Journal of Vegetation Science 6: 487-498, 1995 © IAVS; Opulus Press Uppsala. Printed in Sweden - Composition and ecology of vascular epiphyte communities in Mexico - 487

Composition and ecology of vascular epiphyte communities along an altitudinal gradient in central Veracruz, Mexico

Hietz, Peter1* & Hietz-Seifert, Ursula2

Instituto de Ecología, Aptdo 63, 91000 Xalapa, Veracruz, Mexico; 1*Author for correspondence: present address: Botanisches Institut, Universität für Bodenkultur, Gregor Mendel- Str. 33, A-1180 Wien, Austria; Fax +43 1 3195670; E-mail [email protected] 2Present address: Institut für Pflanzenphysiologie, Universität Wien, Althanstr. 14, A-1091 Wien, Austria

Abstract. Vascular epiphytes were studied in forests at alti- Compared to the amount of information available for tudes from 720 to 2370 m on the Atlantic slope of central ground-rooted vegetation, little is known about the com- Veracruz, Mexico. The biomass of all trees of each species > position of epiphytic communities and the influence of 10 cm diameter at breast height within plots between 625 and environmental variables on the vegetation of the canopy. 1500 m2 was estimated. The number of species per plot ranged No detailed studies have been published on the epi- between 22 and 53, and biomass between 9 and 249 g dry weight/m2. The highest values, both of species and biomass, phytic vegetation of Mexico. Moreover, most studies were found at an intermediate altitude (1430 m). Habitat published so far are descriptive rather than analytical. diversity may contribute to epiphyte diversity in humid for- Numerical methods for classification and ordination, ests, but the importance of this effect could not be distin- now a standard procedure in vegetation analysis, have guished from the influence of climate. A remarkably high successfully been used for epiphytic cryptogams (Kenkel number of bromeliads and orchids grew in relatively dry & Bradfield 1981; Kantvilas & Minchin 1989; Wolf forests at low altitudes. In wet upper montane forests, 1993), but, with few exceptions (van Leerdam et al. bromeliads were replaced by , while orchids were numer- 1990), almost never for vascular epiphytes. ous at all sites, except for a pine forest. The number of A number of studies carried out in tropical moun- epiphytic species and their biomass on a tree of a given site were closely related to tree size. According to Canonical tains (Gradstein & Frahm 1987; Frahm 1990; Wolf Correspondence Analysis, the factor determining the compo- 1993) have shown a strong stratification of cryptogamic sition of the epiphytic vegetation of a tree was altitude and to epiphytes along altitudinal gradients. On the other hand, some extent tree size, whereas tree species had practically no Sugden & Robins’ (1979) paper on epiphytes in the influence. The only trees which had an evidently negative Sierra de Santa Marta and the Serranía de Macuira, effect on epiphytes were pines, which were particularly hostile Colombia, is, to our knowledge, the only detailed ac- to orchids and to a lesser degree to ferns, and Bursera simaruba, count of the distribution of vascular epiphytes. Their which generally had few epiphytes due to its smooth and study suffers to some extent from the rather small plot defoliating bark. 2 sizes (98 and 100 m ) and the fact that many species could not be identified. Keywords: Altitudinal transect; Biomass; Canopy; Diversity; For one of the sites described here, it has been shown Epiphyte; Host specifity. that there are at least two gradients which influence the epiphytic vegetation on the host trees (Hietz & Hietz- Seifert in press). One is the microclimatic gradient from Nomenclature: Sosa & Gómez-Pompa (1994). the humid and shaded stem base to the drier and lighter outer twigs, and the other one along branches of varying Introduction diameters. This study describes the differences amongst epi- The canopy of tropical and subtropical forests has phytic communities for trees at the same location and received much attention within the last two decades for different locations. The importance of climatic vari- (Perry 1984) and has, not without justification, been ables and variables associated with the host trees for the called “the last great unexplored frontier of life on distribution of species, species groups, diversity, and Earth” (Myers 1984). As canopy access techniques were biomass, is examined. In contrast to most other studies, developed and the enthusiasm of biologists increased, we recorded epiphyte biomass, which is a more accurate we began to understand the importance of the canopy measure of the importance of a species than are indi- for the functioning of the forest, and for biodiversity. vidual numbers or frequency. 488 Hietz, P. & Hietz-Seifert, U.

Study sites and Methods Today, most of the original forest has been cleared or replaced by young secondary forest. Old-growth for- The study sites are located in central Veracruz, ests remain only on steep slopes, on soils too rocky for Mexico, close to the state capital Xalapa, along an agriculture, and in a few poorly accessible locations. altitudinal gradient ranging from 720 to 2370 m a.s.l. The six sites selected were old-growth forests with- (Table 1). The exact location of the sites is not revealed out signs of recent human disturbance, except for site 6, in order to protect from collectors, but is available where some big trees were probably felled a few decades upon request to scientists with specific interests. In this ago, and which contained two large rocks, and site 3, area the north-south oriented Sierra Madre Oriental, where goats occasionally graze. Sites 1 and 2 are situated which borders the central Mexican highlands, meets an on rather dry and shallow soils. Consequently, they are east-west oriented chain of tertiary volcanoes (Rzedowski of lower stature and have a lighter canopy than forests on 1986), with Xalapa lying at the foothills of the 4250 m soils commonly found at this altitude, but these have high Cofre de Perote. The humid air from the warm been replaced by fields. Site 3, an almost monospecific Mexican Gulf rises along the Atlantic slopes and causes stand of Pinus pseudostrobus, is exceptional in being high levels of precipitation and humidity and frequent situated on a rocky lava flow with a very shallow soil. At mists. While temperature shows a strong negative linear all other sites, oaks were dominant trees. See Table 1 for correlation with altitude, precipitation is more influ- details and Fig. 2 for some photographs. 2 enced by local climatic conditions and topography, but Plot size was at least 625 m , but larger if the spe- generally is highest around 2000 m (Fig. 1). Monthly cies-area curve (Hopkins 1957) suggested so. For the means of the coldest (December or January) and the purpose of this paper some variation in plot size is not warmest month (May or June) differ by ca. 6 - 7 °C. important; for comparison, species numbers are also 2 75 -80 % of rain falls in the wet season between June given for 625-m subplots in each case (Table 2). and October. All vascular epiphytes were recorded on all trees

