Tree Densities and Sex Ratios in Breeding Populations of Dioecious Central Amazonian

Myristicaceae
Author(s): David D. Ackerly, Judy M. Rankin-De-Merona and William A. Rodrigues
Source: Journal of Tropical Ecology, Vol. 6, No. 2 (May, 1990), pp. 239-248
Published by: Cambridge University Press
Stable URL: http://www.jstor.org/stable/2559274
Accessed: 28-06-2016 16:39 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to
Journal of Tropical Ecology

This content downloaded from 128.110.184.42 on Tue, 28 Jun 2016 16:39:27 UTC
All use subject to http://about.jstor.org/terms
Journal of Tropical Ecology (1990) 6:239-248. With 1 figure

Tree densities and sex ratios in breeding

populations of dioecious Central Amnazonian
Myristicaceae

DAVID D. ACKERLY* ? , JUDY M. RANKIN-DE-MERONAt

and WILLIAM A. RODRIGUES4

* World Wildlife Fund - US, Washington, DC, USA

t Departamento de Ecologia, Instituto Nacional de Pesquisas da Amazonia,
Manaus, Brazil
$ Departamento de Botadnica, Instituto Nacional de Pesquisas da Amazonia,
Manaus, Brazil
? Present address: Department of Organismic and Evolutionary Biology,
Harvard University, 16 Divinity Avenue, Cambridge, Massachusetts,
02138 USA

ABSTRACT. The densities of the breeding populations and the sex of all flowering individuals
were recorded for five dioecious canopy tree species of Central Amazonian Myristicaceae, in
11 study areas of the Minimum Critical size of Ecosystems Project totalling 22.5 ha. Adult
population densities were extremely low, ranging from 0.38 to 1.61 ha-' for the five species
studied. In a 10 ha study plot the mean distance to the nearest flowering conspecific ranged
from 48 to 100 m, while the mean distance to the nearest opposite sex conspecific was 147 m.
The two most abundant species, Iryanthera macrophylla and Virola calophylla, both showed
male-biased sex ratios, of 23:9 and 20:6, respectively. The size class distribution of males,
females and non-flowering individuals in V. calophylla suggests that earlier reproductive matura-
tion of male plants may provide a partial explanation for this bias. In I. macrophylla, since 95%
of the individuals were observed flowering, the observed ratio is representative of the popula-
tion, and may be caused by sex shifts from male to female. The low reproductive densities,
combined with the skewed sex ratios and overlapping generations of these species, create very
small effective breeding populations, placing species such as these at great risk in the face of
deforestation and habitat fragmentation.

KEY WORDS: Amazonia, dioecy, flowering, Minimum Critical Size of Ecosystems Project,
Myristicaceae, sex ratio, tropical trees.

RESUMO. [Densidades de airvores e razao dos sexos em populac6es reprodutivas de especies

dioicas da familia Myristicaceae na Amazonia Central.J

Foi realizado um levantamento da densidade de individuos adultos e do sexo de todos os

individuos reprodutivos, para cinco especies de arvores dioicas da familia Myristicaceae. 0
estudo foi densenvolvido em 22,5 ha distribuidos por onze reservas biologicas na Amazonia
Central dentro do Projeto de Dinaimicas Biol6oicas de Fragmentos Florestais. A densidade das
populaioes foi extremamente baixa, de 0,38 a 1,61 arvores ha', para as cinco espetcies estudadas.
A dist'ancia media ao coespeci-tico mais proximo foi de 48 a 100 m, enquanto a distancia media
ao coespecifico do sexo oposto mais proximo atingiu valores de ate' 147 m. As duas especies
mais abundantes, Iryanthera macrophylla e Virola calophylla, apresentaram razoes de sexos de

(239)

This content downloaded from 128.110.184.42 on Tue, 28 Jun 2016 16:39:27 UTC
All use subject to http://about.jstor.org/terms
 240 DAVID D. ACKERLY et al.

