Conservation Genetics 1: 57–66, 2000. © 2000 Kluwer Academic Publishers. Printed in the Netherlands.

57

Conservation genetics of the endangered conifer Fitzroya cupressoides in Chile and Argentina A.C. Premoli1∗ , T. Kitzberger1 & T.T. Veblen2 1 Centro

Regional Universitario Bariloche, Universidad Nacional del Comahue, Quintral 1250, 8400 Bariloche, Argentina; 2 Department of Geography, University of Colorado, Boulder, CO 80309-0260, USA; ∗ author for correspondence (E-mail: [email protected]) Received 22 December 1999; accepted 19 April 2000

Key words: gene pool, isozymes, Patagonia, protected areas, rare species

Abstract Intraspecific patterns of genetic variation can often be used to identify biogeographic divisions which can be especially useful in the design of conservation strategies. Although abundant empirical evidence exist on the genetic characteristics of plant species from the Northern Hemisphere as well as tropical endangered taxa, this information is particularly limited on threatened species from endemism-rich areas in the southern Andes of Argentina and Chile. The objective of the current study was to analyze the levels and distribution of the isozyme variation in Fitzroya cupressoides (Mol.) Johnst. (Cupressaceae), a rare conifer restricted to temperate rainforests of northern Patagonia, and to evaluate the role of current conservation areas protecting the gene pool of this valuable longlived conifer. Sampling schedules consisted of fresh foliage collected from 30 randomly selected trees at each of 24 different populations located along the geographic range of the species. Extraction of enzymes followed standard procedures and homogenates were loaded in 12% starch gels which were analyzed by horizontal electrophoresis. Eleven enzyme systems were resolved using a combination of four different buffer solutions which yielded information on 21 putative loci, 52% of them were polymorphic in at least one population. Relatively low levels of within-population genetic variability were scored in Fitzroya populations which were approximately half of the typical levels published for gymnosperms (percent of polymorphic loci, P = 23 vs. 53% and expected heterozygosity, HE = 0.077 vs. 0.155 for Fitzroya and other conifers respectively). Substantial between-population variation was detected, and certain individual populations stand out as much more genetically variable than nearby populations, which in turn are located outside protected areas. Our findings suggest that if the objective is to protect key species like Fitzroya, spatially explicit genetic information can be a useful tool to attain this goal.

Introduction Knowledge of distribution patterns of genetic variation in plants is important for informing conservation strategies. The theoretical basis of conservation genetics resides in the fact that preservation of genetic variability is essential to the maintenance of evolutionary potential of natural populations (Frankel and Soulé 1981). In recent years, the protection of genetic diversity within species has become a primary goal of biological conservation, and the use of genetic markers has been suggested as a valuable tool for achieving that goal, particularly of tree species endemic to

temperate forests in South America (Premoli 1998). However, some debate has arisen over the relative importance of ecological and genetic factors in survival of species and populations (Lande 1988; Falk and Holsinger 1991; Schmeske et al. 1994; Hamrick and Godt 1996). Although the persistence of most species over the short term depends upon demographic and environmental threats, genetic variability also needs to be considered in planning effective long-term conservation strategies (Mace et al. 1996). In the present study we analyze the amount and distribution of isozyme variation in Fitzroya cupressoides (alerce), a long-lived tree endemic to

58 temperate forests in southern South America. Isozymes are functionally similar forms of enzymes produced by different gene loci which together with allozymes (subsets of isozymes that represent different allelic alternatives of the same locus) are the most cost-effective method for investigating genetic phenomena at the molecular level (Murphy et al. 1996), and thus they can be used as genetic markers. Fitzroya has been cited as one of the world’s most spectacular examples of the serious impoverishment of forest genetic resources (Veblen et al. 1976). This impoverishment is due not only to the extirpation, or near extirpation of Fitzroya from entire habitats, but also to its slow recovery following logging. Seedling establishment is often scarce or nil after logging, and tree growth rates are exceptionally slow. Thus, after logging removes all mature trees from a site a long time must pass before a new generation of trees reaches sexual maturity. Even in forests unaffected by logging, coarse-scale disturbance events that permit abundant regeneration occur at intervals that typically are much longer than a century. Consequently, an inherently low rate of population turnover in this species makes it vulnerable to anthropogenic genetic impoverishment. This vulnerability, plus continued economic and political pressures to relax logging prohibitions (Lara et al. 1996), means that spatially explicit knowledge of the genetic variability of this species is required for effective planning of conservation strategies. For example, in Chile’s Central Depression only five small stands and eight sites of scattered trees remain from a forest type that formerly covered thousands of hectares (Fraver et al. 1999), and restoration efforts need to be guided by knowledge of how genetically similar these surviving trees are in relation to other potential sources of seed and cuttings. Fitzroya has been recently the subject of a genetic study by Allnutt et al. (1999) using random amplified polymorphic DNA (RAPD) variation, a technique that generates individual fingerprints via the polymerase chain reaction using short, random sequence primers. That study reported certain degree of amongpopulation genetic divergence although a limited number of populations (only 12) and individuals (a total of 89 samples) were analyzed. Thus, we will not only provide information on a wider range of populations using a different marker to be used in conservation efforts, but also further evidence to elucidate the type of polyploidy of alerce which is a tetraploid with 2n = 44 (Hair 1968). Isozyme electrophoresis is used

to distinguish allopolyploids from autopolyploids by the presence of fixed heterozygosity in the former (e.g. Watson et al. 1991) whereas polysomic segregation is expected in the latter (e.g. Soltis and Rieseberg 1986).