Table 1. Characterization of the six forests. According to Rzedowski’s (1986) classification, sites 1 and 2 were studied along an altitudinal transect in central Veracruz. Average temperature calculated with the regression equation from Fig. 1; annual precipitation estimated from the nearest weather station or extrapolated between the two or three closest stations. Sites 1 and 2: mixture of dry oak forest (encinar) with tropical dry forest (selva baja); site 3: pine forest (pinar); sites 4 and 5: mesophilous montane forest (bosque mesófilo de montaña); site 6: transition between mesophilous montane forest and montane pine-oak forest (bosque de pino-encino).

Site Altitude Average temperature Plot size (m2); Stem density/ha; Dominant tree species; Site description Annual precipitation Inclination (°); Basal area (m2/ha); No. of tree species/plot Mist frequency Exposition Canopy height1; Average tree height (m) 1 720 21.9 1500 407 Quercus polymorpha Light, open forest with 1200 ~20 21.1 Bursera simaruba dense undergrowth on dry ± absent SW 13.2 Brahea sp. shallow, calcareous soil 9.0 ~15 2 1000 20.3 676 725 Quercus polymorpha Forest still dry and light in 1300 32 34.0 Persea liebmannii appearance; on calcareous, very rare ~N 15.9 Wimmeria concolor but deeper soil than site 1 11.9 ~14 3 1370 18.2 900 422 Pinus pseudostrobus Open forest on lava flow, 1400 < 5 29.3 Quercus sartorii shallow soil; only one tree, occasional NE 22.0 2 no shrub layer, Quercus 17.8 only two small individuals 4 1430 17.9 625 560 Quercus salicifolia Dense forest on calcareous ~1600 35 75.7 Ostrya virginiana soil; shrub layer not dense frequent W 20.3 9 13.6 5 1980 14.8 625 560 Quercus salicifolia Dense forest on deep, humid, 1850 37 61.5 Fagus grandifolia volcanic soil; shrub layer not very high N 18.5 Carpinus caroliniana dense 11.4 14 6 2370 12.8 900 378 Quercus laurina Forest open due to two large 1500 30 59.4 Pinus patula rocks; shrub layer scarce; very high SW 27.32 Alnus acuminata deep, humid volcanic soil; 16.6 7 somewhat exposed on a ridge

1 Calculated as average height of 20 % of highest trees; 2 21.8 excluding Pinus. - Composition and ecology of vascular epiphyte communities in Mexico - 489

Table 2. Species number, biomass, niche distinctiveness (see text), and diversity indices of the six epiphyte communities studied. Site 123456 Altitude 720 1000 1370 1430 1980 2370 Species per plot 42 40 22 53 39 23 Species per 625 m2 subplot 37 40 21 53 39 23 Estimated epiphytic biomass (g dry weight /m2)30 97 9 249 32 16 Niche distinctiveness 0.208 0.119 0.176 0.232 0.402 0.250 Shannon-Index H' 2.70 1.11 1.99 1.95 1.95 1.84 Simpson-Index D 0.102 0.549 0.178 0.226 0.218 0.228

with a diameter at breast height (DBH) > 10 cm. Tree Consequently, total epiphytic biomass was underesti- number per plot ranged from 34 to 61. Each tree was mated for sites with abundant ferns (site 5), whose divided into six zones: (1) branches < 5 cm diameter; (2) rhizomes may account for a large part of their biomass, branches > 5 and < 20 cm diameter; (3) branches > 20 cm and for sites 4, 5 and 6 with abundant woody hemi- diameter; (4) upper stem; (5) lower stem; and (6) stem epiphytes. Cryptogams were not included in this study; base to 1 m height. No tree at any site had buttresses. they contributed considerably to the total epiphytic Most trees were climbed and observations were made biomass at site 5 only. Dead organic matter, especially by one person in the tree and one person on the ground tree leaves trapped among epiphytes, was conspicuous using binoculars. Vegetative units (individuals, leaves, at site 2, but was not at any site more than a fraction of or shoots) of epiphytes were counted and (for cacti) the total epiphytic biomass. shoot length measured. A number of vegetative units Most individuals, also juveniles, could be identified per species were sampled, dried, and weighed, and total in the field. Only immature individuals of some bro- dry matter of each species within the plots was calcu- meliads were often difficult to distinguish; these were lated per m2 ground area. Only green parts of vascular assigned to groups (e.g. Tillandsia, broad-leaved). Meth- epiphytes were counted, partly because the quantifica- ods are explained in more detail in Hietz & Hietz-Seifert tion of wood and roots was difficult, but also because (subm.). Voucher specimens have been deposited at the green parts, being photosynthetically active, are most herbarium of the Instituto de Ecología at Xalapa (XAL), important for ecosystem studies, since their surface, with some duplicates, especially of bromeliads, at the turnover rates, and biochemical activity are greatest. Institute of Botany, University of Vienna (WU).

