Hou and
Ceriops tagal (Perr.) C.B. Robinson (Rhizophoraceae)
ALURI J. SOLOMON RAJU*, HENRY J. KARYAMSETTY
Department of Environmental Sciences, Andhra University,
Visakhapatnam 530 003, India
Ceriops decandra and C. tagal are evergreen trees in the inner mangrove forests in
Andhra Pradesh, India. Ceriops decandra is primarily a landward species whereas C.
tagal is a seaward and highly salt tolerant species. Ceriops decandra produces flowers
and fruits continuously throughout the year while C. tagal produces flowers and fruits
during winter only. Both species have a mixed mating system with cross-pollination as the
principal system and self-pollination, which is primarily vector-dependent. Ceriops
decandra flowers have a simple pollination mechanism that is adapted for pollination by
daytime foragers, Nomia bees and Odynerus wasps. In contrast, C. tagal flowers have an
elaborate and explosive pollination mechanism. Flies and honey bees trip the flowers and
contribute to explosive pollen release and subsequent self- and cross-pollination. Wind is
also effective in tripping the pollination mechanism but it largely contributes to self-polli-
nation. Bud abortion occurs only in C. decandra. Flower and fruit abortion occurs in both
the species of Ceriops. Abortion at different stages of the reproductive unit has been con-
sidered to be a strategy by the plants to adjust the available maternal resources to the
growing fruits and propagules. The paper provides the basis for further study for the con-
servation and management of Ceriops species in the study areas.
Keywords: Ceriops decandra, Ceriops tagal, pollination, mating, vivipary, mangrove.
Introduction
Limited studies have been devoted to pollination biology and mating systems, and for
mangroves the records of flower visitors and potential pollinators are mostly fragmental
(TOMLINSON 1986; SOLOMON RAJU 1990 a, b; SUBBA REDDI et al. 1995). The knowledge of
these aspects is a prerequisite to provide guidance for the conservation, planning and man-
agement of mangroves in view of their overexploitation throughout their distribution
range.
The genus Ceriops consists of two species, C. decandra and C. tagal; the former more
often grows within the tidal zone mixed with other Rhizophoraceae species while C. tagal
occurs in harder and higher muddy soil of the polyhaline zone, and forms pure stands on
* Corresponding author, e-mail: ajsraju@yahoo.com
ACTA BOT. CROAT. 67 (2), 2008 201
U:\ACTA BOTANICA\Acta-Botan 2-08\Solomon.vp
9. listopad 2008 10:36:15
Color profile: Disabled
Composite 150 lpi at 45 degrees
SOLOMON RAJU A. J., KARYAMSETTY H. J.
better drained soils and shows stunted growth in exposed and highly saline sites (AKSORNKOAE
et al. 1992, SELVAM and KARUNAGARAN 2004, TOMLINSON 1986). These two species are vi-
viparous which means that the seed produces propagules while on the mother plant and
then drops off for dispersal.
In the state of Andhra Pradesh, C. decandra is a common constituent of Krishna and
Coringa mangrove forests while C. tagal is restricted to Krishna mangroves only. In this
paper we report on the ecology of sexual reproduction and seed dispersal in C. decandra
and C. tagal in Krishna and Godavari Mangrove forests respectively.
Materials and methods
Field studies were conducted with twenty five trees of Ceriops decandra (at Coringa)
and with fifteen trees of C. tagal (at Krishna) from flowering through seedling fall, during
2004–2006. Floral structural and functional details as well as nectar characteristics were
examined according to protocols (DAFNI et al. 2005, ROUBIK 1995, BAKER and BAKER
1973). Flower visitors were observed for their role in effecting pollination. Hourly foraging
visits during the entire foraging period were counted on a selected number of branches with
flowering inflorescences. The percentage of foraging visits for each forager species was
calculated based on the daily total visits from five censuses performed on different days.
Ten individuals of each forager species were captured, washed in aniline blue on a glass
slide and the number of pollen grains counted under a microscope to determine their role in
pollen transfer. In case of C. tagal, the role of wind in petal explosion and subsequent polli-
nation was also observed during day time. For this, five easily accessible inflorescences
with newly open flowers exposed to direct sunlight towards the creek in the windward di-
rection were tagged and bagged carefully on the evening of the previous day; bags were re-
moved carefully on the morning of the next day and the tagged inflorescences were ob-
served at close quarters until late evening to note whether petal explosion occurred.
