34 Assembling the Tree of Life Where We Stand at the Beginning of the 21st Century

Back

Joel Cracraft

Michael J. Donoghue

553

Few endeavors in biology, or in all the sciences, can match

our quest to understand the course of life’s history on Earth,

which stretches across billions of years and captures the

descent of untold millions of species. The notion that scientific

inquiry might achieve that goal is little more than a century

and a half old, and yet surprisingly, most of the species

that have appeared on the twigs of the Tree of Life (TOL)

have been put there only in the last decade. The systematists

who have contributed to the chapters in this volume have

collectively contributed a significant step toward a grand

vision of systematic biology: achieving a comprehensive picture

of the TOL is finally within our grasp. Darwin, Haeckel,

Huxley, and the other giants who convinced the world of life’s

long history of change, and built the first scaffold of that

history, might very well say “finally . . . it’s about time”!

That it has taken so long to get to this point is testimony

to the fundamental conceptual and technical challenges that

have faced systematic biologists over the years. For many

decades systematists had no clear theoretical or methodological

idea how to recover life’s history in an objective way. That

challenge, as many of the greatest in the sciences, was met

by deceptively simple logic. Willi Hennig, and the phylogenetic

principles he developed (1950, 1966), quickly formed

the foundation for quantitative, objective methodologies for

comparing the characters of organisms. The technical challenges,

in turn, were met when it became easier to collect new

kinds of data, primarily molecular, and as computational

software and hardware improved to make these comparisons

faster and more efficient.

The last major summary of our knowledge of the TOL—

compiled from the 1988 Nobel symposium titled “The Hierarchy

of Life” (Fernholm et al. 1989)—establishes a point

of comparison with which to understand the intense work

of the past decade. The phylogenetic trees presented in that

volume rarely included more than 15–20 taxa, and data sets

hardly exceeded 100 or 200 characters, most far fewer than

that. Perusal of the journals of that time paints a similar story.

The scientific work summarized here, in contrast, manifests

a huge growth in phylogenetics research. Virtually all

the chapters include taxon and character samples that were

unheard of a mere 10 years ago. Yet, because the focus of

the chapters in this volume is the relationships among the

higher taxa, even these summaries cannot convey the vast

increase in our knowledge that has taken place at all hierarchical

levels. For that, the reader will have to go to specialized

volumes—Benton (1988), Stiassny et al. (1996), Fortey

and Thomas (1998), Littlewood and Bray (2001), and Judd

et al. (2002) are but five examples that have been published

in recent years—as well as to the numerous journals publishing

phylogenetic results in every issue.

Having knowledge of the phylogenetic relationships of

life is crucial if we are to advance societal well-being, including,

importantly, building a sustainable world. In this volume,

the chapters by Yates et al. (ch. 1), Colwell (ch. 2), and

554 Perspectives on the Tree of Life

Futuyma (ch. 3) describe numerous examples of the contributions

that phylogenetic understanding has already made

to science and society. Phylogenetic relationships establish

the framework for all comparative analyses of biological data,

and this hierarchical structure is also a predictive tool that

leads us from those characteristics we now know about organisms

to those we might expect to find in those less known

or newly discovered. Such logic, whether expressed explicitly

or not, underlies the expectation that certain organisms

might harbor pharmacologically important compounds,

might be pathogenetic or toxic, might express agriculturally

important gene products, and so on. Indeed, the use of phylogenetic

knowledge, including analytical methods that have

been developed to solve phylogenetic problems, has grown

so rapidly in recent years that even a single volume devoted

to the subject could not be comprehensive.

The practical outcomes and applications of TOL research

are certainly a clear reason why society should continue to

support a better understanding of phylogenetic relationships

(see ch. 1–3; see also Cracraft 2002). Yet, what drives many

scientists engaged in this effort is the sheer wonder associated

with knowing a chunk of life’s history. To step back and

attempt to grasp the entire history of life on Earth is itself an

almost unimaginable task. Here we are, one species out of

hundreds of millions that have existed since the diversification

of life began several billion years ago, and we are attempting

to see how that history has unfolded. It is difficult enough

to see how we will build the TOL for the living species, let

alone for all those that vanished over the course of time, but

it is an exciting prospect. All people on the planet understand

something about their “genealogical roots,” and that serves

as a crucial metaphor for seeing how human existence and

origins fit into the bigger picture of life’s diversity. This is a

nontrivial exercise, for truly understanding that history is

bound to influence the ethical picture people develop about

the importance of life forms other than our own and how

these have been inextricably linked to our own well-being

over time. Obviously, it is not easy for us to step back from

an anthropocentric view of the world, but a TOL can facilitate

such a perspective.