Fig. 1. Relationships between altitude, temperature (circles) and precipitation (triangles) in the study area of central Veracruz. All climatic stations are located on the east (Atlan- tic) slope, the highlands are much drier. The inset shows the altitudinal profile along 19° 30' N just south of Xalapa and north of the Cofre de Perote summit. Climatic data are from Soto & García (1989). 490 Hietz, P. & Hietz-Seifert, U.

Fig. 2a. Site 1. Rather open forest with many orchids (flowering Encyclia radiata) and narrow-leaved bromeliads (Tillandsia juncea).

Fig. 2b. Site 2. Similar to site 1, but the epiphytic community is strongly dominated by the two bromeliads Tillandsia juncea and T. fasciculata.

Fig. 2a-c. Representative aspects of Fig. 2c. Site 3. Pine forest of very homogeneous structure on a rocky lava the epiphytic communities in the for- flow. Only bromeliads are common. ests studied: Sites 1 - 3.

Data analysis Bursera simaruba, present only at site 2) to 4 for very rough with abundant deep fissures. The influence of The positive correlation between host size and the bark roughness on the number of species per tree and the number of epiphytic species is most obvious (Fig 3). estimated epiphytic biomass was tested using analysis Whereas the relation between DBH or basal area (BA) of covariance (ANCOVA, Zar 1984), with the number and the number of species per tree was non-linear, log of species and log biomass, respectively, as dependent BA and species number were linearly related. Hence, variables, bark roughness class as factor, and log BA as linear regressions between log BA and the number of covariable. ANCOVA assumptions of normal distribu- epiphytic species per tree were calculated. Slopes were tion and homogeneity of variances were not met for site compared by multiple linear regression (Zar 1984) to 2. Although ANCOVA is considered rather robust against test whether trees at various sites and altitudes were violations of these constraints (Zar 1984), we addition- hosts to a significantly different number of species, ally compared slopes of regressions for bark roughness eliminating the fact that the trees were of different size. 1 (Bursera) and 4 (Quercus) for site 2. Four categories of bark roughness were distinguished, Habitat diversity for epiphytes may be calculated as ranging from 1 for very smooth and defoliating (only the distinctiveness of epiphyte assemblages in different - Composition and ecology of vascular epiphyte communities in Mexico - 491

Fig. 2d. Site 4. Very rich in epiphytic species with epiphytic biomass dominated by a number of broad-leaved and narrow-leaved bromeliads.

Fig. 2f. Site 6. Although ferns dominate, orchids (flowering Encyclia vitellina) may form large clus- ters on thick branches.

Fig. 2d-f. Representative aspects of the Fig. 2e. Site 5. At high altitudes, Elaphoglossum glaucum and other epiphytic communities in the six forests ferns are abundant. The only frequent bromeliad is Tillandsia imperialis. studied: Sites 4 - 6.

zones on a tree. Colwell & Futuyma’s (1971) ‘collective altitude by treating it as a covariable to test whether any heterogeneity of resource states’, calculated with the of the variables associated with individual trees were of species biomass values of the six zones distinguished, importance for the composition of the epiphyte commu- was used as a measure of habitat diversity at each site. nity. All parameters were default values of CANOCO Theoretically, the value ranges between 1 (no resource (ter Braak 1987). shared by two species) and 0 (all species taking the same Two commonly used diversity indices were calcu- share of the same resources). lated (Krebs 1989). 1. Shannon-Index: The epiphytic communities were analysed with Ca- nonical Correspondence Analysis (CCA, ter Braak 1987) Hpp′ = ∑ ii∗ ln (1) 10 using log biomass values as species scores and treating where pi is the proportion of a species compared to the each tree as a site or sample. Environmental variables total biomass, is a measure of evenness and declines if tested were: altitude, DBH, the number of epiphyte few species are dominating. 2. Simpson-Index: 10 species, log of epiphyte biomass, and bark roughness. 2 The first run resulted in altitude being the single domi- Dp= ∑ i (2) nant variable. We consequently eliminated the effect of which in itself is a measure of dominance. 492 Hietz, P. & Hietz-Seifert, U.

Fig. 3. Relationships between DBH and the number of epiphytic species per host tree. The most frequent host species or genera are distinguished by symbols. Lines were fitted by linear regression with log BA as independ- ent variable, but for greater clearness, x-axes of the graphs are DBH instead of log BA. The two lines for site 1 correspond to Quercus (bark rough- ness 4) and Bursera (bark roughness 1).