The species Fitzroya is a large long-lived conifer (Cupressaceae) of a monotypic genus that grows in Chile and Argentina. It can reach up to 5 m in diameter and 50 m in height (Veblen et al. 1976; Lara 1991). It is the second oldest living tree in the world after bristlecone pine (Pinus longaeva), and tree-ring chronologies exceeding 3600 years have been developed from trees sampled in Chile (Lara and Villalba 1993). It grows as discontinuous populations in the coastal cordillera and the central depression of Chile and on the western and eastern slopes of the Andes in both Chile and Argentina from ca. 39◦500 to 42◦ 450 S. Generally, it occurs at elevations from ca. 100 to 1200 m, typically on nutrient-poor soils, and where mean annual precipitation ranges from 2000 to over 4000 mm (Veblen et al. 1995). Fitzroya stands have often been described as consisting almost exclusively of large, old trees with scarce or no younger individuals (Veblen et al. 1995 and references therein). This type of stand structure led earlier workers to conclude that the species is a climatic relict not capable of reproducing under current climatic conditions (Kalela 1941; Tortorelli 1956; Schmithusen 1960). However, more recent studies have elucidated the regeneration strategy of this long-lived conifer (Veblen et al. 1995; T.T. Veblen, unpublished data). Fitzroya is highly shade-intolerant and competes poorly with other tree species on sites that are edaphically and climatically favorable. Many of the stands of mature Fitzroya lacking regeneration are remnant populations which established following coarse-scale disturbance by fire, landslides, flood deposition, or volcanic ash deposition (Veblen and Ashton 1982; Lara 1991; Veblen et al. 1995; Fraver et al. 1999; Lara et al. 1999). During long intervals free of such coarse-scale disturbance, there typically is little or no regeneration of Fitzroya at that site. Recent studies have documented adequate Fitzroya regeneration following natural disturbances in various habitats in both Chile and Argentina as long as sites are not logged or subjected to browsing by livestock (Veblen and Ashton 1982; Lara 1991; Donoso et al. 1993; T.T. Veblen, unpublished data).

59 Conservation status of Fitzroya Because of its valuable wood, Fitzroya has been heavily logged since the early European settlement in the 1500s, and logging has eliminated most populations from accessible areas. In many cases post-logging regeneration has been insufficient to maintain the populations (Veblen et al. 1976). Extensive logging of this valuable timber species has extirpated many local populations, and extant stands are restricted to remote areas of difficult access (Veblen et al. 1976, 1995; Lara et al. 1996). Due to the heavy exploitation of alerce, concern about its conservation status led Argentinean authorities to list Fitzroya since 1941 on the Annex to the Convention on Nature Protection and Wildlife Preservation in the Western Hemisphere. In 1969 Chile enacted some protective legislation requiring management measures to assure regeneration after logging (Corporación Nacional Forestal 1974) and in 1976 logging was prohibited (Ministerio de Agricultura 1976). In 1973, Fitzroya was placed on Appendix I of CITES, Convention on International Trade in Endangered Species of Wild Fauna and Flora (1984) which prohibits its commercialization and international trade. Fitzroya’s conservation status under CITES was relaxed in the mid-1980s, but in 1987 it was returned to Appendix I (Lara et al. 1991). In addition, Fitzroya is included under the US Endangered Species Act and the importation of its wood into the United States is prohibited (Threatened Species Conservation Act 1979). Since the late-1980s, there has been very little illegal exploitation of the relatively small Fitzroya populations in Argentina where well over 80% of Fitzroya forests occur within the protected area system (Kitzberger et al., in press). However, enforcement of the logging ban in the much larger Chilean populations, including extensive private holdings, has often been lax, and illegal cutting of Fitzroya continued into the 1990s (Lara et al. 1996). Fitzroya is listed as vulnerable on the International Union for the Conservation of Nature’s (IUCN) Red List (Farjon et al. 1993), which means that although it is considered endangered it is regarded as facing a high risk of extinction in the wild in the medium-term future (IUCN 1994). Fitzroya is protected in several national parks. In Chile, the most extensive of these parks is Alerce Andino National Park at ca. 43◦ S in the Andes, but large areas of Fitzroya forest are also privately owned. In Argentina, most of the extant population of Fitzroya is protected by location in one of three

National Parks (Nahuel Huapi, Lago Puelo, and Los Alerces). Much smaller areas of Fitzroya forests occur in provincial reserves and on privately owned land in Argentina (Kitzberger et al., in press). Although large-scale logging of Fitzroya has been stopped in both Chile and Argentina, small-scale illegal logging activities continue, especially in Chile (Lara et al. 1996). In this study we address the following questions: What are the levels of within-population isozyme variation in a tree species which has suffered dramatic range reduction such as Fitzroya? What is the relative importance to the conservation of the entire gene pool of the species of geographically isolated populations, especially those located outside protected areas? What are the levels of between-population differentiation that could be utilized in restoring degraded populations?

