Annals of Botany 94: 229–232, 2004
doi:10.1093/aob/mch142, available online at www.aob.oupjournals.org
Density Affects Gametophyte Growth and Sexual Expression of Osmunda
cinnamomea (Osmundaceae: Pteridophyta)
Y A O - M O A N H U A N G 1, H S U E H - M E I C H O U 1 and W E N - L I A N G C H I O U 2,*
1
Department of Biology, National Taiwan Normal University, Taipei 106, Taiwan and 2Division of Forest Biology, Taiwan
Forestry Research Institute, Taipei 100, Taiwan
Received: 14 January 2004 Returned for revision: 18 March 2004 Accepted: 20 April 2004 Published electronically: 30 June 2004
Background and aims To understand how gametophyte densities affect the sexual expression and sizes of
Osmunda cinnamomea and to provide information on the density of growth needed to favour successful reproduction, fresh spores were sown at various densities and subsequent gametophyte growth was studied.
Methods Spores were sown and cultured in the laboratory. Subsequent gameophytes at different population
densities were sampled and their sexual expression and sizes were recorded.
Key results One-year-old multispore cultures of the fern O. cinnamomea demonstrated that population density
affected gametophyte growth and sexual expression. As density increased, gametophytes became significantly
smaller and more slender. Female and asexual gametophytes dominated in populations of low and high densities,
respectively. At intermediate population densities, hermaphroditic and male gametophytes were dominant. Female
gametophytes were larger than gametophytes of all other types. Hermaphroditic gametophytes were larger than male
gametophytes, which were larger than asexual gametophytes. Large gametophytes were wide-cordate, whereas
smaller ones tended to be narrow-spathulate.
Conclusions Gametophyte size of O. cinnamomea is negatively related to the population density, which significantly affects gametophytes’ sexual expression. The presence of unisexual and bisexual gametophytes at
intermediate densities indicates that both intergametophytic and intragametophytic selfing may occur.
ª 2004 Annals of Botany Company
Key words: Gametophyte density, Osmunda cinnamomea, sexual expression.
INTRODUCTION
To understand fern reproduction, it is necessary to understand sexual expression and the factors that affect it.
Although homosporous ferns have the potential to be hermaphroditic, many factors may affect their growth and sexual expression (Korpelainen, 1994; Eberle and Banks,
1996; Banks, 1997; Verma and Selvan, 2001). Population
density is one of the factors affecting sexual expression and
growth in gametophyte communities. In overcrowded
populations resources are limited by competition, and
gametophytes are often asexual or male, and narrow
(usually spathulate). Female and hermaphroditic gametophytes, on the other hand, often occur in sparse populations
(Klekowski and Lloyd, 1968; Lloyd and Gregg, 1975;
Cousens, 1979).
Osmunda cinnamomea is found in Siberia, Manchuria,
Korea, Japan and China. It is rare in north-eastern Taiwan
(Shieh, 1994). Artificial propagation ex situ, for example in
botanical gardens, is one of the ways to conserve this rare
fern. Its gametophyte morphology has been described
(Momose, 1967; Nayar and Kaur, 1971). However,
whether its sexual expression and gametophyte sizes are
influenced by population density is unclear. In this study,
the effect of gametophyte density on the sexual expression
and growth of this fern was examined to provide
* For correspondence. E-mail [email protected]
information on the density of growth needed to favour
successful reproduction.
MATERIALS AND METHODS
Fresh spores of Osmunda cinnamomea L. were obtained
from a plant growing in a marsh at Tsao-Pi (24 450 N,
121 330 E), Ilan County, in north-eastern Taiwan. Spores
were sown on eight sheets of membrane filter (pore 045
mm, 47 mm in diameter; Gelman Laboratory). Each sheet
lay on 2–4 cm deep moist Sphagnum moss in a plastic
box (114 · 86 · 65 mm). All cultures were maintained
under white fluorescent illumination at about 24 mmol m–2
s1, 12 h d1 and the temperature was between 20 and
28 C. Because spores disperse randomly in nature,
instead of being controlled at specific densities as in
most experiments, spores in this study were sown randomly and gametophytes were sampled randomly in the
culture. One year later, gametophytes growing at different
population densities (1, 2, 3, 6, 10, 12, 14, 32, 37, 41, 66,
92, 131 and 253 individuals cm2) were sampled. The
length and width of each gametophyte was measured (Fig.
