Download A View from the Tropics - Oxford Academic

2005
POINTS OF VIEW
689
Syst. Biol. 54(4):689–693, 2005
c Society of Systematic Biologists
Copyright ISSN: 1063-5157 print / 1076-836X online
DOI: 10.1080/10635150590950308
Species Concepts, Species Boundaries and Species Identification:
A View from the Tropics
R OHINI B ALAKRISHNAN
Centre for Ecological Sciences, Indian Institute of Science, Bangalore, India;
E-mail: [email protected]
The species has been treated as a fundamental unit in
biology (Hull, 1977) and, more recently, in biodiversity
conservation (Sites and Crandall, 1997). Almost all studies in biology, whether at the level of molecules, cells, individuals or populations, are typically referenced to the
level of the species. In the field of conservation biology,
assessments of biodiversity are made at the level of the
species: typical criteria include species richness, numbers of endemic species, and the number or presence of
endangered species in given areas (Myers et al. 2000).
The accurate identification of species is crucial both to
research in all areas of biology and to biodiversity conservation. It is therefore surprising that, in the field of systematics, species are currently used mostly as terminal
taxa in the reconstruction of phylogenetic trees, whereas
the methods by which they are delimited and identified
receive scant attention (Wiens and Penkrot, 2002).
S PECIES I DENTIFICATION: PRACTICAL PROBLEMS
The taxonomic classification and identification of
species is an endeavour with a long and complicated history (Gilmour, 1951; Ridley, 1986). The Linnaean system
provided a successfully established and universally accepted code of classification and nomenclature that has
served us well over centuries. Whereas the development of these formal rules, together with the existence
of voucher specimens (holotypes) in museum collections
for verification of species identity is of undeniable value,
species identification still remains a difficult problem for
practising taxonomists, particularly in tropical countries.
The reasons for this are manifold and are briefly outlined
below.
The first reason is the inadequacy of currently available taxonomic keys: on the Indian subcontinent, for example, identification keys for most groups of invertebrates have not been updated since the publication of
the ‘Fauna of British India’ series of volumes in the mid1900s. Whereas many of the monographs in the above series are monumental works, the taxonomic keys are often
not good enough to unambiguously identify specimens
to the species level. The reasons for this difficulty include
imprecise character definition and a paucity of illustrations to effectively convey the exact nature of the defined
characters. These, combined with completely hierarchical, dichotomous key structures, do not allow unambiguous identification to the species level. Further, the species
descriptions are not exhaustive and often suffer from an
inconsistent inclusion of characters, even among descriptions of very similar species.
The second reason is the lack of an objective criterion for defining an individual specimen as the standard
reference holotype for a species: holotypes have traditionally been designated by taxonomists without an objective argument and sometimes without examination
of a sufficient number of specimens to take into account
interindividual variability in different characters. For example, the discrimination between some species of crickets (Chopard, 1969) is made on the basis of differences in
the number of tibial spines. There were several cases that
I examined, however, where the difference in spine number between the left and right tibiae of the same specimen exceeded or equaled the variation in spine numbers between the designated species. In the absence of
other distinguishing characters in the keys, it was essentially impossible to assign such specimens to a given
species.
These problems are further compounded for taxonomists in tropical countries by the inaccessibility of
both the taxonomic literature on previous descriptions
and the reference specimens or holotypes of their local
fauna and flora. These are largely available, for historical
reasons, only in museums in Europe and North America.
Added to this is the worldwide decline in the number
of professional taxonomists, particularly for invertebrate
animal groups (Gaston and May, 1992). The impact of
this is becoming severe, with the increase in numbers
of comparative studies in biology that require accurate
species identification. All of the above problems are wellrecognized and have been the subject of much discussion
in recent years (Gaston and May, 1992; Godfray, 2002;
Wheeler et al. 2004).
