Interciencia ISSN: 0378-1844 [email protected] Asociación Interciencia Venezuela

Cornejo-Tenorio, Guadalupe; Ibarra-Manríquez, Guillermo reproductive phenology in a temperate forest of the monarch biosphere reserve, México Interciencia, vol. 32, núm. 7, julio, 2007, pp. 445-452 Asociación Interciencia Caracas, Venezuela

Available in: http://www.redalyc.org/articulo.oa?id=33932704

How to cite Complete issue Scientific Information System More information about this article Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal Journal's homepage in redalyc.org Non-profit academic project, developed under the open access initiative Plant reproductive phenology in a temperate forest of the Biosphere Reserve, Guadalupe Cornejo-Tenorio and Guillermo Ibarra-Manríquez

SUMMARY

Monthly flowering and fruiting observations were recorded for and fruits throughout the year without peaks in any season; and the most dominant species (11 annual herbs, 72 perennial herbs, 3) nearly all had flowers and fruits during the dry season. 21 , and 8 trees) in a temperate forest, during 2004, in the Correlations of the number of flowering species at community Cerro Altamirano Core Zone of the Monarch Butterfly Biosphere level and perennial herbs against rainfall showed a significant Reserve in central Mexico. Intraspecific synchrony in flowering positive relationship. However, a negative relationship was found and fruiting of eight woody species was estimated by monitoring between rainfall and number of fruiting species in shrubs and 20 individuals of each. Flowering and fruiting occurred mainly trees. High reproductive synchrony (>60% of individuals in during the rainy season and at the beginning of the dry season the same phenological phase) was detected in five species. (Jul-Dec) and showed a low degree of seasonality. Reproductive Phenological reproductive patterns in the study area, essential- activity within growth forms occurred in different periods: 1) an- ly a temperate high altitude forest in the tropics, were similar nual and perennial herbs flowered principally during the rainy to those documented for the seasonal lowland tropical forests, season and at the beginning of the dry season, while their fruit- mainly explained by annual rainfall and growth form. ing peaked during the dry season; 2) shrubs produced flowers

ne of the most impor- et al., 1994; Wright and Calderón, 1995; and Janarthanam, 2004). Another impor- tant aspects of phenol- Poulin et al., 1999; Spina et al., 2001; tant aspect is the intraspecific synchrony ogy studies is the search Bolmgren et al., 2003). in reproductive events. A high degree of of factors that explain the phenological An additional recur- synchrony in flowering and fruiting could behavior of species. Rainfall, tempera- ring focus in plant phenology studies is be advantageous for the by increas- ture, soil water availability and photope- the comparison of phenological patterns ing the attraction of pollinators and riod appear to be the main abiotic factors among different growth forms. Several dispersers, or the satiation of seed preda- that trigger flowering and fruiting events studies have found that herbaceous spe- tors (Rathcke and Lacey, 1985; Smith and (van Schaik et al., 1993; Newstrom et al., cies produce flowers and fruits during Bronstein, 1996; Olvera et al., 1997; Kelly 1994; Morellato et al., 2000; Borchert et the rainy season, whereas woody spe- and Sork, 2002). In contrast, asynchrony al., 2004). On the other hand, biotic fac- cies tend to have flowers during the dry could minimize competition for dispersal tors such as fruit type, and season and fruits during the dry or rainy agents, propagule predation and pathogen seed dispersal syndromes are also very seasons (Frankie et al., 1974; Croat, incidence (Rathcke and Lacey, 1985; van important for understanding the flower- 1975; Opler et al., 1980; Sarmiento and Schaik et al., 1993; Poulin et al., 1999). ing and fruiting patterns of plant species Monasterio, 1983; Ibarra-Manríquez et This work describes for (Bawa et al., 1985; Ibarra-Manríquez et al., 1991; Ibarra-Manríquez and Oyama, the first time the reproductive plant phe- al., 1991; Ibarra-Manríquez and Oyama, 1992; Chapman et al., 1999; Batalha and nology in one of the core zones of the 1992; van Schaik et al., 1993; Newstrom Mantovani, 2000; Ramírez, 2002; Joshi Monarch Butterfly Biosphere Reserve in

KEYWORDS / Ecosystem Conservation / Flowering Phenology / Fruiting Phenology / Growth Forms / Intraspecific Synchrony / Received: 04/03/2007. Modified: 05/30/2007. Accepted: 06/06/2007.

Guadalupe Cornejo-Tenorio. M.Sc. in Biological Sciences, Universidad Nacional Autónoma de México (UNAM). Researcher, UNAM, Morelia, México. e-mail: [email protected] Guillermo Ibarra-Manríquez. Ph.D. in Biology, UNAM, Mexico. Investigador, UNAM, Mexi- co Address: Centro de Investigaciones en Ecosistemas, UNAM. Antigua Carretera a Pátzcuaro Nº 8701, Col. San José de la Huerta, 58190 Morelia, Michoacán, México. e-mail: [email protected]

