HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (20) Citing Articles via Google Scholar Google Scholar Articles by Valiente-Banuet, A. Articles by Dávila, P. Search for Related Content PubMed Articles by Valiente-Banuet, A. Articles by Dávila, P. Agricola Articles by Valiente-Banuet, A. Articles by Dávila, P.

The chaparral vegetation in Mexico undernonmediterranean climate: the convergence and Madrean-Tethyan hypothesesreconsidered¹

^a Instituto de Ecología, Universidad NacionalAutónoma de México, Apartado Postal 70–275, UNAM,04510 México, D.F.; and ^b UBIPRO,ENEP-Iztacala, Universidad Nacional Autónoma de México,Apartado Postal 314, México, 54090, Tlalnepantla,México

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
A comparative study between an unburned evergreen sclerophyllousvegetation located in south-central Mexico under a wet-summer climate,with mediterranean regions was conducted in order to re-analyzevegetation and plant characters claimed to converge under mediterraneanclimates. The comparison considered floristic composition,plant-community structure, and plant characters as adaptations tomediterranean climates and analyzed them by means of a correspondenceanalysis, considering a tropical spiny shrubland as the external group.We made a species register of the number of species that resproutedafter a fire occurred in 1995 and a distribution map of the evergreensclerophyllous vegetation in Mexico (mexical) under nonmediterraneanclimates.

The Tehuacán mexical does not differ from the evergreensclerophyllous areas of Chile, California, Australia, and theMediterranean Basin, according to a correspondence analysis, whichordinated the Tehuacán mexical closer to the mediterranean areasthan to the external group.

All the vegetation and floristic characteristics of the mexical, aswell as its distribution along the rain-shadowed mountain parts ofMexico, support its origin in the Madrean-Tethyan hypothesis of Axelrod.Therefore, these results allow to expand the convergence paradigm of thechaparral under an integrative view, in which a general trend to ariditymight explain floristic and adaptive patterns detected in theseenvironments.

Key Words: chaparral • convergence • evergreen • mexical • Mexico • sclerophyll • TehuacanValley • vegetation

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The evergreen sclerophyllous vegetation consists mostly oflow-stature shrubs and trees (1–3 m high) and broad-leafedsclerophyllous species with 40–100% coverage by woodyvegetation. This ecosystem has been associated typically withmediterranean climates with warm, dry summers and cool, wet winters(Griesebach, 1872; Cain, 1950; Naveh, 1967; Specht, 1969; Thrower and Bradbury, 1977; Cody and Mooney, 1978; DiCastri, 1981). It receives different names according to thefive different regions of the world where it grows, including"chaparral" in California, "fynbos" in the CapeProvince of South Africa, "matorral" in Chile,"maquia" in the Mediterranean basin, and"mallee" in the south and south-western Australia (Cody and Mooney, 1978). Although these regionsoccupy <5% of the earth's surface, they harbor almost20% of the known vascular plants of the world (Cowling et al., 1996). These plantcommunities, except the Chilean matorral, have also been traditionallycharacterized by fires. Resprouting ability by means of lignotubers,burls, etc., has been considered an adaptation to these disturbances(Naveh, 1974).

Under the view of the "superorganism" or Clementsianparadigm, this ecological system has been analyzed extensively in acomparative manner for more than three decades. Originally, all thestudies were focused on testing plant community-level convergence,assuming that under a similar climate these communities will evolvetowards convergent solutions, including a maximum efficiency flow ofenergy and nutrients (Barbour and Minnich,1990). Since then, a significant number of studies testingconvergence have been published in the literature, including differentmediterranean-climate regions (Naveh,1967; Mooney and Dunn, 1972;Parsons, 1976; Pignatti and Pignatti, 1985; Cowling and Witkowski, 1994, among others).

Although the convergent viewpoint has played an important role inpromoting comparative studies among mediterranean-type ecosystemsfocused on plant physiology (Mabry and Difeo,1973), plant anatomy (Kummerow,1973), phenology (Kummerow,1983), floristics (Pignatti andPignatti, 1985; Arroyo et al.,1995), life history (Armesto, Vidiella,and Jiménez, 1995; Zedler,1995), seed dispersal (Milewski andBond, 1982: Hoffmann and Armesto,1995), and vegetation (Parsons andMoldenke, 1975; Parsons,1976; Naveh and Whittaker,1979; Cowling and Campbell,1980), this conceptual framework has limited the approach inwhich the sclerophyllous vegetation can be integrally viewed andunderstood. For example, paleobotanical evidence has beenunderestimated, although many studies of the paleobotanical developmentof the chaparral have been described in different publications byAxelrod (1950, 1958, 1973,1975, 1977, 1989),emphasizing that a significant number of chaparral genera arerepresented in Tertiary floras and in many cases the living species arevery similar to the fossils. These records provide reliable evidencethat sclerophyllous vegetation, which is distributed all over the world,constitutes a reminiscence of the Madrean-Tethyan sclerophyllousvegetation that occupied a subhumid belt across much of NorthAmerica-Eurasia region by the middle Eocene. This vegetation hadoriginated from alliances in older laurophyllous forests that adapted toa spreading dry climate (Axelrod,1977). In North America, the Madro-Tertiary Geoflora appearedon the drier borders of the North American tropics by the Middle Eoceneand probably occupied much of the southwestern United States andadjacent Mexico by the end of the Oligocene, and it expanded its rangenorth and southward, as well as east and westward in response toexpanding dry climate conditions during the Miocene epoch (Axelrod, 1958). Similar processes occurred inthe Northern Hemisphere and also explain many of the broad floristicchanges in the actual mediterranean-type ecosystems of the SouthernHemisphere (Rundel [1981] forChilean matorral, but see Arroyo et al.[1995] for a different hypothesis on the origin ofChilean sclerophyllous vegetation; Specht[1981] for Australian Mallee and Linder, Meadows, and Cowling [1992] forSouth African Fynbos).

