Aim Cycads constitute an ancient plant group that is generally believed to disperse poorly. However, one group of cycads (subsection Rumphiae) is thought to have dispersed relatively recently from a Malesian source area westwards to East Africa and eastwards into the Pacific, using a floatation‐facilitating layer in their seeds. We use morphological and allozyme characters to investigate the relationships among the species within this group and to deduce whether the wide distribution was achieved by recent dispersal (as evidenced by high genetic similarity) or more distant vicariance events (high genetic differentiation). Location We examined specimens collected throughout the range of subsection Rumphiae, from East Africa through Southeast Asia to Tonga in the South‐west Pacific. Methods We investigated relationships within subsection Rumphiae of the genus Cycas by analysing 18 variable (11 informative) morphological characters and 22 allozyme loci for seven of the 10 species currently assigned to this taxon. Results Distinctive morphological characters are few and fail to resolve relationships within the group. Allozyme data show that species within this subsection are closely related and suggest that there are two groups within the subsection, one comprising Cycas thouarsii (East Africa) and C. edentata (the Philippines), and the other the remaining species (from Malesia and the Pacific). The Australian species C. silvestris is sister to subsection Rumphiae in the morphological analysis but is closely allied to C. rumphii (nested within the subsection) in the allozyme analysis, suggesting that Rumphiae may be paraphyletic and that characters thought to be taxonomically important may need to be re‐evaluated. Main conclusions Cycads within subsection Rumphiae are closely related, and the wide distribution of this group was probably achieved through relatively recent oceanic dispersal events. Separate events probably account for the dispersal of these cycads into the Pacific and to Africa. The origin and distribution of C. silvestris (Australia) could be explained by a dispersal event from New Guinea or may have resulted from a former land connection between Australia and New Guinea.

Aim Cycads constitute an ancient plant group that is generally believed to disperse poorly. However, one group of cycads (subsection Rumphiae) is thought to have dispersed relatively recently from a Malesian source area westwards to East Africa and eastwards into the Pacific, using a floatation-facilitating layer in their seeds. We use morphological and allozyme characters to investigate the relationships among the species within this group and to deduce whether the wide distribution was achieved by recent dispersal (as evidenced by high genetic similarity) or more distant vicariance events (high genetic differentiation). Location We examined specimens collected throughout the range of subsection Rumphiae, from East Africa through Southeast Asia to Tonga in the South-west Pacific. Methods We investigated relationships within subsection Rumphiae of the genus Cycas by analysing 18 variable (11 informative) morphological characters and 22 allozyme loci for seven of the 10 species currently assigned to this taxon. Results Distinctive morphological characters are few and fail to resolve relationships within the group. Allozyme data show that species within this subsection are closely related and suggest that there are two groups within the subsection, one comprising Cycas thouarsii (East Africa) and C. edentata (the Philippines), and the other the remaining species (from Malesia and the Pacific). The Australian species C. silvestris is sister to subsection Rumphiae in the morphological analysis but is closely allied to C. rumphii (nested within the subsection) in the allozyme analysis, suggesting that Rumphiae may be paraphyletic and that characters thought to be taxonomically important may need to be re-evaluated. Main conclusions Cycads within subsection Rumphiae are closely related, and the wide distribution of this group was probably achieved through relatively recent oceanic dispersal events. Separate events probably account for the dispersal of these cycads into the Pacific and to Africa. The origin and distribution of C. silvestris (Australia) could be explained by a dispersal event from New Guinea or may have resulted from a former land connection between Australia and New Guinea.