23:9 e 20:6 respectivamente, com uma clara predominancia de individuos masculinos. A

distribui;ao de classes de tamanho para ambos os sexos e de indivi'duos nao reprodutivos em V.
calophylla sugere que o amadurecimento precoce dos indivi'duos masculinos seja uma das razoes
para a dominancia dos mesmos. Em I. macrophylla, uma vez que 95% dos individuos flores-
ceram, a raza6 de sexos observada e representitiva da populagao, e pode ser causada por mu-
dangas sexuais de uma fase masculina para uma fase feminina. As baixas densidades de indi-
viduos reprodutivos, aliadas as diferengas nas razoes de sexos e a sobreposi`ao de geraoes,
reduzem muito o tamanho efetivo das populag5es reprodutivas, colocando especies como
estas em alto risco de exting6o, devido ao desmatamento e a fragmentaqao de seus habitats.

INTRODUCTION

Tropical moist forests around the world are characterized by high diversity of
constituent tree species, many of which occur at very low densities. In com-
parison with temperate forests, a large proportion of tropical tree species are
obligate outcrossers (Bawa et al. 1985a, Bullock 1985). The successful repro-
duction of sparsely distributed populations of outbreeding taxa is only possible
due to the widespread reliance on animal pollination (Bawa et al. 1985b).
There are few data on the densities of breeding populations. Most studies of
population dynamics survey large numbers of trees based on a minimum dia-
meter at breast height (DBH), in order to examine patterns of growth and
mortality (e.g. Hubbell & Foster 1986, Lieberman & Lieberman 1987). De-
pending on the species' size at first flowering, only a small portion of surveyed
trees may be reproductive. For example, in Central Amazonia the bat pollina-
ted overstorey species Caryocar pallidum A. C. Smith (Theaceae) has a repro-
ductive density of only 0.33 trees ha-', and less than one quarter of the indi-
viduals >10 cm DBH are reproductive (Rankin-de-Merona & Ackerly 1987).
Many dioecious tropical trees also have biased sex ratios (Opler & Bawa 1978)
which will reduce further the effective population size. Species which occur
at low densities are particularly sensitive to the effects of habitat fragmenta-
tion. The minimum viable population will vary from species to species (Soule
1987) and is particularly hard to estimate for long-lived tree species. Hubbell
& Foster (1986) suggest an absolute minimum density of 0.6 adults ha-' for
outcrossing overstorey tree species on Barro Colorado Island. This estimate is
based on the absence of dioecious species at lower densities; the minimum
viable density for some species is probably higher and will depend on the
pollinator dynamics and other life history attributes.
The mean distance to the nearest flowering conspecific (DNC) provides a
measure of dispersion which is significant from the perspective of animal
pollination (Hubbell 1979), though these distances must be scaled to pollina-
tor flight distances for meaningful interpretation. In Guanacaste Province,
Costa Rica, Bawa & Opler (1977) reported mean distances of 3.9 to 9.8 m for
five dioecious species. In a larger study of 61 tree species in a Guanacaste dry
forest, Hubbell (1979) found that for the 30 most common species the nearest
adult was within 40 m, while for the rarest species this distance increased to
over 100 m. In two Australian rain forest populations of Myristica insipida

This content downloaded from 128.110.184.42 on Tue, 28 Jun 2016 16:39:27 UTC
All use subject to http://about.jstor.org/terms
 Reproductive densities of Amazonian trees 241

R. Br. Armstrong & Irvine (1989a) observed mean distances of roughly 9-

20 m, corresponding to densities of 27 and 19 ha-l, respectively.
This study reports the results of observations on sex ratios and reproductive
densities for five dioecious species of Myristicaceae in Central Amazonia. To
the best of our knowledge these are the first such data from Neotropical
populations of Myristicaceae, an important family of dioecious plants through-
out the tropics, complementing the recently published data of Armstrong &
Irvine (1989a, b) from Australia.
Floral morphology in Myristicaceae conforms closely to the syndrome of
dioecy in tropical trees (as suggested by Bawa 1980, Bawa & Opler 1975). The
flowers are unspecialized, very small (less than 2 mm in breadth) and dull in
colour, yellow-green in Iryanthera and rusty-brown in Virola. They are pro-
bably entomophilous, though the pollinators of the Neotropical taxa are
unknown. Myristica fragrans Houtt., the commercial nutmeg of Asia, and M.
insipida, of Australia, are both beetle pollinated (Armstrong & Drummond
1986, Armstrong & Irvine 1989b) and these authors suggest that the same is
true for all four myristicaceous genera of South and South-east Asia.

