Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Fathoming tropical biodiversity: the continuing discovery of Neotropical mammals

Fathoming tropical biodiversity: the continuing discovery of Neotropical mammals Introduction It remarkable that scientists do not know — even to the nearest order of magnitude — the number of species that share our planet ( May, 1988, 1990 ; Stork, 1999 ). Lay peoples seem willing to concede this ignorance, assuming that it is based mainly on bacteria, nematodes or canopy‐inhabiting beetles simply too diverse for scientists to enumerate. Most find it shocking that this sad state of ignorance on questions of biodiversity also applies to our nearest relatives, the mammals. One has only to examine changing estimates over the past half century of the global number of mammal species to appreciate how tenuous such appraisals actually are ( Fig. 1 ). The rate at which new mammal species are being described is about 10 times the rate at which new bird species are described ( Patterson, 2000 ). Is it possible to simply extrapolate from known diversity patterns to estimate the number of species? A survey of various biases in the Neotropical mammal fauna suggest that such extrapolations are destined to be highly imprecise. 1 Estimates of global mammalian species richness since 1950. Sources: 1, Hvass (1961) ; 2, Davis & Golley (1963) ; 3, Morris (1965) ; 4, Matthews . (1969) ; 5, Vaughan (1972) ; 6, Grzimek (1975) ; 7, Corbet & Hill (1980) ; 8, Honacki . (1982) ; 9, Boitani & Bartoli (1982) ; 10, MacDonald (1984) ; 11, Corbet & Hill (1986) ; 12, Corbet & Hill (1991) ; 13, Wilson & Reeder (1993) ; 14, Nowak (1999) . A linear fit is provided by the equation y = −44809.55 + 24.754 x , indicating 25 species are added each year, but the relationship is hardly linear. METHODS The Neotropical Region is home to roughly a quarter of the world’s mammal fauna ( Nowak, 1999 ), and this fauna offers some understanding of the process by which a diverse and geographically differentiated biota has become known systematically. For the past decade, I have maintained a database of 5100 scientific names proposed for and applied to Neotropical mammals since the time of Linnaeus ( Patterson, 1994, 1996 ). This bibliographic and specimen database details the precise sequence in which all known species became recognized, at least in a formal taxonomic sense, via dates of publication. It also specifies geographical components of this knowledge via the localities of type specimens for these names. I suggest that patterns defined for this set of species will prove to characterize mammals in other zoogeographic regions, and may also apply, with varying degrees of fidelity, to other groups of organisms too poorly known to establish these relationships (cf. Chapman, 1999 ). Of course, the value of these extrapolations must await corroboration by independent analyses — mammals differ from wind‐dispersed weeds or planktonic forms in ways that suggest there may be major differences in associated distributional patterns. RESULTS AND DISCUSSION Perhaps the most obvious trend in the discovery of Neotropical mammals has been in terms of body size: larger species of mammals tend to have been discovered earlier than smaller ones ( Gaston, 1991 ; Gaston & Blackburn, 1994 ; Patterson, 1994 ). Combing original descriptions and secondary literature, especially Eisenberg & Redford (1989, 1999) , Emmons & Feer (1990) , Redford & Eisenberg (1992) , Silva & Downing (1995) , Patterson . (1996) , Reid (1997) and Nowak (1999) , published body mass records were located for 793 species, tallying means or midpoints of ranges. The masses of 245 additional species were estimated using correlations of mass with linear dimensions (head‐and‐body length for terrestrial species and forearm length for bats) among congeneric species and species within some closely allied genera. In this fashion, masses were included for 1040 of the 1197 terrestrial Neotropical mammal species currently recognized (lacking were masses for 84 rodents, 23 bats, 14 opossums, 11 primates, 10 carnivores, six shrews, six ungulates, two edentates and a tapir). Masses were plotted against the year each species was scientifically described Figure 2 . 