S PECIES I DENTIFICATION: CONCEPTUAL AND
M ETHODOLOGICAL PROBLEMS
The process of allocating individuals to a given species
obviously depends on the criteria by which species are
defined and delimited, which are in turn determined by
the concept of what a species is. The species concept
continues to be the subject of much debate (Cracraft,
2000) and, until there evolves a consensus, there can
be no well-defined universal criteria by which species
may be delimited or identified. In the following paragraphs, I therefore examine some of the major species
690
SYSTEMATIC BIOLOGY
concepts and methods of species delimitation implied by
them.
Classical taxonomists classified individuals as members of a species based on a suite of shared morphological characters that were diagnostic and differentiated
them from other such morphologically defined groups.
In recent times, behavioral and ecological characters
have also been increasingly used. Numerical taxonomy
shares the concept of a morphological (or rather, phenetic) species and differs from classical taxonomy mainly
in its methodology: the use of a large number of phenotypic characters, rather than only diagnostic ones, with
equal a priori weighting, and in its emphasis on species
as clusters derived from measures of overall similarity
(Sneath and Sokal, 1973). The phenetic species concept is
practical and applicable to all taxa and therefore of high
operational value.
The biological species concept of Dobzhansky (1937)
and Mayr (1942) defines species based on their ability to
interbreed: species are considered as “natural” entities
distinguishable from other species by the criterion of reproductive isolation and not overall phenotypic similarity. The biological species concept has been criticised for
several reasons, including its lack of universality (for example, inapplicability to asexual taxa: Mishler and Theriot, 2000; Wheeler and Platnick, 2000), lack of a temporal
dimension (Willmann and Meier, 2000), logical fallacy,
and the difficulty of applying it to allopatric populations
(Mallet, 1995). Although the biological species concept
provides an unambiguous criterion for differentiating
and delimiting species, one of the major problems is
the practical impossibility of ascertaining reproductive
isolation between large numbers of populations in the
wild. As a result, most taxonomists, even those that
accept the biological species concept, continue to use
morphology and other phenotypic characters in order to
delineate species boundaries.
Phylogenetic species concepts are increasingly being
proposed as alternatives to the biological species concept. These concepts also have both philosophical and
practical problems: one of the frequently raised philosophical problems is the validity of a phylogenetic approach to units at and below the species level (Mallet,
1995; Willmann and Meier, 2000). Even assuming this
validity, there are several problems with species delimitation and identification using a phylogenetic approach. The various phylogenetic species concepts (de
Quieroz, 1998) may be classified (following Baum and
Donoghue, 1995) into two major types: character-based
and lineage-based.
The character-based species concepts define species
as “the smallest aggregation of populations (sexual)
or lineages (asexual) diagnosable by a unique combination of character states in comparable individuals
(semaphoronts)” (Nixon and Wheeler, 1990). This “pattern cladistic” species concept has both philosophical
and practical flaws that have been convincingly argued
by Baum and Donoghue (1995). The major practical difficulties include determining whether shared traits have
attained fixation in the population (Wiens and Servedio,
VOL. 54
2000), determining how many diagnostic traits to consider and the inapplicability of this approach to most
continuously varying quantitative traits (Willmann and
Meier, 2000).
Lineage-based species concepts include the Hennigian
species concept (itself rooted strongly in the idea of reproductive isolation) which views species as “a temporal series of populations connecting two speciation
events” (quoted from Willmann and Meier, 2000). Operationally, the Hennigian species concept is essentially
a more restricted form of the biological species concept
that can only be tested by reproductive isolation together
with a phylogenetic reconstruction. The “genealogical
species concept” of Baum and Donoghue (1995) defines
a species as a “basal group of organisms all of whose
genes coalesce more recently with each other than with
those of any organism outside the group.” Operationally,
this species concept would involve tracing the genealogy of several different alleles and would require extensive molecular analysis of a large number of genetic loci,
including nuclear and mitochondrial genes. Although
philosophically and methodologically the most attractive of the phylogenetic concepts, it would be highly impractical, time-consuming, and prohibitively expensive
if all species of all taxonomic groups were required to
be delimited in this fashion, particularly in the speciesrich and financial resource-limited environment of the
tropics. In this context, it is worth noting that seven of
the nine methods of delimiting species boundaries that
were recently reviewed by Sites and Marshall (2003)
require molecular data, which would be impractical
for species identification in the assessment of tropical
biodiversity.