Materials and methods Twenty-four naturally occurring populations were sampled throughout the species’ range (Table 1). Populations 1 through 12 are in Argentina on the eastern slopes of the Andes, and populations 13 through 24 are in Chile on the western slopes of the Andes or the coastal cordillera (Figure 1, Table 1). These two groups are subsequently referred to as eastern and western populations. Sampling consisted of collecting fresh leaf material from 30 randomly selected individuals in each population. Approximately 20 cm of twig with needle tissue was collected from randomly scattered trees separated by a minimum of 10 m in order to get a representative sample of each stand. Given that alerce is known to reproduce vegetatively (Veblen and Ashton 1982; Lara 1991), special care was given at the time of leaf collection to avoid sampling of the same individual. In cases of small populations all the individuals in the stand were sampled (e.g. Coastal Cordillera Eastern), whereas more than one population was sampled from locations with relative extensive and continuous stands (e.g. Lago Menéndez). Labeled samples were transported in a portable cooler and in the laboratory stored at 0–5 ◦ C. Fresh foliage was grounded in mortars using liquid nitrogen, and enzymes were extracted with the extraction buffer of Mitton et al. (1979). Homogenates were frozen at –80 ◦ C until electrophoresis was performed. The tissue homogenate was absorbed onto Whatman No. 3 paper wicks,

60 Table 1. Genetic variability measures in populations of Fitzroya cupressoides; 1–12 western and 13–24 eastern populations. N = average number of individuals analyzed per locus and population; A = mean number of alleles per locus; P<0.95 = percent of polymorphic loci (criterion 0.95: a locus is considered polymorphic if the frequency of the most common allele does not exceed 0.95); HO = average observed heterozygosity; HE = average expected heterozygosity, unbiased estimate; CS – Conservation status, NP – National Park (LA: Los Alerces, NH: Nahuel Huapi, AA: Alerce Andino), NR – National Reserve (LL: Llanquihue, VA: Valdivia), PP – Private Property, SL – State Land, NM – Natural Monument (AC: Alerce Costero). Different letters indicate significant differences; P < 0.01, one-way ANOVA among groups of populations (eastern 1–12 and western 13–24) Population

N

A

P<0.95

HO (s.e.)

HE (s.e.)

CS

1. Chucao 2. Lago Men´endez North 3. Lago Men´endez South 4. Lago Esperanza Valley 5. Lago Esperanza High 6. Rio Tigre North 7. Rio Tigre South 8. EL Rayado 9. Laguna C´antaros 10. Puerto Blest 11. Lago Roca 12. Cord´on Serrucho 13. Carretera Austral 14. Cordillera Pelada High 15. Punta Estaquilla 16. Astillero 17. Cordillera Piuch´e, Chilo´e 18. Corral Mixed Forest 19. Llanquihue 20. Coastal Cordillera Eastern 21. Corral Sphagnum Bog 22. Alerce Andino Bog 23. Cordillera Pelada Bog 24. Cordilllera Pelada Piedra Indio

30.0 30.4 18.9 30.0 29.8 19.4 18.3 31.0 30.0 29.9 59.5 30.8 29.7 29.3 19.8 30.7 30.0 29.0 29.7 7.8 29.0 27.1 30.8 30.8

1.6 1.7 1.6 1.6 1.6 1.7 1.6 1.4 1.5 1.3 1.7 1.6 1.2 1.3 1.7 1.6 1.2 1.4 1.2 1.3 1.6 1.3 1.2 1.4

33.3 19.0 19.0 42.9 28.6 42.9 38.1 19.0 23.8 19.0 14.3 14.3 9.5 14.3 47.6 28.6 9.5 19.0 14.3 19.0 28.6 19.0 9.5 23.8

0.100 (0.042) 0.093 (0.042) 0.088 (0.042) 0.102 (0.035) 0.092 (0.040) 0.143 (0.048) 0.113 (0.040) 0.085 (0.041) 0.061 (0.031) 0.069 (0.041) 0.074 (0.035) 0.058 (0.030) 0.039 (0.034) 0.038 (0.023) 0.137 (0.042) 0.093 (0.034) 0.016 (0.011) 0.062 (0.040) 0.049 (0.030) 0.036 (0.025) 0.065 (0.030) 0.073 (0.037) 0.029 (0.019) 0.053 (0.028)

0.104 (0.040) 0.090 (0.039) 0.079 (0.037) 0.104 (0.033) 0.091 (0.037) 0.125 (0.037) 0.119 (0.043) 0.090 (0.044) 0.071 (0.033) 0.069 (0.035) 0.085 (0.036) 0.064 (0.033) 0.037 (0.027) 0.041 (0.023) 0.131 (0.037) 0.102 (0.036) 0.032 (0.026) 0.053 (0.030) 0.055 (0.033) 0.064 (0.037) 0.070 (0.029) 0.064 (0.032) 0.033 (0.020) 0.079 (0.035)

NPLA NPLA NPLA PP PP SL SL SL NPNH NPNH NPNH SL SL NMAC PP PP SL PP NRLL PP NRVA NPAA NMAC NMAC

Species average Eastern populations average Western populations average

28.4 29.8 27.0

1.5 1.6a 1.4b

23.2 26.2a 20.2a

0.074 (0.006) 0.090a (0.006) 0.057b (0.009)

