Growth Rates and Age Estimates of Alsophila setosa Kaulf. in Southern Brazil

2006·5.9 MB·English

by  Jairo L Schmitt

Amorican
Fern Journal 96(4j:103-lll (2006)
Growth and Age
Rates Estimates of Alsophila setosa
Southern
Kaulf. in Brazil
Schmitt
Jairo L.
Centro Universitario FEEVALE, 93510-250 Novo Hamburgo, RS, Brazil
Paulo
G. Windisc:h
Universiclade do Vale do Rio dos Sinos, UNISINGS, PPC-BIOLOGIA,
93022-000 Sao Leopoldo, RS, Brazil
—
Abstract. The tree fern Alsophila sclosa, occurs in primary and secondary forests of southern and
Two
southeastern Brazih populations in secondary forest formations in the northeastern part of
th(i
State of Rio Grande do Sul, in the municipalities of Morro Reuter (45 plants] and Sapiranga (48
Two
plants), were studied to estimate the ages of the plants. approaches were tested, one based on
the total length in relation to the yearly growth rate of the caudex, and the second on the total
number of scars and remains of stipe bases along the caudex in relation to the yearly frond
production. Estimates based on growth rates and total length did not agree with the information
and records of the past land use, whereas frond production over a longer time period presented
acceptable estimates. The development of a new plant formed through vegetative reproduction was
A
observed during three years. brief discussion of the problems of age estimates in tree ferns
is
presented.
Biological data are urgently needed for conservation efforts of endangered
Some
such commercial
species as tree ferns subject to exploitation. species are
the source of fibrous material formed by the adventitious root mass covering
the caudex w^hich used for growing orchids and aroids, while others have
is
plants removed from the field and used in landscaping and decoration. Even
with adherence to the CITES agreements restricting the trade, local use
still
management
represents a major pressure for conservation. Forest plans tend to
exclude tree ferns or inflate the information on their growth and regeneration
potential, while the concerned protection agencies lack proper infornration on
A
growth understanding and
their rates. better of the role of tree ferns the
presence of some epiphytic species urgently needed. In some cases, as
is still
by Coomes
discussed et qL [2005), tree ferns offer conditions for the
woody
and some
germination establishment of vascular seed plants,
initial
thus having an important
role in their regeneration.
many
some
Tree ferns present interesting biological aspects, growing meters
sometimes up
the canopy, while presenting only primary
tall, to forest tissue
woody
and water supply through Their growth, compared some
a tracheids.
to
plants can be considered slow, but the lack of more precise data
field
major problem Ages
constitutes a in establishing the age of individual plants.
of tree ferns tend to be estimated using a) the total caudex length in relation to
the yearly growth rates or b) the yearly frond production in relation to the total
number
of frond vestiges (old fronds, remains of stipe bases or scars) along the
The method
caudex. assumes that the growth constant throughout the
first is
NUMBER
AMERICAN VERN JOURNAL: VOLUML:
104 1)B 4 (20G(i)
life-span, while the initial stages, from the formation of the young spornphyte
gametophyle from The second method cannot
on the are out the estimates.
loft
damage
be applied arborescent species without serious to the plants (leaf
to all
scars covered by adventitious roots and/or remains of stipe bases] while the
up caudex with
initial development, to plants presenting a definite leaf scars,
also out of the estimates.
is left
Age estimates based on the caudex length and yearly growth rates were
total
presented by Conant Cyalhea arhorea Sm. and Alsophila
(1976) for (L.)
bryopbila R. M. Tryon, growing in Puerto Rico; Ortega (1984] for Sphaoropteris
&
M.
Tryon, Venezuela; Brcckle A.
senilis (Klot/sch.) R. in Bittner (1995] for
Domin,
erinaceae (H. Karst.) D. Conant, A. polystichoides (H. Christ] C.
S.
Domin
pinnula and
delgadii Sternb., C. nigripes (C. Chr.] Dcnnin, C. (H. Christ]
Domin,
(Maxon] Costa
tricbiata in Rica.
C.
Seller (1981] estimated the age of Alsophila salvinii Hook., in El Salvador
based on the frond production and the scars along the caudices, pointing out
the exact age could not be established as the frond production rate of very
thai
young plants not known. Tanner (1983] using the same basic method
is
Kuhn
estimated the age of Cyatbea pubescens Melt, ex plants growing in
Ash
Jamaica. (1987] estimated the age of C. bornei (Baker] Copel. plants in Fiji,
more
including information on the spacing of frond scars.
tlie
Walker & Aplet presented age estimates and information on the
(1994]
growth response of the caudex of Cibotiuni glaucum Sm] Hook. & Arn. after
(J.