1), and their genders were recorded. A stereomicroscope
(Leica, Wild M8) and a compound light microscope
(Leitz, Dialux 20) were used to work out the density
and to examine the morphology, sizes and genders of
the gametophytes.
Annals of Botany 94/2, ª Annals of Botany Company 2004; all rights reserved
230
Huang et al. — Density Affects Gametophyte Growth and Sexual Expression
low-density populations and were female or hermaphroditic. Small, usually slender-spathulate gametophytes
occurred in high-density populations and were asexual
(Fig. 4). The mid-sized male gametophytes appeared in
populations of intermediate density.
DISCUSSION
F I G . 1. Measurement of gametophyte width and length in this study.
RESULTS
The chloroplast-bearing spores germinated within 1 week
of sowing. A typical, mature Osmunda cinnamomea
gametophyte was cordate and without trichomes. Initiation
of antheridia and archegonia did not occur until 6 and 9
months after sowing, respectively. Antheridia formed on
the dorsal and ventral surfaces of the wing, while archegonia formed on the ventral surface of the anterior cushion.
Sporophytes were produced 1 year after the spores were
sown.
One year after spores were sown, hermaphroditic, male,
female and asexual gametophytes appeared in the culture
synchronously. Of 17 hermaphroditic gametophytes examined, nine had antheridia containing antherozoids and eight
had empty antheridia. At the density of 1, 2 and 3 individuals cm2 all gametophytes were female, but at 41 individuals cm2 there were no female gametophytes (Fig. 2).
In contrast, the frequency of asexual gametophytes
increased with density. At densities over 66 individuals
cm2, only asexual gametophytes were produced. Male
gametophytes occurred at densities between 12 and 41
individuals cm2, and hermaphroditic gametophytes
occurred at densities between 6 and 37 individuals cm2
(Fig. 2).
Female gametophytes were much larger, in all
respects, than other types of gametophytes (Table 1). Hermaphroditic gametophytes were larger than male gametophytes, which were larger than asexual gametophytes.
Gametophyte size could be estimated by the regression
equation:
A = 17545D14247 ,
r 2 = 095, P < 005
where A = width · length (mm2) and D = individuals cm2.
This equation clearly shows that gametophyte size
decreased significantly as density increased. When the
population density was more than 16 individuals per cm2,
the size of all gametophytes did not exceed 338 mm2 (Fig. 3).
Large, usually wide, cordate gametophytes occurred in
Gametophyte growth and sexual expression are affected by
many factors, including spore size, germination time, nutrition source, growth density and antheridiogen (Schedlbauer,
1976; Nester and Schedlbauer, 1981; Korpelainen, 1994).
The density of cultured gametophytes has been shown to
modify sexual expression in several species of homosporous
ferns (Cousens and Horner, 1970; Rashid, 1970; Lloyd and
Gregg, 1975; Cousens, 1979; Rubin et al., 1985; Warne and
Lloyd, 1987; Greer, 1993). In this study, it was found that
population density affected the sexual expression of
Osmunda cinnamomea. Gametophyte densities above a specific level yield relatively constant sex ratios (Warne and
Lloyd, 1981). In this study, solely asexual gametophytes
were produced when the population density exceeded 66
individuals cm2. The high asexual/sexual percentage (74/
26) of the laboratory-cultured gametophytes and the low
asexual/sexual percentage (20/80) of field gametophytes
of Sadleria cyatheoides and S. pallida (Ranker and Houston,
2002) may partly be explained by the influence of density as
demonstrated in this study.
Female and hermaphroditic gametophytes are usually
larger than male and asexual gametophytes (Korpelainen,
1994). This was true of O. cinnamomea gametophytes. In
this observation, density affected gametophyte size and
shape, and size was closely related to gametophyte gender.
Miller (1968) stated that weak vegetative growth favours
the formation of male and asexual gametophytes, whereas
vigorous growth favours the formation of female and hermaphroditic gametophytes. When gametophytes are sparse,
each gametophyte can obtain more resources than those
gametophytes in dense populations. This may explain, in
part, why the gametophyte size was significantly negatively
correlated with population density.
