Megafaunal extinctions and the disappearance of a specialized wolf ecomorph

The gray wolf (Canis lupus) is one of the few large predators to survive the Late Pleistocene megafaunal extinctions [1]. Nevertheless, wolves disappeared from northern North America in the Late Pleistocene, suggesting they were affected by factors that eliminated other species. Using skeletal material collected from Pleistocene permafrost deposits of eastern Beringia, we present a comprehensive analysis of an extinct vertebrate by exploring genetic (mtDNA), morphologic, and isotopic (delta(13)C, delta(15)N) data to reveal the evolutionary relationships, as well as diet and feeding behavior, of ancient wolves. Remarkably, the Late Pleistocene wolves are genetically unique and morphologically distinct. None of the 16 mtDNA haplotypes recovered from a sample of 20 Pleistocene eastern-Beringian wolves was shared with any modern wolf, and instead they appear most closely related to Late Pleistocene wolves of Eurasia. Moreover, skull shape, tooth wear, and isotopic data suggest that eastern-Beringian wolves were specialized hunters and scavengers of extinct megafauna. Thus, a previously unrecognized, uniquely adapted, and genetically distinct wolf ecomorph suffered extinction in the Late Pleistocene, along with other megafauna. Consequently, the survival of the species in North America depended on the presence of more generalized forms elsewhere.     

Mitochondrial DNA phylogeography and population history of the grey wolf canis lupus

The grey wolf (Canis lupus) and coyote (C. latrans) are highly mobile carnivores that disperse over great distances in search of territories and mates. Previous genetic studies have shown little geographical structure in either species. However, population genetic structure is also influenced by past isolation events and population fluctuations during glacial periods. In this study, control region sequence data from a worldwide sample of grey wolves and a more limited sample of coyotes were analysed. The results suggest that fluctuating population sizes during the late Pleistocene have left a genetic signature on levels of variation in both species. Genealogical measures of nucleotide diversity suggest that historical population sizes were much larger in both species and grey wolves were more numerous than coyotes. Currently, about 300 000 wolves and 7 million coyotes exist. In grey wolves, genetic diversity is greater than that predicted from census population size, reflecting recent historical population declines. By contrast, nucleotide diversity in coyotes is smaller than that predicted by census population size, reflecting a recent population expansion following the extirpation of wolves from much of North America. Both species show little partitioning of haplotypes on continental or regional scales. However, a statistical parsimony analysis indicates local genetic structure that suggests recent restricted gene flow.     

Origin and status of the Great Lakes wolf

An extensive debate concerning the origin and taxonomic status of wolf-like canids in the North American Great Lakes region and the consequences for conservation politics regarding these enigmatic predators is ongoing. Using maternally, paternally and biparentally inherited molecular markers, we demonstrate that the Great Lakes wolves are a unique population or ecotype of gray wolves. Furthermore, we show that the Great Lakes wolves experienced high degrees of ancient and recent introgression of coyote and western gray wolf mtDNA and Y-chromosome haplotypes, and that the recent demographic bottleneck caused by persecution and habitat depletion in the early 1900s is not reflected in the genetic data.     

Genetic and morphometric analysis of sixteenth century <Emphasis Type="Italic">Canis</Emphasis> skull fragments: implications for historic eastern and gray wolf distribution in North America

Resolving the taxonomy and historic ranges of species are essential to recovery plans for species at risk and conservation programs that aim to restore extirpated populations. In eastern North America, planning for wolf population restoration is complicated by the disputed historic distributions of two wolf species: the Old World-evolved gray wolf (Canis lupus) and the New World-evolved eastern wolf (C. lycaon). We used genetic and morphometric data from 4- to 500-year-old Canis samples excavated in London, Ontario, Canada to help clarify the historic range of these two wolf species in the eastern temperate forests of North America. We isolated DNA and sequenced the mitochondrial control region and found that none of the samples were of gray wolf origin. Two of the DNA sequences corresponded to those found in present day coyotes (C. latrans), but morphometric comparisons show an eastern wolf, not coyote, origin. The remaining two sequences matched ancient domestic dog haplotypes. These results suggest that the New World-evolved eastern wolf, not the gray wolf, occupied this region prior to the arrival of European settlers, although eastern-gray wolf hybrids cannot be ruled out. Furthermore, our data support the idea of a shared common ancestry between eastern wolves and western coyotes, and that the distribution of gray wolves at this time probably did not include the eastern temperate forests of North America.     