JUL 2007, VOL. 32 Nº 7 0378-1844/07/07/445-08 $ 3.00/0 445 central Mexico. This reserve is the win- was predicted that 1) flow- ter refuge of the monarch butterfly (Dan- ering and fruiting would aus plexippus L.) and is one of the most be triggered by rainfall, 2) important protected areas of temperate growth forms would show forest in Mexico, in terms of diversity different phenological pat- of vascular plants, area, and its biogeog- terns, and 3) woody spe- raphy, which includes a unique combina- cies would display a pat- tion of northern and southern elements at tern of intraspecific syn- high elevation within the tropics. To date, chrony in flowering and phenological studies in Mexico have been fruiting. conducted mainly on woody species of tropical dry forests and tropical rain-for- Materials and Methods ests (Carabias-Lillo and Guevara, 1985; Bullock and Solís-Magallanes, 1990; Ibar- The study ra-Manríquez et al., 1991; Ibarra-Man- was carried out in one Figure 1. Climatic diagram with data from the Tepuxtepec Me- ríquez, 1992; Ochoa-Gaona and Domín- of the three major core teorological Station, located at Contepec, Michoacan, Mexico guez-Vázquez, 2000; Lobo et al., 2003), zones of the MBBR, the (19º59’S, 100º13’W, 2330masl). The values for mean monthly with few studies in the temperate forests Cerro Altamirano, in the rainfall (curve in gray, 1970-2000), mean monthly temperature (Ramírez and Nepamuceno, 1986; Bello, states of Michoacan and (line, 1970-2000) and monthly rainfall registered in 2004 (black 1994; Olvera et al., 1997). Thus, pheno- Mexico, central Mexico bars) were obtained from the Comisión Nacional del Agua. logical information for Mexican temper- (19°59’42’’-19°57’07’’N and ate species is scarce and only partial data 100°09’54’’-100°06’39’’W), with a surface All observations were made during the can be found in some regional flora or of 588ha and altitudes of 2500-3320masl first week of each month. Each observa- taxonomic monographs where the pheno- (Cornejo et al., 2003). Geologically, this tion site consisted of a 25×4m transect. logical information comes from records in reserve is within the Transmexican Vol- Based on knowledge of the area (Cornejo herbaria specimens rather than periodical canic Belt (Ferrusquía-Villafranca, 1993). et al., 2003), counts were limited to in- field observations. The regional climate is temperate-subhu- clude only the most abundant plant spe- Understanding phenolog- mid, with wet summers C(w1), an aver- cies in the community (Table I): 11 an- ical patterns and the underlying factors is age annual rainfall of 830mm and a mean nual herbs, 72 perennial herbs, 21 shrubs, important in the Monarch Butterfly Bio- annual temperature of 15.7ºC (García, and 8 trees. The presence of open flow- sphere Reserve (MBBR) to help analyze 1981). Rainfall is strongly seasonal, with ers and ripe fruits was recorded only in the wide array of biological processes most precipitation occurring from June to those species with ≥10 adult individuals governing forest functions and structure, September (Figure 1). Vegetation is clas- in at least one of the 35 observation sites. and also to reflect positive or negative in- sified as temperate forest, with two main In the case of herbs, it was sometimes not teractions among species (e.g., dispersal subtypes: Quercus forest at lower alti- possible to define individuals, in which of diaspores, population biology of herbi- tudes and Abies forest at higher altitudes case we recorded flowers or fruits at the vores). Phenological data will also provide (Rzedowski, 1978). The Quercus forest level of ramets in no less than 10 sites. valuable information to design sustainable is a floristically rich formation found at To investigate pheno- plans for the management and conserva- 2500-2900masl. In this forest type the logical synchrony of the dominant woody tion of biodiversity. Specifically, such data most important tree species are Q. casta- species, 160 mature individuals were will allow to recognize keystone fruit re- nea Née and Q. obtusata Humb. & Bon- marked and observed monthly, belonging sources in the plant community and will pl. (Fagaceae), and xalapensis to seven tree species (A. religiosa; Arbu- also be useful in planning restoration ac- Kunth (), while the understory tus tesellata, A. xalapensis, Ericaceae; tions in areas affected by human activi- contains a great diversity of shrubs and Clethra mexicana, Clethraceae; Q. cas- ties (Chapman et al., 1999; Wallace and herbs, predominantly Asteraceae, Lamia- tanea, Q. laurina, and Q. obtusata) and Painter, 2002). Unfortunately, deforesta- ceae and Scrophulariaceae. The Abies for- to one species ( pun- tion in the MBBR is a major problem that est is mostly found above 3000masl, has a gens, Ericaceae). Hereafter, this species includes a diminished natural resource canopy dominated by A. religiosa (Kunth) group was named as tree species. Based base for the local people, as well as eco- Schltdl. & Cham. (), Q. laurina in field experience, dominant species were system degradation associated with the Humb. & Bonpl. (Fagaceae) and Clethra recognized mainly by their numerical broad changes in forest cover (Brower et mexicana DC. (Clethraceae), and an un- abundance. Observations had to be limited al., 2002; Ramírez et al., 2003). derstory of several shrub and herb spe- to this number of species and individuals The purpose of the pres- cies (Asteraceae and Lamiaceae; Cornejo- for each species due to logistical reasons. ent study of the reproductive phenology of Tenorio et al., 2003). However, the distance between individu- 112 plant species in the Cerro Altamirano als of particular species varied from 20 mountain massif in the core zone of the Phenological data and analysis to 50m, depending on their relative abun- MBBR was threefold: 1) to describe phe- dance and local distribution. The presence nological patterns at the community level The flowering and fruit- of flowers and fruits was observed direct- and within growth forms (annual herbs, ing of 112 species were observed in Cerro ly or with the aid of binoculars. Voucher perennial herbs, shrubs, and trees); 2) to Altamirano during one year (Jan-Dec materials of all studied species were de- examine whether or not seasonal varia- 2004), along a transect of approximately posited in the herbaria of Universidad Na- tion in rainfall and temperature was cor- 3km that encompassed a 500m elevational cional Autónoma de México (MEXU) and related with phenophase peaks; and 3) to gradient, from the lower area at 2500masl Instituto de Ecología (IEB), Mexico. estimate the degree of individual repro- to the hill summit at 3000masl. Through- The flowering and fruit- ductive synchrony for important woody out this path observation sites were estab- ing periods of every species were charac- species. Based on preliminary findings, it lished every 100m, for a total of 35 sites. terized according to rainfall seasonality