In recent years, there have been several attempts to
delimit species boundaries based on phylogenetic reconstructions, mostly using mitochondrial DNA haplotypes. Although some of these studies have revealed
concordance with species defined by classical taxonomy,
others have not (Wiens and Penkrot, 2002). More seriously, there have been observations of discord between
phylogenetic trees or phylogenetic species boundaries
based on morphological versus molecular characters,
character versus lineage based reconstructions (Wiens
and Penkrot, 2002) or mitochondrial versus nuclear DNA
(Shaw, 2002). The ability of mitochondrial DNA sequences to accurately reflect species boundaries and
species histories is being increasingly questioned (Sites
and Crandall, 1997; Puorto et al., 2001; Shaw, 2002).
Avise and Wollenberg (1997) have provided a
thoughtful exposition of the connections between population genetics, gene genealogies, and reproductive
isolation. They have pointed out that reproductive
isolation is an important component of all lineagebased phylogenetic species concepts: it provides the
tokogenetic-phylogenetic border of Hennigian cladism,
the restriction or elimination of gene flow between
populations that results in the “independently evolving
lineages” of the evolutionary species concept (Wiley
and Mayden, 2000), and is the basis of the “deep” and
concordant genealogical splits that are the basis of the
2005
POINTS OF VIEW
genotypic exclusivity criterion of the “genealogical
species concept.”
One may then argue that, for sexually reproducing
taxa, reproductive isolation does provide a sound criterion for delimiting species. Of course, several species
maintain their identities in the face of a high level of gene
flow between them (Mallet, 1995). Reproductive isolation may thus be looked upon as a sufficient but not
necessary condition for delimiting species boundaries:
where it does exist, it is likely to unambiguously delimit
species.
If one accepts reproductive isolation as a sufficient
criterion for delimiting species, but is constrained to
delineate species boundaries using morphology, it becomes imperative to examine the concordance between
morphological and “biological” species boundaries. This
is relatively easy in certain groups such as crickets
(Order Orthoptera, Family Gryllidae) because the calling songs of cricket species are often reliable indicators of
reproductively isolated populations (Shaw, 1999). Concordance in the species boundaries provided by song
and morphological characters would imply that phenetic clusters based on morphology correctly reflect the
species boundaries defined by reproductive isolation.
This then validates the use of phenetic clustering and
ordination techniques (Sneath and Sokal, 1973) on morphological and/or song data to unambiguously classify
and identify cricket species. For taxa that do not possess
such behavioural “signatures” of reproductive isolation,
the species boundaries implied by phenetic clustering or
ordination will have to be compared with those implied
by genetic distances (Sites and Marshall, 2003; Hebert
et al., 2003).
T HE FUTURE: A PLURALISTIC APPROACH?
I have illustrated how both conceptual and practical problems have resulted in the lack of accurate and
accessible species identification keys. The philosophical debate on the validity of different species concepts
is unlikely to be resolved in the near future. One may
even dispute the wisdom of imposing a universal species
concept on philosophical or conceptual grounds alone,
given that systematics has two related, yet different,
aims: to reconstruct the historical paths of evolution
and to serve as a frame of reference for all studies in
biology. At the level of the species, at least, the two
may not be easily reconciled. A universal species concept should not be imposed without careful consideration of both objectives: a concept that is philosophically
sound may well make species identification practically
impossible.