Since the seminal model of Mooney and Dunn(1970), proposing that in the mediterranean-type climates,fire, drought, high temperatures, rainfall unpredictability, and mineraldeficiencies have selected resprouting evergreen sclerophyllous shrubsas the predominant growth form, most of the studies were designed totest for this evolutionary convergence. Therefore, the central idea ofthe model relating to the existence of an evergreen sclerophyllousvegetation with mediterranean climate has remained untouched, eventhough much evidence of its presence in nonmediterranean climates allover the world should also be considered to expand the model. Forexample, Muller (1939) described thewestern montane chaparral in the summer rain area of Nuevo Leon (Mexico)where the floristic composition and growth form are strongly similar tothe California chaparral. Axelrod(1975) reported the presence of sclerophyllous taxa in areasof summer and winter precipitation (Arizona-New Mexico, easternMediterranean), summer rain and winter drought (eastern Mexico,northwestern India), and well-distributed rainfall throughout the year(northern coast, Turkey). Accordingly, Vankat(1989) claimed for a revision of the Mooney and Dunn's (1970) long-standingparadigm, after finding different responses in seasonal patterns ofstem-water potentials between Arizona (with summer rainfall) andCalifornia chaparrals. Barbour and Minnich(1990) reviewed the IBP and post-IBP literature on chaparralto assess the degree of convergence among the five mediterranean-typeecosystems and found so many differences among vegetations that theyquestioned the convergence paradigm. In addition, phylogenetical andhistorical effects have also been accounted for the explanation of theevolution of plant traits of mediterranean plants (Blondel and Aronson, 1995). For example,Herrera (1992) found that the lifehistory traits such as flowering biology and seed dispersal of southernSpain are better explained by means of historical processes described bysome authors (i.e., Axelrod, 1975;Pignatti, 1978; Pons, 1981; and Palamarev,1989) than by similar (convergent) selection pressures under aMediterranean climate. Keeley (1995)supports this explanation due to the remarkable similarity in seedgermination, dispersal, and seedling recruitment patterns observed inCalifornia and Mediterranean species of Quercus,Rhamnus, and Prunus.

The presence of the evergreen sclerophyllous vegetation in Mexicounder a tropical climate of summer rains has been reported by differentauthors (Muller, 1939, 1947; Miranda andHernández, 1963; Axelrod,1975, 1989; Rzedowski, 1978). Taking into considerationthe information from these studies, it is possible to assume that thesclerophyllous vegetation in Mexico might be a relict of theMadro-Tertiary Geoflora that constitutes the principal element of theMexican chaparral (named here mexical) in nonmediterranean climates andthat its current patchy distribution probably responds to a gradualtrend toward increased dryness during the Tertiary and Quaternary(Axelrod, 1958). If thesclerophyllous vegetation in Mexico exhibits the same plant charactersassociated with the chaparral vegetation under mediterranean climates,this would allow analysis of what factors have contributed to similarplant traits dominating under different climates. At the same time, andconsidering that fire has not played an important ecological factor inthe Tehuacán mexical (only a 1-ha fire occurring in 1995 has beenreported in the study zone for >50 yr), the resprouting ability ofplants and characters in plants that have been referred as adaptationsto fire need to be reconsidered.

This paper is the first attempt to assess a comparison of theevergreen sclerophyllous vegetation located in south-central Mexico(Tehuacán Valley), with other mediterranean regions of the world.The aim of this study is to test whether the characters at the communityand population level that have been claimed to converge undermediterranean climates all over the world are the same as those in theplant communities and populations of the Tehuacán Valley undernonmediterranean climate. We try to determine common environmentalcharacteristics between mediterranean and nonmediterranean climates toexplain the patterns observed. We consider: (1) floristic compositionand the plant community; (2) plant characters that have been consideredas adaptations to mediterranean climates, including fire as anecological factor; and (3) distribution of the mexical undernonmediterranean climates and a general description of the vegetationincluding dominant woody genera, maximum height of vegetation, presenceof evergreen and sclerophyllous species, altitudinal range, and type ofclimate.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Study site
The semiarid Tehuacán-Cuicatlán valley is locatedbetween the states of Puebla and Oaxaca, Mexico(17°39'–18°53'N,96°55'–97°44'W). It covers 10 000km including several intermountain valleys (Fig. 1). Climate is semiarid with anaverage annual precipitation around 400 mm and a drought period in themiddle of the summer rain season (Fig.2). This region owes its aridity to the rain shadow produced bythe Eastern Sierra Madre (Villaseñor,1990; Dávila et al.,1993).

View larger version (25K): [in this window] [in a new window]   Fig. 1. Location of sites. The area of the Tehuacán Valley is delimited by the continuous line (located between 17°39' and 18°53'N and between 96°55' and 97°44'W). Site 1 = Cerro Viejo and Site 2 = Cerro Zotoltepec. The irregular line represents the state boundary between Puebla and Oaxaca states.

View larger version (32K): [in this window] [in a new window]   Fig. 2. Ombrothermic diagram of the Tehuacán city climate.

Vegetation sampling
At Cerro Viejo and Zotoltepec a total of 15 Canfield lines of 50 meach per locality (Mueller-Dombois andEllenberg, 1974) were used to describe the vegetation. Fromthese lines, the following data were obtained for each woody species: relative cover, maximum and minimum height, and frequency (%) asthe proportion of times each species was found in the 15 lines. Fromthese data the Relative Dominance Index (R.D.I) per species wasobtained, such that R.D.I. = relative cover (%) xfrequency (%) x relative density, where relative density= number of individuals/50 m. A total of three1000-m quadrats (50 x 20 m) were used to determine themean number of species in the Tehuacan mexical, in order to compare themwith other mediterranean regions reported by Cowling et al. (1996).