446 JUL 2007, VOL. 32 Nº 7 (Figure 1) as occurring in the wet (Jun- Table I Sep) or in the dry season (Oct-May). The FLOWERING AND FRUITING PHENOLOGIES OF 112 SPECIES FROM CERRO Pearson’s correlation coefficient (Zar, ALTAMIRANO, ONE OF THE CORE ZONES OF THE MONARCH BUTTERFLY 1999) was used to correlate the number BIOSPHERE RESERVE IN CENTRAL MEXICO of flowering and fruiting species observed Taxa Growth form Flowering Fruiting each month against the monthly rainfall Apiaceae data registered during the study period Donnellsmithia juncea (Humb. & Bonpl.) Math. & Constance Perennial herb 9-10 11 and mean monthly temperatures (Figure Tauschia alpina (Coult. & Rose) Math. Perennial herb 5-8, 10 8-9 Asteraceae 1). The Rayleigh test (Zar, 1999) was Acourtia turbinata (Lex.) Reveal & King Perennial herb 1-3 3 used to assess whether species had flow- Ageratina areolaris (DC.) Gage Shrub 1, 10-12 1, 11-12 ers or fruits uniformly throughout the A. glabrata (Kunth) RM King & H Rob. Shrub 2-4 4-5 year. To calculate the circular statistic pa- A. mairetiana (DC.) RM King & H Rob. var. mairetiana Shrub 1-5, 12 2-6 rameters, months were converted to angles A. pazcuarensis (Kunth) R.M. King. & H. Rob. Perennial herb 10-12 — A. petiolaris (Moc. ex DC.) R.M. King & H. Rob. Shrub 2-4 3-6 (0º for Jan, 30º for Feb, etc.). The Ray- Archibaccharis hirtella (DC.) Heering Shrub 1-3, 11-12 3-5 leigh test (z) determines the significance A. serratifolia (Kunth) S.F. Blake Perennial herb 1-4, 11-12 3-5, 12 of the mean angle (a), which represents Artemisia ludoviciana Nutt. Perennial herb 10-12 12 the period of the year throughout which Baccharis conferta Kunth Shrub 3-6 4-7 B. pteronioides DC. Perennial herb 5-6 7 flowering and fruiting is recorded for Bidens odorata Cav. Annual herb 10-12 10-12 most species. If z is significant for each Cosmos parviflorus (Jacq.) Kunth Annual herb 9 10-11 reproductive event, then these are con- C. scabiosoides Kunth Perennial herb 8-10 — centrated in a specific period of the year, Dahlia coccinea Cav. Perennial herb 8-9 10-11 but if z is not significant, it is concluded D. scapigera (A. Dietr.) Knowles & Westc. Perennial herb 8-9 10-11 Dyssodia papposa (Vent.) Hitchc. Annual herb 9-10 12 that the phenophases were distributed uni- D. pinnata (Cav.) B.L. Rob. var. pinnata Annual herb 10 10-12 formly throughout the year. The degree of Galinsoga quadriradiata Ruiz & Pav. Annual herb 10 10-12 seasonality for the reproductive activity Hieracium dysonymum Blake Perennial herb 3-5 5-6 may be indicated by a vector (R), which Jaegeria hirta (Lag.) Less. Perennial herb 10-11 10-12 Laennecia schiedeana (Less.) G.L. Nesom Annual herb 10-11 12 is a measure of concentration around the Montanoa grandiflora DC. Shrub 9-10 1, 12 mean angle. The value of the R may vary Packera sanguisorbae (DC.) C. Jeffrey Perennial herb 2-6 5-7 between 0 and 1, and a high value indi- Piqueria trinervia Cav. Perennial herb 8-12 10-12 cates seasonal phenological behavior; R Roldana barba-johannis (DC.) H. Rob. & Brettell Shrub 1-2, 12 3-4 >0.75 was considered as a high value for Stevia elatior Kunth Perennial herb 8-9 9-12 S. monardifolia Kunth Perennial herb 1, 8-12 10-12 this variable. S. salicifolia Cav. var. salicifolia Shrub 1, 10-12 1-4, 12 The activity index (Mo- S. serrata Cav. var. serrata Perennial herb 8-12 10-12 rellato et al., 1990; Bencke and Morellato, Tagetes lucida Cav. Perennial herb 8-12 12 2002) was used to estimate the synchrony T. lunulata Ortega Annual herb 9-12 10-12 T. micrantha Cav. Annual herb 8-9 10-12 between individuals of each woody spe- Verbesina oncophora B.L. Rob. & Greenm. Shrub 1, 12 — cies. This index indicates the percentage Viguiera hemsleyana S.F. Blake Perennial herb 9-10 10-11 of individuals during the flowering or Viguiera sessilifolia DC. Perennial herb 9-10 11-12 fruiting peak of each species. It has three Boraginaceae Lithospermum distichum Ortega Perennial herb 6-8 8-9 categories: 1) asynchrony, when >20% of Bromeliaceae the individuals have reproductive struc- Tillandsia andrieuxii (Mez) L.B. Sm. Perennial herb 5-7 5, 8-12 tures; 2) low synchrony, when 20-60% Campanulaceae when >20% of the individuals have them; Diastatea micrantha (Kunth) McVaugh Annual herb 10-12 — and 3) high synchrony, when >60% do Lobelia gruina Cav. Perennial herb 1, 9-12 — Caprifoliaceae so. Symphoricarpos microphyllus Kunth Shrub 7-9 10-12 Caryophyllaceae Results Arenaria bourgaei Hemsl. Perennial herb 5-10 5-12 Cistaceae Helianthemum glomeratum Lag. ex DC. Perennial herb 1-4,8-12 2-5,8-12 Community phenology Clethraceae Clethra mexicana DC. Tree 1-2 3-4 All the species moni- Clusiaceae tored flowered and 82% fruited. About Hypericum silenoides Juss. var. silenoides Perennial herb 8-12 9-12 70% of the species flowered during the Commelinaceae Commelina coelestis Willd. Perennial herb 8-9 10 rainy season and into the beginning of the Gibasis pulchella (Kunth) Raf. Perennial herb 7-10 — dry season (Jul-Dec). The mean angle for Ericaceae flowering corresponds to the beginning of P.D. Sorensen Tree 1-9, 11-12 5-6 September (Figure 2a, Table II). A high A. xalapensis Kunth Tree 3-6 7-12 Arctostaphylos pungens Kunth Shrub 1-12 4-10 proportion of species (73%) produced discolor (Hook.) Diggs subsp. rupestris Shrub 2, 4-9 1-2, 5-6, 11-12 fruits during the dry season (Oct-May), Fabaceae with a maximum activity at the beginning Cologania biloba (Lindl.) G. Nicholson Perennial herb 8-10 — of November (Figure 2b, Table II). Most Desmodium neo-mexicanum A. Gray Perennial herb 9-10 — species produced flowers and fruits in a Lathyrus parviflorus S. Watson Perennial herb 7-9 — Lupinus montanus Kunth Perennial herb 1-3,10-12 1-5 specific period of the year. Nevertheless, Phaseolus pluriflorus Maréchal, Mascherpa & Stainier Perennial herb 8-9 — the low values of R revealed a low degree Fagaceae of reproductive seasonality (Table II). A Quercus castanea Née Tree 3-5 1-4, 12 positive correlation was found among the Q. laurina Humb. & Bonpl. Tree 4-5 12 →