Darwin’s vision had a profound effect on people’s understanding

of themselves. Yet the understanding that went

along with this change in thought is not universally appreciated

even today, despite 150 years of evolutionary thought

and science. The TOL will be a key element in advancing an

expanded vision of life’s history.

The Tree of Life: An Ongoing Synthesis

The chapters in this volume summarize our current understanding

of the phylogenetic history of the major groups of

organisms. It is time to stand back and see the big picture.

Figure 34.1 presents a summary TOL that attempts to provide

an estimate of the interrelationships among the extant

clades of life. Its scope and depth, which is skewed toward

the “higher” eucaryotes, is primarily a function of the coverage

of the chapters in this book, which, in turn, generally

reflect known, described taxonomic diversity. Clearly, many

more groups could have been added to this tree, and numerous

friends working on megadiverse taxa have suggested how

their favorite groups could be expanded. Yet, the best way

for this tree to serve an educational purpose is to limit detail

and to include groups that are familiar to a wide audience.

Conceptually, the tree is constructed as a composite—

constructed by piecing together the trees presented for the

different groups. It is not derived from an analysis of a

“supermatrix.” It attempts to represent relationships that are

moderately to well supported, yet there are unresolved nodes.

Some will see the tree as too conservative and would recommend

resolving certain nodes; others would prefer that more

nodes be depicted as ambiguous. Because the tree is not built

from a data matrix, it is not a rigorous phylogenetic hypothesis

in a traditional sense. Rather, it is a summary of where

we are now and a step in the continuous process of building

a TOL. Importantly, it also stands as a framework for discussing

some of the key problems and controversies raised

in the individual chapters of this volume.

The Basal Clades of Life

It has been standard for a number of years now to recognize

three major basal branches (“domains”) of the TOL, the Bacteria,

the Archaea, and the Eucarya (see Baldauf et al., ch. 4,

and Pace, ch. 5, in this vol.), all of which are generally treated

as monophyletic. A major impediment for understanding the

nature of that monophyly and the relationships among these

groups is, of course, the problem of where to place the root

of the TOL. The present conventional wisdom is that the root

lies along the branch between the Bacteria and the other two

on the basis of evidence presented by duplicated genes

(ch. 4). Some workers, on the other hand, have raised the

issue of lateral gene transfer as possibly confounding the

placement of the root (Doolittle, ch. 6), or that analytical

artifacts such as long-branch attraction can lead to misleading

relationships, which also could affect the placement of

the root (Philippe, ch. 7). Philippe also argues that we have

seen the evolutionary world as proceeding from the simple

to the complex and thus have potentially overlooked the

possibility that “prokaryotic”-like organisms could have been

derived from eucaryotes by simplification. A major concern

for all these scenarios, however, is that given the monophyly

of these three groups, the placement of the root may be unsolvable

because it remains a three-taxon problem.

The trailblazing work of Carl Woese, Norman Pace, and

others to use the small subunit ribosomal RNA (rRNA) gene

to reconstruct life’s earliest branches can truly be said to have

revolutionized our view of the TOL, and at the same time

those data have shaped how the question of basal relationships

has been approached. It is now clear that rRNA seAssembling

the Tree of Life 555

Figure 34.1. A Tree of Life for the major groups of organisms. The relationships shown attempt

to summarize those discussed in the chapters of this book. See discussion in text.

Echinodermata

Hemichordates

Ecdysozoa

Lophotrochozoa

Bacteria

Archaea

Eucarya

Animalia

Metazoa

Fungi

Plantae

Choanozoa

Bilateria

Chordata

Vertebrata

"amitochondriate excavates"