STUDY AREA

This study was conducted in the evergreen lowland rain forest of Central Ama-
zonia in the research plots of the Minimum Critical Size of Ecosystems Project
(MCSE, World Wildlife Fund - US - Instituto Nacional de Pesquisas da Ama-
zonia), located 80 km north of the city of Manaus, Amazonas, Brazil (longitude
590 55' W, latitude 20 24' S) (Lovejoy et al. 1986). The plots vary in size from
1 to 10,000 ha; clear-cutting for cattle pasture has isolated some plots from the
surrounding virgin forest. Annual rainfall totals 2000-3000 mm with a distinct
dry season from June to October (Marques Filho et al. 1981). The soils are pre-
dominantly clay on the uplands, with sand in the stream bottoms.
As part of the tree survey of the MCSE Project, all trees >10 cm DBH have
been labelled and mapped in over 66 ha, comprising roughly 43,000 individuals.
To date more than 500 species have been identified in the surveyed areas
(Rankin-de-Merona et al. in press). This study was carried out in 11 areas
totalling 22.5 ha: four isolated areas of 1 ha each (plots number 2107, 2108,
3114, 1104); one isolated area of 10 ha (1202); and six areas in undisturbed
forest totalling 8.5 ha (1101, 1102, 1103, 1105, 1109, 1201).

METHODS

Three genera of Myristicaceae, Virola, Iryanthera and Osteophloeum, occur in

the reserves of the MCSE Project. Virola and Osteophloeum are large, canopy
trees, the latter occasionally growing to over 30 m in height and 60 cm DBH;
Iryanthera are generally medium-sized trees of the understorey and mid-canopy.

This content downloaded from 128.110.184.42 on Tue, 28 Jun 2016 16:39:27 UTC
All use subject to http://about.jstor.org/terms
 242 DAVID D. ACKERLY et al.

Table 1. Distribution of the numbers of individuals of eight species of Iryanthera (I), Osteophloeum
(0) and Virola (V) in the 11 study areas. The six areas in continuous forest have been combined.

Continuous Isolated 1 ha reserves Isolated

forest 10 ha reserve
Plot no. (6 areas) 1104 2107 2108 3114 1202 Total

Area (ha) 8.5 1 1 1 1 10 22.5

Species:
.L macrophylla 10 4 4 - - 17 35
L elliptica 3 - - 2 1 9 15
Ljuruensis 16 1 3 3 3 10 36
I. aevis 13 1 1 1 - 8 24
L polyneura 12 - - - 3 15
O. platyspermum 6 - - - - 12 18
V. calophylla 20 1 3 3 1 38 66
V. elongata 10 1 - 2 1 25 39
Others 41 3 5 2 2 37 90

Total 131 11 16 13 8 159 338

Density (ha-') 15.4 11 16 13 8 15.9 15.0

Although the family is commonly considered to be entirely dioecious (Hey-

wood 1978), recent work by Rodrigues (1980) has confirmed earlier reports
(Ducke 1950, Smith 1937) that several species of Iryanthera are monoecious.
Four of these, I. laevis Mgf., L juruensis Warb., I. elliptica Ducke, and I. coriacea
Ducke, are found in the MCSE study area.
The 11 study plots contained a total of 338 individuals with DBH >10 cm
belonging to 18 species of Myristicaceae. Of these trees 173 were among the
five dioecious species which flowered during the study; the four monoecious
species had a total of 77 individuals (Table 1, and J. M. Rankin-de-Merona, un-
published data). All individuals wvvere determined to species, and voucher
collections are deposited in the MCSE Project herbarium. Flowering of the
Myristicaceae in the MCSE Project area is concentrated from May to October
(J. M. Rankin-de-Merona, unpublished data). Some species of Iryanthera are
known to flower below 10 cm DBH, but the large areas and low densities of
each species made it impractical to locate all reproductive individuals.
Five visits were made to the study area between June 1985 and January
1986 to verify the reproductive status and determine the DBH of all individuals;
flowers were collected and preserved in alcohol to verify each tree's sex with
the aid of a dissecting microscope. Distances to the nearest flowering con-
specific (DNC) and the nearest flowering conspecific of the opposite sex
(DNCOS) were calculated for the populations of each species within the 10 ha
study area (1202), based on the maps of the tree survey.