JUL 2007, VOL. 32 Nº 7 447 Table I (cont.) number of flowering species and rainfall Q. obtusata Humb. & Bonpl. Tree 3-5 11-12 (r= 0.59, P<0.05), whereas for fruiting Gentianaceae species the correlation was not signifi- Gentiana spathacea Kunth Perennial herb 1-2, 12 2-3 cant (r= -0.44, P>0.05). However, for all Halenia brevicornis (Kunth) G. Don Perennial herb 11 — comparisons performed at the community H. plantaginea (Kunth) Griseb. Perennial herb 11 — Geraniaceae level and within growth forms against Geranium potentillaefolium DC. Perennial herb 6-8 9 mean monthly temperature, there were G. seemanni Peyr. Perennial herb 7-8 9 no significant differences in flowering or Hypoxidaceae fruiting periods. Hypoxis mexicana Schult. & Schult. f. Perennial herb 8-9 9 Iridaceae Sisyrinchium convolutum Nocca Perennial herb 7-9 10 Phenology and growth forms Lamiaceae Lepechinia caulescens (Ortega) Epling Perennial herb 8-9 10 Flowering and fruiting Salvia amarissima Ort. Perennial herb 8-12 — activity in annual and perennial herbs S. elegans Vahl Perennial herb 1-5,10-12 3-5 S. fulgens Cav. Perennial herb 7-12 10-12 occurred mainly during the rainy season S. laevis Benth. Perennial herb 9-12 12 and the beginning of the dry season (Jul- S. lavanduloides Benth. Perennial herb 1, 7-12 2-4, 10-12 Dec). The number of flowering and fruit- S. mexicana L. var. mexicana Perennial herb 8-12 1-4, 10-12 ing annual herbs was not uniformly dis- S. patens Cav. Perennial herb 8-11 — tributed throughout the year (Figure 2c-d, Scutellaria caerulea Sessé & Moc. Perennial herb 7-10 10-11 Stachys parvifolia M. Martens & Galeotti Perennial herb 7-9 8-9 Table II). The degree of seasonality de- Lentiburaliaceae noted by R was high (Table II). The mean Pinguicula moranensis Kunth Perennial herb 5-9 10 Lythraceae Cuphea aequipetala Cav. Perennial herb 7-10 8-11 Onagraceae Fuchsia thymifolia Kunth Shrub 5-9 10-12 Lopezia racemosa Cav. Annual herb 1, 9-12 2 Orchidaceae Corallorrhiza odontorhiza (Willd.) Nutt. Perennial herb 7-8 — Govenia capitata Lindl. Perennial herb 6-9 11-12 Orobanchaceae Conopholis alpina Liebm. Perennial herb 4-5 5-7 Pinaceae Abies religiosa (Kunth) Schltdl. & Cham. Tree 3-5 1-4 Polemoniaceae Loeselia mexicana (Lam.) Brand Perennial herb 1-2, 10-12 — Polygalaceae Monnina ciliolata DC. Shrub 1, 8-9 11-12 Ranunculaceae Clematis dioica L. Shrub 5-10 1 Ranunculus petiolaris Kunth ex DC. var. arsenei (Benson) T Duncan Perennial herb 6-10 9-11 Rhamnaceae Ceanothus coeruleus Lag. Shrub 1, 4-8, 10-12 1 Rosaceae Alchemilla procumbens Rose Perennial herb 7-12 — Rubiaceae Bouvardia longiflora (Cav.) Kunth Shrub 4-10 12 B. ternifolia (Cav.) Schltdl. Perennial herb 4-10 1, 10-12 Crusea longiflora (Willd. ex Roem. & Schult.) WR Anderson Annual herb 8-10 — Galium aschenbornii Nees & S. Schauer Perennial herb 7-9 1-2, 10-12 Salicaceae Salix paradoxa Kunth Tree 3 5 Scrophulariaceae Castilleja tenuiflora Benth. Perennial herb 1-12 2-3, 10 Lamourouxia dasyantha (Cham. & Schltdl.) Ernst Perennial herb 11-12 12 L. multifida Kunth Perennial herb 9-12 10-12 Mecardonia procumbens (Mill.) Small Perennial herb 8-10 10-12 Penstemon roseus (Cerv. ex Sweet) G Don Perennial herb 7-12 1-2, 10-12 Solanaceae Cestrum thyrsoideum Kunth Shrub 1-2, 9-11 — Physalis orizabae Don Perennial herb 7-9 9 Solanum demissum Lindl. Perennial herb 7-8 — Valerianaceae Valeriana barbareifolia M. Martens & Galeotti Perennial herb 7-10 10-12 Verbenaceae Glandularia teucriifolia (M. Martens & Galeotti) Umber Perennial herb 6-9 8-9 V. recta Kunth Perennial herb 8-10 10 Violaceae Viola humilis Kunth Perennial herb 6-9 10 V. painteri Rouse & House Perennial herb 5-9 9 Viscaceae Cladocolea diversifolia (Benth.) Kuijt Shrub 4-5 3-4 Figure 2. Flowering and fruting species through- Phoradendron schumanni Trel. Shrub 5 5-10 out the year in Cerro Altamirano. a-b: commu- nity level (112 species), c-d: annual herbs (11 Months are indicated by numbers. Hyphens or commas between months means continual or intermit- species), e-f: perennial herbs (72 species), g-h: tent phenological activity, respectively. shrubs (21 species), i-j: trees (8 species).

