Stasipatric Speciation Research Articles

Oh, what a tangled web we weave

Oh, what a tangled web we weave

Molecular Phylogenetics of Aotus (Platyrrhini, Cebidae)

The accurate identification of taxa of Aotus is essential for 1) the development of precise biomedical assays, 2) the determination of potential illegal traffic of this genus, and 3) conservation. Although many studies have contributed to what we know about the phylogenetics of Aotus, none used a sufficiently large number of samples to clarify its complexity. To address this need, we sequenced 696 base pairs of the mitochondrial cytochrome-oxidase II gene (mtCOII) in 69 specimens of 7 taxa of Aotus. We also analyzed 8 microsatellite loci in 136 individuals of 6 taxa. In contrast to previous studies, we sampled only wild individuals and have a precise geographical origin for each one. The mtDNA results showed that: 1) the northern gray-necked group of Aotus is genetically more homogeneous than the polyphyletic red-necked group of Aotus; 2) the ancestors of Aotus vociferans seem to be the original species candidate for the current Aotus; 3) Aotus azarae azarae and A. a. boliviensis are the most differentiated taxa, likely a result of extreme genetic drift during stasipatric speciation; 4) the first genetic splits found among taxa of Aotus occurred during the Pliocene (or even Miocene) while the most recent ones happened during the Pleistocene, when forest refugia may have played an important role in speciation. The mean number of microsatellite alleles was 3–5.33 alleles per locus. We found some private alleles that could be useful in helping to identify illegal trade, although a larger sample size is needed to ensure that these alleles are really private to the relevant taxa. These new findings increase our understanding of the phylogeny of Aotus and the level of genetic diversity within different taxa of Aotus.

Chromosomal Speciation Revisited: Modes of Diversification in Australian Morabine Grasshoppers (Vandiemenella, viatica Species Group)

Chromosomal rearrangements can alter the rate and patterns of gene flow within or between species through a reduction in the fitness of chromosomal hybrids or by reducing recombination rates in rearranged areas of the genome. This concept, together with the observation that many species have structural variation in chromosomes, has led to the theory that the rearrangements may play a direct role in promoting speciation. Australian morabine grasshoppers (genus Vandiemenella, viatica species group) are an excellent model for studying the role of chromosomal rearrangement in speciation because they show extensive chromosomal variation, parapatric distribution patterns, and narrow hybrid zones at their boundaries. This species group stimulated development of one of the classic chromosomal speciation models, the stasipatric speciation model proposed by White in 1968. Our population genetic and phylogeographic analyses revealed extensive non-monophyly of chromosomal races along with historical and on-going gene introgression between them. These findings suggest that geographical isolation leading to the fixation of chromosomal variants in different geographic regions, followed by secondary contact, resulted in the present day parapatric distributions of chromosomal races. The significance of chromosomal rearrangements in the diversification of the viatica species group can be explored by comparing patterns of genetic differentiation between rearranged and co-linear parts of the genome.

Stasipatric speciation: resurrecting a system to bury a hypothesis?

The process of speciation remains a fundamental topic in evolutionary biology. Numerous models of speciation have been proposed and they are as diverse and colourful as the scientists who conceived them (Coyne & Orr 2004). One of the more controversial theories has been the 'stasipatric speciation' model, proposed by the pioneering and influential cytogeneticist Michael White and his co-workers (White 1968; White et al. 1967). This is one of a number of speciation models whereby chromosomal rearrangements drive the speciation process. The inspiration for the theory of stasipatric speciation came from White's karyotypic analyses of a group of Australian grasshoppers of the genus Vandiemenella (White et al. 1967) (Fig. 1). It has been exactly three decades since the last scientific publication on this group of grasshoppers, over which time the molecular revolution dramatically altered the landscape of evolutionary genetics. Kawakami and colleagues have successfully resurrected the Vandiemenella system (Kawakami et al. 2009a, 2007) and in this issue they have applied modern molecular-based techniques to reassess the validity of the stasipatric speciation model for this historically important group (Kawakami et al. 2009b).