Chromalveolates

Radiolaria, Cercozoa

Foraminifera

Amoebozoa

Deuterostomia

Opisthokonta

Discricristales

green plants

(Viridophytae)

spirochaetes

green-suphur bacteria

gamma-proteobacteria

beta-proteobacteria

mitochondria

alpha-proteobacteria

plastids

cyanobacteria

Thermus

mycoplasmas

hydrogenobacteria

euryarchaeotes

crenarchaeotes

korarchaeotes

parabasalids

diplomonads

euglenids

trypanosomes

Leishmania

acrasid slime molds

dinoflagellates

apicomplexans

ciliates

water molds (oomycetes)

chrysophytes, brown algae

diatoms

radiolarians

cercomonads

foraminiferans

green algae

charalians

mosses

liverworts

hornworts

lycophytes

ferns, horsetails (monilophytes)

cycads

conifers

Amborella

water lilies

rosids

asterids

cacti (Caryophyllales)

poppies (Ranunculales)

laurels (Laurales)

magnolias (Magnoliales)

lilies (Liliales)

orchids, irises (Asparagales)

palms (Arecales)

grasses (Poales)

red algae

slime molds, lobose amoebae (mycetozoans)

ascomycote fungi

basidiomycote fungi

zygomycote fungi

chytridiomycote fungi

microsporidians

choanoflagellates

ichthyosporeans

silicean "sponges"

calcarean "sponges"

ctenophorans

cnidarians

acoelomorphs

velvet worms (onychophorans)

water bears (tardigrades)

spiders

mites

scorpions

horseshoe crabs (xiphosurans)

ostracods

barnacles

copepods

crabs, lobster, shrimp (decapods)

millipedes

centipedes

dragonflies

cockroaches, mantises, termites

grasshoppers (orthopterans)

true bugs (hemipterans)

beetles (coleopterans)

ants, wasps (hymenopterans)

fleas (siphonapterans)

flies (dipterans)

butterflies, moths (lepidopterans)

nematodes

kinorhynchs, priapulids

chaetognaths

gastrotrichs

rotifers

platyhelminths

phoronids

brachiopods

ectoprocts (bryozoans)

entoprocts

nemerteans

polychaetes, earthworms, leeches (annelids)

chitons (polyplacophorans)

clams (bivalves)

squids, octopuses (cephalopods)

snails (gastropods)

sea cucumbers (holothurians)

sea urchins (echinoids)

brittlestars (ophiuroids)

starfish (asteroids)

acorn worms (hemichordates)

tunicates (urochordates)

lancelets (Cephalochordata)

hagfish (myxinoids)

lampreys (petromyzontids)

sharks, rays (condrichthyians)

perches, silversides (percomorphs)

lizardfish, lancetfish (aulopiforms)

salmon, smelts (protacanthopterygians)

minnows, catfish (otophysians)

eels, morays (elopomorphs)

elephantfish, mooneyes (osteoglossomorphs)

coelacanths (actinistians)

lungfish (dipnoans)

frogs (anurans)

salamanders (caudates)

caecilians (gymnophionans)

side-necked turtles

vertical-necked turtles

snakes

monitor lizards

skinks (scincoids)

iguanas and allies (iguanians)

tuatara (Sphenodon)

crocodyles, alligators

ratites, tinamous (palaeognaths)

pheasants, waterfowl (galloanserans)

other modern birds (neoavians)

platypus, echidna (monotremes)

marsupial mammals

placental mammals

angiosperms

Arthropoda

Gnathostomata

556 Perspectives on the Tree of Life

celled taxa whose relationships to these three clades are still

unresolved; therefore, the tripartite division discussed here

is certainly simplistic.

Plants

The overall backbone of plant phylogeny is moderately well

supported (Donoghue, ch. 33, and Delwiche et al., ch. 9, in

this vol.). The term “algae” has been applied to a diverse array

of unrelated taxa possessing plastids, some of which lie

at the base of the land plants, and indeed from the perspective

of Delwiche et al., the land plants simply comprise a terrestrial

lineage of green algae. Although the relationships

among these algal groups need much further study, current

molecular evidence identifies the Charales as the sister group

of the land plants (embryophytes).

Within the embryophytes, the interrelationships among

the three major groups of nonvascular plants—the liverworts,

hornworts, and mosses—and the vascular plants (tracheophytes)

are still a matter of controversy (Delwiche et al.,

ch. 9). The base of the tracheophyte tree is less controversial,

with lycophytes being the sister group of the rest and

then monilophytes (horsetails and various “fern” groups)

being the sister group of the seed plants (Pryer et al., ch. 10).