Phosphorous
the addition of nutrients, noting that Nitrogen or increased the
caudex growth
growth These authors also verified that the has higher
rate.
when
compared younger
rates in 1000 year old forest plots, to that of forest
Durand and growth and
formations. Goldstein (2001] evaluated relative rates
the reproductive potential in native [Cibotiuin spp.] as well as invasive
Hawaii and
[Spbaevopteris cooperi (Hook, ex F. Muell.] Tryon] tree ferns in
more Arens
verified that the invasive species presents a intensive growth.
Domin,
growth Cyotbea caracasana
(2001) studied rates of (Klotzsch.] at
and
different sites in Colombia, under differing ecological conditions distinct
regeneration
stages.
Among
the tree ferns used for landscaping and decoration in the State of Rio
Grande do Sul, Brazil, Alsophila sctosa Kaulf. (Cyatheaceae] currently being
is
The
extracted from forest remnants (Windisch, 2002]. existing populations are
being reduced by deforestation lumbering and use of the land for
also for
grazing or agriculture (Schmitt and Windisch, 2005). This species grows in
humid
primary and secondary forests in South and Southeastern Brazil,
m
m
between 20 and 1800 elevation. In southern Brazil occurs also in forests
it
humid
wuth Araucaria and Podocarpus well places semi-deciduous
as as in
m
cm
grow up have caudex
10 10
forest formations. Tlie plants to tall, a ca. in
on
diameter (Castony, 1973], with remains of stipe bases (with blackish spines]
the distal part and frond, and scars (from fallen fronds] on the basal portion.
m
The crown up long and has fronds with tripinnate-pinnatisect
to 3
is
laminae (adult plants]. Considering the lack of biological data for A. setosa, as
need more knowledge development, growth
well as the for of tree fern rates
GROWTH
SCHMITT & WINDISCH: RATES ALSOPHILA
IN 105
and comparative age estimations based on different approaches are being
presented.
Material AND Methods
Fieldwork was conducted two semi-deciduous seasonal broadleaved
in
forest formations in northeastern part of the State of Rio Grande do Sul»
tlie
Brazil. The first site is located within the municipality of Morro Renter [29 32'
51^04' W, 700 m] and consists of a forest remnant surrounding a spring
S,
alt.
use water supply, being protected from and from
that in as a local cattle
is still
by The surrounded by
(confirmed
deforestation local residents]. area less
is
homogeneous secondary growth. The second the municipality of
site, at
'38' 00' W, 570 on an abandoned
Sapiranga (29 S, 51 alt. m), is situated field
An
was up photograph
that in agricultural use to the early 1960s. aerial of the
shows was devoid cover
region that this particular area of tree as late as
December of 1964, so that the current secondary forest stand at the time of the
was The two was
observations less than 40 years old. selection of populations
made between two
without intending comparative analyses the populations,
but basically in order guarantee results even predatory extraction of ferns
to
if
or lumbering resulted in the loss of one of the populations during the course of
the study,
was
(NPK) samples from both performed
Basic soil analysis of localities at
the Geosciences Laboratory of the Universidade do Vale do Rio dos Sinos.
Voucher specimens were placed the Universidade do Vale do Rio dos Sinos
at
HASU
(HASU
Herbarium
10176; 9773].
The humid and from
climate sub-tropical (Teixeira et ah, 1986], data the
is
m
mean
nearest meteorological station 127 indicate an average
(Ivoti, alt]
month C warmest
temperature of the coolest of 13.3 average of the
(July],
IX
minimum
month and
of 25. (February], absolute -I'^C (July] absolute
maximum
and
(September January] during the study period. Total
35.7''C
mm
mm, May
was 2138 month being 2000 and
precipitation the driest (40.9 in
mm
maximum
mm],
with September October
78 in 2001], rainfalls in (232.2
(303.9 mm], and January (308.6 mm]. Data for an extended period of time
more
could not be obtained. Frost occurs several times during intense winters.