0.077 (0.006) 0.091a (0.005) 0.063b (0.008)

which were inserted into 12% weight to volume ratio starch gels. Horizontal electrophoresis was conducted using four combinations of gel and electrobuffers. Eleven enzyme systems involving twenty-one putative genetic loci were resolved and consistently scorable. Aconitase (Aco), Isocitrate dehydrogenase (Idh), Malic enzyme (Me1, Me2), and 6-Phosphogluconate dehydrogenase (6Pgd1, 6Pgd2) were assayed using the morpholine-citrate buffer by Ranker et al. (1989); Alcohol dehydrogenase (Adh) and Menadione reductase (Mnr1, Mnr2, Mnr3) were resolved on the Tris-citrate buffer by Poulik (1957);

Malate dehydrogenase (Mdh1, Mdh2), Phosphoglucomutase (Pgm1, Pgm2), and Shikimate dehydrogenase (Skdh1, Skdh2, Skdh3) were assayed using the Histidine-tris buffer by King and Dancik (1983); and Peroxidase (Per1, Per2) and Phosphoglucoisomerase (Pgi1, Pgi2) were resolved on the Tris-borate buffer by O’Malley et al. (1979). For each enzyme system, the loci and alleles were numbered sequentially from the most anodal to the most cathodal. The genetic basis of allozyme patterns was inferred from known subunit compositions and number of isozymes commonly observed in plant species (Soltis and

61

Figure 1. Location of sampled populations of Fitzroya cupressoides. See Table 1 for site names.

Soltis 1989; Murphy et al. 1996). Given that no breeding experiments using individuals with known isozyme phenotypes or progeny tests were performed, genotypes were inferred directly from electromorphs, thus the analyzed genetic loci are considered putative. The following genetic parameters were calculated for each population: percent of polymorphic loci (0.95 criterion: a locus is considered polymorphic if the frequency of the most common allele does not exceed 0.95, P<0.95 ), observed and expected heterozygosity (HO and HE ), and mean number of alleles per

locus (A). Observed and expected heterozygosities for each polymorphic locus were compared by calculating Wright’s fixation index (where F = 1 − HO /HE ) to determine deviations from random-mating expectations. Deviations of F from zero were tested using chi-square tests. Variation among populations for each group of populations was estimated with genetic diversity statistics (Nei 1973). Total genetic diversity (HT ) and mean diversity within populations (HS ) were calculated from polymorphic loci. The proportion of genetic diversity residing among populations (GST ) was determined by

62

GST = (HT − HS )/HT for each locus, and were averaged over all polymorphic loci.

Results and discussion Levels and distribution of genetic variation Eleven of the 21 resolved putative loci (52%) were polymorphic (0.95 criterion) in at least one population; these were: Aco, Idh, Mdh2, Me1, Me2, Per1, Per2, Pgi2, Pgm1, 6Pgd1, and Skdh1. Relatively low levels of within-population isozyme variability were scored in Fitzroya populations. Polymorphism (P<0.95 ) was 23%, observed and expected heterozygosity (HO and HE ) were 0.074 and 0.077, respectively, and on average 1.5 alleles (A) were scored per putative locus (Table 1). These values were approximately half of the typical levels published for gymnosperms, particularly in terms of the percent of polymorphic loci and the expected heterozygosity (P = 53.4% and HE = 0.151 respectively; Hamrick et al. 1992). However, alerce’s levels of isozyme variability were similar to those reported for long-lived woody species with restricted geographic range, which were called endemics by Hamrick et al. (1992). They had on average 1.5 alleles per locus, 26% polymorphism, and 0.056 expected heterozygosity (Hamrick et al. 1992). Fixation indices were significantly different from zero in 34 of 167 (20%) possible estimates, 85% of which (25 of 34) yielded a positive deviation from zero and thus indicated heterozygote deficiencies at these loci. This homozygous excess maybe the result of the applied sampling schedule, resulting in the collection of nearby individuals that were genetically similar. Overall, eastern populations were more variable than western ones; they had significantly greater mean number of alleles per locus and heterozygosity (Table 1). Observed and expected heterozygosity of eastern populations (HO = 0.090 and HE = 0.091) were 40% greater than those measured in western populations (HO = 0.057 and HE = 0.063) and those differences were significant (P < 0.01, one-way ANOVA among groups of populations). Substantial heterogeneity was found in the levels of genetic variation of different populations, detecting hot spots of genetic diversity for both eastern

Figure 2. Total genetic diversity in populations of Fitzroya. Location of analyzed populations are shown by dark circles.

and western populations (Figure 2). Several eastern populations located towards the southern limit of the species in Argentina, such as the Rio Tigre and Lago Esperanza populations, concentrated most of the genetic diversity. These populations had 60% higher polymorphism and about 20% greater heterozygosity (HO and HE ) than populations nearby such as those in Los Alerces National Park (Chucao and Lago Menéndez). Among western populations, most genetic variation was concentrated in the single population of Punta Estaquilla, a coastal population at the southern limit of the continental distribution in Chile. Indicators of genetic variability in this population such as the percent of polymorphic loci and heterozygosity (observed and expected) were more than two-fold greater than the average values for western populations. A nearby population (Astillero), also had significantly greater heterozygosity confirming the existence of the hot spot. Most other western populations had evenly reduced levels of genetic variability (Table 1). Our data supports RAPD information in which some degree of divergence was detected between western and eastern populations. In addition, population at Punta Estaquilla emerges as highly variable (Shannon’s diversity measure, S = 0.611 compared to an average over 12 Fitzroya populations of Spop = 0.547) and Astillero arises as genetically distinct from all the analyzed populations based on an UPGMA dendrogram that analyzed relationships between populations (Allnutt et al. 1999). Other studies on Fitzroya showed some degree of amongpopulation variability based on foliage terpenes of populations analyzed in Chile. Although populations were quite similar to each other based on monoter-