Genetic diversity and genetic structure in Engelmann spruce (Picea engelmannii Parry ex Engelm.) were interpreted with respect to the effects of glacial and interglacial displacement and compared with patterns in blue spruce (Picea pungens Engelm.), which occupies a range well south of the last glacial front. On average, Engelmann spruce populations were polymorphic at 80% of 24 isozyme loci, with 2.4 alleles per locus and expected heterozygosity of 0.255. The respective means for four populations of blue spruce were 42.7%, 1.6 alleles, and 0.138. Of total diversity, 14.7% was among populations in Engelmann spruce and 8.6% in blue spruce. In Engelmann spruce, numbers of alleles increased from south to north. Recent bottlenecks were indicated in the three southernmost populations of Engelmann spruce and in the northernmost population of blue spruce. Cluster analysis divided Engelmann spruce into a northern group and a Southwestern group; however, populations from Utah were distributed among both clusters. Genetic distance was correlated with geographic distance between northern populations but not between Southwestern populations, where drift predominated over gene flow. In two Engelmann spruce populations from Utah, multilocus outcrossing rates were 0.951 (±0.016) and 0.940 (±0.071). At Flys Peak, Arizona, the southernmost location of Engelmann spruce, outcrossing was also high, 0.899 (±0.017). Thus, inbreeding coefficients observed for parental (–0.104 to 0.047) and filial (0.011 to 0.026) generations were low. Although Engelmann spruce seemed genetically robust, the evidence of bottlenecks in the southern extreme of its range suggested future problems in an era of global warming.

Santa Lucia fir ( Abies bracteata ), is a unique fir, the sole member of the subgenus Pseudotorreya . It is a relict of the Miocene broadleaved evergreen sclerophyll forest, and is now restricted to a highly fragmented range in the Santa Lucia Mountains of central coastal California. Expected heterozygosity for 30 isozyme loci in 18 enzyme systems, averaged over six populations that spanned the species’ north–south range, was only 0.036. Despite a fragmented range and isolated populations, differentiation ( F ST ) was only 0.080 for mature trees, and the number of migrants per generation ( Nm ) was 2.88 or 3.83, depending on the method of estimation. F ST for embryos was lower, 0.025, and Nm correspondingly higher, 9.75. Nei’s genetic distances were small and unrelated to geographic distances between populations. The proportion of full seeds per cone was only 0.082–0.488, depending on population, which suggests a high incidence of selfing followed by embryo abortion. However, the level of accumulated inbreeding, F IS , in mature trees was low, only 0.049. By contrast, F IS for embryos was 0.388, which indicates a high proportion of selfed progeny, in agreement with the low seed yields. The difference in inbreeding coefficients between seed trees and their progeny suggest that most inbreds are eliminated before maturity and, therefore, seed production, already low, overestimates the true potential for regeneration of these populations. These results have implications for conservation.

lInstitute of Forest Genetics, Pacific Southwest Research Station, USDA Forest Service, 2480 Carson Road, Placerville, California 95667 USA; 3Departamento Forestal, Universidad Aut6noma Agraria Antonio Narro, Buenavista, Saltillo, Coahuila, MBxico; 4Service de CdnBtique et d'AmBlioration des Arbres Forestieres, Centre National de la Recherche Forestiere, Charia Omar Ibn El Khattab, B.P. 763 Rabat, Morocco; 5Programa Foresral, Institute de Recursos Naturales, Colegio de Postgraduados en Ciencias Agricolas, Montecillo, MBxico, C.P. 56230, MBxico; and Tentro de GenBtica Forestal, Universidad Aut6noma Chapingo, Apartado Postal No. 37, Chapingo, MBxico, C.P. 56230, MBxico

Brewer spruce (Picea breweriana), a relict of the widespread Arcto-Tertiary forests, is now restricted to a highly fragmented range in the Klamath Region of California and Oregon. Expected heterozygosity for 26 isozyme loci, averaged over 10 populations, was 0.121. More notable than the relatively high level of diversity when compared to other woody endemics was the strong decrease in expected heterozygosity with latitude. Differentiation (F(ST)) was 0.152, higher than values for many north temperate conifers with larger distributions. The number of migrants per generation (Nm) was 1.34 or 2.70, depending on the method of estimation. Inbreeding appeared low; F(IS) was only 0.003, in agreement with multilocus population outcrossing rates (t(m)), which were generally well above 0.90. No difference in t(m) was found between isolated vs. clustered trees. However, the number of seeds per cone was greatest in the densest populations; t(m) is a measure of effective outcrossing after mortality in the embryonic stage, whereas a reduced number of seeds per cone indicates self pollination. Selfing increased after logging; outcrossing rate before logging was 0.961 and after logging, 0.756. Despite Brewer spruce's narrow, fragmented distribution, the outlook for its conservation was good, with the exception of possible negative effects of logging.