448 JUL 2007, VOL. 32 Nº 7 Table II during the driest months (Mar-May) and Values of circular statistic analyses for flowering and the mean angle of 86° corresponds to the fruiting species in Cerro Altamirano core zone end of March. The number of fruiting species was distributed homogeneously Rayleigh test (z) Degree of Mean angle seasonality (R) during the annual cycle, with a low de- Flowering gree of seasonality (Figure 2j, Table II). The correlation between the number of Community 32.37 * 0.27 245° Annual herbs 17.35 * 0.80 275° flowering trees and rainfall was not sta- Perennial herbs 50.28 * 0.42 241° tistically significant (r= -0.42, P>0.05), Shrubs 0.02 ns 0.01 --- while for fruiting trees the correlation was Trees 8.49 * 0.54 86° negative (r= -0.64, P<0.05). Fruiting Community 11.32 * 0.16 308° Intraspecific synchrony in tree species Annual herbs 14.88 * 0.74 304° Perennial herbs 12.72 * 0.21 292° Flowering of the eight Shrubs 1.55 ns 0.13 --- Trees 0.91 ns 0.18 --- tree species showed a low synchrony in half of the species and high synchrony * P<0.001. in the other half (Table III). With regard angle for flowering species corresponds ing and fruiting in a specific period of to fruiting, only A. xalapensis was asyn- to the beginning of October, whereas the the year (Figure 2e-f, Table II); although chronic, three species displayed low syn- mean angle for fruiting corresponds to the degree of seasonality denoted by R chrony, and four species were highly syn- the beginning of November. The number was from medium to low (Table II). The chronic. Q. laurina was the only species of flowering and fruiting annual herbs mean angle for flowering species was at that exhibited a low synchrony in the pro- was not correlated with rainfall (r= 0.18 the beginning of September and the mean duction of flowers and fruits; on the con- and r= 0.33, respectively; P>0.05). Simi- angle for fruiting corresponds to the end trary, A. pungens and C. mexicana were larly, perennial herbs also show flower- of October. The number of flowering spe- highly synchronic in both phenophases cies was positive- (Figure 3, Table III). ly correlated with rainfall (r= 0.67, Discussion P<0.05), while the correlation for Community phenology fruiting was not significant (r= - The number of species 0.23, P>0.05). flowering in Cerro Altamirano was posi- Shrubs pre- tively correlated with rainfall. This result sented flowers differs from reports for the arboreal, lia- and fruits all nas and palms species in the tropical low- year long (Fig- land rain forest (Carabias-Lillo and Gue- ure 2g-h, Table vara, 1985; Ibarra-Manríquez et al., 1991; II). Although it Ibarra-Manríquez, 1992; Ibarra-Manríquez was observed that and Oyama, 1992) or tree species in the fruiting had two temperate forest (Ramírez and Nepamu- peaks (Apr and ceno, 1986; Olvera et al., 1997), where this Dec, Figure 2c), the R value (0.13) Table III indicated that Degree of intraspecific this phenophase synchrony* in flowering and was not seasonal. fruiting AS percentage of Only the correla- individuals for eight woody tion between the species of Cerro Altamirano number of fruit- core zone ing shrubs against Species Flowering Fruiting the rainfall was % % statistically sig- Abies religiosa 90 50 nificant (r= -0.62, P<0.05). Arbutus tessellata 80 25 Trees did not 40 15 flower uniformly Arctostaphylos pungens 95 95 throughout the Clethra mexicana 70 70 year and showed Quercus castanea 40 90 a moderate de- Quercus laurina 60 30 gree of season- Quercus obtusata 50 75 ality (Figure 2i, * Synchrony categories: asynchrony (<20% of in- Figure 3. Reproductive phenology in marked individuals of eight woody Table II). A large dividuals with flowers or fruits), low synchrony species in Cerro Altamirano. Shaded circles: flowers, open circles: fruits, number of species (20-60%) and high synchrony (>60%). n= 20 for squares: both structures. produced flowers each species.

JUL 2007, VOL. 32 Nº 7 449 phenophase is associated with the season This pattern could be explained consider- are probably pollinated by diurnal insects of lower precipitation. These discrepancies ing that woody species have a deep root (e.g., bees), presented high synchrony dur- are understandable if it is considered that system that allows them to reach available ing their flowering period, which would 74% of the species included in the present water, or they have water storage struc- allow for the attraction of a higher num- study are herbs. In fact, the results show tures that buffer the negative impact of ber of generalist pollinators (Rathcke and that herbs, shrubs, and trees have different seasonal drought (Sarmiento and Monas- Lacey, 1985; Ims, 1990). phenological patterns, reflecting different terio, 1983). From the eight tree spe- responses to environmental factors, partic- On the other hand, one cies studied, four showed a high fruiting ularly to rainfall seasonality, and probably of the factors proposed to explain flower- synchrony (Table III). It has been widely also due to biotic factors such as pollina- ing activity of trees in the dry season is proposed in the literature that the mast tion and dispersal syndromes (see below). that wind pollinated species need specific fruiting effect leads to satiation of special- environmental conditions (dry and windy ist and generalist predators, which allows Phenology and growth forms weather) for optimum pollen dispersion for a part of the fruit crop to escape pre- (Frankie et al., 1974; Bawa et al., 1985; dation (Rathcke and Lacey, 1985; Crawley, Several studies have Ramírez and Nepamuceno, 1986; Bello, 2000 and references therein). This event found that flowering activity among her- 1994; Olvera et al., 1997; Barnes et al., probably occurred in Q. castanea and Q. baceous species is strongly associated to 1998). This argument is useful to explain laurina. Both oak species had a high fruit the rainy season (Croat, 1975; Sarmiento the present findings, since the wind pol- production during 2004 but also showed and Monasterio, 1983; Batalha and Man- linated trees in the study area (A. reli- many damaged nuts, probably by squirrels tovani, 2000; Spina et al., 2001; Tyler, giosa, Q. castanea, Q. laurina, Q. ob- or mice. Mast fruiting in oak species is a 2001; Ramírez, 2002; Batalha and Mar- tusata and S. paradoxa) flower during widely documented phenological behavior tins, 2004), which agrees with the present the driest months of the year (Mar-May; in other localities of temperate forest and findings. The strong relationship between Figures 1 and 3). Fruiting periodicity de- has been considered as an evolved repro- herbs and the rainy season is due to the pends principally on flowering, but it is ductive strategy, because it is not simply fact that this life form requires high wa- also influenced by environmental condi- a response to weather conditions (Sork et ter availability for their vegetative devel- tions appropriate for fruit development, al., 1993; Kelly and Sork, 2002). opment and reproduction (Janzen, 1967; diaspore dispersal, and seedling establish- Rathcke and Lacey, 1985). Furthermore, ment (Rathcke and Lacey, 1985; Ibarra- Conclusions in the study site most of the perennial Manríquez et al., 1991; van Schaik et al., herbs flower earlier than annual species. 1993). Available information from several The results obtained Ramírez (2002) found the same result and Neotropical regions indicate that during indicate that the reproductive phenol- considered that this may be the result of the dry season the number of species with ogy patterns of the temperate flora of the fact that perennial plants have reserve anemochorous or autochorous diaspores Cerro Altamirano are similar to those structures (rhizomes or tubers) that allow is higher, while species with zoochorous documented for seasonal tropical com- them to start their reproductive activity diaspores seem to produce them most of- munities. Nevertheless, it is necessary before annual plants. The fact that herba- ten in the rainy season (Morellato et al., to perform phenological studies in the ceous species, the most important growth 1990; Ibarra-Manríquez et al., 1991, 2001; other core zones of the reserve, with the form in this plant community, produce Batalha and Mantovani, 2000; Batalha purpose of contrasting the findings and fewer flowers during the dry season rein- and Martins, 2004). These reproductive to obtain long-term data. This last is- forces the argument of Alonso-Mejía et al. patterns were observed in tree species of sue has special significance as temporal (1997), that the potential nectar sources Cerro Altamirano, where anemochorous changes in plant resources profoundly for overwintering monarch be- and autochorous trees have fruits in the affect animals, and also because cycles come increasingly unavailable as the dry dry season (A. religiosa, C. mexicana, Q. of plant reproduction are crucial for an season advances. Based on their earlier castanea, Q. laurina, Q. obtusata, and S. understanding of ecosystem functioning blooming, it is not unexpected that fruit- paradoxa), and zoochorous species fruit (Rathcke and Lacey, 1985; van Schaik et ing herbs may also show a peak approxi- mainly in the wet season (A. tessellata al., 1993; Barnes et al., 1998; Chapman mately two months after the maximum and A. xalapensis). et al., 1999; Poulin et al., 1999; Wal- flowering season is reached, since this pe- lace and Painter, 2002). Furthermore, riod is required for fruit formation. Fruit Intraspecific synchrony in tree species it has been detected that the reproduc- ripening has also been associated with the tive season of particular species may appropriate dispersal season; for instance, Considering the results change through the years and also that diaspores of wind dispersal species ripen of both reproductive cycles, 50% of the many species have multiyear reproduc- during the dry season (Lieberman, 1982; woody species showed high synchrony tive cycles (Frankie et al., 1974; Bawa et Morellato et al., 1990; Ibarra-Manríquez and the other 50% showed low synchrony al., 1985; Ibarra-Manríquez et al., 1991; et al., 1991; Batalha and Martins, 2004), or asynchrony (Table III). Rabinowitz et Newstrom et al., 1994; Chapman et al., a situation that also occurs in several spe- al., (1981) found that in comparison with 1999). In fact, recently (Oct-Dec 2006) a cies of Asteraceae in the study area. insect-pollinated species, flowering phe- high number of reproductive A. religiosa Like other Neotropi- nology in wind-pollinated plants showed trees were detected in each one of the cal localities (Opler et al., 1980; Smith- greater intrapopulation synchronization or three core zones of the MBBR, a condi- Ramírez and Armesto, 1994; Batalha individuals with shorter flowering times. tion never observed along the previous and Mantovani, 2000; Spina et al., 2001; In Cerro Altamirano, wind-pollinated six years of floristic inventory. This situ- Ramírez, 2002; Batalha and Martins, species (A. religiosa and three species ation suggests that A. religiosa could be 2004), flowering and fruiting of shrubs of Quercus) had short periods of flower- considered as a supra-annual flowering was observed during the whole year, even ing, but only A. religiosa was highly syn- species, (sensu Newstrom et al., 1994), though the greatest number of species chronic (Figure 3; Table I). Also, A. tesel- but in order to confirm this hypothesis presented fruits during the dry season. lata, A. pungens, and C. mexicana, which long-term phenological observations (at