Results tween 300 and 500 m we usually found 10 - 15 species, considerably less than at site 1. Floristic diversity and distribution There were many orchids in most forests, contribut- At the six sites between 720 and 2370 m a.s.l., 134 ing between 19 and 45 % to the species richness (Fig. 4). species of vascular epiphytes were found. The species However, only one orchid species was found at site 3. number per site decreased from 42 at site 1 at 720 m to The number of bromeliads was high at sites 1 to 4 23 at site 6 at 2370 m (Table 2). The pine forest at 1370 (between 9 and 17 species) but decreased significantly m, with 22 species, was much poorer than the broad- at higher altitudes. Ferns, on the contrary, increased leaved forest at a comparable altitude, with 53 species. with altitude from two species at site 1 to 22 species, Two days of intensive searching in an area of several belonging to seven families, at site 5. Cacti were richest ha around each plot increased the number of species in species in the drier forests at low elevations (sites 1 found by no more than 15 - 20 % and consequently the and 2) and Peperomia at intermediate altitudes (site 4). plot size is considered adequate for a description of the Araceae, Araliaceae, Clusiaceae, Commelinaceae, Cras- epiphyte communities in the forests studied. Also, spe- sulaceae and Solanaceae were present with no more cies numbers and composition of the plots reported were than two species at any site, except for Araceae with consistent with what we found during many field trips to four species at site 4. Woody hemi-epiphytic trees or locations of comparable altitude in the area. shrubs of the families Araliaceae, Clusiaceae and Solana- Although no single forest plot below 700 m or above ceae were common at sites 4 to 6, but rare at lower 2400 m was studied in detail, frequent field trips and a altitudes. Hemi-epiphytic figs, abundant and diverse in revision of the XAL herbarium gave a fairly accurate the lowland rainforest of southern Veracruz (Ibarra & picture of species richness at these altitudes. 17 species Sinaca 1987), are very rare in the montane forests of the of vascular epiphytes were recorded above 2700 m and area and were not present in any of our plots. nine species above 3000 m. In forests at altitudes be- Table 3 shows the number of species common to two - Composition and ecology of vascular epiphyte communities in Mexico - 493

biomass (Fig. 4). In the forests at higher elevations (sites 5 and 6), biomass was dominated by ferns. Orchids, although present with many species, never contributed more than 13 % (at site 2) to the total epiphytic biomass. Primary hemi-epiphytic trees and shrubs, which germi- nate on their host-trees and later establish contact with the soil by aerial roots, represented about 5 % of the epiphytic biomass at sites 4 and 6, and 21 % at site 5 (mainly due to one large individual of Oreopanax liebmannii). Secondary hemi-epiphytes start as climb- ers and may later loose contact with the soil - although none within our plots had reached that stage. Of this life form only Syngonium neglectum at site 1 was promi- nent, accounting for 3 % of the total epiphytic biomass. However, note that only leaves were counted and that if wood were included, hemi-epiphytic trees at sites 4 to 6 would probably outweigh all other epiphytes. Sites 3 to 6 at intermediate and high altitudes had similar values for evenness (H' ranging from 1.84 to 1.99; Table 2) and dominance (D between 0.178 and 0.228). Site 1 had the highest evenness and lowest domi- Fig. 4. Number of epiphytic species per plot (upper graph) and nance measures, whereas the opposite was true for the 2 estimated epiphytic dry matter per m of the six sites studied. floristically very similar site 2. In that forest two bromeliads, Tillandsia fasciculata and T. juncea, domi- sites and Sørensen’s index (SI, Krebs 1989) of similar- nate the epiphytic vegetation and together they account ity. SI ranges between 0.02 and 0.34, and similarity was for 88 % of the total epiphytic biomass. generally greatest between adjacent altitudes. The larg- est difference was found between sites 4 and 5, which Host specificity also show the largest altitudinal difference. Only one At all sites the number of epiphytic species per tree species, the Phlebodium areolatum, occurred at all was positively correlated with tree size, and the correla- sites, and 75, or more than half of the species, occurred tion was generally stronger at sites with larger species at only one site (see App. 1). numbers (Fig. 3). Species numbers per tree differed among sites, as tested by multiple linear regression. Site Habitat diversity 4 not only had the largest total species number, but also Based on habitat diversity, the six zones from the the largest number of epiphytes on a tree of comparable stem base to branches < 5 cm diameter recorded in each size, with two oaks with a DBH of ca. 75 cm carrying 32 forest differed most at site 5 (Table 2). Niche distinc- species of vascular epiphytes each. Sites 1 and 2 had tiveness of all other sites was considerably lower, with about the same total species number per plot as site 5, lowest values at sites 2 (0.119) and 3 (0.176). Appar- but trees of the same size carried a significantly larger ently there were no species adapted to the stem base in number of species at sites 1 and 2. The forest at the these forests with open canopies and much light reach- highest elevation (site 6) and the pine forest (site 3) had ing the soil. Only one small was found on the stem the lowest number of species per tree. All differences base of one tree at site 2, whereas the stem bases of trees mentioned are significant at p < 0.001. in the other forests with open canopies (sites 1 and 3), A significant (ANCOVA, p < 0.01) effect of bark although without a distinct epiphyte community, were roughness on the number of species and epiphyte bio- by no means devoid of epiphytes. mass could only be shown for site 2, which was the only site with Bursera simaruba, the tree with the smoothest Biomass bark. Also, comparing the slopes of the regression be- Epiphyte biomass was estimated to be 250 and 100 g tween log BA and the number of epiphytic species for dry matter/m2 at sites 4 and 2, respectively, but not more bark roughness 1 (Bursera) and 4 (Quercus) resulted in 2 than 35 g/m at the other sites (Table 2). In forests of low significant differences (p < 0.05). Since the forest at site and intermediate altitudes, bromeliads dominated the 2 was mainly composed of oaks with very rough bark epiphytic community, accounting for between 70 % and Bursera with very smooth bark, the comparison (site 1) and over 99 % (site 3) of the total epiphytic between these two extremes resulted in a significant 494 Hietz, P. & Hietz-Seifert, U.