Numerous populations from six spruce taxa, including four relict endemics, Picea chihuahuana (Chihuahua spruce), P. martinezii (Martinez spruce), P. mexicana (Mexican spruce), and P. breweriana (Brewer spruce), and two widespread species, P. engelmannii (Engelmann spruce) and P. pungens (blue spruce), were compared at homologous isozyme loci to test various hypotheses about their affinities and origins. Each of the species was clearly separated, and Neighbor-Joining and Unweighted Pair Group analyses of Nei's genetic distance grouped all populations within a taxon into their own clusters. Spruces from Flys Peak, Chiricahua Mountains, Arizona, joined a P. engelmannii cluster and were not a bridge to P. mexicana as previously believed. Spruces from Cerro Mohinora, Chihuahua, were clearly P. mexicana, not phantom hybrids of P. chihuahuana and P. pungens. Nuclear random amplified polymorphic DNA and chloroplast simple sequence repeat and cleaved amplified polymorphic genetic markers were compared in a smaller sample of populations, using distance and parsimony approaches. DNA markers, like isozymes, clearly identified spruces from Cerro Mohinora as P. mexicana. In contradiction to the most recent taxonomic treatment, P. chihuahuana and P. martinezii were separated as distinct species by both isozyme and DNA markers, and formed a sister-species group. Picea engelmannii and P. mexicana formed a separate cluster, and the genetic distance between them was similar to values associated with closely related species but greater than distances typical of subspecies or varieties in conifers. Picea pungens, which is so similar to P. engelmannii that the two are frequently misidentified, was clearly distinguished from it, sometimes joining a P. chihuahuana-martinezii group and sometimes a P. engelmannii-mexicana group, depending on analysis. Picea breweriana was well isolated from all other taxa. Both DNA and isozyme phylogenies agreed with results from crossability studies and contradicted intrageneric relationships constructed largely on cone mor- phology.

Numerous populations from six spruce taxa, including four relict endemics, Picea chihuahuana (Chihuahua spruce), P. martinezii (Martinez spruce), P. mexicana (Mexican spruce), and P. breweriana (Brewer spruce), and two widespread species, P. engelmannii (Engelmann spruce) and P. pungens (blue spruce), were compared at homologous isozyme loci to test various hypotheses about their affinities and origins. Each of the species was clearly separated, and Neighbor-Joining and Unweighted Pair Group analyses of Nei's genetic distance grouped all populations within a taxon into their own clusters. Spruces from Flys Peak, Chiricahua Mountains, Arizona, joined a P. engelmannii cluster and were not a bridge to P. mexicana as previously believed. Spruces from Cerro Mohinora, Chihuahua, were clearly P. mexicana, not phantom hybrids of P. chihuahuana and P. pungens. Nuclear random amplified polymorphic DNA and chloroplast simple sequence repeat and cleaved amplified polymorphic genetic markers were compared in ...

Mexican spruce (Picea mexicana Martínez), an endangered species of the highest sky islands in México’s Sierra Madre Oriental and Sierra Madre Occidental, is threatened by fire, grazing, and global warming. Its conservation depends on whether it also is threatened by inbreeding and loss of genic diversity. We used 18 isozyme markers in 12 enzyme systems to assay genic diversity, characterize the mating system, and test for recent bottlenecks in three known populations. Unbiased, expected heterozygosity (He) averaged 0.125. Despite a separation of 676 km between populations in the Sierra Madre Oriental and the Sierra Madre Occidental, Wright’s FST, the proportion of total genic diversity among populations, was only 6.9%. Nei’s genetic distance was 0.001 between the populations in the Sierra Madre Oriental and more than an order of magnitude greater, 0.019, between the Sierra Madre Oriental and Sierra Madre Occidental. However, both values point to relatively recent divergence. Mating systems were predominantly outcrossing, but with significant selfing. Multilocus estimates of selfing varied from 19% to 41%, and the means of single-locus estimates were higher, suggesting that additional inbreeding occurred by mating among relatives. Despite significant inbreeding, observed heterozygosity was as high as or higher than He; Wright’s fixation index, FIS, was −0.107. Under the observed level of selfing, positive values of FIS were expected. Therefore, selection against inbreds and homozygotes must be intense. Cornuet-Luikart tests indicate recent bottlenecks in at least two of the three populations. The results suggest that Mexican spruce is a genetically viable species, and threats are primarily environmental.