Re‐examination of a proposed case of stasipatric speciation: phylogeography of the Australian morabine grasshoppers (Vandiemenella viatica species group)

Karyotypic differences have been used for delimiting populations or species, although whether these mutations provide strong barriers to gene flow between populations and promote speciation remains contentious. In this study, we assessed whether 11 chromosomal races of Australian morabine grasshoppers (Vandiemenella viatica species group) represent genetically distinct populations by analyses of cytological and allozyme (35 loci) data and DNA sequences of the elongation factor-1 alpha (EF-1alpha), anonymous Mvia11, and mitochondrial cytochrome c oxidase subunit I (COI) loci. While the Vandiemenella chromosomal taxa generally represent genetically distinct units, a substantial portion of the total genetic variation in our samples was not explained by the chromosomal variation. Mantel tests indicated that Vandiemenella populations were spatially structured and have maintained gene flow at a local scale within each of the taxa. The group was subdivided into 13 genetic clusters; four chromosomal taxa comprised single exclusive clusters, while others comprised more than one cluster or clusters shared with other taxa. Boundaries of these cryptic population subdivisions correspond with several biogeographical barriers, such as straits, gulfs, the Murray River, and an ancient mega-lake, Lake Bungunnia. The viatica species group was previously proposed to have diversified without major geographical separation based on the stasipatric speciation model; however, the present study suggests the involvement of allopatric fragmentation. Given extensive nonmonophyly of chromosomal taxa and incomplete barriers to gene flow among taxa, all Vandiemenella chromosomal taxa and genetically distinct populations within chromosomal taxa, except Vandiemenella pichirichi, should be regarded as populations of one species: Vandiemenella viatica.

The role of chromosomal rearrangement in mammalian speciation with special reference to Cetacea and Pinnipedia

The karyotypes of 17 cetaceans and 18 pinnipeds are surveyed. Within the Cetacea only two different chromosome numbers have been found, viz. 2n = 42 and 2n = 44. Within the Pinnipedia three different chromosome numbers have been described, viz. 2n = 32, 2n = 34 and 2n = 36. The material indicates that the orders Cetacea and Pinnipedia are characterized by a remarkable stability of their karyotypic patterns. Factors cooperating towards karyotypic stability may be sought in the reproductive biology of these animals — late sexual maturity and small progeny — as well as in their ecology — good mobility in an environment without distinct niches. In such animals, the main mode of speciation should be a allopatric. The orders Insectivora and Rodentia on the other hand which both display great karyotype variability would accordingly represent a different mode of speciation. The species belonging to these orders have early sexual maturity and large progeny; their mobility is often limited and the distribution discontinuous. Consequently, in Insectivora and Rodentia stasipatric speciation should be frequent.

Chromosomal divergence in house mice in the light of palaeontology: A colonization-related event?

Chromosomal divergence in wild populations of the house mouse has been considered a stasipatric speciation process. Numerous Robertsonian (Rb) centric fusions are fixed in the populations of Europe and Tunisia and involve almost all the acrocentric chromosomes of the standard 2n = 40 karyotype. The tempo and mode of fixation of these chromosomal mutations in nature as well as the evolutionary consequences of this differentiation remain for the most part misunderstood. Palaeontological data indicate that the house mouse has colonized (or recolonized) Europe during the Holocene. The dispersion of this commensal species is obviously related to human activity. The colonization of the Mediterranean area can be approximately dated region by region. This shows that chromosomal divergence occurs in very recent populations, i.e. for which the period of colonization is after the Bronze Age. No Rb populations have been observed in regions colonized by the house mouse before this period. This pattern suggests that chromosomal divergence in the house mouse may be related to the colonization wave. Several interpretations are presented to explain the relationship between the distribution and age of Robertsonian populations.

COMPARISON OF POPULATION STRUCTURE IN CHROMOSOMALLY POLYTYPIC AND MONOTYPIC SPECIES OF SCELOPORUS (SAURIA: IGUANIDAE) IN RELATION TO CHROMOSOMALLY-MEDIATED SPECIATION.