Two One was
approaches were used estimate the age of the plants. based
to
on the relationship between the average yearly growth rate in height and the
total length of the caudex. The second was based on the relationship between
number
produced
the average of fronds during the year in relation to the total
number of fronds along the caudex, as determined by counting fronds and their
and by complete
vestiges (remains of stipes scars the fronds after
left
decomposition
of the stipe bases].
m
maximum
up
For frond production analysis, plants to 4 (the size that
tall
could be reached order follow the frond production the apex]
safely in to at
were marked by wooden aluminum ground
placing stakes Avith labels in the
m
up
close to the base of each caudex. Larger plants with caudices to 4.7 tall
can be found in the but were not studied. At Morro Renter 45 plants were
sites,
AMERICAN FERN JOURNAL: VOLUME NUMHl^R
lOG 96 4
[2()()(i]
marked and another 48 Sapiranga. Each caiidex with a crown of fronds was
at
considered be an individuaL Caudox length was measured from ground
to llie
crown
extreme apex
the center of the of fronds.
sin^face to at
thc^
May
observations frond production wore conducted from 2000
of
hiitial fiekl
April 2001, with the height of the caudiccs registered the beginning and
to at
end of period. Annual frond production was determined by marking (with
ihis
a loose loop of synthetic fiber) the youngest expanding crozier on each marked
plant at the beginning of the observation period, this serving as a reference for
new
produced
recognizing the fronds during the study period.
In order to estimate the total numbers of fronds produced at each site, 10
cm
individuals were sampled, leaving out plants with c:audices less than 40
and taHer than 3.5 m; the latter were mostly covered by adventitious roots
tall
When
or epiphytes making counting impossible. a part of the caudex presented
between and
patches of epiphytic cover, the average the preceding the next
cm was number
20 sections used as an estimate of the of scars present
underneath covered
the
part.
In August 2003, additional data set were collected from nine caudices in
each population, removing all the epiphytes in order to obtain a prec:ise count
number
of the total of stipe bases and scars. This aLso allowed the verification
On
of caudex growth in length and frond produc:lion over a three-year period.
that occasion, the development of a young plant, which in October 2000 was
sprouting out of the ground close one of the marked specimens, was
to
new
evaluated. This individual started out from well developed crozier,
a
emerging from the soil and clearly was not a small plantlet formed from
gametophyte.
a
was used between
Pearson's correlation evaluate the relationships the
test to
caudex and number and
length of the the total of scars stipe bases, the length of
and was
the caudex, the yearly frond production. Linear regression analysis
number
used to estimate the of stipe bases and the yearly frond production
relative to caudex length. In the second approach, the age estimate resulted
number
from the equation: total of fronds and vestiges yearly frond
/
production
rate.
Data analysis was perforuKKl as described by Vieira (1980) and Zar (1999)
SPSS
using the 9.0 program installed the Universidade do Vale do Rio dos
at
Sinos data processing
facility.
Results
month
During the 12 period the average caudex growth rate based on the
=
samples from Morro Renter was 14.51 ±11,49) cm.yr" ranging from
(s.d.
,
maximum
a rate of 47 cm.yr"' to three individuals with no noticeable growth
new
as (although fronds were formed). In Sapiranga the average growth rate
^ maximum
was 6.32 ±5.53) cm.yr" with a growth of 22 cm.yr" and
(s.d.
,
showing no The growth
ten plants noticeable length increase. rates of these
<
popidations were There was no
significantly differ(Mit (P 0.001). significant
GROWTH
SCHMITT & WINDISCH: RATES ALSOPHILA
IN 107
m
Table Comparative age estimates of an Alsophila setosa plant with a 4 long candex, using the
1.
= =
one year data (n 45 for Morro Renter, n 48 for Sapiranga).
Estimated age based on Age based on frond
Locality Growlh rate growth rate pi^oduction equation
Morro Renter 14.51 om.yr ^ 27 years 22 years
Sapiranga cm.yr
6.32 63 years 32 years
=
between caudex growth and Morro
correlation length Renter
total in 0.258,
(r
- - = - -
P n Sapiranga P n
0.086, 45] or in 0.166, 0.253,
(r 48).