Table 3. Number of species common to two sites and the grow at average annual temperatures below 10 °C and Sørensen index of similarity. endure regular frosts. Forests at low altitudes receive Site 1 2 3 4 5 6 less rain. As mist hardly ever occurs and high tempera- Species per site 42 40 22 53 39 23 tures cause a high evaporative demand, only species 221 able to cope with a very tight water budget can survive. 0.34 Along the transect from warm and dry to cool and humid 31110 forests, the combination of temperature and water avail- 0.26 0.24 ability is an important factor determining the diversity 4131814 0.22 0.28 0.27 and abundance of epiphytes. It seems to be optimal at 51237 mid-altitudes around 1500 m. 0.02 0.05 0.09 0.13 Gentry & Dodson (1987) suggest that humid forests 6112414may achieve a finer niche partitioning and thus a higher 0.03 0.03 0.08 0.10 0.31 diversity because a more constant environment may favour within-community microhabitat specialization by epiphytes. Also, as humid forests are denser, com- difference. No significance was found for the other sites. posed of more layers of trees and shrubs, they have more The pine forest at site 3 had the lowest number of pronounced microclimatic gradients than dry forests of epiphyte species per tree, but not all systematic groups low stature and with open canopies. Therefore, there were poorly represented. Whereas only one single or- may not only be a finer niche partitioning and speciali- chid was found, this site had the largest number of zation because of the constant environment, but wider bromeliads (17, plus another three species found on gradients of microclimates and substrates may also re- pines outside the plot studied). Ferns were very sparse sult in an absolutely larger number of niches and thus compared to other sites, and no other epiphytes were contribute to species richness in humid forests. Gentry present within our plot. & Dodson (1987) found 41 species of understory spe- Ordination of single host trees by CCA showed that cialists among epiphytes in a very humid forest at Rio altitude was the most important variable determining Palenque, Ecuador, but none in a semi-deciduous moist the composition of the epiphytic communities. The forest at Jauneche. These findings are corroborated by eigenvalue of the first axis, which was highly correlated results from our study. The forest with the highest with altitude, was 0.844. When the effect of altitude was habitat diversity (site 5), which was probably the most eliminated by treating it as a covariable, the resulting humid forest in the transect, had a very distinct commu- first axis had an eigenvalue of only 0.367, but was still nity of epiphytes restricted to the lower stem and stem significant (p < 0.01) according to the Monte Carlo base (Hietz & Hietz-Seifert in press). Generally, habitat permutation test (ter Braak 1987), and was best corre- diversity was higher in the more humid forests with a lated with the number of epiphytic species per tree. dense canopy cover than in dry forests with open cano- pies. This may be very important for species richness in the forests studied, but it is not possible to separate this Discussion effect from the influence of climate. Altitude is a measure which is easy to obtain, but Diversity difficult to interpret as it stands for a complex combina- Humid montane forests in South and Central America tion of climatic variables to which species may respond. have been found to be among the richest in vascular From studies of a uni-dimensional gradient it is not epiphytes and are often richer than lowland forests at possible to evaluate the influence of each of these vari- comparable latitudes (Gentry & Dodson 1987). ables. Consequently, for most plants found on the warm In the study area, the largest species number and and dry but not on the cool and wet side of the transect, biomass of vascular epiphytes were found at intermedi- it is not clear whether the low temperature, the high ate altitudes, although at 2000 - 2400 m precipitation and humidity, or other possible factors limit their distribu- especially the frequency of mist (a most important water tion. Only some species of Tillandsia with an atmos- source for epiphytes) are higher. However, low average pheric habit (atmospheric bromeliads have very narrow temperatures at these altitudes apparently exclude many leaves and do not form water reservoirs) occur in warm species. Above 2300 m the number of species decreases and dry forests of the transect and in the cool and dry considerably, probably because of episodic sub-freez- forests of the central highlands. These species appear to ing temperatures. In the area of Xalapa, only 17 species be limited by too high humidity. of vascular epiphytes were found at altitudes above The species numbers reported here are considerably 2700 m and 9 above 3000 m, where they must be able to lower than those found in very humid forests of the - Composition and ecology of vascular epiphyte communities in Mexico - 495 northern Andean region and Costa Rica. Bøgh (1992) variation of epiphytes among different trees in the same 2 reported 104 species of vascular epiphytes on 175 m forest, and among different sites at various altitudes. and Gentry & Dodson (1987) 127 species on 1000 m2; Ordination of single trees by CCA showed that alti- Ingram & Nadkarni (1993) found 65 species on one tude was the most important factor determining the single tree in Costa Rica. Our species numbers are well composition of the epiphyte communities. Most species within the range found in other lowland and montane show a rather narrow altitudinal distribution and of the forests from Central and South America (Sugden & sites with a vertical distance of 250 m or more, none Robins 1979; Kelly 1985; ter Steege & Cornelissen were found to share more than 50 % of their species. 