Sceloporus grammicus has been used as an example of stasipatric speciation, and some models of chromosomal speciation have assumed that the fission rearrangements which distinguish the chromosome races within this complex are negatively heterotic in the heterozygous state. Consequently, such rearrangements are assumed to be fixed by drift in small, inbred populations. Allelic variation at 19 presumptive gene loci was examined in 13 samples of a chromosomally conservative congener, Sceloporus graciosus, and was compared to an equivalent data set for the chromosomally polytypic S. grammicus. Several estimates of population variability and structure were calculated, and results were inconsistent with models of extreme population subdivision and/or frequent bottlenecks and extinction-colonization events in S. grammicus. For example, mean level of heterozygosity and mean number of alleles per locus were significantly higher in S. grammicus than in S. graciosus, while the average genetic distance (D) was significantly higher in S. graciosus. These results are considered in light of expected "Wrightian" population genetic parameters, and we conclude that S. grammicus does not seem to have the population structure necessary for the fixation of strongly negatively heterotic rearrangements.

PRIMARY AND SECONDARY TRANSITION ZONES IN SPECIATION AND POPULATION DIFFERENTIATION: A PHYLOGENETIC ANALYSIS OF RANGE EXPANSION.

Transition (hybrid, tension, intergradation and contact) zones are the focus of attention of many evolutionary studies. Concepts of speciation and population evolution are dependent on whether the zone is thought to be due to the differentiation of the populations in situ (i.e., primary zones) as in area effect, clinal and stasipatric speciation (Endler, 1977; White, 1978) or due to secondary contact of populations (i.e., secondary zones) as in allopatric speciation or differentiation. It has been pointed out (Endler, 1977) that a major logical flaw in many such studies is that one cannot distinguish between a primary and secondary original event by the nature of the zone. Irrespective of the truth of this it is obviously logically desirable to hypothesize the origin of a zone independent of its nature. Using grass snake populations as an example, this paper describes some ideas

QUANTITATIVE ANALYSIS OF KARYOTYPE ALTERATION AND SPECIES DIFFERENTIATION IN MAMMALS.

Recent advances in mammalian cytogenetics have revealed that (1) most morphologically distinctive species are karyotypically differentiated (see the chromosome monographs by Hsu and Benirschke, 1967-1975), and (2) there are some species which include karyotypically highly diversified local populations, for example, Mus musculoides (Jotterand, 1972; Matthey, 1973), Mus poschiavinus (Gropp et al., 1972), Spalax ehrenbergi (Wahrman et al., 1969), and Thomomys bottae grahamensis (Patton, 1970). These accumulated data seem to be consistent with the stasipatric speciation model proposed by White (1968, 1978) based on the karyotype analysis of morabine grasshoppers. According to the model, chromosome rearrangements are considered to have played an essential role in speciation as one of the isolation mechanisms. On the other hand, there are also indisputable reports that morphologically distinct species sometimes have apparently identical karyotypes, for example, in Canis (Wurster and Benirschke, 1968), Macaca (Egozcue, 1975), and Myotis (Baker, 1970), suggesting speciation without visible karyotype differentiation. This conflicting evidence may be reconciled, however, by postulating two types of speciation, one with and one without karyotype differentiation. In this case, the role of karyotype alteration in species differentiation should be solved in a quantitative framework. I attempt in the present paper to investigate quantitatively the relationship between karyotype alteration and species differentiation in mammals by analyzing the number of karyotypes and species per genus. ANALYSIS

REDUCTION OF GENE FLOW DUE TO THE PARTIAL STERILITY OF HETEROZYGOTES FOR A CHROMOSOME MUTATION. I. STUDIES ON A "NEUTRAL" GENE NOT LINKED TO THE CHROMOSOME MUTATION IN A TWO POPULATION MODEL.