In case of C. decandra, twenty five mature buds, five each from five trees were used for
each mode of pollination – autogamy, geitonogamy and xenogamy (DAFNI et al. 2005).
Twenty inflorescences each (tagged) were used to record bud, flower and fruit abortion
rates and natural fruit set. Four inflorescences each from five trees were selected for this
purpose. Twenty inflorescences of C. tagal, five each from four trees, were used for
fruit/seedling production to judge whether breeding system is vector-dependent. Ten
tagged inflorescences, five each from two trees were used to note flower abortion rate.
Sixty four open and tagged inflorescences from ten trees were used to record natural fruit
set and fruit abortion at the site. In both species, the colour was used as an indicator of
healthy and unhealthy propagules with the former’s colour changing from green to reddish
brown or dark and from green to yellow in the latter type. The percentage of such
propagules is noted as fruit abortion rate.
Fruit seedling development and characteristics were also investigated in detail by fol-
lowing fruited flowers until propagules dropped. Visual observations on the dispersal and
establishment of propagules at the study sites were made to understand the planting strat-
egy. Some hypocotyls about to fall were tagged on the mother plants of both species in a
particular area. Field studies were made in different areas for their occurrence and subse-
quent settlement.
202 ACTA BOT. CROAT. 67 (2), 2008
U:\ACTA BOTANICA\Acta-Botan 2-08\Solomon.vp
9. listopad 2008 10:36:15
Color profile: Disabled
Composite 150 lpi at 45 degrees
REPRODUCTIVE ECOLOGY OF CERIOPS SPECIES
Results
Both the species of Ceriops are evergreen shrubby tree species. Flowering and fruiting
occurs throughout the year in C. decandra but C. tagal flowers seasonally, during Novem-
ber-February. The associate plants of C. tagal such as Aegiceras corniculatum, Lumnitzera
racemosa and Bruguiera gymnorhiza flower simultaneously. The last species is a year-long
bloomer.
In both the species, the flowers are born in condensed short-stalked cymes. An inflores-
cence produces on average 15.73 flowers (range 5–31) in C. decandra and on average 11
flowers (range 7–17) in C. tagal. The flowers are sessile and odourless in C. decandra but
are short-stalked and fragrant in C. tagal. In both the species, they are small, white,
cup-shaped, bisexual and zygomorphic. The sepals are five, small, light green, valvate en-
closing the inner parts until anthesis and not reflexed after anthesis. The petals are five,
free, two-lobed, and alternating with the sepals but they are light green in C. decandra and
white and pubescent in C. tagal. In C. decandra, the petals are inter-locked marginally by
basal short hairs and this circumstance produces a short corolla tube crowned by the series
of clavate filamentous appendages. In C. tagal, the lower margins of adjacent petals are
held together by patches of tightly intertwining, helically coiled hairs. Each petal has three
distinct clavate appendages on its distal margins. In both the species, the stamens are 10,
five of them antisepalous, five others antipetalous and all ten inserted on the rim of the
calyx cup. In C. decandra, each stamen extends beyond the height of the stigma; anthers
are longer than filaments. Disc within the stamen ring is well developed and anther lobes
enclose the base of the thick filaments. In C. tagal, each petal encloses the antipetalous sta-
men and an adjacent antisepalous stamen; the two stamens remain in the petal under ten-
sion enclosed above by the clavate appendages even after anthesis. In both the species, the
ovary is semi-inferior, three-carpelled and three-locular with a total of six ovules. Style is
slender and terminated into minute separate stigmatic lobes. The stigma stands at the height
of the stamens. In C. tagal, disc within the stamen ring is well developed and anther lobes
enclose the base of the thick filaments.
The mature buds open during 04.30–11.00 h in C. decandra and during 16.30–18.00 h
in C. tagal. In both, the calyx lobes separate at anthesis and diverge to expose the petals. In
both species, anther dehiscence occurs at anthesis. In both the stigma attains receptivity on
the second day and remains receptive up to six days. However, in C. tagal, peak receptivity
occurs from third to fifth day; the period of quick release of bubbles from the stigma on hy-
drogen peroxide application was taken as the duration of peak stigma receptivity. During
this period, the white petals turn red gradually from the top to the base. Nectar is produced
in trace amount in C. decandra but it is 5.65 ± 1.0 mL per flower in C. tagal. In C. tagal, the
nectar sugar concentration is 35–50% and the common sugars include fructose, sucrose
and dextrose with the first relatively more dominant. The nectar contains 12 amino acids
which include tyrosine, glycine, methionine, proline, lysine, aspartic acid, glutamic acid,
serine, cysteine, alanine, threonine and arginine. Of these, glycine, serine, cysteine, alanine
and threonine are relatively dominant. Amino acids such as phenylalanine, valine, leucine,
iso-leucine, tryptophan and histidine were not found in the nectar. In both the species, the
petals and stamens fall off on the seventh day of flower life. The sepals are persistent, be-
come warty and spiny gradually and remain on the plant even after the shedding of
propagules.