Structural comparison with Mediterranean-typevegetation
The characters of the dominant woody species from differentMediterranean-climate areas were collected from the Specht (1988) databook. The importance of thespecies within their communities studied was confirmed from thefollowing studies: Wilson and Vogl(1965), Hanes (1977), andMooney et al. (1977) for theCalifornian chaparral; Mooney et al.(1977) and Rundel (1981) forthe Chilean matorral; Beadle (1981) andSpecht (1981) for the Australian mallee;and Tomaselli (1981), Quezel (1981), and Romaneand Terradas (1992) for the Mediterranean Basin vegetationrepresented in France, Spain, and Greece. Data on South Africa fynbosare lacking. Fourteen characters regarding plant form, photosyntheticorgans, and leaf characteristics were used in the comparative analysis.These three main groups characterize typical mediterranean traits. Everycharacter was scored as binary to avoid the low frequency of rarecharacters. Characters and character states were scored as follows: (1)renewal buds {1 = microphanerophyte, 0 =nanophanerophyte}, (2) plant height {1 = 25–100cm, 0 = 100–1000 cm}, (3) crown diameter {1= <100 cm, 0 = >100 cm}, (4) photosyntheticorgans {1 = leaves, 0 = phyllodes, cladodes, bothleaves and stems absent}, (5) leaf size {1 =subleptophyll to nanomicrophyll (<0.10–12.25 cm),0 = microphyll to mesophyll (>12.25–180.3cm), (6) leaf length {1 = <1–2 cm, 0= 2–20 cm}, (7) leaf width {1 =<1–5 mm, 0 = 5–50 mm}, (8) leaf angle{1= mainly horizontal, <45° with respect to thehorizon, 0 = mainly vertical, >45° with respect to thehorizon}, (9) leaf margin {1 = entire, 0 =serrate/toothed, lobed/deeply dissected, rolled,recurved/revolute, grooved/incurved}, (10) leaf consistency{1 = malacophyll, 0 = semisclerophyll,sclerophyll}, (11) leaf tomentosity {1 = nonhairy, 0= lower side hairy, upper side hairy, both sides hairy},(12) leaf seasonality {1 = evergreen, 0 = summer andwinter deciduous}, (13) leaf color {1 = all green, 0= all glaucous, all white, green and white, glaucous andwhite}, (14) stem number {1 = single, 0 =several}.

The role of the characters was weighted by means of the number ofspecies scored in each area and then ordinated by a SimpleCorrespondence Analysis (Greenacre,1984). As a reference, the spiny shrubland vegetation adjacentto mexical was considered in the analysis, which is typical of aridclimates, located near the Tehuacán mexical, at18°20'N, 97°27'W, with an altitude ranging from 1380to 1800 m a.s.l. (Osorio et al., 1996)and an average annual rainfall around 425 mm (Dávila et al., 1993; Valiente, 1991). The inclusion of thisexternal group into the analysis was considered necessary to test thesimilarity with the Tehuacan mexical situated within the same regionjust below the altitudinal belt of the sclerophyllous vegetation. Atotal of 121 dominant species were included in the analysis: 28 for theTehuacán region, 18 for the Tehuacán spiny shrubland, 23for the Californian chaparral, 13 for the Chilean matorral, 24 for theMediterranean Basin, and 15 for the Australian mallee. The analysis wasrun in SAS version 6.03 (SAS Institute, Cary, NC).

In each of the localities, coal remnants were searched in soilprofiles (up to 10 m deep) in order to detect wildfire evidence in thepast. Resprouting ability was assessed by unearthing the roots andexamining for presence of lignotubers, burls, or rhizomes. In additionto this, field observations were made in order to ensure that disturbedplants resprouted. These included a record of the species thatresprouted after the 1-ha fire that occurred in1995.

The distribution of themexical
A distribution map of the mexical was undertaken, using bibliographicinformation published mostly in Mexico for the last 50 yr. In addition,the information contained in vegetation maps (scale1:1 000 000) published by INEGI(1981) was also used to construct a 1:4 000 000map showing the distribution of the mexical.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The sclerophyllous vegetation of the Tehuacan Valley consists of 225species of seed plants. In both study sites, the plant coverage is100%, although in some places in the same localities, it is~40–60%. Vegetation is predominantly constituted byshrubs (47%), ephemerals (39.2%), camaephites(10.8%), and trees (7.41%) with heights ~2 m tall,although some tree species can reach up to 4 m tall. In both sites,Quercus sebifera Trel. (Fagaceae) is the dominant species andmost of the species observed are evergreen and sclerophyllous (Table 1), whose leaf angles withrespect to the horizontal axis are >45°. Plants mostly haveseveral stems that originate from the base or close to it. The dataobtained shows that 96.4% of the species resprout. All thesecharacters (leaf angles, basal branching, and resprouting), are similarto the mediterranean plant traits considered in the comparison(Table 2).

Some of the genera of the mexical have the same species or closerelatives of those present in the mediterranean-type ecosystems ofCalifornia, i.e., Arbutus, Ceanothus, Garrya,Juniperus, and Rhus (Table 3). In addition, the mexicalhas some genera that are present in different mediterranean ecosystemsof Chile, the Mediterranean Basin, or Australia (i.e., Rhus,Juniperus, Arbutus, Comarostaphylis (=Arctostaphylos), Quercus, Salvia,Acacia, Stevia, Lithospermum, Linum,Aristida, etc. (Table4).

Vegetation structure: comparison withmediterranean-climate areas
The first axis of the correspondence analysis carried out todiscriminate among the geographical areas studied explains 57.9%of the variance, whereas the second axis indicates 16% (Fig. 3). The main plant charactersexplaining the first axis are (1) leaf seasonality, (2) leafconsistency, and (3) leaf angle. Along this axis, the Tehuacánmexical is ordinated closer to the mediterranean-climate areas than tothe tropical external group or spiny shrubland. The character statesthat are common to the mexical and the mediterranean-climatic areas are(1) evergreen, (2) sclerophyllous or semisclerophyllous, (3) leaves withnearly vertical angles. The mediterranean-climatic areas that are moresimilar in this axis to the Tehuacán mexical are the Chileanmatorral and the Californian chaparral.

View larger version (18K): [in this window] [in a new window]   Fig. 3. Correspondence analysis graph of the structural comparative analysis between the Tehuacán mexical and the mediterranean-type ecosystems of California, Chile, Australia, and the Mediterranean Basin. The Tehuacán spiny shrubland was used as an outgroup. The first axis explains 58% of variance and the second axis explains 16% of variance. See Materials and Methods for codes of characters and character states.