1989). This is noteworthy, since our study area is situ- In contrast to the very distinct epiphytic vegetation ated close to the northern limit of the neotropics, and at different sites there was little variation among trees at receives only moderate rainfall. Still more remarkable is one site. Generally, big trees carried more biomass and the large number of species in comparatively dry forests more species, and some species seemed to be restricted at low altitudes. Mexico is not only an evolutionary to big trees. A significant effect of bark roughness on the centre of cacti and other desert succulents, but also of number of epiphytes could only be demonstrated for the genus Tillandsia, and especially the subgenus Bursera simaruba, whose very smooth and defoliating Tillandsia. This subgenus shows a strong tendency to- bark had a negative effect on colonization by epiphytes. wards an atmospheric habit, low stomata/trichome ra- Bursera was generally little colonized by all groups. tios (Winkler 1986) and crassulacean acid metabolism Pinus spp. are unsuitable hosts for some groups of (Medina 1974), adaptations useful for plants in environ- epiphytes but not for others, as almost no orchids, but ments with scarce and insecure water availability. Simi- abundant bromeliads were found in the pine forest at site larly, the number of tree species found by Lott et al. 3. Pine bark is fissured and would easily trap seeds of (1989) in a dry forest in Jalisco, eastern Mexico, receiv- orchids as well as those of other species, but phenolic or ing less than 800 mm of rain, was considerably higher resinous substances are probably unfavourable to their than that reported from dry forests elsewhere in Central growth. The chemical composition of the substrate may and South America. Like in the case of woody plants affect orchids more than other groups, as orchids de- (Rzedowski 1986), the high diversity of epiphytes pend on mycorrhizal fungi, and pine bark and needles adapted to drought in Mexico seems to be the result of are slow to decompose. Bromeliad roots, on the other the extensive and isolated dry and semi-dry forests in hand, serve mainly as holdfasts and the bulk of nutrients Mexico which caused a radiation of these groups. is absorbed by specialized trichomes on the leaves Our own collections and a revision of the herbarium (Benzing 1970). They should, therefore, be more inde- produced some 200 species of flowering epiphytes from pendent of substrate chemistry. 2 an area of ca. 4000 km , ranging from sea level to over We found no evidence of any epiphyte displaying a 4000 m. Compared to these figures, the 100 species specific preference for a certain host species, although (ferns excluded) found in six plots with a total of ca. 0.5 species represented by only one or few individuals and ha is a large number. It suggests that several compara- at one site, only might be found on a single tree species. tively small but carefully selected preserved areas may If a tree was a suitable host, this was due first of all to its contribute much to protect the biodiversity of an area. size and to some extent to its bark roughness. Oaks, To protect a population, preserves would, of course, often the biggest trees with strongly structured bark, are have to be much larger than the 625-m2 plots of this particularly good hosts. We could not confirm Frei & study in which we often found only one or a few indi- Dodson (1972), who found Q. peduncularis to be un- viduals of a species. Data on the abundance of indi- suitable for colonization by orchids because of phenolic vidual species within forests and their distribution in a bark substances. Among the 10 or more oak species we wider area as those presented here, provide important studied in these and other Mexican forests, including a information for the design of size and location of pre- coffee plantation with Q. peduncularis as shade trees, serves. In addition, more information on the demogra- all were suitable hosts, and densely colonized, also by phy and growth of epiphytes, which is very scarce in the orchids. However, determining Mexican oaks (with a literature, is necessary as survival and reproduction total of some 250 species) is difficult, their systematic determine the fate of populations. Such data are cur- relations are insufficiently known, and it cannot be ruled rently being obtained in the area of this study. out that the oak we determined as Q. peduncularis belonged to another species than Frei & Dodson’s, or Species distribution was at least genetically sufficiently remote to show An analysis of site 5 showed that the position within different levels of phenolic substances. Besides, the the canopy and the trees is of importance for the compo- production of bark substances may be influenced by sition of the epiphyte community. Here we describe the environmental factors. 496 Hietz, P. & Hietz-Seifert, U.