ACTA BOT. CROAT. 67 (2), 2008 203
U:\ACTA BOTANICA\Acta-Botan 2-08\Solomon.vp
9. listopad 2008 10:36:15
Color profile: Disabled
Composite 150 lpi at 45 degrees
SOLOMON RAJU A. J., KARYAMSETTY H. J.
In C. decandra, the fruit set is 1% in bagged flowers, 40% in geitonogamy and 92% in
xenogamy. Bud and flower abortion rates are 42% and 31% respectively. Natural fruit set is
15%. Fruit abortion rate is 3%. In C. tagal, the fruit set rate in bagged flowers is 3% and
16.3% in open-pollinations. Bud abortion is absent. Flower abortion rate is 42%. Fruit
abortion rate is 2%.
The pollinated flowers produce mature fruits in eight weeks in C. decandra and four
weeks in C. tagal. In both, each fruit invariably produces a single seed. In C. decandra,
fruits are light green, ovoid, conical and blunt apically. In C. tagal, fruits are conical by the
extrusion of the upper part of the ovary, surface brown and roughened. In both, the fruits are
distinct with a five-lobed persistent calyx. The embryo has no dormancy and penetrates
through the seed coat and the fruit pericarp and grows to a considerable size into a spin-
dle-shaped hypocotyl structure before dispersal while still attached to the maternal parent.
In C. decandra, the hypocotyl grows upright and takes three months before detachment
from the fruit. It is slender, clearly ribbed, angular, sulcate, 15 cm long and broadened at the
lower end. Healthy hypocotyls are entirely green but occasionally purple on one side. A
small number of hypocotyls are yellow and drop off prematurely. Visual observations
showed that healthy hypocotyls dropped when due, floated in tidal waters and finally set-
tled in areas far away from the mother plant. In C. tagal, the cotyledonary yellow cylindri-
cal collar (1cm long) appears from the fruit about 10 days prior to detachment of the hypo-
cotyls. The hypocotyl is 26 cm long, distinctly ridged and hangs downwards. It is initially
green, after the development of collar, it shows a gradual colour change to brownish purple
from hypocotyl end to plumule. Visual observations indicated that the mature hypocotyls
separate from fruit, leaving the latter attached to the mother plant. The detached hypocotyls
were found to settle in the vicinity of the mother plant.
Nomia bees (Halictidae) and Odynerus wasps (Vespidae) were the foragers of C.
decandra. They foraged during 10.30–16.30 h, the hourly foraging frequency of individual
species (Fig. 1). The bees collected both pollen and nectar while the wasps collected only
nectar. In so doing, they contacted the stamens first and then the stigma effecting pollina-
tion. During pollen collection, the bees rotated around the flower to collect pollen from the
anthers. They took 2–5 seconds to collect the forage from each flower. Their body
washings for pollen revealed that they carry 145 to 1,570 pollen grains (Tab. 1) and hence
have an important role in pollination. In addition to this pollen, they also carried pollen
loads in pollen baskets present on their legs. Odynerus wasps while probing the flower for
nectar, contacted the stamens and stigma effecting pollination. Their body washings indi-
No. of flower visits/tree
20
Nomia
15 Odynerus
10
0
10.30 11.30 12.30 13.30 14.30 15.30 16.30
Foraging time in hours
Fig. 1. Hourly foraging activity of Nomia bee and Odynerus wasp on C. decandra
204 ACTA BOT. CROAT. 67 (2), 2008
U:\ACTA BOTANICA\Acta-Botan 2-08\Solomon.vp
9. listopad 2008 10:36:15
Color profile: Disabled
Composite 150 lpi at 45 degrees
REPRODUCTIVE ECOLOGY OF CERIOPS SPECIES
Tab. 1. Pollen carrying capacity of pollinator insects of Ceriops decandra and C. tagal.
Insect species Sample size Range Mean ± S.D.