450 JUL 2007, VOL. 32 Nº 7 least 12 years; Newstrom et al., 1994) REFERENCES Ferrusquía-Villafranca I (1993) Geology of are needed. Mexico: a synopsis. In Ramamorthy TP, Another important point Alonso-Mejía A, Rendón-Salinas E, Montesinos- Bye R, Lot A, Fa J (Comp.) Biological di- Patino E, Brower LP (1997) Use of lipid re- versity of Mexico: origins and distribution. to extend phenological studies to other ar- serves by monarch butterflies overwintering Oxford University Press. New York, USA., eas of the reserve is that the floristic com- in Mexico: implications for conservation. pp. 54-60. position of each core zone (Cerro Altami- Ecol. Appl. 7: 934-947. Frankie GW, Baker HG, Opler PA (1974) Com- rano, Chincua-Campanario-Chivati, and Barnes BV, Zak DR, Denton SR, Spurr SR (1998) parative phenological studies of trees in trop- Cerro Pelón) is very particular and each Forest Ecology. Wiley. New York, USA. 774 ical wet and dry forests in the lowlands of should be considered as a distinct plant pp. Costa Rica. J. Ecol. 62: 881-919. community. The floristic inventory carried Batalha MA, Mantovani W (2000) Reproductive García E (1981) Modificaciones al sistema de clasificación climática de Köeppen. (para out in these sanctuaries reached around 650 phenological patterns of cerrado plant spe- cies at the Pé de Gigante reserve (Santa Rita adaptarlo a las condiciones de la República species (2000-2006), of which only ~14% do Passa Quatro, SP, Brazil): a comparison Mexicana). 3ª ed. Mexico DF, Mexico. 243 (92 species) was shared in all three areas between the herbaceous and woody floras. pp. (Ibarra-Manríquez, unpublished data). In Rev. Bras. Biol. 60: 129-145. Ibarra-Manríquez G (1992) Fenología de las pal- fact, Cerro Altamirano area has 213 species Batalha MA, Martins FR (2004) Reproductive mas de una selva cálido húmeda de México. (33%) registered exclusively in its forests. phenology of the cerrado plant community in Bull. Inst. Fr. Ét. And. 21: 669-683. Consequently, phenological information Emas Nacional Park (central Brazil). Aust. J. Ibarra-Manríquez G, Oyama, K (1992) Ecological Bot. 52: 149-161. correlates of reproductive traits of Mexican must be generated for other species that in- rain forest trees. Am. J. Bot. 79: 344-356. habit these temperate forests together with Bawa KS, Bullock SH, Perry DR, Coville RE, Grayum MH (1985) Reproductive biology of Ibarra-Manríquez G, Sánchez-Garfias B, Gon­ monarch butterflies. For implementing res- tropical lowland rain forest trees II. Pollina- zález-García L (1991) Fenología de lianas y toration actions to recover disturbed areas tions systems. Am. J. Bot. 72: 346-356. árboles anemócoros en una selva cálido húm- near overwintering sites, or for degraded Bello GMA (1994) Fenología y biología del de- eda de México. Biotropica 23: 242-254. ground recovery, the information obtained sarrollo de cinco especies de Quercus, en Ibarra-Manríquez G, Martínez-Ramos M, Oyama should be a guide to know the appropriate Paracho y Uruapán, Michoacán. Ciencia For- K (2001) Seedling functional types in a low- timing for collecting mature of sev- estal 75: 3-40. land rain forest in Mexico. Am. J. Bot. 88: 1801-1812. eral species (e.g., A. religiosa, Ceanothus Bencke CSC, Morellato LPC (2002) Comparação de dois métodos de avaliação da fenologia coeruleus, Lupinus spp., Quercus spp.). Ims RA (1990) The ecology and evolution of re- de plantas, sua interpretação e representação. productive synchrony. Trends Ecol. Evol. 5: Finally, it would be ad- Rev. Bras. Bot. 25: 269-275. 135-140. visable that the reproductive phenology Bolmgren K, Eriksson O, Linder HP (2003) Con- Janzen DH (1967) Synchronizacion of sexual re- of plants in temperate trasting flowering phenology and species production of trees within the dry season in be addressed in a more comprehensive richness in abiotically and biotically pollinat- Central America. Evolution 21: 620-637. ed angiosperms. Evolution 57: 2001-2011. way, where phenological patterns could Joshi VC, Janarthanam MK (2004) The diversity be related to other reproductive attri- Borchert RS, Meyer A, Felger RS, Porter-Bolland of life-form type, habitat preference and phe- butes, such as pollination syndromes and L (2004) Environmental control of flower- nology of the endemics in the Goa region of ing periodicity in Costa Rican and Mexican the Western Ghats, India. J. Biogeogr. 31: seed dispersal or sexual systems (mon- tropical dry forests. Glob. Ecol. Biogeogr. 1227-1237. oecious, dioicious or hermaphroditic). A 13: 409-425. Kelly D, Sork VL (2002) Mast seeding in peren- better understanding of phenological pat- Brower LP, Castilleja G, Peralta A, López-Gar- nial plants: why, how, where? Annu. Rev. terns at both the level of species and of cía J, Bojórquez-Tapia L, Díaz S, Melgarejo Ecol. Syst. 33: 427-447. D, Missrie M (2002) Quantitative changes ecosystems is crucial for the management Lieberman D (1982) Seasonality and phenology in forest quality in a principal overwinter- and long term conservation of ecosystems in a dry tropical forest in Ghana. J. Ecol. ing area of the Monarch Butterfly in Mexico, 70: 791-806. (Newstrom et al., 1994; Joshi and Janar- 1971-1999. Cons. Biol. 16: 346-359. Lobo JA, Quesada M, Stoner KE, Fuchs EJ, thanam, 2004). A better habitat manage- Bullock SH, Solís-Magallanes A (1990) Phenol- Herrerías-Diego Y, Rojas J, Saborio G ment of existing overwintering sites and ogy of canopy trees of a tropical deciduous (2003) Factors affecting phenological pat- forest in Mexico. Biotropica 22: 22-35. buffer areas in the MBBR is a critical el- terns of bombacaceous trees in seasonal for- ement for the preservation of the mass of Carabias-Lillo J, Guevara SS (1985) Fenología de ests in Costa Rica and Mexico. Am. J. Bot. wintering aggregations of monarch butter- una selva tropical húmeda. In Gómez-Pompa 90: 1054-1063. A, del Amo SR (Eds.) Investigaciones sobre flies in Mexico, an exceptional biological Morellato LPC, Leitão-Filho HF, Rodrigues RR, la regeneración de selvas altas en , Joly CA (1990) Estratégias fenológicas de phenomenon highly threatened by human México. Vol. II, Alambra. México DF, Méxi- espécies arbóreas em floresta de altitude na activities (Brower et al., 2002; Ramírez et co. pp. 27-66. Serra do Japi, Jundiaí, SP. Rev. Bras. Biol. al., 2003). Chapman CA, Wranghman RW, Chapman LJ, 50: 149-162. Kennard DK, Zane AE (1999) Fruit and Morellato LPC, Talora DC, Takahasi A, Bencke flower phenology at two sites in Kibale Na- Acknowledgments CC, Romera EC, Zipparro VB (2000) Phe- tional Park, Uganda. J. Trop. Ecol. 15: 189- nology of Atlantic rain forest trees: a com- 211. The authors acknowl- parative study. Biotropica 32: 811-823. Cornejo-Tenorio G, Casas A, Farfán B, Villase- edge the careful and extensive revision Newstrom LE, Frankie GW, Baker HG, Colwell ñor JL, Ibarra-Manríquez G (2003) Flora y RK (1994) Diversity of long-term flowering and edition of this paper by Lincoln vegetación de las zonas núcleo de la Reserva patterns. In McDade LA, Bawa KS, Hespen- B. Brower, and thank Ellen Andresen de la Biosfera Mariposa Monarca, México. heide HA, Hartshorn GS (Eds.) La Selva. Bol. Soc. Bot. Mex. 73: 43-62. and William C. Burger for criticism of Ecology and Natural History of a Neotropi- earlier version of the manuscript, Juan Crawley MJ (2000) Seed predators and plant cal Rain Forest. University of Chicago Press. Martínez-Cruz, Fernando Pineda-García, population dynamics. In Fenner M (Ed.) Chicago, IL, USA. pp. 142-160. Seeds: the ecology of regeneration in plant Miguel Ángel Salinas-Melgoza and Ro- Ochoa-Gaona S, Domínguez-Vázquez G (2000) communities. 2nd ed. CABI. Wallingford, UK. Distribución y fenología de la flora leñosa de berto Sáyago-Lorenzana for their help in pp. 167-182. the field work. The first author received Chajul, Selva Lacandona, , México. Croat TB (1975) Phenological behavior of habit Brenesia 54: 1-24. a scholarship from the Consejo Nacional and habitat classes on Barro Colorado Island Olvera VM, Figueroa RBL, Moreno GS, Solís- de Ciencia y Tecnología (CONACyT; Nº (Panama Canal Zone). Biotropica 7: 270- Magallanes A (1997) Resultados preliminares 277. 181848). de la fenología de cuatro especies de encino