Antheridium formation may consume energy and reduce
gametophyte growth (Miller, 1968; Näf, 1979). Antheridiogen, a hormone promoting antheridium formation, affects
size and size-influenced sexual expression in Dryopteris
filix-mas (Korpelainen, 1994). Potential vegetative growth
is diverted to antheridium formation by antheridiogen in
polypodiaceous ferns (Näf, 1956). However, further studies
are needed to determine whether antheridiogen is present
in O. cinnamomea. Chemicals other than antheridiogen
may also affect gametophyte growth. For example, the
gametophytes of some ferns in the Polypodiaceae may produce growth inhibitors (Chiou and Farrar, 1997). Reciprocal
allelopathy has been demonstrated between O. cinnamomea
and Dryopteris intermedia, but gametophyte growth was
not retarded by spraying with the supernatant of the same
species (Petersen and Fairbrothers, 1980). However, in this
study, it is unclear whether the growth of gametophytes in
high-density populations was inhibited by allelochemicals.
231
Huang et al. — Density Affects Gametophyte Growth and Sexual Expression
100
100
80
60
40
60
40
20
20
0
0
0
20
40
60
80
100
0
20
40
60
80
100
100
Male
80
Frequency (%)
Frequency (%)
100
60
40
20
0
Asexual
80
Frequency (%)
Frequency (%)
Female
Hermaphroditic
80
60
40
20
0
60
80
20
40
Density (individuals cm–2)
0
100
0
20
40
60
80
Density (individuals cm–2)
100
F I G . 2. Gender frequencies of Osmunda cinnamomea grown at different gametophyte densities.
T A B L E 1. Growth characteristics of Osmunda cinnamomea gametophytes of different genders (average – s.d.)
Gender
Number of samples
Width (mm)
Length (mm)
Width · length
Asexual
Male
Female
647
033 – 039
055 – 033
029 – 065
40
119 – 040
148 – 036
183 – 081
31
436 – 153
486 – 250
2445 – 2040
Gametophyte size (mm2)
200
.
A = 175.45D–1 4247
r2 =0.95 P< 0.05
150
100
50
0
0
50
100
150
200
Density (individuals cm–2)
250
300
F I G . 3. The relationship between Osmunda cinnamomea population
density and average gametophyte size after 1 year in culture. A, size; D,
density.
Different patterns of sexual expression may provide
evidence for different mating systems (Klekowski, 1972;
Pajarón et al., 1999). Unisexual gametophytes favour outcrossing, whereas bisexual ones provide the opportunity
Hermaphroditic
17
274 – 098
246 – 105
744 – 557
for intragametophytic selfing (Schneller et al., 1990; Haufler,
2002). The occurrence of unisexual (female and male) and
bisexual gametophytes of O. cinnamomea indicates that both
intergametophytic and intragametophytic selfing may produce sporophytes, although possible limitations of genetic
load on the latter have not been tested. In nature, most fern
spores are dispersed only a limited distance, with very few or
evenonlysinglespores travelling longerdistances(Pecketal.,
1990). The habitat in this study is fully covered by a Sphagnum
mat. Dispersed spores of O. cinnamomea may either gather on
the Sphagnum mat or float further and separate to various
densities. Gametophytes growing in different densities will
express different sexualities and may produce sporophytes
through different mating systems accordingly.
These results were derived from a single plant and thus may
not represent the species as a whole. However, this study does
indicate the potential that ex situ conservation of this species
may be achievable by sowing spores sparsely, as it will lead to
subsequent development of gametophytes of both sexes and
promote successful sporophyte production.
232
Huang et al. — Density Affects Gametophyte Growth and Sexual Expression
F I G . 4. Two types of Osmunda cinnamomea gametophytes: (A) female gametophytes are wide-cordate; (B) asexual gametophytes are spathulate.
ACKNOWLEDGEMENTS
We thank Drs Donald R. Farrar, Tom A. Ranker and Liz
Sheffield for their valuable comments. The assistance of
Dr Alan Warneke in editing the manuscript is gratefully
acknowledged.
LITERATURE CITED
Banks JA. 1997. The transformer genes of the fern Ceratopteris simultaneously promote meristem and archegonia development and repress
antherdia development in the developing gametophyte. Genetics
147: 1885–1897.
Chiou WL, Farrar DR. 1997. Antheridiogen production and response
in Polypodiaceae species. American Journal of Botany 84:
633–640.
Cousens MI. 1979. Gametophyte ontogeny, sex expression, and genetic load
as measures of population divergence in Blechnum spicant. American
Journal of Botany 66: 116–132.