Reviving the African Wolf Canis lupus lupaster in North and West Africa: A Mitochondrial Lineage Ranging More than 6,000 km Wide

The recent discovery of a lineage of gray wolf in North-East Africa suggests the presence of a cryptic canid on the continent, the African wolf Canis lupus lupaster. We analyzed the mtDNA diversity (cytochrome b and control region) of a series of African Canis including wolf-like animals from North and West Africa. Our objectives were to assess the actual range of C. l. lupaster, to further estimate the genetic characteristics and demographic history of its lineage, and to question its taxonomic delineation from the golden jackal C. aureus, with which it has been considered synonymous. We confirmed the existence of four distinct lineages within the gray wolf, including C. lupus/familiaris (Holarctic wolves and dogs), C. l. pallipes, C. l. chanco and C. l. lupaster. Taxonomic assignment procedures identified wolf-like individuals from Algeria, Mali and Senegal, as belonging to C. l. lupaster, expanding its known distribution c. 6,000 km to the west. We estimated that the African wolf lineage (i) had the highest level of genetic diversity within C. lupus, (ii) coalesced during the Late Pleistocene, contemporaneously with Holarctic wolves and dogs, and (iii) had an effective population size of c. 80,000 females. Our results suggest that the African wolf is a relatively ancient gray wolf lineage with a fairly large, past effective population size, as also suggested by the Pleistocene fossil record. Unique field observations in Senegal allowed us to provide a morphological and behavioral diagnosis of the African wolf that clearly distinguished it from the sympatric golden jackal. However, the detection of C. l. lupaster mtDNA haplotypes in C. aureus from Senegal brings the delineation between the African wolf and the golden jackal into question. In terms of conservation, it appears urgent to further characterize the status of the African wolf with regard to the African golden jackal.     

Genetic outcomes of wolf recovery in the western Great Lakes states

Conflicting interpretations of the influence of coyote hybridization on wolf recovery in the western Great Lakes (WGL) states have stemmed from disagreement over the systematics of North American wolves. Questions regarding their recovery status have resulted. We addressed these issues with phylogenetic and admixture analysis of DNA profiles of western wolves, WGL states wolves and Wisconsin coyotes developed from autosome and Y-chromosome microsatellites and mitochondrial DNA control region sequence. Hybridization was assessed by comparing the haplotypes exhibited by sympatric wolves and coyotes. Genetic variability and connectivity were also examined. These analyses support the recognition of Canis lycaon as a unique species of North American wolf present in the WGL states and found evidence of hybridization between C. lupus and C. lycaon but no evidence of recent hybridization with sympatric coyotes. The recolonized WGL states wolves are genetically similar to historical wolves from the region and should be considered restored.     

A signal for independent coastal and continental histories among North American wolves

Relatively little genetic variation has been uncovered in surveys across North American wolf populations. Pacific Northwest coastal wolves, in particular, have never been analysed. With an emphasis on coastal Alaska wolf populations, variation at 11 microsatellite loci was assessed. Coastal wolf populations were distinctive from continental wolves and high levels of diversity were found within this isolated and relatively small geographical region. Significant genetic structure within southeast Alaska relative to other populations in the Pacific Northwest, and lack of significant correlation between genetic and geographical distances suggest that differentiation of southeast Alaska wolves may be caused by barriers to gene flow, rather than isolation by distance. Morphological research also suggests that coastal wolves differ from continental populations. A series of studies of other mammals in the region also has uncovered distinctive evolutionary histories and high levels of endemism along the Pacific coast. Divergence of these coastal wolves is consistent with the unique phylogeographical history of the biota of this region and re-emphasizes the need for continued exploration of this biota to lay a framework for thoughtful management of southeast Alaska.     

Going coastal: shared evolutionary history between coastal British Columbia and Southeast Alaska wolves (Canis lupus)

Background

Many coastal species occupying the temperate rainforests of the Pacific Northwest in North America comprise endemic populations genetically and ecologically distinct from interior continental conspecifics. Morphological variation previously identified among wolf populations resulted in recognition of multiple subspecies of wolves in the Pacific Northwest. Recently, separate genetic studies have identified diverged populations of wolves in coastal British Columbia and coastal Southeast Alaska, providing support for hypotheses of distinct coastal subspecies. These two regions are geographically and ecologically contiguous, however, there is no comprehensive analysis across all wolf populations in this coastal rainforest.