Testing the concordance between species boundaries
implied by different species concepts and using different data sets should be a prioritized research theme in
systematics. Concordance between boundaries obtained
using different species concepts or different data sets
within a particular taxon, say an order or family, would,
in the context of that taxon, relegate the debate on species
concepts to the realm of theory and philosophy and enable the construction of identification keys. Currently,
691
there are too few studies on the problem of species delimitation to know whether there are taxon-specific differences in levels of concordance or discordance of species
boundaries implied by different data sets and different
species concepts. The majority of studies examining this
question have been carried out on reptiles, where discordant species boundaries appear to be common (Sites and
Crandall, 1997; Puorto et al., 2001; Wiens and Penkrot,
2002). It is quite possible that this will not be the case for
insect taxa, where distinct and stereotyped genitalic and
behavioral characters are often good indicators of reproductive isolation (Otte, 1994; Shaw, 1999 for crickets).
The recent demonstration (Hebert et al., 2003, 2004) of
the efficacy of DNA “barcoding” using the cytochrome c
oxidase I (COI) gene in correctly identifying specimens
of a large number of species of birds and moths provides hope for an unambiguous and standard method of
species identification at the molecular level. In addition,
the studies by Hebert et al. (2003, 2004) also demonstrate,
for North American birds and for three moth families, the
concordance in species boundaries implied by molecular phylogenetic versus classical taxonomic methods
based on morphological and behavioral characters. If
this turns out to be the case for a large number of taxa
and for groups of closely related species, it opens up
the possibility to create databases that allow identification using independent data sets. For example, parallel
identification keys based on molecular, morphological
and/or behavioral data may be developed. Even if DNA
barcoding is highly accurate for species identification,
it may be too expensive to be used in biodiversity surveys, where the number of specimens to be identified can
be very large. Further, field identification and noninvasive sampling, two increasingly important requirements
in biodiversity surveys, are not always possible using
molecular methods. Field identification methods such
as acoustic sampling (Riede, 1993) for taxa such as birds,
frogs, and crickets, and visual sampling based on diagnostic morphological characters are rapid and inexpensive and can be developed to be accurate. Identification
keys using nonmolecular data will therefore remain crucial to the process of species identification, particularly
in the tropics.
The problem of accessibility to taxonomic information
is beginning to be addressed via the development of
Internet-accessible taxonomic databases (Godfray, 2002;
Mallet and Wilmott, 2003). For example, in the case of
orthopteran insects, the Orthoptera Species File Online
(Otte and Naskrecki, 1997) is an invaluable first step in
this direction. Nonetheless, it is not adequate for the purpose of species identification. What are urgently needed,
in addition, are Web-based identification keys for different taxonomic groups based on an extensive and detailed
examination of several characters of a sufficient number
of individuals of each species. These should be verified
against the established holotypes for historical continuity of nomenclature but one would like to envisage the
consequent freedom from conventional holotype referencing for users of such databases, whether based on
molecular or morphological characters. The ‘holotype’
692
SYSTEMATIC BIOLOGY
in the new system would be a virtual one: the centroid
of the cluster used to define the species in the database.
Such a population-based approach has been previously
advocated (Mayr, 1976). The current insistence on species
verification using inaccessible type specimens is a major
impediment to rapid and accurate species identification,
particularly for those working in the tropics.
The development of accurate Web-based keys for several thousands of taxonomic groups using different data
sets is a formidable task and requires both an increase
in the number of professional taxonomists specializing
in different taxa in tropical and temperate countries, as
well as research into the bioinformatic methods required
to construct optimal key structures, in terms of efficiency
and accuracy. A conspicuous feature of the majority of
keys available today, whether on- or off-line, is the absence of any measure of validation: all keys should ideally be validated by probing them with specimens of
known identity to arrive at the probability of correct
identification and assignment (Sneath and Sokal, 1973).