63 Table 2. Comparison of mean genetic diversity in its components total (HT ), within (HS ), and among (GST ) populations of F. cupressoides with other conifers. Values are based on polmorphic loci only (13 for Fitzroya). Averages were generated on 24 populations of Fitzroya. Standard errors are in parentheses Group of populations

HT

HS

GST

Reference

Fitzroya cupressoides

0.132 (0.049)

0.122 (0.044)

0.079 (0.017)

This study

Gymnosperms

0.281 (0.010)

0.255 (0.010)

0.073 (0.010)

Hamrick et al. (1992)

Table 3. Comparison of within-population levels of isozyme variation in the three species of the Cupressaceae endemic to temperate forests of South America. Data on other gymnosperms are also provided. Npops, N, and N loci represent total number of analyzed populations, individuals, and loci, respectively; A – mean number of alleles/locus; P∗ – percent of polymorphic loci in at least one population under the 0.95 criterion; HO – observed heterozygosity Species

Npops

Austrocedrus chilensis Austrocedrus chilensis Fitzroya cupressoides Pilgerodendron uviferum

1 15 24 20

Other gymnosperms

8.9

N 120 245 682 598 ∼200

N loci

A

P∗

HO

Reference

12 13 21 14

1.75 2.08 1.50 1.19

41 77 52 78

0.071 0.138 0.074 0.024

Ferreyra et al. (1996) Calculated from Pastorino and Gallo (1998) This study Premoli, unpublished data

17.3

1.83

71

0.151

Hamrick et al. (1992), Review 213 species

pene data, some geographic heterogeneity was found among different populations in their sesquiterpene profiles (Cool et al. 1991). Analyses of genetic diversity using polymorphic loci (13 out of the 21 total loci) indicated that total genetic diversity of Fitzroya (HT = 0.132) was half of that found in other conifers or woody endemics (HT = 0.281 and 0.267 respectively; Hamrick et al. 1992). However, as in other conifers, most of the total genetic diversity of Fitzroya is found within populations (92%) with only 8% distributed on average among different populations (Table 2). These results are consistent with those reported using RAPDs, with most of the observed variation (85%) found in the within-population component (Allnutt et al. 1999). Isozyme information on the other two endemic species of Cupressaceae (Austrocedrus chilensis (Don) Flor. et Boutleje and Pilgerodendron uviferum (Don) Flor. both are monospecific genera, such as Fitzroya) from southern Chile and Argentina, allows meaningful among-species comparisons on the levels of genetic variability. Austrocedrus and Pilgerodendron are more widespread than Fitzroya; they occur over a latitudinal range of 10 and 15◦ respectively, whereas

the overall latitudinal range of Fitzroya is less than 4◦ (Pilgerodendron occurs from 39◦ 300 to 55◦ 300 S whereas Austrocedrus is found from 34◦ 450 to 44◦ in Chile and from 36◦ 300 to 43◦ 350 latitude S in Argentina; Veblen et al. 1995). Previous studies on Austrocedrus yielded either similar (Ferreyra et al. 1996) or greater (Pastorino and Gallo 1998) levels of allozyme variability than those found for Fitzroya (Table 3). Moreover, results from an isozyme survey on 18 populations indicate that Pilgerodendron, has less genetic variation than the more geographically restricted Fitzroya or Autrocedrus (A.C. Premoli, unpublished data; Table 3). As a result, these preliminary data do not support the generalization that species with greater geographical ranges tend to have higher levels of genetic variability (Hamrick and Godt 1989; Hamrick et al. 1992). The explanation for that generalization, is that greater gene flow in more widespread plant species with historically larger population sizes favors genetic diversity in comparison with small isolated populations in which genetic drift may erode diversity. Although Pilgerodendron has an overall greater geographic range, it often consists of smaller and relatively more scattered populations than

64 either Fitzroya or Autrocedrus. Thus, factors other than geographic range need to be considered when predicting the levels of genetic variation of rare woody species endemic to the southern Andes, such as population size, life-history characteristics, habitat requirements as well as present and past distributions of suitable habitats. For example, the genetic distinction and greater diversity of eastern populations of Fitzroya appears to reflect the location of forest refugia on the eastern side of the Andes during the last Pleistocene glaciation (Premoli 1998; Premoli et al., in press). Furthemore, the levels of isozyme variability found in Fitzroya may also be due to its ploidy level which in turn correlates with individual genetic heterozygosity (e.g. Soltis and Rieseberg 1986). Fitzroya is a tetraploid with 44 chromosomes whereas Austrocedrus and Pilgerodendron are both diploids with 22 chromosomes (Hunziker 1961; De Azkue 1982) and thus possible heterozygotes in Fitzroya (AAAa, Aaaa, Aaaa) exceed those expected in Pilgerodendron (Aa only). In addition, the isozyme diversity of Fitzroya could be a result of new mutations, or novel intra-genic recombinations that may accumulate in polyploids because of the presence of duplicated genes that buffer the effects of deleterious recessive mutations. Banding patterns found for the analyzed isozymes showed evidence of the tetrasomic inheritance patterns indicative of autotetraploids (A.C. Premoli, unpublished data). Both balanced and unbalanced heterozygotes (corresponding to equal and unequal dosages of different alleles) were present. No evidence of fixed heterozygotes was scored for any of the analyzed enzymes as might be seen in allotetraploids. Thus, the evidence found in this study suggest Fitzroya as an autopolyploid. Implications for the conservation of Fitzroya Greater levels of genetic variability in eastern populations measured by higher number of alleles per locus, degree of polymorphism, and heterozygosity highlights the importance of these populations as a reservoir of genetic diversity. This is particularly important because of the greater conservation attention and funding naturally devoted to the much larger Fitzroya populations in Chile. The lack of accurate information on the geographic distribution of rare South American temperate conifers (IFS 1998) has also impeded the identification of isolated populations that could be genetically distinct from the main populations and generally hindered overall