Host specificity of vascular epiphytes has been dis- niche breadth and overlap. Ecology 52: 567-576. cussed by several authors (Went 1940; Benzing 1990; Daniels, J.D. & Lawton, R.O. 1991. Habitat and host prefer- Daniels & Lawton 1991), but unambiguous evidence is ences of Ficus crassiuscula, a neotropical strangling fig of scarce. Within a site, the size of the host tree is the most the lower-montane rain forest. J. Ecol. 79: 129-141. important factor determining the number of epiphytes Frei, J.K. & Dodson, C.H. 1972. The chemical effect of certain bark substances on the germination and early growth of found on it. Therefore, studies evaluating the effect of epiphytic orchids. Bull. Torrey Bot. Club 99: 301-307. different host tree species on epiphytes by comparing Frahm, J.-P. 1990. The ecology of epiphytic bryophytes on the proportion of trees colonized, remain inconclusive Mt. Kanabalu, Sabah (Malaysia). Nova Hedwigia 51: 121- as long as tree size is not included in statistical tests. Our 132. results show that within a forest, describing canopy Gentry, A.H. & Dodson, C.H. 1987. Diversity and biogeogra- height strata or branch size classes can explain more of phy of neotropical vascular epiphytes. Ann. Mo. Bot. the epiphytes’ distribution than distinguishing single Gard. 74: 205-233. trees or tree species. Gradstein, S.R. & Frahm J.-P. 1987. Die floristische Höhen- If an epiphyte appears to show some preference for a gliederung der Moose entlang des BRYOTROP-Transektes certain host species, this preference is often shared by in NO Peru. Beih. Nov. Hedw. 88: 105-113. Hietz, P. & Hietz-Seifert, U. In press. Structure and ecology of most other species present, indicating that the general epiphyte communities of a cloud forest in central Veracruz, suitability of a tree for epiphyte colonization rather than México. J. Veg. Sci. a particular relation between two species is responsible, Hopkins, B. 1957. The concept of minimal areas. Ecology 45: which is a considerable difference. A genuine host 441-449. specificity of an epiphyte would include that one host Ibarra, G. & Sinaca, S. 1987. Listados florísticos de México. species is frequently colonized by one species and less VII Estación de Biología Tropical Los Tuxtlas, Veracruz. often by others and that the supposed host-specific UNAM, México. epiphyte is less abundant on other phorophytes in the Ingram, S.W. & Nadkarni, N.M. 1993. Composition and dis- area, which are suitable for other epiphytes. There are tribution of epiphyte organic matter in a neotropical cloud probably many examples of hosts which tend to be more forest, Costa Rica. Biotropica 25: 370-383. Kantvilas, G. & Michin, P.R. 1989. An analysis of epiphytic or less suitable for colonization because of their size, lichen communities in Tasmanian cool temperate rainfor- bark structure and chemistry, persistent leaf bases, or est. Vegetatio 84: 99-112. other features. These hosts are generally suitable or Kenkel, N.C. & Bradfield, G.E. 1981. Ordination of epiphytic unsuitable either for all species of an area or for certain bryophyte communities in a wet-temperate coniferous groups like bromeliads, orchids, or stranglers. A genu- forest, South-Coastal British Columbia. Vegetatio 45: 147- ine host specificity of single species appears to be rare. 154. Kelly, D.L. 1985. Epiphytes and climbers of a Jamaican rain forest: vertical distribution, life forms and life histories. J. Acknowledgements. Thanks are due to M. A. Soto Arenas, Biogeogr. 12: 223-241. UNAM, Mexico, M. Palacios-Rios, Xalapa, and W. Till, Krebs, C.J. 1989. Ecological methodology. Harper & Row, Vienna, who helped to identify orchids, ferns, and bromeliads, New York, NY. respectively. L. Mucina, Vienna, G. Williams-Linera, Xalapa, Lott, E.J., Bullock, S.H. & Solís-Magallanes, J.A. 1989. Flo- and an anonymous reviewer gave useful comments on the ristic diversity and structure of upland and arroyo forests manuscript, and H. Hurtl and U. Roy-Seifert helped with the of coastal Jalisco. Biotropica 19: 28-235. language. We are grateful to the Instituto de Ecología, Xalapa, Medina, E. 1974. Dark CO2 fixation, habitat preference and and its staff for their hospitality and co-operation. U. H.-S. was evolution within the Bromeliaceae. Evolution 28: 677- supported by an academic exchange program of the Mexican 686. foreign ministry. Myers N. 1984. The primary source. W.W. Norton, New York, NY. Perry, D.R. 1984. The canopy of the tropical rain forest. Sci. References Amer. 251: 138-147. Rzedowski, J. 1986. Vegetación de México, 3rd ed. Editorial Benzing, D.H. 1970. Foliar permeability and the absorption of Limusa, Mexico, DF. minerals and organic nitrogen by certain tank bromeliads. Sosa, V. & Gómez-Pompa, A. 1994. Flora de Veracruz, Bot. Gaz. 131: 23-31. Listado de especies. Instituto de Ecología, A.C., Xalapa, Benzing, D.H. 1990. Vascular epiphytes. Cambridge Univer- México and University of Riverside, Riverside, CA. sity Press, Cambridge. Soto, M. & García, E. 1989. Atlas climatico del estado de Bøgh, A. 1992. Composition and distribution of the vascular Veracruz. Instituto de Ecología, A. C., Xalapa. epiphyte flora of an Ecuadorian montane rain forest. Sugden, A.M. & Robins, R.J. 1979. Aspects of the ecology of Selbyana 13: 25-34. vascular epiphytes in Colombian cloud forests, I. The Colwell, R.K. & Futuyma, D.J. 1971. On the measurement of distribution of the epiphyte flora. Biotropica 11: 173-188. - Composition and ecology of vascular epiphyte communities in Mexico - 497 ter Steege, H. & Cornelissen, J.H.C. 1989. Distribution and Went, F.W. 1940. Soziologie der Epiphyten eines tropischen ecology of vascular epiphytes in lowland rain forest of Urwaldes. Ann. Jard. Bot. Buitenz. 50: 1-98. Guyana. Biotropica 21: 331-339. Winkler, S. 1986. Differenzierungen und deren Ursachen ter Braak, C.J.F. 1987. CANOCO - a FORTRAN program for innerhalb der Bromeliaceen. Beitr. Bid. Pflanzen 61: 283- canonical community ordination by [partial] [detrended] 314. [canonical] correspondence analysis, principal compo- Wolf, J.A.D. 1993. Ecology of epiphytes and epiphyte commu- nents analysis and redundancy analysis (version 2.1). nities in montane rain forests, Colombia. Ph.D. Thesis, TNO Institute of Applied Computer Science, Wageningen. Univ. Amsterdam. van Leerdam, A., Zagt, R.J. & Veneklaas, E.J. 1990. The Zar, J.H. 1984. Biostatistical analysis. 2nd ed. Prentice-Hall, distribution of epiphyte growth-forms in the canopy of a Englewood Cliffs, NJ. Colombian cloud-forest. Vegetatio 87: 59-71. Received 23 June 1994; Revision received 6 February 1995; Accepted 2 March 1995.