Methodology/Principal Findings

By combining mitochondrial DNA datasets from throughout the Pacific Northwest, we examined the genetic relationship between coastal British Columbia and Southeast Alaska wolf populations and compared them with adjacent continental populations. Phylogenetic analysis indicates complete overlap in the genetic diversity of coastal British Columbia and Southeast Alaska wolves, but these populations are distinct from interior continental wolves. Analyses of molecular variation support the separation of all coastal wolves in a group divergent from continental populations, as predicted based on hypothesized subspecies designations. Two novel haplotypes also were uncovered in a newly assayed continental population of interior Alaska wolves.

Conclusions/Significance

We found evidence that coastal wolves endemic to these temperate rainforests are diverged from neighbouring, interior continental wolves; a finding that necessitates new international strategies associated with the management of this species.     

周口店变异狼(Canis lupus variabilis)在家畜狗(Canis familiaris)祖先系列中的位置

本文作者对周口店第一、第三及第十三地点的一种化石犬类——变异狼(Canis lupus uariadilis)进行了观察和测量。根据其头骨大小、形态特征以及与中国早期人类共生的情况来看,认为它有可能是从驯化的野生狼导致家畜狗出现的一种祖先类型。     

Y‐chromosome evidence supports widespread signatures of three‐species Canis hybridization in eastern North America

There has been considerable discussion on the origin of the red wolf and eastern wolf and their evolution independent of the gray wolf. We analyzed mitochondrial DNA (mtDNA) and a Y‐chromosome intron sequence in combination with Y‐chromosome microsatellites from wolves and coyotes within the range of extensive wolf–coyote hybridization, that is, eastern North America. The detection of divergent Y‐chromosome haplotypes in the historic range of the eastern wolf is concordant with earlier mtDNA findings, and the absence of these haplotypes in western coyotes supports the existence of the North American evolved eastern wolf (Canis lycaon). Having haplotypes observed exclusively in eastern North America as a result of insufficient sampling in the historic range of the coyote or that these lineages subsequently went extinct in western geographies is unlikely given that eastern‐specific mtDNA and Y‐chromosome haplotypes represent lineages divergent from those observed in extant western coyotes. By combining Y‐chromosome and mtDNA distributional patterns, we identified hybrid genomes of eastern wolf, coyote, gray wolf, and potentially dog origin in Canis populations of central and eastern North America. The natural contemporary eastern Canis populations represent an important example of widespread introgression resulting in hybrid genomes across the original C. lycaon range that appears to be facilitated by the eastern wolf acting as a conduit for hybridization. Applying conventional taxonomic nomenclature and species‐based conservation initiatives, particularly in human‐modified landscapes, may be counterproductive to the effective management of these hybrids and fails to consider their evolutionary potential.     

Genetic assessment of the Iberian wolf Canis lupus signatus captive breeding program

The main goal of ex situ conservation programs is to improve the chances of long term survival of natural populations by founding and managing captive colonies that can serve as a source of individuals for future reintroductions or to reinforce existing populations. The degree in which a captive breeding program has captured the genetic diversity existing in the source wild population has seldom been evaluated. In this study we evaluate the genetic diversity in wild and captive populations of the Iberian wolf, Canis lupus signatus, in order to assess how much genetic diversity is being preserved in the ongoing ex situ conservation program for this subspecies. A sample of domestic dogs was also included in the analysis for comparison. Seventy-four wolves and 135 dogs were genotyped at 13 unlinked microsatellite loci. The results show that genetic diversity in Iberian wolves is comparable in magnitude to that of other wild populations of gray wolf. Both the wild and the captive Iberian wolf populations have a similarly high genetic diversity indicating that no substantial loss of diversity has occurred in the captive-breeding program. The effective number of founders of the program was estimated as ∼ ∼16, suggesting that all founders in the studbook pedigree were genetically independent. Our results emphasize also the genetic divergence between wolves and domestic dogs and indicate that our set of 13 microsatellite loci provide a powerful diagnostic test to distinguish wolves, dogs and their hybrids.     