RESULTS

The densities of the reproductive populations ranged from 0.38 to 1.61 ha'
(Table 2). These values represent conservative estimates based only on those

This content downloaded from 128.110.184.42 on Tue, 28 Jun 2016 16:39:27 UTC
All use subject to http://about.jstor.org/terms
 Reproductive densities of Amazonian trees 243

Table 2. Reproductive densities and sex ratios in the five dioecious species which flowered during the
study. Nt indicates total sample size; Nf is the number which flowered. Densities (ha-') are based only on
the reserves in which the species occurs (see Table 1). Undet. indicates individuals which flowered but
were not sexed. Chi-square test is used to test for significant deviation of sex ratios from 1:1. (*P< 0.05;
**P <0.01; NS = not significant.)

Density
Species Nt Nf (ha-') Male Fem. Undet. M:F ratio x2

1. macrophylla 35 33 1.61 23 9 1 2.55 6.12*

I. polyneura 15 12 0.65 7 5 - 1.40 0.32NS
0. platyspermum 18 7 0.38 2 5 - 0.40 1.28NS
V. calophylla 66 28 1.24 20 6 2 3.33 7.52**
V. elongata 39 14 0.65 7 6 1 1.17 0.08NS

plots in which the species occurred (see Table 1). Iryanthera macrophylla
(Benth.) Warb., and probably also Iryanthera polyneura Ducke, flower below
10 cm DBH, so the tree survey sample is an underestimate of the entire repro-
ductive population.
Overall, the two species with the largest reproductive populations, L macro-
phylla and Virola calophylla Warb., both had significant male-biased sex ratios
of 23:9 and 20:6, respectively (Table 2). In the 10 ha plot (1202), the largest
contiguous area sampled, the corresponding figures are 11:5 and 13:6 (Table 3).
The size class distributions of the two sexes (Figure 1) show that males of V.
calophylla flower at smaller sizes than the females. The smallest flowering
male had a diameter of 11.7 cm, compared with 15.7 cm for the smallest female.
This distribution suggests that all reproductive females in the population were
included in this sample. Males and females of I. macrophylla occur in all size
classes. Two smaller flowering individuals encountered in the study area, of
7.2 and 9.3 cm DBH, were both male. In both species, sampling of smaller size
classes would probably include more males, thus increasing the male-biased sex
ratio.
One important difference between the two species is the proportion of non-
flowering and effectively sterile trees. Since only 26 of 67 individuals of V.
calophylla flowered during the study, the sex of roughly two-thirds of the
trees of this species is unknown. Two of these trees, one male and one female,
are known to have flowered in past years but did not do so in 1985. In con-
trast, all but two individuals of I. macrophylla were observed in flower, so the
observed sex ratio represents the actual ratio for this portion of the population.
The mean distances to the nearest conspecific for the dioecious species ranged
from 48 to 100 m, while for the monoecious species they ranged from 66 to
136 m (Table 3). Of more significance for the pollination of dioecious species
are the distances to the nearest conspecific of the opposite sex, which range
from 57 to 147 m for these species. The difference between these two values
is greatest, as would be expected, in species with a skewed sex ratio. In V.
calophylla, which has twice as many males as females the DNCOS is twice as
large as the DNC. For Virola elongata (Benth.) WVarb., with four males and

This content downloaded from 128.110.184.42 on Tue, 28 Jun 2016 16:39:27 UTC
All use subject to http://about.jstor.org/terms
 244 DAVID D. ACKERLY et al.