2 Varying body mass of Neotropical mammals as a function of publication date. The plotted relationship is y = 192.523 − 0.1964x + 0.000051x 2 and all three coefficients are highly significant ( P < 0.001). Overall, the relationship between body mass ( y ) and year of publication ( x ) can be described by a second‐order polynomial, y = 192.523 − 0.1964 x + 0.000051 x 2 , where all three regression coefficients are highly significant ( P << 0.0001). This relationship indicates that, at the time of the 10th edition of Systema Naturae ( Linnaeus, 1758 ), the average Neotropical mammals known weighed nearly 2.5 kg. By 1800 the expected size of species being described had fallen to 494 g, reaching 56 g in 1900 and 45 g by 1940. In 2000, predicted mass of new Neotropical mammal species had actually climbed to 65.6 g. The sharp early decline in body sizes of mammals parallels a comparable decline among British beetles over the same period, but beetles show no signs of a quadratic relationship ( Gaston, 1991 ). The recent upswing in body size of mammal species now being described is at least partially attributed to a spate of discoveries of new primate species, averaging one per year over the last decade ( Patterson, 2000 ; Rylands ., 2000 ). If most new species are small, it is because most species of mammals are small — Pine (1994) has estimated that three species of large mammals are described each decade. Although year of publication accounts for significant variation in body mass, this correlation is unlikely to be a direct effect. It is true that larger mammals are more conspicuous and more useful in terms of food, clothing, and/or artefacts, making them harder to overlook. On the other hand, they typically have lower abundances and densities ( Damuth, 1987 ; Blackburn , 1993 ), which make them harder to sample. However, larger species also tend to have larger home ranges ( Damuth & MacFadden, 1990 ) and geographical ranges ( Rapoport, 1982 ; Brown & Maurer, 1987 ), which increase the probability that a given sampling locality will be included in a species’ geographical range. For example, significantly higher fractions of South America’s edentate and carnivore faunas occur in Surinam and Cayenne, French Guiana (provenances of many early names) than of rodents or marsupials — those groups initially appeared most diverse to early cataloguers, until the diversity of rodents eclipsed all other groups of mammals in 1820 ( Patterson, 1994 ). Unfortunately, quantitative estimates of range size are lacking for examining this variable as covariate of body size. Geographic biases are also known to shape accumulation of new species. In some poorly known plant or invertebrate groups, ‘hotspots’ of species richness may correspond to sampling points for a single productive worker or the coverage of their latest monograph ( Steyskal, 1965 ; White, 1975 ; Gaston & May, 1992 ). This obvious source of heterogeneity does not now apply to Neotropical mammals. However, temporal and spatial heterogeneity in sampling is still evident: there is a discernible tendency for species described in the first century following Linnaeus to originate in areas along coastlines or navigable waterways, and for more recently described species to hail from interior basins and cordilleras ( Patterson, 1994 ). Such efforts have led to uneven sampling of South America’s principal regions of endemism. A flurry of new species and genera now being described from species‐poor arid habitats proves that earlier diversity estimates were grossly incomplete (e.g. Braun & Mares, 1995 ; Mares ., 2000 ). Both sorts of heterogeneity invalidate attempts to estimate diversity by way of asymptotes. Changing systematic practices and concepts also influence species accumulation. During the height of colonial expansion by European and later North American powers, scientists devoted their careers to the cataloguing of collections and the naming of novelties. Without peer in this regard was Oldfield Thomas of the British Museum, who between 1880 and 1929 described no fewer than 848 Neotropical mammals (and some 3000 mammals worldwide; Hill, 1990 ). No one today has access to a comparable network of collectors, nor is any modern worker encouraged to be so devoted to such basic, descriptive research; indeed, some systematists clearly disdain this role ( Renner & Ricklefs, 1994 ). Small wonder that the average species described a century ago was based on a specimen collected only 2 years earlier; today, this lag between collecting a new Neotropical mammal and formally describing it has grown exponentially to more than 14 years ( Patterson, 1994 ). Mayr (1969) noted that classifications pass through various stages of maturation, with corresponding activities and responsibilities of allied systematists: certainly the emphasis of 19th and early 20th century workers was on ‘alpha taxonomy’— the description of new species and their arrangement into genera — whereas modern workers devote far more attention to ‘beta taxonomy’— assessing their inter‐relationships. The political and