Weeping piñon (Pinus pinceana) has a restricted and fragmented range, trees are widely scattered within populations, and reproduction is limited. Nevertheless, genetic diversity was high; based on 27 isozyme loci in 18 enzyme systems, unbiased expected heterozygosity averaged 0.174. Differentiation also was high (F(ST) = 0.152), reflecting isolation between southern, central, and northern fragments of the range. Among populations in the northern fragment, F(ST) was only 0.056, and the number of migrants per generation (Nm) was 4.21, which should preclude fixation. Nm between central and southern populations or between them and populations in the northern fragment was lower, 0.99-1.66, indicating a degree of genetic isolation. Multilocus outcrossing rates (t(m)) ranged from 0.836 in the south to 0.897 in the north. Therefore, selfing is low but statistically significant. The equilibrium inbreeding coefficient (F(e)) calculated from t(m) was in good agreement with observed inbreeding coefficients, suggesting that weeping piñon may be near equilibrium with respect to inbreeding and selection against selfed trees. Weeping piñon was variable at all loci polymorphic in maxipiñon (Pinus maximartinezii) and, therefore, qualifies as a possible progenitor of maxipiñon. Because of the high level of diversity, reasonable levels of gene flow within the northern fragment of weeping piñon's range, high rates of outcrossing, and, perhaps, only weak selection against inbred trees, protection in reserves would be a viable option for conservation.

Martínez spruce (Picea martínezii T.F. Patterson) is a conifer currently passing through a bottleneck, reduced to a few relict populations totaling less than 800 trees. We used isozyme markers to analyze the mating system and survey the level of genic diversity in two populations. The mating system was characterized by a high frequency of selfing. The multilocus outcrossing rates (tm) and 95% confidence intervals were only 0.399 (0.197 < tm< 0.601) for the smallest population and 0.589 (0.475 < tm< 0.703) or 0.685 (0.465 < tm< 0.905), depending on year, for the largest. These are among the lowest rates of outcrossing observed in conifers. The fixation indices for the two populations were -0.058 and 0.121, less than expected for such high levels of selfing. Expected heterozygosity, unbiased He, based on 22 loci in 13 enzyme systems, was 0.121 and 0.101 in the two populations. The proportion of the total genic diversity between populations, FST, was 2.4%. Nm, the number of migrants per generation, was about 1.00 or 10.17, depending on the method of estimation. The time since the two populations were isolated was estimated from Nei's genetic distance as only 150 to 15 000 years, which is consistent with a hypothesis of population collapse during late Pleistocene or Holocene warming. We discuss the implications for conservation.

Maxipiñon (Pinus maximartinezii Rzedowski), which is confined to a single population of approximately 2000 to 2500 mature trees, covers about 400 ha in southern Zacatecas, Mexico. Genetic diversity measured by expected heterozygosity was 0.122, which is moderate for pines. However, percentage polymorphic loci was low, 30.3%. The fixation index (F) of 0.081 indicated only slight heterozygote deficiency. Mating system analysis indicated a significant but low level of selling; the multilocus outcrossing rate, tm , was 0.816. The mean of single locus estimates, ts , was smaller (0.761), perhaps suggesting mating among relatives, although the difference between tm and ts was not statistically significant. The most striking features of maxipiñon's genetic structure were that no polymorphic locus had more than two alleles and most alleles at polymorphic loci were at intermediate frequencies. This is in contrast to other pines, which often have three to five or more alleles at some loci and in which the distribution of allele frequencies is U-shaped, most alleles being present at frequencies less than 10% or greater than 90%. A population with only two alleles per locus and at intermediate frequencies could occur if the population had been reduced to an extreme bottleneck and then expanded rapidly before random drift modified allele frequencies. A novel origin from a hybridization event would also explain the results. Significant gametic disequilibrium was detected at several pairs of loci in both maternal and paternal gametes. The presence of disequilibrium is in agreement with an origin from an extreme bottleneck, perhaps even a single seed. Furthermore, it demands that the event be relatively recent. The number of generations, as calculated from the observed mean disequilibrium, suggested that maxipiñon derived from an extreme bottleneck four to five generations ago, which is less than 1000 years in this species.