Geographical patternsof the mexical
The distribution of the mexical is highly concentrated along thedifferent mountain chains of the country (Fig. 4), including the Mexican statesof Aguascalientes, Coahuila, Chihuahua, Durango, México,Guanajuato, Guerrero, Hidalgo, Jalisco, Nuevo León, Oaxaca,Puebla, Querétaro, San Luis Potosí, Tamaulipas, Tlaxcala,and Zacatecas. In all cases, the mexical is located along therain-shadowed (dry) aspects of the mountain chains, which include theSierra Madre Oriental, Sierra Madre Occidental, Neovolcanic belt, andOaxacan mountains. Altitudinally the mexical ranges between 1700 and2800 m a.s.l. (Fig. 4;Table 5). In all cases,the mexical occupies an intermediate position between oak and pineforests (above), and the xerophitic communities (below). The mexicalranges from arid to dry-temperate climates (Table 5), which corresponds to typesBs and Cw, respectively, according to García (1973). Thevegetation for all the states where descriptions are available(Table 5) is dominated byevergreen, sclerophyllous shrubs with few tree representatives. Finally, a significant number of genera are common among the differentmexical localities and with different mediterranean-type ecosystems ofthe world (Table5).

View larger version (28K): [in this window] [in a new window]   Fig. 4. Distribution of the mexical (dotted marks) under wet-summer climates in México. Altitude lines are in m a.s.l. The black delimited zone in the North West of Mexico indicates the chaparral distribution under mediterranean climate.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The correspondence analysis shows two different groups. The firstcorresponds to the areas of evergreen sclerophyllous vegetation (Chileanmatorral, Tehuacán mexical, Californian chaparral, Australianmallee, and Mediterranean Basin vegetation) and the second to theTehuacan spiny shrubland. Both groups have been differentiated on thebasis of leaf seasonality, consistency, and angle. In all fiveevergreen, sclerophyllous vegetation areas of the analysis, thecharacter states "evergreen" and"sclerophyllous" and "vertical leaf angles" weredominant. In contrast, in the Tehuacan spiny shrubland the characterstates "deciduous," "malacophyllous," and"horizontal leaf angles" were the dominant traits. Two ofthese discriminatory characteristics (evergreen and sclerophyllous)correspond to those found by Barbour and Minnich(1990) as consistently similar among the fivemediterranean-type ecosystems. The other factor, steep leaf angles, hasbeen reported as an adaptation of Mediterranean sclerophyllous plants toreduce radiation absorption during extended drought periods (Ehleringer and Comstock, 1989;Valiente-Banuet et al., unpublished data). Within chaparral ecosystems,limited water represents a major stress affecting plant performance andsince leaf temperatures usually increase plant water stress, steep leafangles contribute to a reduction in leaf solar radiation absorption(Smith and Ullberg, 1989). In the Tehuacan mexical, 93% of thespecies present leaf angles ranging between 45° and 90° withrespect to the horizontal axis, which suggests thatevergreen-sclerophyllous species are responding mainly to waterstress.

Considering the second axis in the correspondence analysis, thenumber of stems is a significant discrimination factor. This charactercan be considered as an architectural characteristic related toresprouting ability, and it is represented by a significant number ofspecies in the Chilean matorral not so frequently observed in othermediterranean regions. Mediterranean-type ecosystems have been matchedwith natural fires and therefore typical features of shrubs, such aslignotubers and burls at the root crown/stem base, have beenassociated with postfire resprouting ability (James, 1984; Kummerow andEllis, 1989). These woody structures and resprouting abilityare present in the shrubs of the Tehuacan mexical where there is noevidence of periodic fire, indicating that their presence can be betterconsidered as an ancient pre-adaptation to fire inherited from ancestorsbelonging to laurophyllous forests (Axelrod,1975). Actually, the fact that in the Tehuacán mexicala significant number of species resprouted after a fire in 1995 suggestsa similar response to fire when compared with mediterranean zones(Keeley, 1992). Particularly,Ceanothus gregii, a common species in the California chaparral,recruits seedlings after fire both in California (Keeley, 1992) and in the Tehuacánmexical. Therefore, resprouting ability has recently been considered notas an adaptation, but a pre-adaptation to fire with an evolutionaryorigin based on the response of the plants to herbivory (López-Soria and Castell, 1992). Forthis reason, resprouting ability has been associated with ancient woodygroups abundant in Chile, under a nonextreme climate (Arroyo et al., 1995). Mexical vegetation,developed probably under the mildest climate among all the studiedsites, maintains also many old woody tropical lineages (i.e.,Acacia, Amelanchier, Arbutus,Bursera, Ceanothus, Cercocarpus,Comarostaphylis, Garrya, Karwinskia,Leucaena, Litsea, Quercus, Rhus,Satureja, etc.), which are fossil representatives of thepre-Pliocene period. Mexical vegetation contains the highest percentageof resprouters (94.7%; present study), followed by Chile(75.6%; García-Fayos, unpublished data), whereas the moreextreme climate areas of California and Mediterranean Basin only contain50 and 64% of resprouter taxa, respectively (Hanes, 1971; García-Foyas, 1991). These datasupport the hypothesis that under less extreme climates, the maintenanceof older woody resprouter lineages takes place. This hypothesis isreinforced by the fact that in the other mediterranean-type ecosystems,belonging to the Mediterranean Basin, the older tropical woody lineagessuffered extinction processes, whereas the group of seeders diversified(Herrera, 1992).

An evolutionary convergence explanation can be claimed at this point,without invoking group selection. However, the results of this studyshow that the same traits that have been thought to converge undermediterranean climate are also developed among plants under tropical andnonmediterranean climate. Consequently, it seems that the presentclimate is not an important factor in the evolution of the plant traitsstudied, which is also supported by the fact that plants of the Tehuacanspiny shrubland living under the same tropical climate of theTehuacán mexical, but in drier areas and below the 1900 m ofaltitude, differ completely in these characteristics. In contrast, theTehuacan mexical plants are much more similar to mediterranean-typevegetation, even though they are established in a tropical climate.