The primary role of chromosome mutations that cause partial sterility of the heterozygote in triggering some processes of speciation is supported by an increasing amount of data. This type of speciation seems to have played a particularly important role in animal species with limited vagility. Among the more extensively studied examples are the Australian Morabinae (White, 1968, 1974), the complex Didymuria violescens (Bocquet, 1953; Lecher, 1964, 1967, 1968), the isopods Jaera (Lecher and Prunus, 1972) and the Mexican lizards of the complex Sceloporus grammicus (Hall and Selander, 1973). Besides these exemplary cases, there are numerous other situations of closely related species with different karyotypes in numerous groups of animals which have been interpreted in terms of chromosomal speciation mechanisms (see White, 1978a, 1978b, for a complete review). The mechanisms of chromosomal speciation are open to discussion. The stasipatric speciation model, proposed and developed by White (1968, 1978a, 1978b), envisages the production of a heterozygote mutant, the fixation of the mutation by genetic drift and the subsequent spread of the mutant homozygote population due to systematic factors advantageous to the mutation. Clarification and quantification of the processes and mechanisms involved with regard to sequence and hierarchy are necessary: besides further knowledge of individual examples, theoretical models are important to explain the onset, fixation (see Hedrick, 198 1), spread and accumulation of chromosome mutations and their role in reproductive isolation. We address here the efficiency, as a mechanism of reproductive isolation, of the partial sterility of heterozygotes for chromosome mutations. In this paper we quantify the efficiency of a single chromosome mutation that causes partial sterility of the heterozygote in reducing the gene flow. We use a model similar to the loci systems on which an extensive literature exists (see Hedrick et al., 1978), but which scarcely addresses the problem of chromosomal speciation. The model consists of two populations of the same size which come into contact and are initially monomorphic, one with a normal form and the other a mutated form of one specific chromosome. These populations differ in frequency for a pair of selectively neutral alleles, not linked to the chromosome involved in the mutation. We propose a model with non-overlapping generations based on the following parameters: 1) constant symmetrical migration rate between the two populations in the diploid stage (m); and 2) reduction of fitness, caused by the partial sterility of the heterozygote for the chromosome mutation (s). Our model will be based upon the following quantities: 1) the difference in frequency of the mutated chromosome in populations 1 and 2 at the nth generation (p ,n p2(n)); 2) the difference in frequen-

Rate of Accumulation of Reproductive Isolation by Chromosome Rearrangements

In populations with small or moderate effective population size, in the absence of strong homozygote advantage or drive, chromosomally induced isolation can occur only by the fixation of many weakly underdominant rearrangements. For these populations the minimal speciation time is an increasing function of Ne/u, the ratio of the effective population size and the spontaneous rearrangement rate. With moderate Ne and either strong rearrangement homozygote selective advantage or drive, isolation can occur by fixation of a few strongly underdominant rearrangements. Under these circumstances, speciation occurs at a more rapid rate, provided that such rearrangements occur with sufficient frequency. Finally, for large Ne, only rearrangements which are meiotically driven can become fixed, so that isolation times reflect the average strength of drive. A somewhat counterintuitive result for meiotic drive is that in certain cases the minimal speciation times decrease as the effective population size increases. The implications of these results for stasipatric speciation are that mildly to strongly underdominant rearrangements (i.e., those with h > .01) can contribute to hybrid sterility only in very small populations (Ne < 50) which are likely to have a high extinction probability and suffer inbreeding depression, or in moderate to large populations which have a high spontaneous occurrence rate of driven rearrangements. Even in populations with a high occurrence of driven rearrangements, only those rearrangements with D > h (the drive coefficient) can contribute to increasing hybrid sterility. Thus, the conditions under which the stasipatric model of speciation will operate in a reasonable time are fairly strict and speciation by fixation of a large number of weakly underdominant (cryptic) rearrangements is often more likely.

Corella species in the American Pacific Northwest: distinction of C. inflata Huntsman, 1912 from C. willmeriana Herdman, 1898 (Ascidiacea, Phlebobranchia)

The solitary phlebobranch ascidian Corella inflata, originally described by Huntsman in 1912 but later synonymized with C. willmeriana Herdman, 1898, is now reinstated as a separate species based on comparative morphology of adults, sperm, and tadpoles. Corella inflata is probably largely or completely self-fertilizing, broods its eggs and embryos until well after hatching of the tadpole larvae, and is confined in its distribution to shallow-water (0.1–18 m depth) areas of northwestern Washington and British Columbia. Corella willmeriana appears to be both self- and cross-fertilizing, does not brood its eggs, and is more widespread in distribution, occurring from southern Alaska to southern California and at least as deep as 50 m; it is rarely encountered at the sea surface. The two Corella species differ morphologically from the European Corella parallelogramma and the Japanese C. japonica and C. japonica asamushi. Differential spawning latencies are discussed and the possibility that C. inflata evolved through stasipatric speciation is proposed.