Molecular genetics of the most endangered canid: the Ethiopian wolf Canis simensis

The world's most endangered canid is the Ethiopian wolf Canis simensis , which is found in six isolated areas of the Ethiopian highlands with a total population of no more than 500 individuals. Ethiopian wolf populations are declining due to habitat loss and extermination by humans. Moreover, in at least one population, Ethiopian wolves are sympatric with domestic dogs, which may hybridize with them, compete for food, and act as disease vectors. Using molecular techniques, we address four questions concerning Ethiopian wolves that have conservation implications. First, we determine the relationships of Ethiopian wolves to other wolf-like canids by phylogenetic analysis of 2001 base pairs of mitochondrial DNA (mtDNA) sequence. Our results suggest that the Ethiopian wolf is a distinct species more closely related to gray wolves and coyotes than to any African canid. The mtDNA sequence similarity with gray wolves implies that the Ethiopian wolf may hybridize with domestic dogs, a recent derivative of the gray wolf. We examine this possibility through mtDNA restriction fragment analysis and analysis of nine microsatellite loci in populations of Ethiopian wolves. The results imply that hybridization has occurred between female Ethiopian wolves and male domestic dogs in one population. Finally, we assess levels of variability within and between two Ethiopian wolf populations. Although these closely situated populations are not differentiated, the level of variability in both is low, suggesting long-term effective population sizes of less than a few hundred individuals. We recommend immediate captive breeding of Ethiopian wolves to protect their gene pool from dilution and further loss of genetic variability.     

Phylogeography of northern Dolly Varden Salvelinus malma malma based on analysis of mitochondrial DNA

The northern Dolly Varden, Salvelinus malma malma, is a typical representative of arctic fauna distributed in northeastern Asia and northwestern North America. Because its spawning habitats were affected by Pleistocene glacial advances over most of its natural range, S. m. malma is among the most interesting objects of phylogeographic and microevolutionary studies. We reconstructed the genealogy of mtDNA haplotypes from 27 Alaskan and Asian populations to study the influence of glacial and geological vicariance events on the contemporary population genetic structure, phylogeographic subdivision and distribution of the northern Dolly Varden. Analysis of restriction site states in three PCR‐amplified mtDNA regions (ND1/ND2, ND5/ND6, Cytb/D‐loop; 47% of the mitochondrial genome) resolved 75 haplotypes in 436 fish. Similar patterns of subspecific variation apparent from hierarchical diversity and nested clade analyses of mtDNA haplotypes identify weak spatial differentiation and low levels of divergence. Our results suggest that (1) demographic history has been influenced by historical range expansions and recent isolation by distance, (2) present populations from Asia and North America were colonized from one main Beringian Refugium, and (3) that this taxon's ancestral population probably experienced a bottleneck in the Beringian Refugium during the late Pleistocene (Wisconsin) glacial period.     

Detection of an East European wolf haplotype puzzles mitochondrial DNA monomorphism of the Italian wolf population

Southern European wolves suffered from reiterated population declines during glacial periods and historically due to human persecution. Differently from other European wolf populations, a single mitochondrial DNA (mtDNA) control region haplotype (W14) has been so far described in the Italian wolves, although no intensive genetic sampling has ever been conducted in historical source populations from central and southern Italy. Using non-invasive genetic techniques, we report the occurrence of an unexpected mtDNA haplotype (W16) in the wolf population of the Abruzzo, Lazio and Molise National Park (PNALM), central Italy. This haplotype, detected in three out of 90 faecal samples from the PNALM, was previously reported in wolves from the North Carpathians, Slovakia and the Balkans only. Microsatellite analysis and molecular sex determination confirmed that the W16 samples belonged to three distinct wolves. Although alternative explanations can be formulated for the origin of this mtDNA haplotype in the otherwise monomorphic Italian wolf population, assignment procedures indicated the likely admixed ancestry of one W16 sample with East European wolves. Anthropogenic introgression with dogs has been detected in the Italian wolf population using nuclear DNA microsatellites, but no population-wide genetic survey had previously reported a mtDNA control region variant in Italian wolves. Our findings strongly suggest that, in addition to wolf × dog hybridization, captive-released wolves or wolf × dog hybrids may successfully interbreed with wolves in the wild, and that human-mediated introgression may occur even in well established protected areas.     