The mexical vegetation can be considered as a relict of theMadro-Tertiary Geoflora, whose patchy distribution might be the resultof the expanding dry climate during the Miocene epoch (Axelrod, 1958). For example, Axelrod (1975) reports genera such asAcacia, Gochnatia, Baccharis,Satureja, Stevia, Lithospermum,Yucca, Rhus, Juniperus, Arbutus,Comarostaphylis, Quercus, Garrya,Salvia, Ceanothus, etc., which are present both inMexico and California, and some of them are also present in theMediterranean Basin and the Chilean matorral. The presence of thesegenera in both zones with summer-wet climates of Mexico and themediterranean regions of the world supports the origin of the mexical inthe Madrean-Tethyan sclerophyllous vegetation, which occupied a subhumidbelt across much of North-America-Eurasia by the middle Eocene andoriginated from alliances in older laurophyllous forests that adapted tospreading into dry climate (Axelrod,1977). Actually, the mexical is distributed entirely along thedry parts of the mountain chains produced by the rain shadow of theEastern and Western Sierras Madre and Neovolcanic belt, as well as theOaxacan mountains. In all the cases, below the altitudinal limit of themexical, different types of xerophitic shrublands are found, whereasabove the mexical, oak and pine forests are always present if thealtitude is above 2500 m. This patchy distribution of the mexical alongthe main mountains of Mexico perhaps had a wider distribution thantoday. Indeed, during the Pliocene, California chaparral had a widerdistribution than it does today, occupying areas now desert or coveredwith broadleaved evergreen forests (Axelrod,1973). The gradual development of regional differences in thedistribution of seasonal rainfall and in temperature relations as drierclimates developed (Axelrod, 1973)probably accounted for its actual patchy and relict distribution alongthe Mexican mountains. Undoubtedly, the recent formation of importantmountain chains such as the Sierra Madre Occidental, no more than 5million years B.P. (Cserna, 1989),provided areas for spread of mexical. Indeed, the fossil recordindicates that the summer rainfall regime from Arizona to Texas andsouthward into Mexico approximates the conditions under whichchaparral-type vegetation occurred during most of its recorded history,which can be traced back into the Oligocene (Axelrod, 1973). Therefore, the mediterraneanclimate is not old (Axelrod, 1973),and therefore sclerophyllous species that now typify the mediterraneanareas are survivors of a richer flora that persisted in Mexico undersummer-wet climates. This accounts for the phenological patternsreported for mediterranean regions. Plant phenology is one of theprocesses that have been thought to converge among themediterranean-climate areas. In response to hot and dry summers andmild and wet winters, plant communities around the mediterraneanecosystems show similar phenological patterns consisting of a markedflowering peak at the spring season (Arroyo,1988). This indicates that species responded to amediterranean climate that developed gradually since the late Cenozoic(Axelrod, 1973). In contrast, themexical community has shown a different phenological pattern. Data(unpublished) taken over one year revealed that a marked flowering peakis shown neither in the spring nor in any other season. In contrast tothe marked seasonality in mediterranean communities, mexical showed aconstant flowering percentage across the four seasons.

The prevalence of Tertiary genera in Mexico, which now are extinct insome mediterranean regions of the Mediterranean Basin (Herrera, 1992), probably is due to more benignand nonextreme climatic conditions in Mexico, such as has been suggestedfor Chile by Arroyo et al. (1995). Ifso, it might help to explain both the high and similar plant diversityencountered in 1000 min the Tehuacan mexical (90 ± 6species), with respect to the Chilean matorral (100 ± 15species), and the prevalence of resprouter species in both sites. AsHerrera (1992) pointed out, theextinction of these taxa allowed the diversification of seeders duringQuaternary. In Mexico, nothing is known about diversification of taxaduring the Quaternary, however, it is possible that if this phenomenonreally occurred, the high diversity shown in the Tehuacán mexicalcan be related to the presence of a mixture of Tertiary and Quaternaryspecies, with the predominance of species from the Tertiary.

In summary, the similarities found in this comparative study betweenthe mexical and the mediterranean regions indicate that the mexicalrepresents the same vegetation that traditionally has been consideredonly associated with mediterranean climates. In addition to classicalviews of evergreenness and sclerophylly as adaptations to environmentalconstraints, such as oligotrophy, water stress, and herbivory, thisstudy allows us to expand the convergence paradigm. Therefore, theseecological systems should be better understood under the integrativeview of the Madrean-Tethyan hypothesis proposed by Axelrod (1958, 1975) in which a paleoclimatical trend toaridity might explain many of the floristic and ecomorphologicalpatterns detected in these environments. At the same time, once themediterranean climate developed gradually after the late Cenozoic in thefive different mediterranean regions, taxa under a summer-wet climategradually adapted physiologically to the newconditions.

	FOOTNOTES

 
¹ The authors thank D. Axelrod, J. Rzedowski, P. García-Fayos, F. Lloret, and one anonymous reviewer for comments on the manuscript; Leopoldo Valiente and Ignacio Méndez for their help in statistical analysis at the IIMAS. This study was supported by a grant provided by the Dirección General de Asuntos del Personal Académico de la UNAM (DGAPA-IN208195).

⁴ Author for correspondence.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Armesto, J. J., P. E. Vidiella, and H. E.Jiménez. 1995. Evaluating the causes andmechanisms of succession in the Mediterranean regions in Chile andCalifornia. In M. K. T. Arroyo, P. H. Zedler, and M. A. Fox[eds.], Ecology and biogeography of mediterraneanecosystems in Chile, California and Australia, 418–434. Springer-Verlag, New York, NY.

Arroyo,J. 1988. Fenología de la floración enespecies del matorral del sur de España. Lagascalia 15:593–606.

Arroyo, M. T. K., L. Cavieres, C.Marticorena, and M.Muñoz-Schick. 1995. Convergence in themediterranean floras in Central Chile and California: insights fromcomparative biogeography. In M. K. T. Arroyo, P. H. Zedler, andM. A. Fox [eds.], Ecology and biogeography ofmediterranean ecosystems in Chile, California and Australia,43–88. Springer-Verlag, New York, NY.

Axelrod, D. I. 1950. Theevolution of desert vegetation in western North America. Publicationof the Carnegie Institution of Washington 590:215–306.

———. 1958. Evolution of theMadro-Tertiary Geoflora. Botanical Review 24:433–509.