Non-Allopatric Speciation in Animals

Futuyma, D. J., and G. C. Mayer (Department of Ecology and Evolution, State University of New York, Stony Brook, New York 11794 and Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138) 1980. Non-allopatric speciation in animals. Syst. Zool., 29:254-271.-Major recent challenges to the view that animal speciation is usually allopatric are reviewed, and are found unconvincing, either because of their theoretical implausibility or because of insufficient evidence. Special attention is given to the theory of stasipatric speciation, and to purported cases, especially in tephritid fruit flies, of sympatric speciation associated with a shift to a new host. Stasipatric speciation is unlikely under population genetic theory; moreover, chromosome rearrangements probably seldom facilitate speciation. In the tephritid genus Rhagoletis, the archetypal case of sympatric speciation, there is little or no evidence of genetic divergence or of sympatric speciation. The conditions under which host-associated sympatric speciation might occur are so exacting as to be met by very few species. [Rhagoletis; Didymuria; Vandiemenella; Mus; Sceloporus; speciation; sympatric speciation; parapatric speciation; allopatric speciation; chromosome races; isolating mechanisms; host selection.] is in the nature of science that once a position becomes orthodox it should be subjected to criticism .... It does not follow that, because a position is orthodox, it is wrong .... -Maynard Smith, 1976 Until recently, it has been widely accepted that speciation in animals usually proceeds by the differentiation of geographically isolated populations (e.g., Dobzhansky, 1970; Lewontin, 1974; Rosenzweig, 1975; Dobzhansky et al., 1977). The reasons for this belief were developed in extenso by Mayr (1942, 1963, 1970), who argued that other modes of speciation were so unlikely as to be insignificant. Since 1963, there have been several important challenges to this view (e.g., Maynard Smith, 1966; Bush, 1975a; White, 1978a). Because White's (1978a) book, devoted largely to demonstrating the prevalence of non-allopatric speciation, seems likely to have an important impact on the climate of opinion on this topic, it is useful at this time to review the extent to which the theory and evidence amassed since 1963 warrant a major change in our views of animal spe-

MECHANISMS OF FIXATION AND ACCUMULATION OF CENTRIC FUSIONS IN NATURAL POPULATIONS OF MUS MUSCULUS L. I. KARYOLOGICAL ANALYSIS OF A HYBRID ZONE BETWEEN TWO POPULATIONS IN THE CENTRAL APENNINES.

Chromosomal mutations causing partial sterility of the heterozygote are decisive in speciation processes in animals of limited vagility (e.g., Orthoptera and small mammals). In this regard the stasipatric speciation model (production of a heterozygous mutant, fixation in a small colony and subsequent spread of the homozygous mutant population) was proposed (see White, 1968, 1973, 1975, 1977), as opposed to the hypothesis of allopatric speciation (e.g., Key, 1974). Rodents offer several suitable characteristics for this type of study: 1) A variety of chromosomal rearrangements are involved: pericentric inversions, as in Spalax (Soldatovic and Savic, 1967; Wahrman et al., 1969; Lay and Nadler, 1972), Thomomys (Patton and Dingman, 1970), Perognathus (Patton, 1969), Peromyscus (Hsu and Arrighi, 1968) and, above all, centric fusions-and to a lesser extent, fissions-in Spalax (Wahrman et al., 1969; Lay and Nadler, 1972), Thomomys (Thaeler, 1968a; Nevo et al., 1974), Perognathus (Patton, 1969), Gerbillus (Wahrman and Gourevitz, 1973), Acomys (Wahrman and Goitein, 1972), Geomys (Selander et al., 1974) and Leggada (Matthey, 1970). 2) Different steps in the speciation process are evident: among different species of the same genus, among sibling species, among incipient species and among interbreeding populations. In Thomomys (Thaeler, 1968a, 1968b; Patton and Dingman, 1970;