25-

Fertile, sex unknown

20-

DSterile
5- 15
o Male

10 E Female

10 20 30 10 15 20 25

Diameter (cm) Diameter (cm)

Virola caloiphyl la I ryanthera macrophyl la

Figure 1. Size class distributions of all flowering and non-flowering individuals (> 10 cm DBH) of I. macro-
phylla and V. calophylla in 22.5 ha of forest

Table 3. Sex ratios and nearest neighbour distances in 10 contiguous hectares of forest, plot 1202. DNC
is the average distance in metres to the nearest conspecific in flower, DNCOS is the average distance to
the nearest flowering conspecific of the opposite sex. The last three species are monoecious, so only
DNC is given.

Species Nt Nf Male Female M:F ratio DNC (m) DNCOS (m)

I. macrophylla 17 16 11 5 2.20 49* 75*

O. platyspermum 12 6 2 4 0.33 100 147
V. elongata 25 9 4 5 0.80 52 57
V. calophylla 38 19 13 6 2.16 48 97
I. elliptica 9 4 136
I. juruensis 10 10 66
I. laevis 8 7 69

* L macrophylla flowers below 10 cm DBH, so these are overestimates.

five females, the two distances are almost the same. Due to the low densities
of flowering individuals, the trees growing near the edge could not be excluded
from this analysis, so the mean distances reported here are probably slight
overestimates.

DISCUSSION

The estimates of reproductive population density obtained here appear to be

roughly similar to the minimum viable density of 0.6 adults ha-1 suggested by
Hubbell & Foster (1986). Although there is considerable uncertainty in these
results because of the small sample sizes, it is important to recognize that these
samples represent the breeding populations of the species in a fairly large area

This content downloaded from 128.110.184.42 on Tue, 28 Jun 2016 16:39:27 UTC
All use subject to http://about.jstor.org/terms
 Reproductive densities of Amazonian trees 245

of forest (22.5 ha). Six of the ten additional species of Myristicaceae, which did
not flower during this study, have a total of six or fewer individuals >10 cm
DBH in the same area, and so their reproductive densities must be considerably
lower. The distances to the nearest conspecific we observed are 10-20 times
higher than the corresponding results reported by Bawa & Opler (1977) for
dioecious species in Costa Rica, roughly five times greater than those in Myris-
tica insipida (Armstrong & Irvine 1989a), and similar to the values for the rare
species in Hubbell's (1979) study.
The significance of the low density from the plant's perspective depends on
the flight distances of the pollinators. Myristica insipida, of Australia, has
flowers with similar morphology and is pollinated by small beetles 1-3 mm
in length. Based on experimental pollination studies, fruit and seed set in this
species is apparently pollination limited (Armstrong & Irvine 1989a, b). The
pollinators of the Neotropical species are probably fairly small insects as well,
which may not travel great distances, thus limiting gene flow through pollen
transfer. The population genetic consequences of short distance pollen dis-
persal will depend on the effectiveness of gene flow through seed dispersal as
well (Levin & Kerster 1974). From the pollinator's viewpoint, these species
may form a guild, with fairly short distances between functionally similar
flowering individuals, though they belong to several species. The simultaneous
flowering of several species will increase pollinator attraction at the expense
of pollen load dilution, a possible case of Mullerian mimicry among plants.
The small flower size, and the height and wide spacing of these trees renders
study of these questions exceedingly difficult.
The male-biased sex ratios observed here are consistent with those reported
for several other tropical tree species (Armstrong & Irvine 1989a, Opler & Bawa
1978, Rathore 1969, Styles 1972) and reinforce the conclusion that male-
biased ratios are more common in woody plants and perennials (Lloyd 1973).
The mechanism of sex determination in the Myristicaceae is not clearly under-
stood, though it is of considerable practical importance in the cultivation of
commercial nutmeg, Myristica fragrans. Flach (1966) reports that seedlings
raised in plantations segregate into a 50:50 ratio of male and female, suggesting
a simple Mendelian mechanism. He adds, however, that other authors have
reported a bias towards females in populations growing in good conditions (cf.
Freeman et al. 1980).
The size difference between the smallest flowering males and females for V.
calophylla (Figure 1) suggests that males may start flowering earlier, a possible
explanation for biased sex ratios, as suggested for several Costa Rican species
(Opler & Bawa 1978) and for M. insipida (Armstrong & Irvine 1989a). Dif-
ferences between the sexes in the frequency of flowering could also contribute
to the observed sex ratios in this species, as reported for Guarea rhopalocarpa
Radlk. in Costa Rica (Bullock et al. 1983). Since nearly two-thirds of the
individuals of V. calophylla did not flower during this study, the actual sex
ratio of the population is not known, and may be close to 1:1.