———. 1973. Historyof the mediterranean ecosystem in California. In F. Di Castriand H. A. Mooney [eds.], Mediterranean type ecosystems,origin and structure, 225–284. Springer-Verlag, New York,NY.

———. 1975. Evolution andbiogeography of Madrean-Tethyan sclerophyll vegetation. Annals ofthe Missouri Botanical Garden 62: 280–334.

———. 1977. Outlinehistory of California vegetation. In M. G. Barbour and J.Major [eds.], Terrestrial vegetation of California,139–193. John Wiley, New York, NY.

———. 1989. Age andorigin of chaparral. In S. C. Keeley [ed.], TheCalifornia chaparral: paradigms reexamined, Science Series 34,7–19. Natural History Museum of Los Angeles County, Los Angeles,CA.

Barbour, M. G., and R. A.Minnich. 1990. The myth of chaparral convergence.Israel Journal of Botany 39: 453–463. [ISI]

Beadle, N. C. W. 1981. Themallee and marlock communities. In G. Fisher [ed.],Vegetation of Australia, 340–359. Stuttgart, WestGermany.

Blondel, J., and J.Aronson. 1995. Biodiversity and ecosystem function inthe Mediterranean Basin. Human and non-human determinants. InG. W. Davis and D. M Richardson [eds.], Mediterranean-typeecosystems: the function of biodiversity, 43–119. Springer-Verlag,Heidelberg.

Brunet, J. 1967. Geologicstudies. In D. S. Byers [ed.], The prehistory of theTehuacan Valley, Environment and subsistence, vol. 1, 66–90.Robert Peabody Foundation. University of Texas Press, Austin,TX.

Cain, S. A. 1950. Lifeformsand phytoclimate. Botanical Review 16: 1–33. [ISI]

Cody, M. L., and H. A.Mooney. 1978. Convergence versus nonconvergence inmediterranean-climate ecosystems. Annual Review of Ecology andSystematics 9: 265–321. [CrossRef][ISI]

Cowling, R. M., and B. M.Campbell. 1980. Convergence in vegetation structure inthe mediterranean comunities of California, Chile and South Africa.Vegetatio 43: 191–197. [CrossRef][ISI]

———, W. P. Rundel, B. B.Lamont, K. M. Arroyo, and M. Arianoutsou. 1996. Plantdiversity in mediterranean-climate regions. Trends in Ecology andEvolution 11: 362–366. [CrossRef]

———, and E. T. F.Witkowski. 1994. Convergence and nonconvergence ofplant traits in climatically and edaphically matched sites inMediterranean Australia and South Africa. Australian Journal ofEcology 19: 220–232.

Cruz-Cisneros, R., and J.Rzedowski. 1980. Vegetación de la Cuenca delRío Tepelmeme, Alta Mixteca, Estado de Oaxaca, México. Anales de la Escuela Nacional de Ciencias Biológicas deMéxico 22: 19–84.

Dávila, A. P., R. J. L.Villaseñor, L. R. Medina, R. A. Ramírez, T. A. Salinas,J. Sánchez-Ken, and L. P.Tenorio. 1993. Listados florísticos deMéxico. X. Flora del Valle de Tehuacán-Cuicatlán.Instituto de Biología Universidad Nacional Autónoma deMéxico, México.

de Cserna, Z. 1989. Anoutline of the geology of Mexico. In A. W. Bally and A. R.Palmer [eds.], The geology of North America—an overview,vol. A, 233–264, The Geology of North America. Geological Societyof America, Boulder, CO.

di Castri,F. 1981. Mediterranean-type shrublands of the world.In F. Di Castri, D. W. Goodall, and R.L. Specht[eds.], Ecosystems of the world: mediterranean-typeshrublands, vol. 11, 1–52. Elsevier, Amsterdam.

Ehleringer, J. R., and J. P.Comstock. 1989. Stress tolerance and adaptive variationin leaf absorptance and leaf angle. In S. C. Keeley[ed.], The California chaparral, paradigms reexamined, ScienceSeries 34, 21–24. Natural History Museum of Los Angeles County,Los Angeles, CA.

García,E. 1988. Modificaciones al sistema declasificación climática de Köppen. Instituto deGeografía, Universidad Nacional Autónoma de México,México.

García, M.A. 1983. Estudio ecológico-florístico deuna porción de la Sierra de Tamazulapan, Distrito de Teposcolula,Oaxaca, México. Tesis de Licenciatura, Facultad de Ciencias,Universidad Nacional Autónoma de México,México.

García-Fayos,P. 1991. La vegetación silicícola de laSierra Calderona (Comunidad Valenciana). Lazaroa 12:317–332.

González-Quintero,J. 1968. Tipos de vegetación del Valle deMezquital, Hidalgo. Departamento de Prehistoria, Instituto Nacional deAntropología e Historia, México.

Greenacre, M.J. 1984. Theory and applications of correspondenceanalysis. Academic Press, London.

Grisebach, A. 1872. Dievegetation der Erde nach ihrer klimatischen Anordnung. Engelmann,Leipzig.

Hanes, T.L. 1971. Succession after fire in the chaparral ofsouthern California. Ecological Monographs 41:27–52.[CrossRef]

———. 1977. CaliforniaChaparral. In M. G. Barbour and J. Major [eds.],Terrestrial vegetation of California, 417–470. John Wiley, NewYork, NY.

Hennessy, J. T., R. P. Gibbens, J. M.Tromble, and M. Cárdenas. 1983. Water propertiesof caliche. Journal of Range Management 36:723–726.[CrossRef][ISI]

Herrera, C.M. 1992. Historical effects and sorting processes asexplanations for contemporary ecological patterns: character syndromesin mediterranean woody plants. American Naturalist 140:421–446.[CrossRef][ISI]

Hiriart, V.P. 1981. Vegetación y fitogeografía de laBarranca de Tolantongo, Hidalgo, México. Tesis Licenciatura,Facultad de Ciencias, Universidad Nacional Autónoma deMéxico, México.