Unfortunately, the construction and validation of identification keys is not currently a popular research area
in systematics, as a glance at the publication titles of
most journals in the area will reveal: the majority of articles deal with intricacies of phylogenetic reconstruction (mostly above the species level) or, in more specialised journals, with species descriptions. The difficulty
in publishing studies that explore the problem of identification in journals that specialize in systematics can
only worsen the current situation with regard to species
identification.
The problem of species identification should be placed
in the context of the current global biodiversity crisis
(Wheeler et al., 2004). In order to decide conservation
priorities, biologists in tropical countries are increasingly
called upon by governmental and legal bodies to carry
out rapid assessments of species diversity in different
areas. Accurate assessments require the availablity of
rapid, inexpensive, accessible, and accurate methods of
species identification, that are unavailable given the current state of the art in systematics.
Debates and decisions on species concepts and optimal methods for species delimitation cannot and should
not be divorced from the practical issue of species identification upon which they impact so heavily. There is
need as never before for a dialogue between systematists, evolutionary biologists, and conservation biologists
on the one hand, and between systematists in tropical and temperate countries on the other. The problems
faced by biologists in the tropics with regard to accessing
information on their native flora and fauna should also
be a cause for concern among the global community of
systematists and inspire action to remedy the situation,
given that these problems impact directly on biodiversity
conservation.
R EFERENCES
Avise, J. C., and K. Wollenberg 1997. Phylogenetics and the origin of
species. Proc. Natl. Acad. Sci. USA 94:7748–7755.
VOL.
54
Baum, D. A., and M. J. Donoghue. 1995. Choosing among alternative
“phylogenetic” species concepts. Syst. Bot. 20:560–573.
Chopard, L. 1969. The Fauna of India and the adjacent countries:
Orthoptera, Vol. 2. (R. B. Seymour Sewell, ed.). Baptist Mission Press,
Calcutta.
Cracraft, J. 2000. Species concepts in theoretical and applied biology:
A systematic debate with consequences. Pages 3–14 in Species concepts and phylogenetic theory (Q. D. Wheeler and R. Meier, eds.),
Columbia University Press, New York.
de Queiroz, K. 1998. The general lineage concept of species, species criteria, and the process of speciation. Pages 57–75 in Endless forms:
Species and speciation (D. J. Howard and S. H. Berlocher, eds.).
Oxford University Press, New York.
Dobzhansky, T. 1937. Genetics and the origin of species. Columbia University Press, New York.
Gaston, K. J., and R. M. May. 1992. Taxonomy of taxonomists. Nature
356:281–282.
Gilmour, J. S. L. 1951. The development of taxonomic theory since 1851.
Nature 168:400–402.
Godfray, H. C. J. 2002. Challenges for taxonomy. Nature 417:17–19.
Hebert, P. D. N., A. Cywinska, S. L. Ball, and J. R. de Waard. 2003.
Biological identification through DNA barcodes. Proc. R. Soc. Lond.
B 270:313–321.
Hebert, P. D. N., L. Y. Stoeckle, T. S. Zemlak, and C. M. Francis.
2004. Identification of birds through DNA barcodes. PloS Biology 2:
10e312.
Hull, D. L. 1977. The ontological status of species as evolutionary
units. Pages 91–102 in Foundational problems in the special sciences
(R. Butts and J. Hintikka, eds.). D. Reidel Publishing Company,
Dordrecht, Holland.
Mallet, J. 1995. A species definition for the Modern Synthesis. Trends
Ecol. Evol. 10:294–299.
Mallet, J., and K. Willmott. 2003. Taxonomy: Renaissance or Tower of
Babel? Trends Ecol. Evol. 18:57–59.
Mayr, E. 1942. Systematics and the origin of species from the viewpoint
of a zoologist. Columbia University Press: New York.
Mayr, E. 1976. Typological versus population thinking. Pages 26–29 in
Evolution and the diversity of life. Harvard University, Cambridge,
Massachusetts.