Relationships within the monilophytes, and especially at the

base of the clade that includes the modern seed plants, are

not entirely resolved. Within the latter group, which contains

some 300,000 species, the angiosperms comprise the most

diverse clade. The phylogenetic unity of the clade that includes

the extant “gymnosperms” is still questionable, and

the sister group of all the angiosperms has not yet been identified

with confidence.

The angiosperms (flowering plants) are by far the dominant

group of land plants, and their interrelationships have been

the subject of a large number of morphological and molecular

systematic studies over the last decade. Soltis and colleagues

(ch. 11) have been important contributors to this effort. They

note that relationships at the base of the angiosperms are moderately

well understood. One of the more remarkable findings

to emerge in recent years is that Amborella trichopoda of New

Caledonia is the only living representative of the sister group

of all other angiosperms, and the next branch contains the water

lilies. The three largest clades within the core angiosperms—

monocots, magnoliids, and the eudicots—are well defined, but

their relationships to one another and to several other smaller

clades remain unresolved (ch. 11).

Fungi

In recent years fungi have emerged as the sister group to the

animals (see Baldauf et al., ch. 4, and Eernisse and Peterson,

ch. 13, in this vol.). It is also becoming increasingly apparent

that they will eventually be seen as one of the most diverse

groups on Earth. The large-scale phylogenetic structure of the

fungi has become clearer with the addition of sequence data,

quences alone cannot resolve the branching order among

bacterial lineages to a convincing degree (Pace, ch. 5).

Bacterial relationships have been strongly influenced by

decades of attempts to classify using phenetic data sets of a

small number of key “characters” (e.g., gram-positive vs.

gram-negative staining). This approach is bound to create

some nonmonophyletic taxa. Bacterial systematists have also

classified these taxa at high taxonomic rank (subkingdoms,

divisions, phyla) on the basis of distinctiveness, and that tradition

has continued as genetically distinct forms have been

discovered from environmental samples. As Pace (ch. 5) describes,

there are two main groups of Archaea, the crenarchaeotes

and the euryarchaeotes. The third group shown on

figure 34.1, the korarchaeotes, is only represented by environmental

rRNA gene sequences and is of uncertain status (see

also Baldauf et al., ch. 4).

Are viruses life, or not, and what has been their history?

These are the subjects of chapter 8 by David Mindell and his

colleagues. Although the topic of viral phylogeny was not the

subject of a talk in the New York symposium Assembling the

Tree of Life, its inclusion in this volume was deemed important

for understanding the full panoply of biotic history.

Mindell et al. show that viruses have arisen multiple times,

and they summarize what we understand of their evolutionary

relationships. Importantly, they also discuss how phylogenetics

and its methodology can be applied to issues of

human health.

Basal Eucarya

The base of the eucaryotic tree is very uncertain, with candidate

groups being the parabasalid + diplomonad clade or

discricristates, among others (Baldauf et al., ch. 4). Some

would argue (Philippe, ch. 7) that the basal position of such

taxa as parabasalids or diplomonads is probably a longbranch

artifact. Their basal position seems reasonable at first

glance, because it has been thought they branched off before

the acquisition of the bacterial precursors of mitochondria.

It is now known, however, that these “amitochondriate

excavates” have some mitochondrial genes in their nuclear

genomes. “Basal” eucaryotes remain one of the most unexplored

regions of the TOL, and inasmuch as some groups

are apparently very diverse, numerous candidates for the

basal eucaryotic divergences are likely to emerge as new data

are acquired.

There are three large monophyletic clades of eucaryotes,

the green plants (upwards of 500,000 species), fungi (around

60,000 described), and animals (more than one million described).

It is widely accepted that millions of species of fungi

and animals remain to be discovered and described, whereas

plant diversity has been more completely characterized. One

of the more interesting phylogenetic findings of recent years

is that the fungi and animals are sister taxa relative to other

organisms (Opisthokonta; see Baldauf et al., ch. 4). It is important

to note, however, that there are numerous singleAssembling

the Tree of Life 557

and it is now accepted that the two great groups of terrestrial

fungi, the ascomycotes and basidiomycotes, are monophyletic

and sister taxa (Taylor et al., ch. 12). As Taylor and colleagues

note, relationships within these two diverse groups are still in

need of considerable study. The base of the fungal tree is also

poorly understood and is occupied by lineages usually assigned

to two more obscure groups, the zygomycotans and chytridiomycotans,

both of which may be nonmonophyletic.