Hoffman, A. J., and J. J.Armesto. 1995. Modes of seed dispersal in themediterranean regions in Chile, California, and Australia. InM. K. T. Arroyo, P. H. Zedler, and M. A. Fox [eds.],Ecology and biogeography of mediterranean ecosystems in Chile,California and Australia, 289–310. Springer-Verlag, New York,NY.

INEGI. 1981. Cartas 1:1 000 000 de usode suelo y vegetación. Instituto Nacional de EstadísticaGeografía e Informática, México.

James, S. 1984. Lignotubersand burls—ir structure, function and ecological significance inmediterranean ecosystems. Botanical Review 50:225–266.

Keeley, J.E. 1992. Recruitment of seedlings and vegetativesprouts in unburned chaparral. Ecology 73:1194–1208.[CrossRef][ISI]

———. 1995. Seed-germinationspatterns in fire-prone mediterranean-climate regions. In M. T.K. Arroyo, P. H. Zedler, and M. A. Fox [eds.], Ecology andbiogeography of mediterranean ecosystems in Chile, California andAustralia, 239–273. Springer-Verlag, New York, NY.

Kummerow,J. 1973. Comparative anatomy of sclerophylls ofmediterranean climatic areas. In F. di Castri and H. A. Mooney[eds.], Mediterranean-type ecosystems: origin andstructure, 157–167. Springer-Verlag, New York, NY.

———. 1983. Comparativephenology of Mediterranean-type plant communities. In F. J.Kugger, D. T. Mitchell, and J. U. M. Jarvis [eds.],Mediterranean-type ecosystems, 300–317. Springer-Verlag, New York,NY.

———, and B. A.Ellis. 1989. Structure and function in chaparralshrubs. In S. C. Keeley [ed.], The Californiachaparral: paradigms reexamined, Science Series 34, 140–150.Natural History Museum of Los Angeles County, Los Angeles,CA.

Le Sueur, H. 1945. Theecology and the vegetation of Chihuahua, México, north parallel28. University of Texas Publication 4521:1–92.

Linder, H. P., M. E. Meadows, and R. M.Cowling. 1992. History of the Cape flora. InR. M. Cowling [ed.], The ecology of fynbos: nutrients, fireand diversity, 113–134, Oxford University Press,Oxford.

López-Soria, L., and C.Castell. 1992. Comparative genet survival after fire inwoody mediterranean species. Oecologia 91:493–499.[CrossRef][ISI]

Mabry, T. M., and D. R. Jr.Difeo. 1973. The role of the secondary plant chemistryin the evolution of the mediterranean scrub vegetation. In F.di Castri and H. A. Mooney [eds.], Mediterranean-typeecosystems: origin and structure, 129–155. Springer-Verlag, NewYork, NY.

Martin, P. S. 1958. Abiogeography of reptiles and amphibians in the GómezFarías region, Tamaulipas, México. MiscellaneousPublication of the Museum of Zoology, University of Michigan 101:1–102.

Milewski, A. V., and W. J.Bond. 1982. Convergence of myrmecochory inmediterranean Australia and South Africa. In R. C. Buckley[ed.], Ant-plant interactions in Australia, 89–98. Junk,The Hague.

Miranda, F. 1948. Datossobre la vegetación de la Cuenca Alta del Papaloapan. Analesdel Instituto de Biología, Universidad Nacional Autónomade México 19: 333–364.

———. 1952. Lavegetación de Chiapas. Ediciones del Gobierno del Estado. TuxtlaGutiérrez, México.

———, and X. E.Hernández. 1963. Los tipos de vegetaciónen México y su clasificación. Boletin de la SociedadBotánica de México 28: 29–179.

Mooney, H. A., and E. L.Dunn. 1970. Convergent evolution ofmediterranean-climate evergreen sclerophyll shrubs.Evolution 24: 292–303.

———, and———. 1972. Land-use history of Californiaand Chile as related to the structure of the sclerophyll scrubvegetations. Madroño 21:305–319.

———, J. Kummerow, A. W. Johnson, D. J. Parsons, S. C. Kelley, A. Hoffmann, R. Y. Hays, J.Giliberto, and C. Chu. 1977. The producers—irresources and adaptative responses. In H. A. Mooney[ed.], Convergent evolution in Chile and California:mediterranean climate ecosystems, 85–143. Dowden, Hutchinson andRoss, Stroudsburg, PA.

Mueller-Dombois, D., and H.Ellenberg. 1974. Aims and methods of vegetationecology. John Wiley&Sons, New York, NY.

Muller, C.H. 1939. Relations of the vegetation and climatic typesin Nuevo Leon, México. American Midland Naturalist 21:687–729.[CrossRef]

———. 1947. Vegetation andclimate of Coahuila, México. Madroño 9:1–32.

Naveh,Z. 1967. Mediterranean ecosystem and vegetation typesin California and Israel. Ecology 48:445–459.

———. 1974. Theevolutionary significance of fire in the Mediterranean region. Vegetatio 29: 199–208. [CrossRef]

———, and R. H.Whittaker. 1979. Structural and floristic diversity ofshrublands and woodlands in northern Israel and other Mediterraneanareas. Vegetatio 41: 171–190. [CrossRef][ISI]

Osorio, O., A. Valiente-Banuet, P.Dávila, and R. Medina. 1996. Tipos devegetación y diversidad ß en el Valle de Zapotitlánde las Salinas, Puebla, México. Boletin de la SociedadBotánica de México 59: 35–58.

Palamarev,E. 1989. Paleobotanical evidences of the Tertiaryhistory and origin of the mediterranean sclerophyll dendroflora.Plant Systematics and Evolution 162: 93–107. [CrossRef][ISI]

Parsons, D.J. 1976. Vegetation structure in the mediterraneanscrub communities of California and Chile. Journal of Ecology64: 435–447. [CrossRef]

———, and A. R.Moldenke. 1975. Convergence in vegetation structurealong analogous climatic gradients in California and Chile.Ecology 56: 950–957. [CrossRef][ISI]

Pignatti, E., and S.Pignatti. 1985. Mediterranean type vegetation of SWAustralia, Chile and the Mediterranean Basin, a comparison. AnnaliDi Botanica 43: 227–243.