Mishler, B D., and E. C. Theriot. 2000. A critique from the Mishler and
Theriot phylogenetic species concept: Monophyly, apomorphy, and
phylogenetic species concepts. Pages 119–132 in Species concepts and
phylogenetic theory (Q. D. Wheeler and R. Meier, eds.). Columbia
University Press, New York.
Myers, N., R. A. Mittermeier, C. G. Mittermeier, G. A. B. da Fonseca,
and J. Kent. 2000. Biodiversity hotspots for conservation priorities.
Nature 403:853–858.
Nixon, K. C., and Q. D. Wheeler. 1990. An amplification of the phylogenetic species concept. Cladistics 6:211–223.
Otte, D. 1994. The crickets of Hawaii: Origin, systematics and evolution. The Orthopterists’ Society, Academy of Natural Sciences of
Philadelphia, Pennsylvania, USA.
Otte, D., and P. Naskrecki. 1997. Orthoptera Species File Online.
http://viceroy.eeb.uconn.edu/Orthoptera
Puorto, G., M. da Graca Salamao, R. D. G. Theakston, R. S. Thorpe,
D. A. Warrell, and W. Wüster, 2001. Combining mitochondrial DNA
sequences and morphological data to infer species boundaries: Phylogeography of lanceheaded pit vipers in the Brazilian Atlantic forest, and the status of Bothrops pradoi (Squamata: Serpentes: Viperidae). J. Evol. Biol. 14:527–538.
Ridley, M. 1986. Evolution and classification. Longman, London and
New York.
Riede, K. 1993. Monitoring biodiversity: Analysis of Amazonian rainforest sounds. Ambio 22:546–548.
Shaw, K. L. 1999. A nested analysis of song groups and species boundaries in the Hawaiian cricket Genus Laupala. Mol. Phylogenet. Evol.
11:332–341.
Shaw, K. L. 2002. Conflict between nuclear and mitochondrial DNA
phylogenies of a recent species radiation: What mitochondrial DNA
reveals and conceals about modes of speciation in Hawaiian crickets.
Proc. Natl. Acad. Sci.USA 99:16122–16127.
Sites, J. W., Jr., and K. A. Crandall. 1997. Testing species boundaries in
biodiversity studies. Cons. Biol. 11:1289–1297.
2005
POINTS OF VIEW
Sites, J. W., Jr., and J. C. Marshall. 2003. Delimiting species: A Renaissance issue in systematic biology. Trends. Ecol. Evol. 18:462–470.
Sneath, P. H. A., and R. R. Sokal. 1973. Numerical taxonomy. W. H.
Freeman and Company, San Francisco.
Wheeler, Q. D., and N. I. Platnick. 2000. A critique from the Wheeler and
Platnick phylogenetic species concept perspective: Problems with alternative concepts of species. Pages 133–145 in Species concepts and
phylogenetic theory (Q. D. Wheeler and R. Meier, eds.). Columbia
University Press, New York.
Wheeler, Q. D., P. H. Raven, and E. O. Wilson. 2004. Taxonomy: Impediment or expedient? Science 303:285.
Wiens, J. J., and T. A. Penkrot. 2002. Delimiting species using DNA
and morphological variation and discordant species limits in spiny
lizards (Sceloporus). Syst. Biol. 51:69–91.
693
Wiens, J. J., and M. R. Servedio. 2000. Species delimitation in systematics: Inferring diagnostic differences between species. Proc. R. Soc.
Lond. B 267:631–636.
Wiley, E. O., and R. L. Mayden. 2000. The evolutionary species concept. Pages 70–89 in Species concepts and phylogenetic theory (Q. D.
Wheeler and R. Meier, eds.). Columbia University Press, New York.
Willmann, R., and R. Meier. 2000. A critique from the Hennigian species
concept perspective. Pages 101–108 in Species concepts and phylogenetic theory (Q. D. Wheeler and R. Meier, eds.). Columbia University
Press, New York.