This content downloaded from 128.110.184.42 on Tue, 28 Jun 2016 16:39:27 UTC
All use subject to http://about.jstor.org/terms
246 DAVID D. ACKERLY et al.

In contrast to V. calophylla, the observed sex ratio of I. macrophylla repre-

sents the actual ratio of the sampled population. This species flowers below
10 cm DBH, however, so the overall size class distribution for the reproductive
population is not known. The lack of non-reproductive individuals (Figure 1)
eliminates the possibility of differences in frequency of flowering as a cause
of the sex ratio bias and suggests that differential maturation is not solely res-
ponsible for the observed ratio. The increasing frequency of females in the
large' size classes argues against differential mortality as a cause. We suggest two
potential causes of this bias: (1) a bias in seed and or seedling populations, or
(2) the possibility that the species is not truly dioecious, but diphasic (Lloyd &
Bawa 1984), switching from male to female with increasing size, possibly due
to improved canopy position (see Charnov 1982, Freeman et al. 1980, Lloyd &
Bawa 1984). The latter hypothesis is strengthened by two lines of circumstan-
tial evidence. First, in the 'dioecious' Myristica fragrans all apparently 'male'
trees, upon close inspection, are found to bear some female flowers and fruit,
and the number of fruit produced increases with increasing size (Flach 1966).
It is possible, however, that this is the result of artificial selection, since the
incidence of female flowers on male trees in the wild is exceedingly low (J.
Armstrong pers. comm.). Second, monoecy is present in several Iryanthera
species, as noted in the introduction, and Rodrigues (1980) suggested that it
represents an intermediate stage in the evolution of dioecy in the Myristicaceae.
We observed a small individual of the monoecious I. juruensis which produced
only male flowers, while larger ones produce both male and female flowers.
Similarly, I. macrophylla may produce male flowers when small, and female
flowers later, thus appearing dioecious at any given moment.
The wide spacing and low densities observed in these populations carry
important consequences for the pollination biology, genetic structure and
conservation of these trees. As noted above, the average distance to the nearest
conspecific of the opposite sex is as long as 147 m for some species in this
study area. Flight distances of up to 1 km will bring a pollinator to only a few
conspecifics of any given tree, so the size of the pollination neighbourhood will
be quite small. The transformation of Amazonia from a continuous habitat to a
patchwork of forest fragments will reduce such gene flow and even isolate
some species in precariously small populations. Franklin (1980) has hazarded
the guess that an effective population of 500 individuals is the minimum for
long-term conservation of genetic resources, though no one number will apply
in all cases (Gilpin & Soule 1986). For the most common species studied here,
an area of roughly 200 ha is necessary to set aside a population of this size.
Uneven sex ratios, however, and the long, overlapping generations of tropical
trees considerably reduce effective population size (Lande & Barrowclough
1987) and correspondingly increase the minimal area requirement (Soule
1980). For the rarer species this requirement will be orders of magnitude
greater, and will only be achieved in the largest forest preserves.

This content downloaded from 128.110.184.42 on Tue, 28 Jun 2016 16:39:27 UTC
All use subject to http://about.jstor.org/terms
 Reproductive densities of Amazonian trees 247

ACKNOWLEDGEMENTS

The WWF-INPA Minimum Critical Size of Ecosystems Project provided funding

and logistical support for this research. J. de D. F. Farias, L. G. Vasconcelos,
and J. F. Menezes aided in collection of the data. We thank F. A. Bazzaz, A.
Moreira, C. Samper and V. Viana for comments on earlier drafts of the manu-
script, and A. Moreira and V. Viana for translation of the abstract. This is pub-
lication number 55 in the technical series of the MCSE Project.