socio‐economic frameworks for the countries of Central and South America have changed dramatically as mammalian classifications have matured. Since biodiversity inventories are partly directed by and responsive to these pressures, it is not surprisingly that the locus of alpha taxonomic research has also shifted. Tracing the affinities of taxonomists by the museums in which they deposited type specimens (and where this information is lacking by the institutional periodicals in which they published their descriptions), a clear trend emerges. The early phases of species description were dominated by continental European authors and museums, whose roles and activity were succeeded in turn by people and institutions in Great Britain, North America, and finally in South America ( Fig. 3 ). This trend is sufficiently well developed that South American museums and authors dominate the description of species in the final quarter of the 20th century. While partly attributable to laws requiring that newly collected holotypes be deposited in national collections, this trend reflects the growing activity and involvement of scientists in biodiversity countries in basic descriptive research. 3 The shifting locus of descriptive taxonomy. Since Linnaeus, museums and authors in continental Europe, Great Britain, North America and South America have taken turns in leading the inventory of Neotropical mammals. ‘Lit.’ refers to species descriptions that were based on bibliographic sources rather than type specimens, an inadmissible practice today. Although the roles of continental, British and North American scientists and institutions in describing biodiversity have shifted over time, all share an enduring commitment to biodiversity inventories through their maintenance of encyclopedic collections and the support of revisionary studies (‘beta taxonomy’) on them. As one prominent example, few new Neotropical mammals are described without prior comparisons directly or indirectly with the type collections of the Natural History Museum, London, U.K. Perhaps the most unexpected pattern to emerge from recent trends in the description of Neotropical mammals has to do with taxonomic synonyms. Synonyms arise when two or more different names are applied to the same species, a post facto evaluation that is possible only during the course of taxonomic revisions. Thus, one could argue that any current estimate of diversity is an overestimate that must be discounted by the number of names that will eventually be shown to be synonyms (e.g. Gaston & Mound, 1993 ). In fact, most newly recognized species of Neotropical mammals were named previously, and had been treated as synonyms until recent collections or analyses proved them to be distinct. For every newly discovered species to be trapped or captured in the forests and fields of the Neotropics, three more are discovered in the drawers of museum collections or on the benches of molecular biology laboratories ( Patterson, 1996 ). For Neotropical mammals, three times as many names came out of synonymy since 1982 as became synonyms. Continued morphological study, higher‐resolution genetic analyses and a shift toward a phylogenetic species concept (and away from polytypic species) seem jointly responsible for this trend to resurrect synonyms ( Patterson, 2000 ; see also Williamson, 1999 ). Although these patterns in species discovery might permit more accurate extrapolations, there are no real shortcuts to describing and analysing biodiversity. No numerical or spatial extrapolations will obviate the expensive and time‐consuming steps of fully documenting and describing biodiversity. Diversity scientists of all stripes must acknowledge that their work rests on taxonomic foundations; in turn, these footings depend on the scientific collections that document biodiversity and the systematic concepts that are used to interpret them. Living terrestrial mammals have attracted a disproportionate degree of systematic attention, constituting a system of checks‐and‐balances on our knowledge of their diversity. If such powerful biases pervade their discovery, our understanding of diversity among invertebrates, marine forms, and in the fossil record must be even more incomplete than commonly acknowledged. Acknowledgments I thank colleagues and co‐workers too numerous to mention for enlightening discussions of descriptive systematics and its role in the biodiversity crisis. Jack Fooden, Scott Lidgard, Dave Richardson, Doug Stotz and an anonymous reviewer offered instructive suggestions on expressing these relationships. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Diversity and Distributions Wiley