Cousens MI, Horner Jr HT. 1970. Gametophyte ontogeny and sex expression in Dryopteris ludoviciana. American Fern Journal 60: 13–27.
Eberle JR, Banks JA. 1996. Genetic interactions among sexdetermining genes in the fern Ceratopteris richardii. Genetics 142:
973–985.
Greer GK. 1993. The influencce of soil topography and spore-rain density
on gender expression in gametophyte populations of the homosporous
fern Aspidotis densa. American Fern Journal 83: 54–59.
Haufler CH. 2002. Homospory 2002: an odyssey of progress in pteridophyte
genetics and evolutionary biology. BioScience 52: 1081–1093.
Klekowski Jr EJ. 1972. Evidence against genetic self-incompatibility
in the homosporous fern Pteridium aquilinum. Evolution 26: 66–73.
Klekowski Jr EJ, Lloyd RM. 1968. Reproductive biology of the Pteridophyta. I. General considerations and a study of Onoclea sensibilis L.
Botanical Journal of the Linnean Society 60: 315–324.
Korpelainen H. 1994. Growth, sex determination and reproduction of
Dryopteris filix-mas (L.) Schott gametophytes under varying nutritional conditions. Botanical Journal of the Linnean Society 114:
357–366.
Lloyd RM, Gregg TL. 1975. Reproductive biology and gametophyte
morphology of Acrostichum danaeifolium from Mexico. American
Fern Journal 65: 105–120.
Miller JH. 1968. Fern gametophytes as experimental materials. Botanical
Review 34: 361–440.
Momose S. 1967. Prothallia of the Japanese ferns (Filicales). Tokyo:
University of Tokyo Press.
Näf U. 1956. The demonstration of a factor concerned with the initiation of
antheridia in polypodiaceous ferns. Growth 20: 91–105.
Näf U. 1979. Antheridiogens and antheridial development. In: Dyer AF,
ed. The experimental biology of ferns. London: Academic Press,
435–470.
Nayar BK, Kaur S. 1971. Gametophytes of homosporous ferns. Botanical
Review 37: 295–396.
Nester JE, Schedlbauer MD. 1981. Gametophyte development in Anemia
mexica Klotzsch. Botanical Gazette 142: 242–250.
Pajarón S, Pangua E, Garcı́a-Álvarez L. 1999. Sexual expression
and genetic diversity in populations of Cryptogramma crispa
(Pteridaceae). American Journal of Botany 86: 964–973.
Peck JH, Peck CJ, Farrar DR. 1990. Influences of the history attributes on
formation of local and distant fern population. Amercian Fern Journal
80: 126–142.
Petersen RL, Fairbrothers DE. 1980. Reciprocal allelopathy between the
gametophytes of Osmunda cinnamomea and Dryopteris intermedia.
American Fern Journal 70: 73–78.
Ranker TA, Houston HA. 2002. Is gametophyte sexuality in the laboratory
a good predictor of sexuality in nature? American Fern Journal 92:
112–118.
Rashid A. 1970. In vitro studies on sex expression in Lygodium flexuosum.
Phytomorphology 20: 255–261.
Rubin G, Robson DS, Paolillo Jr DJ. 1985. Effects of population density on
sex expression in Onoclea sensibilis L. on agar and ashed soil. Annals of
Botany 55: 205–215.
Schedlbauer MD. 1976. Fern gametophyte development: controls of
dimorphism in Ceratopteris thalictroides. American Journal of Botany
63: 1080–1087.
Schneller JJ, Haufler CH, Ranker TA. 1990. Antheridiogen and natural
gametophyte populations. American Fern Journal 80: 143–152.
Shieh WC. 1994. Osmundaceae. In: Editorial Committee of the Flora of
Taiwan, eds. Flora of Taiwan I, 2nd edn. Taipei: Sandos Chromagraph
Printing Company, 80–87.
Verma SC, Selvan PM. 2001. Intraspecific variation in spore-size of homosporous ferns and its implications on fern mating systems. Bionature
21: 1–9.
Warne TR, Lloyd RM. 1981. Inbreeding and homozygosity in the
fern, Ceratopteris pteridoides (Hooker) Hieronymus (Parkeriaceae).
Botanical Journal of the Linnean Society 83: 1–13.
Warne TR, Lloyd RM. 1987. Gametophyte density and sex expression in
Ceratopteris. Canadian Journal of Botany 65: 362–365.