Basal Animals

Animals are taken here to include the choanoflagellates and

their sister group, the metazoans (see Eernisse and Peterson,

ch. 13 in this vol.). Eernisse and Peterson review the evidence

showing that animal and metazoan monophyly has become

increasingly well established in recent years, but that relationships

at the base of the Metazoa have been in a state of

flux, particularly when it comes to those organisms typically

called “sponges.” Traditional classifications using morphological

data recognized a monophyletic Porifera, but molecular

data have led to the conclusion that siliceous sponges

branched off first, followed by the calcareous sponges, the latter

of which are the sister group to the eumetazoans (ch. 13).

Relationships among the major clades of metazoans—ctenophorans,

cniderians, placozoans, and eumetazoans—also remain

uncertain because of conflicts among data sets (see ch. 13

for details)

Bilaterians

The monophyletic bilaterians are composed of three main

groups, the ecdysozoans, lophotrochozoans, and deuterostomes,

and more and more evidence is pointing to the conclusion

that acoelomorph flatworms are their sister group (see

Eernisse and Peterson, ch. 13, and Littlewood et al., ch. 14,

in this vol.). Intense examination of the monophyly of these

groups and the interrelationships of their included taxa has

essentially revolutionized our view of bilaterian evolution

over the last decade by eliminating the simplistic aceolomate

to pseudocoelomate to coelomate description of phylogenetic

history. Although the monophyly of ecdysozoans, lophotrochozoans,

and deuterostomes—particularly the latter—is

increasingly accepted (at least for the “core” taxa of the first

two), their interrelationships are controversial, as is the

placement of a number of small, morphologically disparate

metazoan groups often classified at the phylum level

(Littlewood et al., ch. 14, discuss no less than 15 “phyla”).

Therefore, a major question is whether there exists an

ecdysozoan + lophotrochozoan clade—thus implying the

classical protostome–deuterostome dichotomy.

Lophotrochozoans

As reviewed by Eernisse and Peterson (ch. 13 in this vol.),

the interrelationships among lophotrochozoan taxa are exceedingly

complex and contentious due to conflicts in data,

especially morphological versus molecular. Several groups are

regularly recognized: (1) the lophophorates, encompassing

brachiopods and phoronids; (2) the trochozoans, including

the annelids and mollusks, and their allies (see fig. 34.1);

and (3) the platyzoans (rotifers, platyhelminths, and others).

The latter two groups have traditionally been clustered

in the Spiralia on the basis of possessing spiral cleavage and

a trochophore larva, although it is entirely possible that

lophophorates are within the trochozoans.

The two great groups of lophotrochozoans are sister taxa,

the annelids (Siddall et al., ch. 15) and the mollusks (Lindberg

et al., ch. 16). Within the former, leeches and earthworms

are related, but the sister group of leeches within the

earthworms is still uncertain. Morphological and molecular

data conflict on annelid relationships, along with those of

sipunculans, relative to the diverse marine polychaete worms

(ch. 15). Clearly much more work will be required to resolve

the history of these groups.

The interrelationships of the major clades of mollusks are

moderately well accepted (Lindberg et al., ch. 16; see also fig.

34.1), with cephalopods and gastropods being sister taxa and

related to bivalves and chitons at the base of the tree. All these

groups have a deep evolutionary history, with considerable

fossil diversity, and an integrated picture of their phylogeny

will significantly advance paleontology. Not unexpectedly,

the interrelationships of the recent molluscan biota are comparatively

poorly understood given their extensive diversity.

Ecdysozoans

Different lines of evidence point to the ecdysozoans being a

natural group (summarized in Eernisse and Peterson, ch. 13

in this vol.), yet many questions remain about their interrelationships,

reflected in the unresolved tree in figure 34.1.

Four ecdysozoan clades are now generally accepted (ch. 13

and 14): (1) the panarthropods; (2) nematodes and nematomorphs;

(3) the kinorhynchs, priapulids, and loriciferans;

and (4) chaetognaths. The latter two groups have low diversity,

but the nematodes are thought to be the most numerically

abundant metazoans on Earth, and they undoubtedly

have a tremendous undescribed diversity greatly exceeding

the 25,000 or so species already named. Littlewood and colleagues

(ch. 14) briefly note recent progress on the phylogenetics

of this group.