Sympatric wolf and coyote populations of the western Great Lakes region are reproductively isolated

Interpretation of the genetic composition and taxonomic history of wolves in the western Great Lakes region (WGLR) of the United States has long been debated and has become more important to their conservation given the recent changes in their status under the Endangered Species Act. Currently, the two competing hypotheses on WGLR wolves are that they resulted from hybridization between (i) grey wolves (Canis lupus) and western coyotes (C. latrans) or (ii) between grey wolves and eastern wolves (C. lycaon). We performed a genetic analysis of sympatric wolves and coyotes from the region to assess the degree of reproductive isolation between them and to clarify the taxonomic status of WGLR wolves. Based on data from maternal, paternal and bi‐parental genetic markers, we demonstrate a clear genetic distinction between sympatric wolves and coyotes and conclude that they are reproductively isolated and that wolf–coyote hybridization in the WGLR is uncommon. The data reject the hypothesis that wolves in the WGLR derive from hybridization between grey wolves and western coyotes, and we conclude that the extant WGLR wolf population is derived from hybridization between grey wolves and eastern wolves. Grey‐eastern wolf hybrids (C. lupus × lycaon) comprise a substantial population that extends across Michigan, Wisconsin, Minnesota and western Ontario. These findings have important implications for the conservation and management of wolves in North America, specifically concerning the overestimation of grey wolf numbers in the United States and the need to address policies for hybrids.     

Genetic nature of eastern wolves: Past, present and future

Eastern North American wolves have long been recognized as morphologically distinct from both coyotes and gray wolves. This has led to questions regarding their origins and taxonomic status. Eastern wolves are mainly viewed as: (1) a smaller subspecies of gray wolf (Canis lupus lycaon), potentially the result of historical hybridization between gray wolves (C. lupus) and red wolves (C. rufus), (2) a hybrid, the result of gray wolf (C. lupus) and coyote (C. latrans) interbreeding, or (3) a distinct species, C. lycaon, closely related to the red wolf (C. rufus). Although debate persists, recent molecular studies suggest that the eastern wolf is not a gray wolf subspecies, nor the result of gray wolf/coyote hybridization. Eastern wolves were more likely a distinct species, C. lycaon, prior to the eastward spread of coyotes in the late 1800s. However, contemporary interbreeding exits between C. lycaon to both C. lupus and C. latrans over much of its present range complicating its present taxonomic characterization. While hybridization may be reducing the taxonomic distinctiveness of C. lycaon, it should not necessarily be viewed as negative influence. Hybridization may be enhancing the adaptive potential of eastern wolves, allowing them to more effectively exploit available resources in rapidly changing environments.     

Phylogeography of northern Dolly Varden Salvelinus malma (Salmoniformes: Salmonidae) from Asia and North America: An analysis based on the mitochondrial DNA genealogy

Spawning in habitats affected by Pleistocene glacial advances over most of its natural range, northern Dolly Varden Salvelinus malma malma typifies Arctic fauna distributed in northeastern Asia and northwestern North America. We reconstructed a genealogy of mtDNA haplotypes from 27 Alaskan and Asian populations to study the influence of historical events on the phylogeography and contemporary population genetic structure. Analysis of molecular variance partitioned most of the mtDNA variability to the intrapopulation component (72.5%) with much reduced differences between populations (21.1%) and regions (6.4%). Similar patterns of variation apparent from hierarchical diversity and nested clade phylogeographical analysis (NCPA) of mtDNA haplotypes identify weak spatial differentiation and low levels of divergence. These findings suggest (1) that demographic history has been influenced by historical range expansions and recent isolation by distance, (2) that present populations from Asia and North America were colonized from one main Beringian Refugium, and (3) that this taxon’s ancestral population probably experienced a bottleneck in the Beringian Refugium during the late Pleistocene (Wisconsin) glacial period. The genealogical and NCPA analyses, and mismatch distribution of S. m. malma mtDNA haplotypes do not confirm the assumptions about presence of the two refugia on the territories of the Beringian Land, in which allopatric S. m. malma ancestral populations evolved, and independent origin of the Sea of Okhotsk populations.     