JUL 2007, VOL. 32 Nº 7 451 en Cerro Grande, Reserva de la Biosfera Si- Ramírez MI, Azcárate JG, Luna L (2003) Effects Spina AP, Ferreira WM, Leitão Filho HF erra de Manantlán. Biotam 9: 7-18. of human activities on monarch butterfly (2001) Floração e síndromes de dispersão Opler PA, Frankie GW, Baker HG (1980) Com- habitat in protected mountain forests, Mexi- de uma comunidade de floresta de brejo na parative phenological studies of treelet and co. Forestry Chron. 79: 242-246. região de Campinas (SP). Acta Bot. Bras. shrub species in tropical wet and dry forests Rathcke B, Lacey EP (1985) Phenological pat- 15: 349-368. in the lowlands of Costa Rica. J. Ecol. 68: terns of terrestrial plants. Annu. Rev. Ecol. Tyler G (2001) Relationships between climate and 167-188. Syst. 16: 179-214. flowering of eight herbs in a Swedish decidu- Poulin B, Wright SJ, Lefebvre G, Calderón O Rzedowski J (1978) Vegetación de México. Li- ous forest. Ann. Bot. 87: 623-630. (1999) Interspecific synchrony and asynchro- musa. Mexico DF, Mexico. 432 pp. Wallace RB, Painter RLE (2002) Phenological ny in the fruiting phenologies of congeneric Sarmiento G, Monasterio M (1983) Life forms patterns in a southern Amazonian tropical bird-dispersed plants in Panama. J. Trop. and phenology. In Bourlière F (Ed.) Tropical forest: implications for sustainable manage- Ecol. 15: 213-227. savannas. Elsevier. Amsterdam, The Nether- ment. Forest Ecol. Manag. 160: 19-33. Rabinowitz D, Rapp JK, Sork VL, Rathcke BJ, lands. pp. 79-104. Wright SJ, Calderon O (1995) Phylogenetic pat- Reese GA, Weaver JC (1981) Phenological Smith CM, Bronstein JL (1996) Site variation in terns among tropical flowering phenologies. properties of wind- and insect-pollinated reproductive synchrony in three Neotropical J. Ecol. 83: 937-948. prairie plants. Ecology 62: 49-56. figs. J. Biogeogr. 23: 477-486. van Schaik CP, Terborgh JW, Wright SJ (1993) Ramírez N (2002) Reproductive phenology, life- Smith-Ramírez C, Armesto JJ (1994) Flowering The phenology of tropical forests: adaptative forms, and habitats of the Venezuelan Cen- and fruiting patterns in the temperate rain- significance and consequences for primary tral Plain. Am. J. Bot. 89: 836-842. forest of Chiloé, Chile - ecologies and cli- consumers. Annu. Rev. Ecol. Syst. 24: 353- Ramírez GJA, Nepamuceno MF (1986) Fenología matic constraints. J. Ecol. 82: 353-365. 377. de tres especies de coníferas de la región de Sork VL, Bramble J, Sexton O (1993) Ecology of Zar JH (1999) Biostatistical analysis. Pren- los Altos de Chiapas. Ciencia Forestal 60: mast-fruiting in three species of North Ameri- tice Hall. Upper Saddle River, NJ, USA. 21-50. can deciduous oaks. Ecology 74: 528-541. 663 pp.