The arthropods—insects (Hexapoda); centipedes and

millipedes (Myriapoda); crabs, crayfish, and their allies (Crustaceans);

and the spiders and allies (Chelicerata)—include a

number of megadiverse clades, especially the mites, spiders,

and insects, and together they represent roughly 60% of the

known species diversity on Earth. Wheeler and colleagues

(ch. 17 in this vol.) describe the complex problem of deciphering

relationships among the major groups of arthropods,

the conflicting topologies implied by different data sets, and

the fact that inclusion of fossil taxa in total evidence analy558

Perspectives on the Tree of Life

dered considerable debate (Whiting, ch. 21 in this vol.). The

tree shown in figure 34.1 includes only five holometabolic

clades whose relationships appear with regularity on trees

using both morphology and/or molecules (ch. 21), but

many other smaller clades are omitted. Clearly, the complexity

of the vast taxonomic and morphological diversity

of this group will feed controversy for many years, and it

seems that considerable data will be required to resolve

these long-standing phylogenetic questions.

Deuterostomia

The third great group of the bilaterians is the Deuterostomia

(or Deuterostomata), which includes the echinoderms and

hemichordates (ambulacrarians), on the one hand, and their

sister group, the chordates. Until recently, the boundaries

of the deuterostomes were ambiguous, but morphological

and molecular work has clearly eliminated lophophorates,

ectoprocts, and chaetognaths from the clade and established

the remainder as a monophyletic group (see Eernisse and

Peterson, ch. 13 in this vol.).

Ambulacraria

Smith and colleagues (ch. 22 in this vol.) review recent advances

in hemichordate and echinoderm phylogenetics. The

former group is small in terms of diversity, and relationships

within the group have still not been deciphered satisfactorily.

Echinoderms are also not especially diverse, having only

about 6000 extant species, but they possess an extensive fossil

record and are among the best known marine organisms. As

Smith and colleagues detail, relationships among the major

monophyletic groups are moderately well supported on both

molecular and morphological grounds (fig. 34.1).

Chordata and Vertebrata

The overall pattern of chordate phylogeny is moderately well

corroborated by both morphological and molecular data (fig.

34.1; see Rowe, ch. 23 in this vol.). The tunicates and lancelets

are the successive sister taxa to the craniates (hagfish +

vertebrates). For many years the hagfish and lampreys (fig.

34.1) were grouped together as the agnathans, but the preponderance

of evidence does not favor this, especially the

morphological and developmental data. Rowe notes in his

review, however, that some molecular data find a monophyletic

Agnatha; therefore, the problem needs further attention

using combined data sets, and fossils as well as extant taxa.

Moving up the vertebrate tree, the next node subtends

the sharks and allies (Chondrichthyes) and all other vertebrates

(Osteichythes), which are together termed the Gnathostomata.

The Osteichythes, in turn, are subdivided into the

sarcopterygians (coelacanths, lungfish, and tetrapods) and

the actinopterygian fishes. Stiassny and colleagues (ch. 24)

lead us through the world of things called “fishes,” in their

ses often has dramatic effects on phylogenetic inferences.

Although most of the evidence clusters crustaceans, myriapods,

and hexapods together (as the Mandibulata because

they possess mandibles) to the exclusion of the chelicerates,

resolving relationships among the mandibulates has not been

straightforward (ch. 17).

The higher level relationships of the chelicerates are

moderately well supported, with mites and spiders being

sister taxa and related to scorpions and their allies, and those

three, in turn, are the sister group of the horseshoe crabs (fig.

34.1; Coddington et al., ch. 18). Over the past decade, relationships

among the spiders have received considerable attention,

and they are the best understood of the chelicerates,

whereas relationships among the diverse clades of mites remain

very poorly resolved (ch. 18).

As reviewed by Schram and Koenemann (ch. 19 in this

vol.), the monophyly of the crustaceans has been contentious,

with morphological data tending to support monophyly and

some molecular data sets denying it. Even in this volume,

differences of interpretation exist: Schram and Koenemann

(ch. 19) question monophyly, whereas the analyses of

Wheeler and colleagues (ch. 17) generally find a monophyletic

Crustacea. Many of these differences, and those in the

literature, come down to apparent conflicts between molecules

and morphology, to alternative interpretations of

morphological characters, especially those of fossils, and to

which clade is to be called Crustacea. There is relatively little

argument (see fig. 34.1; see also ch. 19), however, that the

core crustacean clades are monophyletic and related to one

another, especially the maxillopods (copepods, barnacles,

ostracods) and the malacostracans (crabs, shrimps, and

allies).