A
between caudex and
strong positive correlation the length of the the
number of scars and stipe bases was observed both in Morro Renter = 0.856,
[r
= = - - =
P n and Sapiranga P n The
0.02, 10) in (r 0.825, 0.03, 10). yearly
showed
frond production also a strong positive correlation with caudex length
= - - =
both in Morro Renter 0.845, P 0.02, n 10) and Sapiranga 0.876, P
(r (r
= =
n Based on
0.01, 10). these strong direct correlations, a linear regression
was performed number
analysis in order to obtain equations to estimate the of
scars per total caudex length and to predict yearly frond production. The
number
of scars/stipe bases in relation to the length of the caudex (H), in Morro
=
Renter was estimated -12.115 + 0.9H and the yearly frond production
as: y
= ^
by y —0.773 + 0.042H. In Sapiranga the equations were respectively y
-
47.149 + 0.577H and -0.0138 + 0.0218H.
y
m
Approximate
age estimates for hypothetical plants with 4 long caudicos
using the approach caudex length divided by the average growth
first (total
rates) arc presented in Table In the case of the Sapiranga the resulting
1. site,
age estimate does not agree with the age of the
forest.
number
For based on
estimating age the total of fronds in relation to the
Morro would
yearly frond production, for Renter plants the formula be
(-12.115 + 0.9 H)/(-0.773 + 0.042H), whereas for Sapiranga would be
it
(47.149 + 0.577H)/(-0.0138 + 0.0218H). These indicated lower
results ages
than those using the approach (Table
first
1).
The was
analysis of the preliminary data indicated that bias introduced by
a
the variation in growth pattern during the phases of development. Also
initial
the frond production rate in a particular year does not reflect the average
new
growth conditions throughout the life span of a given plant. This led to a
marked
survey of the plants standing in the localities in August 2003. Data
still
caudex and number were
for length of scar/stipe bases obtained for a three year
period for nine plants in each population. The sampled plants in Morro Renter
produced an average 8.66 fronds. yr~^ while those from Sapiranga 6.92
m
Age
based on
fronds. yr"^. estimates these data 4 hypothetical plants
for tall
shown
Table
are in
2.
NPK
As
to the basic nutrients in the soil, no significant differences in were
=
found between populations based on P Phospho-
(Nitrogen: 0.981,
t-tests
= =
rous: P 0.440 and Potassium: P Samples within each did show
0.282).
site
among
heterogeneity
collecting points.
VOLUME NUMBER
AMERICAN FERN JOURNAL:
108 DG 4 (200G)
m
Agti ostimatos using the total numb(M^ of fronds produced in two plants with 4 tall
2.
l\\i![,i-;
=
caudices and the avorago (n 9 in eac;h population] yearly frond prodnntinn year period).
('A
Average
Sears/stipes in a 4 ni yearly frond
long caudex production years) Estimated age
Lot:ality
[3
Morro 244 8.66 fronds-yr ^ 28 vears
R(Milei
"
238 6.92 fronds-yr 34 vears
LSapiranga
Tho rapid growth of vegetatively produced plants illustrated by the single
is
individual, initially observed in October 2000 at the Morro Renter site,
represented by a crozier emerging from the ground. In February 2001
its
m
cm
caudex was cm. By April had reached 10 and bore a crown of ca 3.5
7
it
cm
diameter with fronds up to 230 long. By September 2003, this plant had
cm
70 long caudex.
a
Discussion
The average growth rates observed for Alsophila setosa are within the range
of growth rates reported by other authors for other species of the genus
The "no growth some with reduced
(Table noticeable" status of plants
3).
may
caudex growth be due the presence of unexposed young croziers
to [still
covered by a mass of scales] at the tip of the caudex at the time of the first
measurement. The differences in the growth rates between sites, in the present
and
study, could be related different canopies different successional stages
to
of the surrounding vegetation. Arens (2001] presented some interesting data on
Cyathaa caracasana, growing in different habitats in Colombia, indicating a life
and
history with periods of rapid growth, spore productioti, establishment in
forest gaps, alternating with low growth rates and persistence in the
imderstory, and recorded higher growth rates in open habitats.
Slightly lower growth rates were observed in Sapiranga, where the forest
is
3^ounger and the tipper canopy more continuotis. Alsophila setosa an
is
canopy seems
understory plant, so that the height of the to influence the height
increase once the fronds reach that level. In our experience, this species never
Taull: Growlh of caudex in different species of genus Alsophila.
3. tlio tlie
Growth
rate
Source
S[)ecies (cm-yr~^) Locality Forest
A Isoph ila hryoph ila 5.0 Porto Rico Primary Conant ('Ul7(i)
Alsophila sahinii 6.9 El Salvador Secondary Seiler (1^95)
&
Alsophila crinacea 13.6 Costa Rica Primary BiltnEM' Breckle [1995)
&
Alsophila 18.8 Costa Rica Primary Biltner Breckle (1995)
polysiichoides
Alsophila setosa 14.51 Brazil Secondar} Present study
Alsophila safosa 6.32 Brazil Set:ondar} l-*resent stud)'
GROWTH
SCHMITT RATES ALSOPHILA
& VVINDISCH: IN 109
crown
surpasses the upper tree canopy. Also different heights of the of fronds
may low damage
exposure temperatures. Frost
lead differences in to to
to
young croziers has been reported by Schmitt and Windisch (2001). Although
show
the macro-nutrients (N, P, K] did not significant differences, considering
the different histories of soil usage in both plots, differences in micronutrients
may
occur between the study
sites.