evaluation of conservation strategies. In conjunction with the present study we have produced the first detailed map (scale 1:100.000) of Fitzroya in Argentina showing a greater number and extent of stands than previously believed (Kitzberger et al. in press). In Argentina, Fitzroya populations are distributed from 40◦ 570 4500 to 42◦ 450 2700 S and they occupy approximately 20,625 ha, 85% of which are within protected areas (Kitzberger et al., in press). This information is consistent to that produced by a recent vegetation mapping (scale 1:500.000) of key species, including Fitzroya, in the Valdivian Rainforest Eco-Region, as part of the Global 200 iniciative. Whereas most of Fitzroya forests occur in Chile (280.137 ha representing 96% of the total extent of Fitzroya forests) only 9% of them (25.624 ha) are under protection within the ecoregion. In contrast, 12.246 ha are found in Argentina where 82% (9.997 ha) are protected (Fundación Vida Silvestre 1999). Conserving species populations (sensu Rojas 1992) ultimately is aimed at the maintenance of a species’ ability to respond to new evolutionary challenges under the face of changing environments. This principle deserves particular attention in assessing the conservation status of Fitzroya. Although Fitzroya is internationally protected under the CITES and some of its populations are included in protected areas in Argentina and Chile, the results of the present study indicate that representation of the species’ gene pool could be substantially improved through inclusion of additional populations. Fitzroya populations with the greatest genetic diversity are not included in protected areas in either Chile or Argentina. Eastern populations located to the north and outside of Los Alerces National Park (Río Tigre and Lago Esperanza) deserve particular attention. Although they had the highest observed and expected heterozygosity and were among the populations with greatest polymorphism (Table 1) they are at present outside protected areas. Fitzroya populations at the southern distributional limits on both slopes of the Andes as well as coastal populations in Chile may have been the locations of Plesitocene forest refugia from which other populations originated (Premoli 1998; Allnutt et al. 1999). These more ancient gene pools may deserve high priority in conservation planning. Urgent attention should be also given to genetically rich coastal populations in Chile located outside of conservation units where 45% of the native forest has been replaced by plantations of exotic trees (Lara and Veblen 1993). In southern Chile the co-occurrence of centers of high

65 genetic diversity of key species like Fitzroya with areas of rapid deforestation and different land owners poses a particular challenge to the preservation of the gene pool of unique species and the conservation of biodiversity as suggested by Armesto et al. (1999). Intraspecific variation has increasingly been accepted as a focus for conservation and molecular approaches could be of value to conservation efforts by providing a tool for measuring genetic diversity. Genetic information potentially relevant in practical conservation management has been provided by other case studies in plants (Rieseberg et al. 1989; Ricci and Eaton 1997; Hogbin and Peakall 1999). The detection of centers of genetic variability could be particularly useful for the design of conservation strategies in long-lived species such as Fitzroya on which experimental studies are often difficult. Given the genetic divergence between western and eastern populations, local genetic stocks should be used for ex situ conservation and for restoration programs currently being developed (SUCRE 1998; Fraver et al. 1999) and which are urgently needed, particularly in the central depression and coastal Chile. The current system of protected areas in southern Chile has recently been assessed as inadequate in protecting biodiversity (Armesto et al. 1999). Simple land area under protection was judged as an ineffective measure of biodiversity protection, and, instead it was recommended that areas of high biodiversity and endemism be targeted in revising national systems of nature protection. Our findings also show that geographical patterns of genetic variation within key species such as Fitzroya need to be explicitly considered in conservation strategies, which in combination with ecological and demographic information as suggested by Schemske et al. (1994) will efficiently contribute to the long-term preservation of such unique taxa.

Acknowledgements Funding was provided by a National Geographic Society, grant No. 5620/96. Field work was facilitated by Administración de Parques Nacionales (Argentina), Corporación Nacional Forestal (Chile), Servicio Forestal Andino (Bolsón, Río Negro), and in particular by C. Martin, E. Koninenberg, and F. Mendoza Escalas. For research assistance and sample collection we thank C. Aravena, C. Brewer, C. Chehebar, L. Davy, S. Fraver, F. Kitzberger, D. Lorenz, A. Newton,

F. Premoli, and P. Soares. For laboratory assistance we are grateful to C. Souto and M. Caldiz. We thank A. Kremer and two anonymous reviewers for insightful comments and suggestions. A.C. Premeli and T. Kitzberger were supported by CONICET and Universidad del Comahue.