LITERATURE CITED

ARMSTRONG, J. E. & DRUMMOND, B. A. 1986. Floral biology of Myristica fragrans Houtt. (Myristi-
caceae), the nutmeg of commerce. Biotropica 18:32-38.
ARMSTRONG, J. E. & IRVINE, A. K. 1989a. Flowering, sex ratios, pollen-ovule ratios, fruit set, and
reproductive effort of a dioecious tree, Myristica insipida (Myristicaceae), in two different rain forest
communities. American Journal of Botany 76:74-85.
ARMSTRONG, J. E. & IRVINE, A. K. 1989b. Floral biology of Myristica insipida (Myristicaceae), a
distinctive beetle pollination syndrome. American Journal of Botany 76:86-94.
BAWA, K. S. 1980. Evolution of dioecy in flowering plants. Annual Review of Ecology and Systematics
11:15-39.
BAWA, K. S. & OPLER, P. A. 1975. Dioecism in tropical forest trees. Evolution 29:167-179.
BAWA, K. S. & OPLER, P. A. 1977. Spatial relationships between staminate and pistillate plants of
dioecious tropical forest trees. Evolution 31:64-68.
BAWA, K. S., PERRY, D. R. & BEACH, J. H. 1985a. Reproductive biology of tropical lowland rain
forest trees. I. Sexual systems and self-incompatibility mechanisms. American Journal of Botany
72:331-345.
BAWA. K. S., PERRY, D. R., BULLOCK, S. H., COVILLE, R. E. & GRAYUM, M. H. 1985b. Repro-
ductive biology of tropical lowland rain forest trees. II. Pollination mechanisms. American Journal
of Botany 72:346-356.
BULLOCK, S. H. 1985. Breeding systems in the flora of a tropical deciduous forest in Mexico. Bio-
tropica 17:287-301.
BULLOCK, S. H., BEACH, J. H. & BAWA, K. S. 1983. Episodic flowering and sexual dimorphism in
Guarea rhopalocarpa in a Costa Rican rain forest. Ecology 64:851-861.
CHARNOV, E. L. 1982. The theory of sex allocation. Princeton University Press, Princeton, NJ.
DUCKE, A. 1950. Plantas novas ou poucas conhecidas da Amazonia. Boletim tecnico do InstitutoAgro-
n6mico do Norte 19:3-42.
FLACH, M. 1966. Nutmeg cultivation and its sex problem. Mededelingen van de Landbouwhogeschool
Wageningen 66-1:1-85.
FRANKLIN, I. R. 1980. Evolutionary change in small populations. Pp. 135-149 in Souk, M. E. & Wil-
cox, B. A. (eds). Conservation biology: an evolutionary-ecological perspective. Sinauer Assoc., Sunder-
land, Massachusetts.
FREEMAN, D. C., HARPER, K. T. & CHARNOV, E. L. 1980. Sex change in plants: old and new obser-
vations and hypotheses. Oecologia 42:222-232.
GILPIN, M. E. & SOULE, M. E. 1986. Minimum viable populations: Processes of species extinction.
Pp. 19-34 in Soule, M. E. (ed.). Conservation biology: The science of scarcity and diversity. Sinauer
Assoc., Sunderland, Massachusetts.
HEYWOOD, V. H. (ed.). 1978. Flowering plants of the world. Oxford University Press, Oxford.
HUBBELL, S. P. 1979. Tree dispersion, abundance, and diversity in a tropical dry forest. Science 203:
1299-1 309.
HUBBELL, S. P. & FOSTER, R. B. 1986. Commonness and rarity in a neotropical forest: implications
for tropical tree conservation. Pp. 205-231 in Souk, M. E. (ed.). Conservation biology: The science of
scarcity and diversity. Sinauer Assoc., Sunderland, Massachusetts.
LANDE, R. & BARROWCLOUGH, G. F. 1987. Effective population size, genetic variation, and their use
in population management. Pp. 87-123 in Soule, M. E. (ed.). Viable populations for conservation.
Cambridge University Press, Cambridge.
LEVIN, D. A. & KERSTER, H. W. 1974. Gene flow in seed plants. Evolutionary Biology 7:139-220.