C. decandra
Nomia sp. 10 145–1570 1132 ± 414
Odynerus sp. 10 80– 540 330 ± 59
C. tagal
Chrysomya megacephala 10 79– 147 112 ± 21
Apis cerana indica 10 231– 413 320 ± 68
Apis florea 10 230– 406 310 ± 54
cated the presence of pollen grains ranging from 80 to 540 (Tab. 1). Both bees and wasps
frequently moved between trees of C. decandra in quest of more forage and this inter-tree
movement was considered important for cross-pollination.
In Ceriops tagal, the flowers were tripped by the honeybees, Apis cerana indica and A.
florea and the fly, Chrysomya megacephala during daytime. Honeybees foraged for pollen
and nectar occasionally while the fly foraged for the nectar consistently until the floral
source was exhausted (Fig. 2). All the three species while probing for nectar caused the vio-
lent release of stamens from the petals. In effect, the pollen from the already dehiscent an-
thers was ejected forcibly and deposited on the underside of the foraging bee or fly. Body
washings revealed the presence of pollen grains ranging from 231 to 413 per bee and from
79 to 147 per fly (Tab. 1). In case of bees, they also carried pollen loads in pollen baskets on
their legs. A single foraging visit of the bee or fly did not result in the explosion of all five
petals. The honey bees were found to concentrate principally on B. gymnorhiza and A.
corniculatum while the fly species exclusively on C. tagal.
20
Number of foraging
Chrysomya
megacephala
15
visits/tree
Apis cerana indica
10 Apis florea
0
00
00
0
:0
:0
:0
:0
:0
:0
:0
:0
:0
8:
9:
10
11
12
13
14
15
16
17
18
Foraging time in hours
Fig. 2. Hourly foraging activity of insects on Ceriops tagal.
The explosion of petals in open flowers of C. tagal was also triggered by the action of
wind. As the plant grows in the seaward zone, high winds are characteristic of the site, as
our experience substantiated. Of the twenty four flowers observed, four flowers exhibited
petal explosion triggered by wind action indicating that 16.6% of flowers may achieve pol-
lination by wind action. Neither in such flowers did the explosion of all five petals take
place at one time.
ACTA BOT. CROAT. 67 (2), 2008 205
U:\ACTA BOTANICA\Acta-Botan 2-08\Solomon.vp
14. listopad 2008 10:04:21
Color profile: Disabled
Composite 150 lpi at 45 degrees
SOLOMON RAJU A. J., KARYAMSETTY H. J.
Discussion
Ceriops decandra is a year-long bloomer with alternate flowering and fruiting phases
whereas C. tagal is strictly a winter bloomer with fruit and hypocotyl maturation during
summer season. In C. decandra, the floral characteristics suggest a simple floral mecha-
nism. The short basal hairs of petal edges have no evident function in C. decandra but these
hairs are well developed and help to propagate explosive pollen release effectively in the
fragrant flowers of C. tagal in which petal-stamen configuration is elaborate and special-
ized (JUNCOSA and TOMLINSON 1987). Also, petal clavate appendages in C. decandra have
abundant xylem elements with a significant reservoir of water and hydathodes at or near
the termini; they have functional significance for flower function under extreme water
pressure deficits during the day in mangrove swamps. The abundant xylem is absent in
petal appendages in C. tagal and Kandelia candel, which are pollinated at night, or early in
the morning (JUNCOSA and TOMLINSON 1987). Therefore, the abundant xylem and hyda-
thodes and their function in petal appendages of C. decandra suggest that the latter is
adapted for pollination during the day.
In Ceriops tagal, the floral characteristics suggest an elaborate and specialized floral
mechanism. The petals require a delicate external touch for the explosive release of sta-
mens. The helically coiled hairs at the lower margins of the petals help to propagate explo-
sive pollen release effectively (JUNCOSA and TOMLINSON 1987). The petal clavate append-
ages lack hydathodes and abundant xylem which are characteristically present and have a
role in flower function under extreme water pressure deficits during the day in C. decandra
(JUNCOSA and TOMLINSON 1987). Such a state may make appendages very light and provide
necessary trigger for petal explosion by delicate touch by the forager in C. tagal.
Trigonid bees and small insects usually visit C. decandra flowers (JUNCOSA and
TOMLINSON 1987). Wasps and flies have been previously found to be suitable for pollina-
tion in C. decandra (TOMLINSON 1986). In the present study, Nomia bees and Odynerus
wasps were consistently frequent pollinators of C. decandra flowers. Their foraging fre-
quency and efficiency together characterize them as prime pollinators (VAZQUEZ et al.