References Allnutt TR, Newton AC, Lara A, Premoli AC, Armesto JJ, Vergara R, Gardner M (1999) Genetic variation in Fitzroya cupressoides (alerce) a threatened South America conifer. Mol. Ecol., 8, 975– 987. Armesto JJ, Rozzi R, Smith-Ramírez C, Arroyo MT (1999) Conservation targets in South American temperate forests. Science, 282, 127–128. Convention on International Trade in Endangered Species of Wild Fauna and Flora (1984) Publication 50 CFR 23.23. Department of the Interior, U.S. Fish and Wildlife Service. Cool LG, Power AB, Zavarin E (1991) Variability of foliage terpenes of Fitzroya cupressoides. Biochem. Sys. Ecol., 19, 421–432. Corporación Nacional Forestal (1974) Legislación Forestal. Corporación Nacional Forestal, Santiago, Chile. De Azkue D (1982) Los cromosomas de Pilgerodendron uviferum (Don) Flor. Darwiniana, 24, 19–22. Donoso C, Sandoval V, Grez R, Rodríguez J (1993) Dynamics of Fitzroya cupressoides forests in southern Chile. J. Veg. Sci., 4, 303–312. Falk DA, Holsinger KE (1991) Genetics and Conservation of Rare Plants. Oxford University Press, Oxford, UK. Farjon A, Page CN, Schellevis N (1993) A preliminary world list of threatened conifer species. Biodiv. Cons., 2, 304–326. Ferreyra LI, Latino A, Calderón A, Cardenal CN (1996) Allozyme polymorphism in Austrocedrus chilensis (D. Don) Florin et Boutelje from Patagonia, Argentina. Silvae Genet., 45, 61–64. Frankel OH, Soulé ME (1981) Conservation and Evolution. Cambridge University Press, Cambridge, UK. Fraver S, Gonzalez ME, Silla F, Lara A (1999) Composition and structure of remnant Fitzroya cupressoides forests of southern Chile’s Central Depression. J. Torr. Bot. Soc., 126, 49–57. Fundación Vida Silvestre (1999) Mapeo de la Eco-Región de los Bosques Valdivianos. Boletín Técnico, 51. Buenos Aires. Hair, JB (1968) The chromosomes of the Cupressaceae. 1. Tetraclineae and Actinostrobeae (Callitroideae). N. Z. J. Bot., 6, 277–284. Hamrick JL, Godt MJW (1989) Allozyme diversity in plant species. In: Plant Population Genetics, Breeding, and Genetic Resources (eds. Brown AHD, Clegg MT, Kahler AL, Weir BS), pp. 43–63. Sinauer Associates, Sunderland, MA. Hamrick JL, Godt MJW (1996) Conservation genetics of endemic plant species. In: Conservation Genetics, Case Histories from Nature (eds. Avise JC, Hamrick JL), pp. 281–304. Chapman & Hall, New York. Hamrick JL, Godt MJW, Sherman-Broyles SL (1992) Factors influencing levels of genetic diversity in woody plant species. New Forest, 6, 95–124. Hogbin PM, Peakall R (1999) Evaluation of the contribution of genetic research to the management of the endangered plant Zieria prostrata. Cons. Biol., 13, 514–522. Hunziker JH (1961) Estudios cromosómicos en Cupresus y Libocedrus (Cupressaceae). Rev. Invest. Agríc., 15, 169–185.