Assessing the prevalence of hybridization between sympatric Canis species surrounding the red wolf (Canis rufus) recovery area in North Carolina

Predicting spatial patterns of hybridization is important for evolutionary and conservation biology yet are hampered by poor understanding of how hybridizing species can interact. This is especially pertinent in contact zones where hybridizing populations are sympatric. In this study, we examined the extent of red wolf (Canis rufus) colonization and introgression where the species contacts a coyote (C. latrans) population in North Carolina, USA. We surveyed 22,000km(2) in the winter of 2008 for scat and identified individual canids through genetic analysis. Of 614 collected scats, 250 were assigned to canids by mitochondrial DNA (mtDNA) sequencing. Canid samples were genotyped at 6-17 microsatellite loci (nDNA) and assigned to species using three admixture criteria implemented in two Bayesian clustering programs. We genotyped 82 individuals but none were identified as red wolves. Two individuals had red wolf mtDNA but no significant red wolf nDNA ancestry. One individual possessed significant red wolf nDNA ancestry (approximately 30%) using all criteria, although seven other individuals showed evidence of red wolf ancestry (11-21%) using the relaxed criterion. Overall, seven individuals were classified as hybrids using the conservative criteria and 37 using the relaxed criterion. We found evidence of dog (C. familiaris) and gray wolf (C. lupus) introgression into the coyote population. We compared the performance of different methods and criteria by analyzing known red wolves and hybrids. These results suggest that red wolf colonization and introgression in North Carolina is minimal and provide insights into the utility of Bayesian clustering methods to detect hybridization.     

Wolves in the Great Lakes region: a phylogeographic puzzle

Empirical studies demonstrate that natural hybridization in animals is more common than thought so far ( Mallet 2005 ), particularly among species that originated recently through cycles of population contraction–expansion arising from climate changes over the last glacial period, the Pleistocene. In addition, the post‐glacial global growth of human populations has fostered anthropogenic hybridization events, mediated by habitat changes, the persecution of large predators and the introduction of alien species ( Allendorf et al. 2001 ). The Canis lineage shows cases of both natural and anthropogenic hybridization, exacerbating the controversy about the number of species that should be formally validated in the taxonomic lists, the evolutionary role of genetic introgression and the ways to manage hybrids with invading wild or domesticated populations. The study by Wheeldon et al. (2010) , published in this issue of Molecular Ecology, adds a new piece to the intricate puzzle of evolution and taxonomy of Canis in North America. They show that sympatric wolves (C. lupus) and coyotes (C. latrans) are not (extensively) hybridizing in the western North American Great Lakes region (GLR). Widespread hybridization between coyotes and a genetically distinct, but closely related, wolf‐like population (the eastern wolf) occurred in the northeastern regions of North America. In Wheeldon et al.’s (2010) opinion, these data should prove definitely that two different species of wolf (the western gray wolf C. lupus and the eastern wolf C. lycaon) and their hybrids are distributed across the GLR.     

Activity patterns of gray wolves housed in small vs. large enclosures

Free-ranging gray wolves (Canis lupus) generally inhabit large home ranges, yet they are housed in a variety of restricted spaces when in captivity. There is continual debate as to whether space restrictions alter a wolf's behavior. The purpose of these studies was to remotely measure and then compare the amount and frequency of activity of gray wolves housed in small, artificial enclosures vs. large, more natural enclosures. Test animals comprised three adult wolves housed in kennels and three and four wolves housed in separate natural enclosures. Kenneled wolves had 2.8 m² of surface area per wolf, and wolves in natural enclosures had 466.6 m² (South Pack) and 349.9 m² (North Pack) per wolf. Wolves were fitted with radiotelemetry collars containing activity sensors. Activity data were recorded every 20 min for 57 continuous hr. The amount of activity for each wolf was calculated using areas under the curve (AUCs), and the frequency of activity was analyzed by spectral analysis. There was no difference (P ≥ 0.22) in AUCs between kenneled wolves (1.399 ± 0.214 x 10⁵ radians) and South Pack wolves (1.564 ± 0.139 X 10⁵ radians) or North Pack wolves (1.617 ± 0.192 x 10⁵ radians). All three groups had similar peak spectral values at frequencies that were close to daily cycles (i.e., ω = 0.12–0.17 cycles per unit time). Peaks in coherence near the dominant spectral frequency were most significant between the natural enclosures and the least significant between the kenneled wolves and the South Pack wolves. Based on these criteria of activity and under these circumstances, enclosure size appeared to have no effect on wolf activity. However, small sample sizes and variation in the data do not make these results definitive. © 1996 Wiley-Liss, Inc.