Fathoming tropical biodiversity: the continuing discovery of Neotropical mammals

Diversity and Distributions , Volume 7 (4) – Jul 1, 2001

Loading next page...
 
/lp/wiley/fathoming-tropical-biodiversity-the-continuing-discovery-of-QDIG6U22D0

References (47)

Publisher
Wiley
Copyright
Copyright © 2001 Wiley Subscription Services, Inc., A Wiley Company
ISSN
1366-9516
eISSN
1472-4642
DOI
10.1111/j.1472-4642.2001.00109.x
Publisher site
See Article on Publisher Site

Abstract

Introduction It remarkable that scientists do not know — even to the nearest order of magnitude — the number of species that share our planet ( May, 1988, 1990 ; Stork, 1999 ). Lay peoples seem willing to concede this ignorance, assuming that it is based mainly on bacteria, nematodes or canopy‐inhabiting beetles simply too diverse for scientists to enumerate. Most find it shocking that this sad state of ignorance on questions of biodiversity also applies to our nearest relatives, the mammals. One has only to examine changing estimates over the past half century of the global number of mammal species to appreciate how tenuous such appraisals actually are ( Fig. 1 ). The rate at which new mammal species are being described is about 10 times the rate at which new bird species are described ( Patterson, 2000 ). Is it possible to simply extrapolate from known diversity patterns to estimate the number of species? A survey of various biases in the Neotropical mammal fauna suggest that such extrapolations are destined to be highly imprecise. 1 Estimates of global mammalian species richness since 1950. Sources: 1, Hvass (1961) ; 2, Davis & Golley (1963) ; 3, Morris (1965) ; 4, Matthews . (1969) ; 5, Vaughan (1972) ; 6, Grzimek (1975) ; 7, Corbet & Hill (1980) ; 8, Honacki . (1982) ; 9, Boitani & Bartoli (1982) ; 10, MacDonald (1984) ; 11, Corbet & Hill (1986) ; 12, Corbet & Hill (1991) ; 13, Wilson & Reeder (1993) ; 14, Nowak (1999) . A linear fit is provided by the equation y = −44809.55 + 24.754 x , indicating 25 species are added each year, but the relationship is hardly linear. METHODS The Neotropical Region is home to roughly a quarter of the world’s mammal fauna ( Nowak, 1999 ), and this fauna offers some understanding of the process by which a diverse and geographically differentiated biota has become known systematically. For the past decade, I have maintained a database of 5100 scientific names proposed for and applied to Neotropical mammals since the time of Linnaeus ( Patterson, 1994, 1996 ). This bibliographic and specimen database details the precise sequence in which all known species became recognized, at least in a formal taxonomic sense, via dates of publication. It also specifies geographical components of this knowledge via the localities of type specimens for these names. I suggest that patterns defined for this set of species will prove to characterize mammals in other zoogeographic regions, and may also apply, with varying degrees of fidelity, to other groups of organisms too poorly known to establish these relationships (cf. Chapman, 1999 ). Of course, the value of these extrapolations must await corroboration by independent analyses — mammals differ from wind‐dispersed weeds or planktonic forms in ways that suggest there may be major differences in associated distributional patterns. RESULTS AND DISCUSSION Perhaps the most obvious trend in the discovery of Neotropical mammals has been in terms of body size: larger species of mammals tend to have been discovered earlier than smaller ones ( Gaston, 1991 ; Gaston & Blackburn, 1994 ; Patterson, 1994 ). Combing original descriptions and secondary literature, especially Eisenberg & Redford (1989, 1999) , Emmons & Feer (1990) , Redford & Eisenberg (1992) , Silva & Downing (1995) , Patterson . (1996) , Reid (1997) and Nowak (1999) , published body mass records were located for 793 species, tallying means or midpoints of ranges. The masses of 245 additional species were estimated using correlations of mass with linear dimensions (head‐and‐body length for terrestrial species and forearm length for bats) among congeneric species and species within some closely allied genera. In this fashion, masses were included for 1040 of the 1197 terrestrial Neotropical mammal species currently recognized (lacking were masses for 84 rodents, 23 bats, 14 opossums, 11 primates, 10 carnivores, six shrews, six ungulates, two edentates and a tapir). Masses were plotted against the year each species was scientifically described Figure 2 . 