Population genetic parameters were estimated for six populations of Cronartium ribicola in western North America from British Columbia to the southern Sierra Nevada, and two outgroups from eastern North America, using isozyme, random amplified polymorphic DNA (RAPD) and restriction fragment length polymorphism (RFLP) markers on cultured haploid clones. Diversity was low, with only 8% polymorphism in the 212 markers identified. Each polymorphic locus had only two alleles, except for an RFLP marker in the ribosomal DNA complex with multiple alleles, that resulted from variable numbers of tandem repeats. Expected heterozygosity within populations, estimated from diploid teliospores, was only 0.025. The three types of markers were highly consistent with each other for these parameters. Yet, populations were highly differentiated; the proportion of the total variation attributable to differences among populations was 0.205. Multivariate statistical analysis as well as different clustering algorithms based on contrasting evolutionary assumptions (drift, mutation) all showed similar relationships and differences among populations. Genetic distances were not associated with geographic distances; western populations within a few kilometres of each other were often more distant from each other genetically than they were from eastern populations across the continent. The lack of pattern over the landscape of this metapopulation is consistent with aspects of the life cycle and epidemiological behavior of the pathogen, in which genetic drift appears to play a major role.

Fragmentation and reduction in population size are expected to reduce genetic diversity. However, examples from natural populations of forest trees are scarce. The range of Chihuahua spruce retreated northward and fragmented coincident with the warming climate that marked the early Holocene. The isolated populations vary from 15 to 2441 trees, which provided an opportunity to test whether census number is a good predictor of genetic diversity. Mean expected heterozygosity, He , based on 24 loci in 16 enzyme systems, was 0.093 for 10 sampled populations, which is within the range reported for conifers. However, estimates varied more than twofold among populations and He was closely related to the logarithm of the number of mature trees in the population (rHe,N = 0.93). Diversity among populations, FST , was 24.8% of the total diversity, which is higher than that observed in almost all conifer species studied. Nei's genetic distance, D, was not related to geographic distance between populations, and D̄ was 0.033, which is higher than estimates for most wide-ranging species. Most populations had excess homozygosity and the fixation index, FIS , was higher than that reported for all but one species of conifer. Nm, the number of migrants per generation, was 0.43 to 0.76, depending on estimation procedure, and is the smallest observed in conifers. The data suggest that populations of Chihuahua spruce have differentiated by drift and that they are effectively isolated. The results illustrate how a combination of paleontological observation and molecular markers can be used to illuminate recent evolutionary events. Multilocus estimates of outcrossing for two small populations were zero (complete selfing) and 0.153, respectively, which are in striking contrast to the near complete outcrossing observed in most conifers. The high fixation index and a high proportion of empty seeds (45%) suggest that inbreeding may be a serious problem for conservation of Chihuahua spruce.