This content downloaded from 128.110.184.42 on Tue, 28 Jun 2016 16:39:27 UTC
All use subject to http://about.jstor.org/terms
 248 DAVID D. ACKERLY et al.

LIEBERMAN, D. & LIEBERMAN, M. 1987. Forest tree growth and dynamics at La Selva, Costa Rica
(1969-1 982). Journal of Tropical Ecology 3:347-358.
LOVEJOY, T. E., BIERREGAARD, R. O., RYLANDS, A. B., MALCOLM, J. R., QUINTELA, C. E.,
HARPER, L. H., BROWN, K. S., POWELL, A. H., POWELL, G. V. N., SCHUBART, H. 0. R. &
HAYS, M. 'B. 1986. Edge and other effects of isolation on Amazon forest fragments. Pp. 257-285 in
Soule', M. E. (ed.). Conservation biology: The science of scarcity and diversity. Sinauer Assoc., Sunder-
land, Massachusetts.
LLOYD, D. G. 1973. Sex ratios in sexually dimorphic Umbelliferae.-Heredity 31:239-249.
LLOYD, D. G. & BAWA, K. S. 1984. Modification of the gender of seed plants in varying conditions.
Evolutionary Biology 17:255-338.
MARQUES FILHO, A. de O., RIBEIRO, M. de N. G., SANTOS, H. M. dos & SANTOS,J. M. dos. 1981.
Estudos climatologicos da Reserva Florestal Ducke - Manaus - AM. IV. Precipitaggo. Acta Amazonica
11:759-768.
OPLER, P. A. & BAWA, K. S. 1978. Sex ratios in some tropical forest trees. Evolution 32:812-821.
RANKIN-de-MERONA, J. M. & ACKERLY, D. D. 1987. Estudos populacionais de arvores em florestas
fragmentadas e as implicagoes para conservagao in situ das mesmas na floresta tropical de Amazonia
Central. Instituto de Pesquisas e Estudos Florestais da Escola Superior de Agricultura 'Luiz de Quei-
roz' Unzversidade de Sao Paulo, No. 35:47-59, Piracicaba, Brazil.
RANKIN-de-MERONA, J. M., PRANCE, G. T., SILVA, M. F., RODRIGUES, W. A. & UEHLING, M.
Resultados preliminares de um levantamento florestal de 31 ha de terra firme na Amazonia Central:
descri;ao geral da vegetasao e dados taxonomicos. Acta Amazonica, in press.
RATHORE, J. S. 1969. Distribution patterns of male and female plants of Diospyros melanoxylon Roxb.
in the forests of Sagar, M. P. Indian Forester 96:701.
RODRIGUES, W. A. 1980. Revisao taxonomica das especies de Virola Aublet (Myristicaceae) do Brasil.
Acta Amazonica 10, Suppl.: 1-127.
SMITH, A. C. 1937. The American species of Myristicaceae. Brittonia 2:393-510.
SOUL?, M. E. 1980. Thresholds for survival: maintaining fitness and evolutionary potential. Pp. 151-
170 in Soule, M. E. & Wilcox, B. A. (eds). Conservation biology: an evolutionary-ecological per-
spective. Sinauer Assoc., Sunderland, Massachusetts.
SOULE, M. E. (ed.). 1987. Viable populations for conservation. Cambridge University Press, Cambridge.
STYLES, B. T. 1972. The flower biology of the Meliaceae and its bearing on tree breeding. Silvae Gene-
tica 21:175-182.

Accepted 6 September 1989

This content downloaded from 128.110.184.42 on Tue, 28 Jun 2016 16:39:27 UTC
All use subject to http://about.jstor.org/terms