2 Varying body mass of Neotropical mammals as a function of publication date. The plotted relationship is y = 192.523 − 0.1964x + 0.000051x 2 and all three coefficients are highly significant ( P < 0.001). Overall, the relationship between body mass ( y ) and year of publication ( x ) can be described by a second‐order polynomial, y = 192.523 − 0.1964 x + 0.000051 x 2 , where all three regression coefficients are highly significant ( P << 0.0001). This relationship indicates that, at the time of the 10th edition of Systema Naturae ( Linnaeus, 1758 ), the average Neotropical mammals known weighed nearly 2.5 kg. By 1800 the expected size of species being described had fallen to 494 g, reaching 56 g in 1900 and 45 g by 1940. In 2000, predicted mass of new Neotropical mammal species had actually climbed to 65.6 g. The sharp early decline in body sizes of mammals parallels a comparable decline among British beetles over the same period, but beetles show no signs of a quadratic relationship ( Gaston, 1991 ). The recent upswing in body size of mammal species now being described is at least partially attributed to a spate of discoveries of new primate species, averaging one per year over the last decade ( Patterson, 2000 ; Rylands ., 2000 ). If most new species are small, it is because most species of mammals are small — Pine (1994) has estimated that three species of large mammals are described each decade. Although year of publication accounts for significant variation in body mass, this correlation is unlikely to be a direct effect. It is true that larger mammals are more conspicuous and more useful in terms of food, clothing, and/or artefacts, making them harder to overlook. On the other hand, they typically have lower abundances and densities ( Damuth, 1987 ; Blackburn , 1993 ), which make them harder to sample. However, larger species also tend to have larger home ranges ( Damuth & MacFadden, 1990 ) and geographical ranges ( Rapoport, 1982 ; Brown & Maurer, 1987 ), which increase the probability that a given sampling locality will be included in a species’ geographical range. For example, significantly higher fractions of South America’s edentate and carnivore faunas occur in Surinam and Cayenne, French Guiana (provenances of many early names) than of rodents or marsupials — those groups initially appeared most diverse to early cataloguers, until the diversity of rodents eclipsed all other groups of mammals in 1820 ( Patterson, 1994 ). Unfortunately, quantitative estimates of range size are lacking for examining this variable as covariate of body size. Geographic biases are also known to shape accumulation of new species. In some poorly known plant or invertebrate groups, ‘hotspots’ of species richness may correspond to sampling points for a single productive worker or the coverage of their latest monograph ( Steyskal, 1965 ; White, 1975 ; Gaston & May, 1992 ). This obvious source of heterogeneity does not now apply to Neotropical mammals. However, temporal and spatial heterogeneity in sampling is still evident: there is a discernible tendency for species described in the first century following Linnaeus to originate in areas along coastlines or navigable waterways, and for more recently described species to hail from interior basins and cordilleras ( Patterson, 1994 ). Such efforts have led to uneven sampling of South America’s principal regions of endemism. A flurry of new species and genera now being described from species‐poor arid habitats proves that earlier diversity estimates were grossly incomplete (e.g. Braun & Mares, 1995 ; Mares ., 2000 ). Both sorts of heterogeneity invalidate attempts to estimate diversity by way of asymptotes. Changing systematic practices and concepts also influence species accumulation. During the height of colonial expansion by European and later North American powers, scientists devoted their careers to the cataloguing of collections and the naming of novelties. Without peer in this regard was Oldfield Thomas of the British Museum, who between 1880 and 1929 described no fewer than 848 Neotropical mammals (and some 3000 mammals worldwide; Hill, 1990 ). No one today has access to a comparable network of collectors, nor is any modern worker encouraged to be so devoted to such basic, descriptive research; indeed, some systematists clearly disdain this role ( Renner & Ricklefs, 1994 ). Small wonder that the average species described a century ago was based on a specimen collected only 2 years earlier; today, this lag between collecting a new Neotropical mammal and formally describing it has grown exponentially to more than 14 years ( Patterson, 1994 ). Mayr (1969) noted that classifications pass through various stages of maturation, with corresponding activities and responsibilities of allied systematists: certainly the emphasis of 19th and early 20th century workers was on ‘alpha taxonomy’— the description of new species and their arrangement into genera — whereas modern workers devote far more attention to ‘beta