First submitted 20 May 2004; reviews returned 6 August 2004;
final acceptance 14 December 2004
Associate Editor: Adrian Paterson
Syst. Biol. 54(4):693–695, 2005
c Society of Systematic Biologists
Copyright ISSN: 1063-5157 print / 1076-836X online
DOI: 10.1080/10635150590950344
Counterexample to a Claim About the Reconstruction of Ancestral Character States
B RIAN LUCENA1 AND D AVID HAUSSLER 2
1
2
Division of Computer Science, University of California, Berkeley, California, USA;
E-mail: [email protected]
Howard Hughes Medical Institute, Department of Biomolecular Engineering, University of California, Santa Cruz, California, USA;
E-mail: [email protected]
Since Pauling and Zuckerkandl first suggested it
more than 40 years ago, the idea of reconstructing ancestral proteins and DNA sequences from the
information contained in sequences of present day
species has held considerable fascination (Pauling and
Zuckerkandl, 1963). Such reconstructions can provide
a unifying framework for understanding the molecular origins and evolution of key components in living
organisms. However, only recently has it become relatively straightforward to perform such reconstructions
and then test the reconstructed molecules functionally
in the lab. Now there is a surge of activity in this area.
Ancestral protein sequences for rhodopsin (Chang et al.,
2002), ultraviolet vision gene SWS1 (Shi and Yokoyama,
2003), ribonucleases (Jermann et al., 1995; Zhang and
Rosenberg, 2002), Tu elongation factors (Gaucher, 2003),
and steroid receptors (Thornton et al., 2003) have been
reconstructed and tested (see the reviews by Chang
and Donoghue [2000] and Thornton [2004]). In addition,
DNA from the common ancestor of placental mammals
has been reconstructed for a megabase-sized region containing the cystic fibrosis gene (Blanchette et al., 2004.) for
several families of transposons (Adey et al., 1994; Ivics
et al., 1997; Smit and Riggs, 1996; Jurka, 2002), and for
some complete small genomes like HIV (Hillis et al.,
1994). In the latter case the predicted ancestral sequences
were compared to the known ones, so the accuracy of the
resonstruction could be measured directly. However, in
other cases theoretical results (Yang et al., 1995; Schultz
et al., 1996; Schultz and Churchill, 1999) and computer
simulations (Zhang and Nei, 1997; Blanchette et al., 2004;
Cai et al., 2004) are required to assess the accuracy of the
reconstructed sequence.
In these investigations it has been observed that the
topology of the phylogenetic tree relating the present
day species to the target ancestral species can affect the
accuracy obtainable in reconstruction of the target ancestral character states. Simulations show that a starlike
phylogeny, i.e., a rapid radiation of many different lineages from a common target ancestor, such as occured
in the radiation of placental mammals, allows the ancestral character states of that target ancestor to be more
accurately reconstructed than those of more recent ancestors in parts of the tree where speciation events are
more regularly spaced (Blanchette et al., 2004.). More
generally, it has been claimed that the star phylogeny
always “represents the best case for ancestral character
state reconstruction, because each observation is conditionally independent and yields maximum information
about the ancestor” (Schultz et al., 1996), see also Schultz
and Churchill (1999). Here we show that the actual situation is more complex and depends on the branch lengths.
This complexity occurs even for the simplest evolutionary model with only one parameter: the Poisson model
(known also as the Neyman r -state model, generalized
Jukes-Cantor model, and the Potts model), where the parameter determines the rate of substitution and all substitutions are equally likely.
Consider the tree topology shown in Figure 1, where
A represents the common ancestor of three present day
species, designated by C1 , C2 , and C3 . B2 represents the
common ancestor of C2 and C3 whereas B1 represents
the ancestor of C1 at the same moment in evolutionary
time as B2 . This tree contains a subtree with the simplest
star topology, the two-leaf subtree shown in bold in Figure 1b, and it also contains a subtree with the simplest