66 IFS, International Foundation for Science (1998) Recent Advances in biotechnology for tree conservation and Management, Proceedings of an International Workshop. Universidade Federal de Santa Catarina. Florianópolis, Santa Catarina, Brasil. IUCN (1994) Red List Categories, IUCN Species Survival Commission. IUCN, Gland, Switzerland. Kalela EK (1941) Über die Holzarten und die durch die klimatischen Verhältnisse verursachten Holzartenwechsel in den Wäldern Ostpatagoniens. Annales Academiae Scientarium Fennicae, Series A, 2, 5–151. King JN, Dancik BP (1983) Inheritance and linkage of isozymes in white spruce (Picea glauca). Can. J. Genet.Cytol., 5, 430–436. Kitzberger T, Pérez A, Iglesias G, Premoli AC, Veblen TT. Distribución y estado de conservación del alerce [Fitzroya cupressoides (Mol. Johnst)] en Argentina. Bosque, in press. Lande R (1988) Genetics and demography in biological conservation. Science, 241, 1455–1460. Lara A (1991) The Dynamics and Disturbance Regimes of Fitzroya Cupressoides Forests in the South-central Andes of Chile. PhD thesis, University of Colorado, Boulder, USA. Lara A, Veblen TT (1993) Forest plantations in Chile: A successful model? In: Afforestation Policies, Planing, and Progress (ed. Mather A), pp. 118–139. Belhaven Press, London. Lara A, Villalba R (1993) A 3620-year temperatyre record from Fitzroya cupressoides tree rings in southern South America. Science, 260, 1104–1106. Lara A, Donoso PJ, Cortés M (1991) Development of Conservation and Management Alternatives for Native Forest in South-central Chile. Final Report Project 3181 WWF-US/CODEF, Santiago, Chile. Lara A, Donoso C, Aravena JC (1996) La conservación del bosque nativo en Chile: problemas y desafíos. In: Ecología de los bosques nativos de Chile (eds. Armesto JJ, Villagrán C, Arroyo MK), pp. 335–362. Editorial Universitaria, Santiago, Chile. Lara A, Fraver S, Aravena JC, Wolodarsky-Franke A (1999) Fire and the dynamics of Fitzroya cupressoides (alerce) forests of Chile’s Cordillera Pelada. Ecoscience, 6, 100–109. Mace GM, Smith TB, Bruford MW, Wayne RK (1996) An overview of the issues. In: Molecular Genetic Approaches in Conservation (eds. Smith TB, Wayne RK), pp. 3–24. Oxford University Press, New York. Ministerio de Agricultura (1976) Decreto Supremo 490 que declara Monumento Natural al Alerce. Santiago, Chile. Mitton JB, Linhart YB, Sturgeon KB, Hamrick JL (1979) Allozyme polymorphisms detected in mature needle tissue of ponderosa pine. J. Heredity, 70, 86–89. Murphy RW, Sites Jr. JW, Buth DG, Haufler CH (1996) Proteins: Isozyme electrophoresis. In: Molecular Systematics (eds. Hillis DM, Moritz C, Mable BK), pp. 51–120. Sinauer Associates, Sunderland, MA. Nei M (1973) Analysis of gene diversity in subdivided populations. Proc. Natl. Acad. USA, 70, 3321–3323. O’Malley DM, Allendorf FW, Blake GM (1979) Inheritance of isozyme variation and heterozygosity in Pinus ponderosa. Biochem. Genet., 17, 233–250. Pastorino MJ, Gallo LA (1998) Inheritance of isozyme variants in Austrocedrus chilensis (D. Don) Florin et Boutelje. Silvae Genet., 47, 15–20. Poulik MD (1957) Starch gel electrophoresis in a discontinuous system of buffers. Nature (London), 180, 1477–1479.

Premoli AC (1998) The use of genetic markers to conserve endangered species and to design protected areas of more widespread species. In: Proceedings of an International Workshop: Recent Advances in Biotechnology for Tree Conservation and Management (ed. International Foundation for Science), pp. 157–171. Universidade Federal de Santa Catarina, Santa Catarina, Brasil. Premoli AC, Kitzberger TT, Veblen TT. Isozyme variation and recent biogeographical history of the long-lived conifer Fitzroya cupressoides. J. Biogeogr., in press. Ranker TA, Haufler CH, Soltis PS, Soltis DE (1989) Genetic evidence for allopolyploidy in the neotropical fern Hemionitis (Adiantaceae) and the reconstruction of an ancestral genome. Sys. Bot., 14, 439–447. Ricci M, Eaton L (1997) Do all existing Sophora toromiro descend from one individual? Biodiv. Cons., 6, 1697–1702. Rieseberg LH, Zona S, Aberbom L, Martin TD (1989) Hybridization in the island endemic, Catalina Mahogany. Cons. Biol., 3, 52–58. Rojas M (1992) The species problem and conservation: what are we protecting? Cons. Biol., 6, 170–178. Schemske DW, Husband BC, Ruckelhaus MH, Goowillie C, Parker IM, Sishop JG (1994) Evaluating approaches to the conservation of rare and endangered plants. Ecology, 75, 584–606. Schmithüsen J (1960) Die Nadelhölzer in den Waldgesellschaften der südlichen Anden. Vegetatio, 9, 313–327. Soltis DE, Rieseberg LH (1986) Autopolyploidy in Tolmiea menziesii (Saxifragaceae): genetic insights from enzyme electrophoresis. Am. J. Bot., 73, 310–318. Soltis DE, Soltis PS (1989) Isozymes in Plant Biology. Discorides Press, Portland, Oregon. SUCRE: Sustainable use, conservation and restoration of native forests in southern Mexico and south-Central Chile (1998) In: INCO-DC International Cooperation with Developing Countries: Practical Information and Programmes, Funded Projects, Agriculture and Natural Resources (ed. European Comission), pp. 182. Belgium. Available at http://helios.bto. ed.ac.uk/ierm/research/sucre/sucre.htm. Threatened Species Conservation Act (1979) Determination that Fitzroya cupressoides is a threatened species. Federal Register, 44, 64730–64733. Tortorelli LA (1956) Maderas y Bosques Argentinos. Editorial Acme, Buenos Aires, Argentina. Veblen TT, Ashton DH (1982) The regeneration status of Fitzroya cupressoides in the Cordillera Pelada, Chile. Biol. Cons., 23, 141–161. Veblen TT, Delamstro RN, Schlatter J (1976) The conservation of Fitzroya cupressoides and its environment in southern Chile. Environ. Cons., 3, 291–301. Veblen TT, Burns BR, Kitzberger T, Lara A, Villalba R (1995) The ecology of the conifers of southern South America. In: Ecology of Southern Conifers (eds. Enright NJ, Hill RS), pp. 120–155. Melbourne University Press, Carlton, Victoria, Australia. Watson L.E, Elisens WJ, Estes JR (1991) Electrophoretic and cytogenetic evidence for allopolyploid origin of Marshallia mohrii (Asteraceae). Am. J. Bot., 78, 408–416.