taxonomy’— assessing their inter‐relationships. The political and socio‐economic frameworks for the countries of Central and South America have changed dramatically as mammalian classifications have matured. Since biodiversity inventories are partly directed by and responsive to these pressures, it is not surprisingly that the locus of alpha taxonomic research has also shifted. Tracing the affinities of taxonomists by the museums in which they deposited type specimens (and where this information is lacking by the institutional periodicals in which they published their descriptions), a clear trend emerges. The early phases of species description were dominated by continental European authors and museums, whose roles and activity were succeeded in turn by people and institutions in Great Britain, North America, and finally in South America ( Fig. 3 ). This trend is sufficiently well developed that South American museums and authors dominate the description of species in the final quarter of the 20th century. While partly attributable to laws requiring that newly collected holotypes be deposited in national collections, this trend reflects the growing activity and involvement of scientists in biodiversity countries in basic descriptive research. 3 The shifting locus of descriptive taxonomy. Since Linnaeus, museums and authors in continental Europe, Great Britain, North America and South America have taken turns in leading the inventory of Neotropical mammals. ‘Lit.’ refers to species descriptions that were based on bibliographic sources rather than type specimens, an inadmissible practice today. Although the roles of continental, British and North American scientists and institutions in describing biodiversity have shifted over time, all share an enduring commitment to biodiversity inventories through their maintenance of encyclopedic collections and the support of revisionary studies (‘beta taxonomy’) on them. As one prominent example, few new Neotropical mammals are described without prior comparisons directly or indirectly with the type collections of the Natural History Museum, London, U.K. Perhaps the most unexpected pattern to emerge from recent trends in the description of Neotropical mammals has to do with taxonomic synonyms. Synonyms arise when two or more different names are applied to the same species, a post facto evaluation that is possible only during the course of taxonomic revisions. Thus, one could argue that any current estimate of diversity is an overestimate that must be discounted by the number of names that will eventually be shown to be synonyms (e.g. Gaston & Mound, 1993 ). In fact, most newly recognized species of Neotropical mammals were named previously, and had been treated as synonyms until recent collections or analyses proved them to be distinct. For every newly discovered species to be trapped or captured in the forests and fields of the Neotropics, three more are discovered in the drawers of museum collections or on the benches of molecular biology laboratories ( Patterson, 1996 ). For Neotropical mammals, three times as many names came out of synonymy since 1982 as became synonyms. Continued morphological study, higher‐resolution genetic analyses and a shift toward a phylogenetic species concept (and away from polytypic species) seem jointly responsible for this trend to resurrect synonyms ( Patterson, 2000 ; see also Williamson, 1999 ). Although these patterns in species discovery might permit more accurate extrapolations, there are no real shortcuts to describing and analysing biodiversity. No numerical or spatial extrapolations will obviate the expensive and time‐consuming steps of fully documenting and describing biodiversity. Diversity scientists of all stripes must acknowledge that their work rests on taxonomic foundations; in turn, these footings depend on the scientific collections that document biodiversity and the systematic concepts that are used to interpret them. Living terrestrial mammals have attracted a disproportionate degree of systematic attention, constituting a system of checks‐and‐balances on our knowledge of their diversity. If such powerful biases pervade their discovery, our understanding of diversity among invertebrates, marine forms, and in the fossil record must be even more incomplete than commonly acknowledged. Acknowledgments I thank colleagues and co‐workers too numerous to mention for enlightening discussions of descriptive systematics and its role in the biodiversity crisis. Jack Fooden, Scott Lidgard, Dave Richardson, Doug Stotz and an anonymous reviewer offered instructive suggestions on expressing these relationships.

Journal

Diversity and DistributionsWiley

Published: Jul 1, 2001

There are no references for this article.