Stands of canyon live oak (Quercus chrysolepis, Fagaceae) are maintained for fuelwood, fire management, recreation, and as habitat for wildlife. Information about the link between the oak's reproductive ecology and its extent of genetic diversity is important in developing land management policies that will maintain the long-term viability of populations. Basal sprouting is the primary means of reproduction following fire or cutting, and stands frequently include groups of visibly connected trees in a clustered distribution that suggests cloning. We determined the extent to which clusters of trees were clonal and defined the spatial pattern and diversity of genotypes for six populations across nearly the entire east-west extent of the San Bernardino Mountains in southern California. We mapped over 100 trees at each of five sites and genotyped each tree for allozymes at seven polymorphic loci. We identified clones using these multilocus genotypes and detected an average of 34.4 ± 7.3 (SD) clones per site, most of which had unique genotypes. In general, clustered trees belong to single clones and most clones consist of few trees (mean = 3.4 ± 0.6 trees per clone). However, clone size increased significantly with increased individual heterozygosity, suggesting that selection may favor highly heterozygous clones. Clonal diversity and evenness were high relative to reports for most other clonal species; an average of 97% of clones had distinct genotypes, and Simpson's index of diversity averaged 0.95 ± 0.02. Population genetic analyses of 319 clones from six sites revealed high genetic diversity within sites (mean HS = 0.443). Only a small proportion of the total genetic diversity was explained by variation among sites (mean GST = 0.018), which is consistent with high gene flow among sites (Nm = 9.5). We found no significant substructure among plots within sites, and fixation indices within sites were generally small, suggesting that either little inbreeding occurs, and/or few inbred progeny survive. However, spatial autocorrelation analysis of clones indicated fine-scale genetic structure at distances under 4 m, possibly due to limited seed dispersal. Our data suggest that guidelines for seed collection of canyon live oak for use in restoration can be specified in a manner similar to that recommended for conifer species within the region studied.

Lack of genetic markers has hindered the study of the mating system of Cronartium ribicola, an exotic forest pathogen Meeting natural and cultivated white pines throughout North America. Isozymes, randomly amplified polymorphic DNA (RAPDs), and restriction length polymorphisms (RFLPs) were used to study the mating system of this rust. Heterozygosity (outcrossing) in diploid telia was demonstrated by analysis of cultures derived from the meiotic products (basidiospores) of individual telia. Families of basidiospores cultured from single telia were used to test for Mendelian segregation and for conformance of loci to Hardy–Weinberg equilibrium. A total of 18 polymorphic loci were identified with the three marker systems. All except for three RAPD loci showed Mendelian segregation in the single-telium families. To quantify the level of outcrossing, gene and genotype frequencies were calculated for families from a single population. Up to 24 families were surveyed with isozymes, 14 with RAPDs, and 18 with RFLPs. Except for one isozyme locus (MPI) in one sample, all 14 loci tested with these families were in Hardy–Weinberg equilibrium, indicating random mating. Further studies, with a different sample from the same population, showed all three isozyme loci to be in Hardy–Weinberg equilibrium. The three marker systems were consistent as to the amount of variation detected. Resistance selection and breeding programs must consider the implications of genetic recombination that outcrossing affords the rust. Keywords: isozymes, RAPDs, RFLPs, Hardy–Weinberg equilibrium, white pine blister rust.

Population genetic parameters were estimated for six populations of Cronartium ribicola in western North America from British Columbia to the southern Sierra Nevada, and two outgroups from eastern North America, using isozyme, random amplified polymorphic DNA (RAPD) and restriction fragment length polymorphism (RFLP) markers on cultured haploid clones. Diversity was low, with only 8% polymorphism in the 212 markers identified. Each polymorphic locus had only two alleles, except for an RFLP marker in the ribosomal DNA complex with multiple alleles, that resulted from variable numbers of tandem repeats. Expected heterozygosity within populations, estimated from diploid teliospores, was only 0.025. The three types of markers were highly consistent with each other for these parameters. Yet, populations were highly differentiated; the proportion of the total variation attributable to differences among populations was 0.205. Multivariate statistical analysis as well as different clustering algorithms based on contrasting evolutionary assumptions (drift, mutation) all showed similar relationships and differences among populations. Genetic distances were not associated with geographic distances; western populations within a few kilometres of each other were often more distant from each other genetically than they were from eastern populations across the continent. The lack of pattern over the landscape of this metapopulation is consistent with aspects of the life cycle and epidemiological behavior of the pathogen, in which genetic drift appears to play a major role.