Considering the aerial photograph taken in 1964 of the area in Sapiranga site,
much
more
32-34
the estimate of years for the taller plants realistic that the
is
The
one obtained only using the yearly growth rate (63 years). use of the frond
production data, especially obtained over a longer period of time, produces
if
and would mean
acceptable values (compare Tables This that the tallest
1
2).
m) would same
plant the locaUty be about the age as the forest stand.
at (4.7
However, using the formulas applied in the example in Table to a plant
1
m
than 4 the resulting age would have lower value than the one obtained
taller
m
for a 4 plant, as the sample includes a larger number of younger plants with
were
different growth patterns. Attempts exclude this bias not successful
to
due the size of the samples in the different age brackets of the studied plants,
and due development.
surely also to differences in their initial stages of
method number
Tanner indicated the using of fronds in
(1983) that total
more growth
relation to the yearly frond production gives realistic values for in
m
plants of lower stature, but less realistic values for caudices greater than 1
tall.
Two
needed
of the variables affecting these estimates the time to establish
is
the sporophyte from spore and the growth up to the stage where a caudex with
noticeable scars formed and preserved throughout the life of the plant.
is
such
Another variable the reproductive origin of the plants. In species as
is
and
Alsophila with both vegetative sexual reproduction, age estimates
setoso,
may The
stems produced by the two reproductive modes. rapid
differ for
growth of the vegetatively formed individual striking and indicates a stron
is
potential leaving the shaded ground cover by the rapid development in the
for
stages of these plants.
first
on
Based on the yearly frond production rates and the total frond vestiges
m
Alsophila
the caudex, Seller (1981) estimated the age of 4.6 salvinii
tall
m
Tanner Cyothea puhescens with
sporophytes 55 years; (1983) estimated 9
at
m
Ash
hornei
tall plants at 150 years old; (1987) estimated 5.5 tall plants of C.
being between 80 and 105 years. The following age estimates based on
as
may
yearly growth rates and total length of the caudices be found in the
m
growth
cm.yr"^
Alsophila bryophyla, in length,
literature: 5 tall, 5
considered be 100 years Cyathca arborea, m, 28.6 cm.yr"\ 18 years
to old; 5
(Conant,1976); Sphaeroptcris senilis, 2 m, 4 cni.yr"^, 50 years (Ortega, 1984);
A. erinacea, m, 13.6 cm.yr~\ 37 years; and A. polystichoides, 5 m,
5
and
18.8 cm.yr^^, 27 years (Bittncr Brcckle, 1995).
and
There on gametoph^^tic
a lack of field data the stages of
is first
developmemt poor understanding
sporophytic these species resulting in a
for
The
of the complete cycle. extraction of plants for cultivation represents
life
may
and
the removal of material that take decades to be formed, the original
AMERICAN FERN JOURNAL: VC3LUME NUMBER
96
11(} 4 (20()n]
may
growth conditions gametophytos and young sporophytcs no longer be
for
However,
available in the site. in species with vegetative propagation, such as
new
produced manner may
Alsopbilci selosa, individuals in this ocx:upy
tlie
available niches and perpetuate the presence of the species. These young
growth become
individuals plants rapid through the herbaceous
after a layer,
many
and
part of the stand, of the smaller plants considered the present
in
had
study probably This problems
this origin. presents additional for age
we may
estimates because not be differentiating between sporophytes of
and growth
different origins different characteristics.
Our some
data suggest that of the previously published age estimates should
be reconsidered, taking into account the different methods and problems
ontlined above. Longer observation periods, especially in areas of less uniform
climates are highly desirable. In general, our growth rate results arc similar to
those for other species of tree ferns in different regions, especially those of
Cyathea camcasana
(Arens, 2001] Alsopbilci bryophila (Conant, 1976), A.
and and
salvinii, A. erinciceci A. polystichoides (Bittner Breckle, 1995).