2005). Further, the anthesis process is gradual and the accumulation of new flowers with
pollen and nectar begins to appear from late morning. Hand-pollination tests show that it is
an obligate outcrosser. The protandry, the long period of stigma receptivity and long flower
life substantiate this; pollination by bees and wasps favours outcrossing (TOMLINSON 1986).
The pollen recovered from the body washings of these foragers suggests that the latter ef-
fect pollination. The commonness of C. decandra at the study site provides ample opportu-
nities for out-crossing effectively and the genetic variation thus achieved would permit the
species to survive well in the harsh mangrove environment.
Night-flying insects are probable pollinators in Ceriops tagal (TOMLINSON 1986).
Moths visit the flowers for nectar and bees may also visit during daytime. In this study, we
suggest that characteristics such as the absence of nectar guides, and tubular corolla, begin-
ning of nectar secretion prior to anthesis, moderate nectar volume with high sugar concen-
tration having the dominance of hexose sugars, and shallow nature of flowers are suitable
for foraging by short-tongued bees and flies (BAKER and BAKER 1983, CRUDEN et al. 1983).
Apis bees and Chrysomya flies in terms of their foraging efficiency and frequency are the
prime pollinators (VAZQUEZ et al. 2005). Chrysomya flies frequent the flowers of C. tagal
daily effecting pollination but they have limited pollen transport capacity; this however, is
206 ACTA BOT. CROAT. 67 (2), 2008
U:\ACTA BOTANICA\Acta-Botan 2-08\Solomon.vp
9. listopad 2008 10:36:16
Color profile: Disabled
Composite 150 lpi at 45 degrees
REPRODUCTIVE ECOLOGY OF CERIOPS SPECIES
compensated by their numbers and could bring about substantial geitonogamy and xeno-
gamy (FAEGRI and VAN DER PIJL 1979). The close proximity between trees of C. tagal at the
study site also facilitates xenogamy. Pollination also occurs sporadically in nature due to
wind action.
Amino acids are the second most abundant class of compounds after sugars to be found
in nectar (GARDENER and GILLMAN 2002). Their concentrations in nectars are considerably
lower than sugar concentrations. But even the slightest concentrations are important nutri-
tionally and also contribute to the »taste« of the nectar (BAKER and BAKER 1983). In this
context, the floral nectar in C. tagal plays a crucial role in adding taste and supplying cer-
tain amino acids to both honeybees and flies.
Both species of Ceriops show a mixed mating system with cross-pollination as the prin-
cipal mode. C. tagal with a small number of individuals at the study site can produce off-
spring with autogamy if it fails to attract potential and adequate pollinators. C. decandra
shows a high percent of bud and flower abortion while in C. tagal, there is no bud abortion
but it displays a high percent of flower abortion. The bud or flower abortion could be due to
defective nature of the flowers and also to facilitate adjustment of the available resources to
growing healthy fruits and seedlings in order to prevent premature fruit abortion. In both
species, fertilized flowers resort to ovule abortion and always only one ovule produces seed
as against the actual number of six ovules per flower; this may enable the plant to save re-
sources and use them to produce 1-seeded viable fruits. A few propagules in both species
lack green pigment, and have been referred to as »albino« forms (ALLEN and DUKE 2006).
These propagules are non-viable, cannot photosynthesize and die after depleting reserves if
settled in the habitat. The hypocotyl grows upward in C. decandra in which the flowers are
sessile while in C. tagal, it grows downward which seems to be because of stalked flowers
and the greater weight of the hypocotyls. This is an important field characteristic feature to
distinguish C. tagal from C. decandra. In C. tagal, the cotyledonary yellow cylindrical col-
lar emerges from the fruit about 10 days prior to the detachment of the hypocotyl while this
structure is entirely absent in C. decandra.
C. tagal with epigeal seed germination, elongated and pointed hypocotyls with straight
curvature (CLARKE et al. 2001) fall freely from the mother plant and plant themselves into
the mud at the same site during low tide period. The hypocotyls if fallen during high tides
float to another site for settlement. But, field studies do not show settlement of hypocotyls
away from the mother plants suggesting that C. tagal uses self-planting strategy only. This
is further substantiated by McGUINNESS (1997) who also reported that hypocotyls of C.
tagal in northern Australia dispersed very short distances. Therefore, the study suggests
that C. tagal in an undisturbed and human-free site is almost unable to add new plants. In C.
decandra, the fallen hypocotyls float and settle in different disturbed and undisturbed areas
far away from the mother plants. This type of settlement suggests that C. decandra uses
stranding strategy for establishment.