AN INTRODUCTION TO EVOLUTIONARY THEORY

speciation schemes to the measles: all of us become infected when we are young but almost everyone recovers from the disease. I think the situation is different today. No matter what your age, I predict that White's provocative treatment of this subject, combined with the related but really quite different phenomenon of stasipatric speciation, will cause considerable sympatric itching. No one can afford to dismiss as unimportant or exceptional a scheme which appears to be reasonable, genetically impeccable and which is supported by considerable evidence. With modern genetic techniques, judicious use of experiments and sound theoretical formulations, new perspectives on species formation are coming. White suggests that a lot will happen in the next thirty years. Willard Gibbs once wrote: One of the principal objects of theoretical research . . . is to find the point of view from which the subject appears in its greatest simplicity. Realistically, although perhaps unfortunately, speciation processes look exceedingly complex, no matter what the viewpoint.

Chromosomal and morphometric variation in the gekko Diplodactylus vittatus (Gray)

Chromosomal and morphometric analysis has shown that there are five distinct races of Diplodactylus vittatus: 2n = 38 EA, 2n = 34 SA, 2n = 36 WA, 2n = 38 WA A and 2n = 38 WA B. The 2n = 34 SA, and 2n = 36 WA races have evolved independently in Eastern and Western Australia and are interpreted as the products of chromosomal fusions. Each of the five races is morphologically distinguishable, has an extensive and discrete distribution and is chromosomally constant throughout its range; each is regarded as a biological species. The present-day distribution suggests that the fusion races may have evolved by stasipatric speciation.

Morphometric Differentiation in the Viatica Group of Morabine Grasshoppers (Orthoptera, Eumastacidae)

Atchley, W. R., and J. Cheney (Depts. of Biology and Statistics, Texas Tech Univ., Lubbock, Texas, 79409, and Dept. of Genetics, Univ. of Melbourne, Parkville, Victoria, Australia). 1974. Morphometric differentiation in the Viatica group of morabine grasshoppers (Orthoptera, Eumastacidae). Syst. Zool. 23:400-415.-Morphometric differentiation in 10 egg guide characters was examined in 13 chromosomal races and species within the Viatica group of morabine grasshoppers. Considerable divergence was found to have occurred among races within each provisional species and among the various provisional species. The extent of multivariate phenetic overlap among races within a species ranges from 28% between the 2n = 19 and 2n = 17 races of viatica to 13% among the three races of P24. The interracial relationships among the taxa as reflected by canonical variate analyses of the 13 taxa and of the 8 taxa (races pooled) generally gave close correspondence to the known genetic and geographic affinities among the taxa. Factor analysis of the pooled data followed by analysis of variance and a posteriori testing of the factor scores provided evidence of intertaxa differentiation in the shape of the egg guide. Again, a close correspondence between phenetic variation, genetic divergence and ecogeographic separation was noted. Finally, the extent of multivariate phenetic overlap was found to decrease as a function of phylogenetic distance with the phylogenetically most primitive races having the greatest overlap with the other taxa and the most derived forms being phenetically the most discrete. [Morphometric differentiation; multivariate statistics; speciation; morabine grasshoppers.] The Viatica group of morabine grasshoppers (Orthoptera, Eumastacidae) is a cluster of morphologically very uniform species and races occurring in southeastern Australia (Fig. 1). Since these insects are wingless, they exhibit very low vagility and tend to form local colonies between which migration occurs only rarely. The members of the Viatica group have been the subject of critical cytogenetic studies by White and co-workers (White, Carson, and Cheney, 1964; White, et al., 1967; White, et al., 1969; White, 1968 and 1970) who have shown these species to be of considerable importance in speciation theory. According to the interpretation of White, et al. (1967) and White (1968) genetic divergence among several of the included taxa has been achieved by a mechanism which they have referred to as stasipatric speciation. Most of the taxa in this species group, which cannot be assigned to any of the four described genera of Morabinae, are at present taxonomically undescribed. Therefore, most of the taxa are designated with P (for provisional species) and an alphanumerical code, these being the symbols used in the Australian National Insect Collection, Canberra. The morphological uniformity among the provisional species of the Viatica group belies the fact that several have become differentiated into two or more races by combinations of translocations, pericentric inversions and other gross chromosomal rearrangements. In all, 13 distinct taxa are known in the Viatica group and are included in this report. In spite of the considerable evidence available on cytogenetic and ecogeographic separation in these forms, critical studies on morphometric divergence have until recently been lacking. Blackith and Blackith (1969) examined morphometric shape variation in the Viatica group species, but their study was not concerned with measuring