AcKNOU'LI'nCMI'NTS
The support and provided by Lfniversidado do Valo do Kio dos Sinos IJNISINUS.
facilitios
-
-
CkMdro Univcrsitario FEEVv'VEE, the Brazilian Research Clouncil Conselho Nacional do
CNPq made
Desonvolvimenlo Cientffico o Tecnologioo this study possible, Lucas Sclnuitt and
-
welcome Anonymous
Christina ScliniilL provided help in die field. reviewers provid(Hl stinndating
sn[>g(;stions for die improvemtinl of mamiscript.
tin;
Literature CriTC
AkI'NS, N. C. 2001. Variation in performance of the tree fern Cyalhon carcicasaud [(iyatheaceae)
sucessional mosaic Andean cloud Amer. 88:545-551.
ac:ross a in forest.
F3()t.
J.
Demography bonwi
Asn, 1987. of Cyalhea (Cyatheaceat;], a tropical tree-fern in Fiji. Austral.
J. J.
35:331-342.
Bot.
BiTTNi-R, and S. W. BrixikM':. 1995. The growth rate and age of tree fern trunks in r(dation to
J.
Amer. Fern 85:37-42.
habitats.
J.
Cc)NAN'[', D. S. 1976. Ecogeographic and systematic studies in American Cyatheaceae. Ph.D. Thesis,
Harvard Cambridge, Mass.
Uni\'ersity,
Cof)Mi-:s, D. A., K. R. Allen, VV. A. Blnili-v. E. E. Burrows. C. D. Canham. L. Facam. D. M. I'oksyth. A.
dAXK^LA-ArcANTAR. R. E. W. A. RuscoE, D. A. Wakole, D. Wilson and E. F. VVKiciir.
PakI'I'I"!",
J.
2005. Tlui hare, the tortoise and the t:rocodile: the ecology of angiosperm dominance, conifer
persistence and fern fih(M^ing. Ecology 93:918—935.
J.
DriRAND, L. Z. and G. rjO[j)S-n:[N. 2001. Growth, leaf charaf^tiM-istic:s, and sjiore produc:tion in native
and invasive tree ferns in Hawaii. Amer. Fern 91:25-35.
J.
A
Gas'I'ony. G. 1973. revision of tlie fern genus Neph(}l(}(i. Contr. Gray Herb. 203:81-148.
J.
OKria;A, F. 1984. Notas sohre hi auloecologia de SphaoropUnis senilis (KE) Tryon (Cyatheaceae) o,n
-
Parqu(! Natiional Avila Venezuela. 12:31-53.
el Pittieria
el
ScHMiT'i-, E. and P. G. VViNnrsf:n. 2001. Pr(^juizos cansados pela geada no desenxolvimento de
J.
Alsophihi srtosa Kanlh (Pteridoph}1a. Cyatheaceae]. Revista dn esludos Centre Universi-
/
FcGvale 24:79-88.
tario
SciiMn-r. L. and P. G, WiNnuscn. 2005. Aspectos ecologicos de Alsophila sctosa Kaulf.
J.
(Cyatheaceae, Pteridophyhi) no sul do Brasil. Acta Bot. Bras. 19:8G1-8G7.
GROWTH
SCIIMITT WINDISCH: RATES ALSOPHILA
& IN 111
SeilF'R, R. L. 1981. Loaf turnnver rates and natural history of the Central American tree fern
Alsophila Amor. Fern 71:75-81.
salvinii.
J.
Seill:r, R. L. 1995. Verification of estimated growth rates in the tree fern Alsophila salvinii. Amer.
Fern 85:96-97.
J.
Tanner, E. V. 1983. Leaf demography and growtli of tree-fern Cyolhaa pubescens Mett. Kx Knhn
J.
in Jamaica. Bot. Linn. Soc. 87:213-227.
J.
Teixeira, M. B., a. B. Copra Nfto, U. Pastore and A. L. R. Rangel Fieiicl 1986. Vegotagao. Pp. 541-
De
620, in Levantaniento Recursos Ncituruis. Volumes 33. TBGE, Rio de Janeiro.
ViEiRA, S. 1080. Introclugao a Bioostatfstica, 3"^ ed. C^ampus, Rio de Janeiro.
Walker, L. R. and G. H. Aplet. 1994. Growth and fertihzation responses of Hawaiian tree ferns.
Biotropica 26:378-383.
VViNPisc^H, P. G. 2002. Fern consc^rvation in BraziL Fern Gaz. 16:295-300.
4* Upper
Zar, H. 1999. Bioestatistical analysis, edn. Prentice-Hall, Saddle River.
J. '