Acknowledgements
The financial support received through a Major Research Project [No.22/14/2004-CS
(M)] from the Ministry of Environment and Forests, Government of India, New Delhi is
gratefully acknowledged. We thank Dr. S. Purnachandra Rao, SRA (CSIR) for his field as-
sistance.
ACTA BOT. CROAT. 67 (2), 2008 207
U:\ACTA BOTANICA\Acta-Botan 2-08\Solomon.vp
9. listopad 2008 10:36:16
Color profile: Disabled
Composite 150 lpi at 45 degrees
SOLOMON RAJU A. J., KARYAMSETTY H. J.
References
AKSORNKOAE, S., MAXWELL, G., HAVANOND, S., PANICHSUKO, S. 1992: Plants in man-
groves. Chalongrat Co., Bangkok, Thailand.
ALLEN, J. A., DUKE, N. C. 2006: Bruguiera gymnorrhiza (large-leafed mangrove): Species
profiles for Pacific island agroforestry. Permanent Agriculture Resources, Holualoa,
Hawai. http://www.traditionaltree.org.
NAKER, H. G., BAKER, I. 1973: Some anthecological aspects of the evolution of nectar pro-
ducing flowers particularly amino acid production in nectar. In: HEYWOOD, V. H. (ed.),
Taxonomy and ecology, 243–264. Academic Press, New York.
BAKER, H. G., BAKER, I. 1983: A brief historical review of the chemistry of floral nectar. In:
BENTLEY B., ELIAS, T. (eds.), The biology of nectaries, 126–152. Columbia University
Press, New York.
CLARKE, P. J., KERRIGAN, R. A., WESTPHAL, C. J. 2001: Dispersal potential and early
growth in 14 tropical mangroves: do early life history traits correlate with patterns of
adult distribution? J. Ecol. 89, 648–659.
CRUDEN, R. W., HERMANN, S. M., PETERSON, S. 1983: Patterns of nectar production and
plant-pollinator coevolution. In: BENTLEY, B., ELIAS, T. (eds.), The biology of nectaries,
80–125. Columbia University Press, New York.
DAFNI, A., KEVAN, P. G., HUSBAND, B. C. 2005: Practical pollination biology. Enviroquest
Ltd., Cambridge, Ontario, Canada.
FAEGRI, K., VAN DER PIJL, L. 1979: The principles of pollination ecology. Pergamon Press,
New York.
GARDENER, M. C., GILLMAN, M. P. 2002: The taste of nectar – a neglected area of pollina-
tion ecology. Oikos 98, 552–557.
JUNCOSA, A. M., TOMLINSON, P. B. 1987: Floral development in mangrove Rhizophora-
ceae. Am. J. Bot. 74, 1263–1279.
MEF, 1987: Mangrove status report. Ministry of Environment and Forests. Government of
India, New Delhi.
McGUINNESS, K. A. 1997: Dispersal, establishment and survival of Ceriops tagal propa-
gules in a north Australian mangrove forest. Oecologia 198, 80–87.
ROUBIK, D. W. 1995: Pollination of cultivated plants in the tropics. FAO Agricult. Serv.
Bull. 118, 1–196.
SELVAM, V., KARUNAGARAN, V. M. 2004: Coastal wetlands: Mangrove conservation and
management. Orientation guide 1. Ecology and biology of mangroves. M.S. Swami-
nathan Research Foundation, Chennai.
SOLOMON RAJU, A. J. 1990a: Observations on the floral biology of certain mangroves. Proc.
Indian Nat. Sci. Acad. B56, 367–374.
SOLOMON RAJU, A. J. 1990b: The explosive pollination mechanism and mating system of
the weedy mint, Hyptis suaveolens (Lamiaceae). Pl. Sp. Biol. 5, 235–241.
SUBBA REDDI, C., SOLOMON RAJU, A. J., REDDY, S. N. 1995: Pollination ecology of Avicen-
nia officinalis L. (Verbenaceae). J. Palynol. 31, 253–260.
TOMLINSON, T. B. 1986: The botany of mangroves. Cambridge University Press, New York.
VAZQUEZ, D. P., MORRIS, W. F., JORDANO, P. 2005: Interaction frequency as a surrogate for
the total effect of animal mutualists on plants. Ecol. Lett. 8, 1088–1094.
208 ACTA BOT. CROAT. 67 (2), 2008
U:\ACTA BOTANICA\Acta-Botan 2-08\Solomon.vp
9. listopad 2008 10:36:16