Appendix 1. Occurrence of epiphytes in six forest sites near Xalapa, Mexico. Figures are number of trees occupied by species at each site. + indicates that probably more trees were occupied by this species but that juveniles could not be identified. Abbreviated life forms: H-cr = herbaceous long creeping; H-co = herbaceous compact or with very short-creeping forming dense stands; H- pe = herbaceous pendent or scandent; Sf = suffrutescent; He-tr = hemi-epiphytic tree; He-cl = hemi-epiphytic climber; Ta = tank forming rosette; At = atmospheric bromeliad; Ta-at = tank atmospheric intermediate with narrow leaves and small axillary tanks.

Site Life form 1 2 3 4 5 6 Site Life form 1 2 3 4 5 6 Number of trees per site Number of trees per site Pteridophyta 61 49 38 35 35 34 Vittariaceae Anthorphyum ensiforme H-co 1 Vittaria graminifolia H-co 2 Aspleniaceae Asplenium cuspidatum H-co 5 Spermatophyta A. monanthes H-co 1 2 Araceae Grammitidaceae Anthurium schlechtendalii H-co 3 Grammitis cf. prionodes H-co 1 A. scandens H-pe 5 11 Hymenophyllaceae Philodendron advenae He-cl 2 Hymenophyllum cf. crispum H-cr 1 Philodendron sp. He-cl 1 H. polyanthos H-cr 5 Syngonium neglectum He-cl 25 1 H. thunbrigense H-cr 2 Araliaceae Trichomanes sp. 01 H-cr 8 Oreopanax capitatus He-tr 7 T. reptans H-cr 21 O. flaccidus He-tr 2 Lomariopsidaceae O. liebmannii He-tr 1 7 1 Elaphoglossum glaucum H-co 21 Bromeliaceae E. petiolatum H-co 16 Aechmea bracteata Ta 2 Lycopodiaceae Catopsis mooreniana Ta 1 Lycopodium taxifolium H-pe 2 C. nutans Ta 6 3+ 21 L. dichotomum H-co 1 C. paniculata Ta 5 C. sessiliflora Ta 3+ 22 Tillandsia butzii At 5 7 27 Campyloneurum angustifolium H-co 2 T. capitata Ta-at 1 C. xalapense H-co 9 T. concolor Ta-at 2 crassifolium H-co 4 T. fasciculata Ta-at 19 28 19 3 Pecluma spp. H-co 1 1 15 T. filifolia At 11 Phlebodium areolatum H-cr 1 1 6 14 2 5 T. foliosa Ta-at 6 Pleopeltis angustata H-cr 4 T. ghiesbreghtii Ta 10 P. crassinervata H-cr 1 22 5 T. gymnobotrya Ta 2 1 2+ P. mexicana H-cr 20 27 T. heterophylla Ta 1+ Polypodium arcanum H-cr 10 2 T. imperialis Ta 23 P. eatonii H-cr 1 T. ionantha At 43 4 4 P. fraternum H-cr 4 12 T. juncea At 22 32 28 17 P. furfuraceum H-co 27 13 2 14 T. kirchhoffiana Ta-at 26 P. lepidotrichum H-cr 2 7 T. limbata Ta-at 11 3 P. montigenum H-cr 5 T. lucida At 1+ P. plebeium H-cr 1 21 T. multicaulis Ta 2+ 23 P. plesiosorum H-cr 1 26 T. polystachya Ta-at 13 P. polypodioides H-cr 6 1 T. pseudobailey At 2 P. puberulum H-cr 19 14 T. punctulata Ta-at 8 17 P. triseriale H-cr 4 T. recurvata At 9 2 3 498 Hietz, P. & Hietz-Seifert, U.

App. 1, cont.

Site Life form 1 2 3 4 5 6 Site Life form 1 2 3 4 5 6 Number of trees per site Number of trees per site

T. schiedeana At 33 28 35 18 Piperaceae T. streptophylla Ta-at 3 2 Peperomia sp. 01 H-co 1 10 T. tricolor Ta-at 1 2 2 P. deppeana H-co 13 2 T. usneoides At 5 5 7 19 P. galioides H-co 23 7 T. violacea Ta 1+ P. glabella H-co 4 22 T. viridiflora Ta 29 P. aff. quadrifolia H-co 17 P. pseudoalpina H-co 17 Cactaceae P. quadrifolia H-co 29 16 Epiphyllum phyllanthus H-pe 7 7 P. reflexa H-co 3 10 Hylocereus undatus H-pe 9 Rhiposalis baccifera H-pe 12 5 2 Solanaceae Selenicereus cf. coniflorus H-cr 1+ Juanulloa mexicana He-tr 2 S. testudo H-cr 4 Solandra maxima He-tr 3

Clusiaceae Clusia sp. He-tr 10

Commelinaceae Gibbasia cf.geniculata H-cr 5 1

Crassulaceae Echeveria rosea Sf 12 2 Sedum bottieri Sf 2 S. dendroideum Sf 4

Orchidaceae Acineta barkeri H-co 3 Brassia verrucosa H-co 4 Dichaea neglecta H-cr 9 Encyclia cochleata H-co 6 4 E. livida H-co 3 E. ochracea H-co 2 3 2 E. polybulbon H-cr 3 8 E. radiata H-co 4 E. vittelina H-co 1 3 Epidendrum longipetalum H-co 2 3 E. polyanthum H-co 3 Gongora galeata H-co 1 Isochilus unilaterale H-co 7 7 8 Jacqueniella teretifolia H-co 12 Laelia anceps H-co 8 9 Lemboglossum ehrenbergii H-co 15 3 Lepanthes avis H-co 11 L. moorei H-co 2 L. schiedei H-co 4 Lycaste aromatica H-co 6 Maxillaria densa H-pe 6 2 M. meleagris H-co 1 M. variabilis H-pe 1 4 Nageliella purpurea H-co 8 6 Nidema boothii H-co 1 Notylia barkeri H-co 21 2 Oncidium cebolleta H-co 6 O. maculatum H-co 10 2 O. stramineum H-co 17 1 Ornithocephalus inflexus H-co 5 Pleurothallis pubescens H-co 1 P. schiedei H-co 1 P. tribuloides H-co 1 2 P. tubata H-co 4 2 Restrepiella ophiocephala H-co 3 Rhnycholaelia glauca H-co 6 Scaphyglottis livida H-co 4 6 Stanhopea oculata H-co 2 Stelis sp. 01 H-co 4 1 Stelis sp. 02 H-co 1 Vanilla insignis He-cl 1