Morphometrics, Evolution, and Cytotaxonomy of Mainland Bats of the Genus Macrotus (Chiroptera: Phyllostomatidae)

Davis, Brent L., and Robert J. Baker (Dept. Biology, and The Museum, Texas Tech University, Lubbock, Texas) 1974. Morphometrics, Evolution, and Cytotaxonomy of Mainland Bats of the Genus Macrotus (Chiroptera: Phyllostomatidae). Syst. Zool. 23:26-39.-Chromosomal data reveal the existence of two mainland species of Macrotus that are parapatric in distribution. Individuals can be identified to species by both chromosomal features and cranial morphology. A stepwise multiple discriminant analysis and canonical variate analysis show that the two species are morphometrically divergent. The distribution of these two taxa is proof that parapatry occurs in species with high vagility. If chromosomal data had not been available or if chromosomal divergence had not accompanied speciation in Macrotus, this unique pattern of distribution probably would not have been detected. Allopatric, stasipatric, and centrifugal speciation are considered in light of the presently available data for Macrotus. [Macrotus; morphometrics; cytotaxonomy.] Geographic variation in chromosome number is rather uncommon among bats of the family Phyllostomatidae. Of the 77 species of phyllostomatids thus far karyotyped (Baker, 1973), four species show geographic variation-Uroderma bilobatum (Baker et al., 1972), Macrotus waterhousii (Nelson-Rees, et al., 1968), and Micronycteris hirsuta and Vampyressa pusilla (Baker et al., 1973). Chromosomal variation in Macrotus, which previously was thought to have two chromosomal races, is of the Robertsonian type (the fundamental number remains constant but the diploid number varies). The cytotype with a diploid number (2N) of 46 and a fundamental number (FN) of 60 has been reported as having a geographic range from Alamos, Sonora, south through Morelos, and Guerrero (Nelson-Rees et al., 1968). The 2N = 40, FN = 60 cytotype had a known distribution from Carbo, Sonora, north to Arizona and California. Alamos (the most northern published record for the 2N = 46 cytotype) is 335 kilometers south of Carbo (the most southern reported locality of the 2N = 40 cytotype). Variation in diploid number, as understood at the outset of this study, was confined to the subspecies M. w. californicus (Anderson, 1968, 1969; Nelson-Rees, et al., 1968). The primary reason for studying the cytogenetics and morphometrics of Macrotus was to determine the mechanism of evolution and the degree of hybridization between the two cytotypes. Robertsonian variation is the most likely mechanism by which a chromosomal change of this nature could occur (Nelson-Rees et al., 1972). If centric fusions or fissions are the mechanism by which the cytotypes evolved, loss or gain of genetic material probably would have occurred in the heterochromatic region (Jackson, 1971) and a cross between the two cytotypes should be successful, at least on a chromosomal basis of homology. Because of the relative ease with which specimens of Macrotus can be obtained and because the species is chromosomally variable, detailed studies were initiated. Several possibilities are obvious from the outset. First, the two cytotypes could be freely interbreeding. If this were the case, the genetics of the system could be examined by using chromosome number, morphology, and meiotic studies. Second, the two could be sympatric and produce no hybrids. Third, the two could have allopatric distributions. Our studies have revealed a parapatric distribution (as discussed for mammals by Vaughan, 1967) with no phenetic inter-

The Concept of Stasipatric Speciation

The Concept of Stasipatric Speciation