Fenología reproductiva de las plantas de un bosque templado de la Reserva de la Biosfera Mariposa Monarca, México Guadalupe Cornejo-Tenorio y Guillermo Ibarra-Manríquez

RESUMEN

Para documentar la fenología reproductiva de las especies más flores y frutos a lo largo del año, sin máximo en alguna época importantes (11 hierbas anuales, 72 hierbas perennes, 21 arbustos particular, y iii) la mayoría de los árboles concentraron su activi- y 8 árboles) del bosque templado en la zona núcleo Cerro Altami- dad reproductiva en la época de menor precipitación. El número rano, Reserva de la Biosfera Mariposa Monarca, México, se reali- de especies en floración a nivel comunitario y de hierbas perennes zaron observaciones mensuales durante 2004. La sincronía intra- se correlacionó positivamente con la precipitación, y el número de específica de floración y fructificación se estimó en ocho especies especies arbustivas y arbóreas en fructificación mostró una corre- leñosas por medio de la observación de 20 individuos por especie. lación negativa con la precipitación. Se determinó una alta sincro- La floración y fructificación ocurrió principalmente durante la- es nía reproductiva (>60% de los individuos en una fase fenológica tación de lluvias e inicios de la estación seca (jul-dic), con baja específica) en cinco de las especies arbóreas. Los patrones feno- estacionalidad. Las formas de crecimiento mostraron diferencias lógicos reproductivos en el área, un bosque templado de elevada temporales en su actividad reproductiva: i) las hierbas anuales y altitud en una zona tropical, fueron similares a los documentados perennes florecieron principalmente durante la estación de lluvias para bosques tropicales estacionales de bajas altitudes, y explica- e inicios de la seca, mientras que la mayoría de especies con fru- dos principalmente por la precipitación total anual y la forma de tos fue observada en la estación seca; ii) los arbustos presentaron crecimiento de las especies.

FENOLOGIA REPRODUTIVA DAS PLANTAS DE UM BOSQUE TEMPERADO DA RESERVA DA BIOSFERA MARIPOSA MONARCA, MÉXICO Guadalupe Cornejo-Tenorio e Guillermo Ibarra-Manríquez

RESUMO

Para documentar a fenologia reprodutiva das espécies mais flores e frutos ao longo do ano, sem máximo em alguma época importantes (11 ervas anuais, 72 ervas perenes, 21 arbustos e em particular, e iii) a maioria das árvores concentraram sua ati- 8 árvores) do bosque temperado na zona núcleo Cerro Altami- vidade reprodutiva na época de menor precipitação. O número rano, Reserva da Biosfera Mariposa Monarca, México, se reali- de espécies em floração a nível comunitário e de ervas perenes zaram observações mensais durante 2004. Estimou-se a sincro- se correlacionou positivamente com a precipitação, e o número nia intra-específica na floração e frutificação em oito espécies de espécies arbustivas e arbóreas em frutificação mostrou uma lenhosas por meio da observação de 20 indivíduos por espécie. correlação negativa com a precipitação. Determinou-se uma alta A floração e frutificação ocorreram principalmente durante a es- sincronia reprodutiva (>60% dos indivíduos em uma fase fenoló- tação de chuvas e inícios da estação seca (jul-dez), com baixa gica específica) em cinco das espécies arbóreas. Os patrões fe- estacionalidade. As formas de crescimento mostraram diferenças nológicos reprodutivos na área, um bosque temperado de eleva- temporais na sua atividade reprodutiva: i) as ervas anuais e pe- da altitude em zona tropical, foram similares aos documentados renes floresceram principalmente durante a estação de chuvas e para bosques tropicais estacionais de baixas altitudes, e expli- inícios da seca, enquanto que a maioria de espécies com fru- cados principalmente pela precipitação total anual e a forma de tos foi observada na estação seca; ii) os arbustos apresentaram crescimento das espécies.

452 JUL 2007, VOL. 32 Nº 7