Pignatti,S. 1978. Evolutionary trends in Mediterranean flora andvegetation. Vegetatio 37: 175–185. [CrossRef][ISI]

Pons, A. 1981. The history ofthe mediterranean shrublands. In F. Di Castri, D. W. Goodall,and R. L. Specht [eds.], Mediterranean type shrublands,Ecosystems of the world 11, 131–138. Elsevier,Amsterdam.

Puig, H. 1970. Notas acerca dela flora y la vegetación de la Sierra de Tamaulipas. Analesde la Escuela Nacional de Ciencias Biológicas deMéxico 17: 37–49.

Quero, J. H. 1977. Lavegetación de las serranías de la Cuenca Alta delrío de la Laja, Guanajuato. Anales del Instituto deBiología, Universidad Nacional Autónoma deMéxico 47: 73–79.

Quezel, P. 1981. Floristiccomposition and phytosociological structure of sclerophyllous matorralaround the Mediterranean. In F. Di Castri, D. W. Goodall, andR. L. Specht [eds.], Mediterranean type shrublands, Ecosystemsof the world 11, 107–121. Elsevier, Amsterdam.

Rojas-Mendoza,P. 1965. Generalidades sobre la vegetación delestado de Nuevo León y datos acerca de su flora. Ph.D.dissertation, Facultad de Ciencias, Universidad Nacional Autónomade México.

Romane, F., and J. Terradas.[eds.]. 1992. Quercus ilex L. ecosystems:function, dynamics and management. Kluwer, Dordrecht.

Rundel, P. W. 1981. TheMatorral zone of central Chile. In F. Di Castri, D. W. Goodall,and R. L. Specht [eds.], Mediterranean type shrublands,Ecosystems of the world 11, 175–201. Elsevier,Amsterdam.

Rzedowski,J. 1954. Vegetación del Pedregal de San Angel,Distrito Federal, México. Anales de la Escuela Nacional deCiencias Biológicas de México 8:59–129.

———. 1966. Vegetacióndel estado de San Luis Potosí. Acta CientíficaPotosina 5: 5–291.

———. 1978. Vegetacion deMéxico. Ed Limusa. México D. F.

———, G. Guzmán,C. Hernández, and R.Muñiz. 1964. Cartografía de lavegetación de la parte norte del Valle de México. Anales de la Escuela Nacional de Ciencias Biológicas deMéxico 13: 31–57.

———, and R. MacVaugh. 1966. La vegetación de Nueva Galicia. Contributions of the University of Michigan Herbarium 9:1–123.

Shreve,F. 1939. Observations of the vegetation of Chihuahua. Madroño 5: 1–13.

Specht, R. L. 1969. Acomparison of the sclerophyllous vegetation characteristic ofmediterranean type climates in France, California, and SouthernAustralia. Australian Journal of Botany 17:277–292.[CrossRef]

———. 1981. Malleeecosystems in southern Australia. In F. Di Castri, F., D. W.Goodall, and R. L. Specht [eds.], Mediterranean typeshrublands, Ecosystems of the world 11, 203–231.Elsevier, Amsterdam.

———. 1988. Mediterranean-typeecosystems a data source book, Tasks for vegetation science 19.Kluwer, Dordrecht.

Thrower, N. J. W., and D. E.Bradbury. 1977. Chile-California mediterranean scrubatlas: a comparative analysis. Dowden, Hutchinson and Ross, Stroudsburg,PA.

Tomaselli, R. 1981. Mainphysiognomic types and geographic distribution of shrub systems relatedto mediterranean climates. In F. Di Castri, D. W. Goodall, andR. L. Specht [eds.], Mediterranean type shrublands,Ecosystems of the world 11, 95–106. Elsevier,Amsterdam.

Turner, I.M. 1996. Sclerophylly: primarily protective.Functional Ecology 8: 669–675. [CrossRef]

Valiente, B.L. 1991. Patrones de precipitacion en el ValleSemiárido de Tehuacán, Puebla, México. Tesis deLicenciatura. Facultad de Ciencias UNAM. México D.F.

Vankat, J. L. 1989. Waterstress in chaparral shrubs in summer-rain versus summer-droughtclimates—Whither the mediterranean-type climate paradigm?In S. C. Keeley [ed.], The California chaparral:paradigms reexamined, Scientific Series 34, 117–124. NaturalHistory Museum of Los Angeles County, Los Angeles, CA.

Villaseñor, J.L. 1990. The genera of Asteraceae endemic toMéxico and adjacent regions. Aliso 12:685–692.

Wilson, R. C., and R. J.Vogl. 1965. Manzanita chaparral in the Santa AnaMountains, California. Madroño 18:47–62.

Zedler, P. H. 1995. Aresome plants to burn? Trends in Ecology and Evolution 10:393–395.

———, and C. A. Zammit. 1989. A population-based critique of conceptsof change in the chaparral. In S. C. Keeley [ed.],The California chaparral: paradigms reexamined, Scientific Series 34,73–83. Natural History Museum of Los Angeles County, Los Angeles,CA.

This article has been cited by other articles:

				A. Valiente-Banuet, A. V. Rumebe, M. Verdu, and R. M. Callaway Modern Quaternary plant lineages promote diversity through facilitation of ancient Tertiary lineages PNAS, November 7, 2006; 103(45): 16812 - 16817. [Abstract] [Full Text] [PDF]

				G. Mansion and L. Zeltner Phylogenetic relationships within the New World endemic Zeltnera (Gentianaceae-Chironiinae) inferred from molecular and karyological data Am. J. Botany, December 1, 2004; 91(12): 2069 - 2086. [Abstract] [Full Text] [PDF]

				F. Lloret, M. Verdú, N. Flores-Hernández, and A. Valiente-Banuet Fire and resprouting in Mediterranean ecosystems: insights from an external biogeographical region, the mexical shrubland Am. J. Botany, December 1, 1999; 86(12): 1655 - 1661. [Abstract] [Full Text]

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (20) Citing Articles via Google Scholar Google Scholar Articles by Valiente-Banuet, A. Articles by Dávila, P. Search for Related Content PubMed Articles by Valiente-Banuet, A.