Arguably, the greatest challenge to the TOL—as we currently

understand organic diversity—is the relationships

within the hexapods, or insects and their allies. The vast diversity

of forms creates multiple challenges for understanding

insect history. Willmann (ch. 20 in this vol.) presents a

summary of the complexities of hexapod phylogeny and how

viewpoints have shifted over time, and Whiting (ch. 21) discusses

phylogenetic relationships within the most diverse

clade of hexapods, the holometabolic insects. Arguments over

insect relationships exemplify the debates in other groups—

molecules versus morphology, fossil versus extant taxa. The

overall structure of the insect tree, however, is remarkably

consistent from one study to the next (ch. 20): aside from a

number of basal groups, most insects can be clustered into

the Pterygota (those with wings), at the base of which are the

mayflies and dragonflies (whether sister taxa or not is in dispute)

and their great sister group, the Neoptera.

No less than 80% of the insects are found in the

neopteran group, the Holometabola—those insects with

complete metamorphosis. This generally accepted monophyletic

group contains the most familiar of the insects—

beetles, butterflies, wasps, flies, and so forth—yet the

interrelationships of these well-defined groups have engenAssembling

the Tree of Life 559

case, chondrichthyans, actinopterygians, and the “fishlike”

sarcopterygians. Chondrichthyans are easily divided into

elasmobranchs (sharks, rays) and chimaeras, but relationships

within the former clade are still uncertain. Morphological

data recognize two basal sister taxa (galeomorphs and

squalomorphs) and support a moderately resolved phylogeny

within the latter; the conflict comes with some emerging

molecular data that is said to question the monophyly of

the rays and sharks (ch. 24). Within sarcopterygians, resolution

of the coelacanth–lungfish–tetrapod trichotomy has

been contentious. Stiassny et al. remain agnostic on this issue,

whereas Rowe (ch. 23) resolves this in favor of lungfish

+ tetrapods while noting that the debate continues.

The actinopterygian fishes are the most diverse group of

vertebrates and have a huge diversity of forms, so relationships

have generally been difficult to resolve. Most of the actinopterygian

nodes on figure 34.1 are based on morphological

data, as Stiassny and colleagues (ch. 24) note, but new molecular

data are being generated at a rapid rate. Although the

interrelationships of these major groups might be generally

accepted, phylogenetic understanding within most of them has

a long way to go, especially given their high diversity.

It has long been accepted that amphibians are at the base

of the tetrapod tree and are the sister group to all other vertebrates,

which are grouped together as the Amniota (Rowe,

ch. 23 in this vol.). Living amphibia are clearly monophyletic,

and the relationships among the three clades have long been

accepted (Cannatella and Hillis, ch. 25). Thus, caecilians are

the sister group of the salamanders and frogs. Relationships

within the three living taxa, especially within salamanders and

frogs, are greatly unsettled.

The amniotes, so named because they share an amniote

egg, are divided into two major clades, the Reptilia—including

turtles, lepidosaurs (snakes, lizards, tuataras), and

archosaurs (crocodiles and birds)—and the Mammalia

(Rowe, ch. 23, and Lee et al., ch. 26, in this vol.). Higher level

relationships within the reptiles have been particularly contentious.

Crocodiles and birds go together on all trees, but

the turtles, tuatarans, and snakes and lizards sort out in different

ways depending on the data set. There are significant

conflicts across and within data sets that leave these relationships

unresolved. In contrast to this somewhat dismal situation,

Lee and colleagues (ch. 26) show that higher level

relationships within turtles and within the lizards and snakes,

for example, are becoming better understood (fig. 34.1), although

at lower taxonomic levels many gaps in our knowledge

still exist.

Higher level relationships within living birds remain perhaps

the least understood of all the major groups of tetrapods

(Cracraft et al., ch. 27). The basal split between the

tinamous and ratites (paleognaths) and all other birds

(neognaths), and then within the neognaths between the

galliforms–anseriforms and all others (Neoaves), are well

supported by various data. Phylogenetic pattern among the

traditional neoavian “orders,” on the other hand, are largely

unresolved. The reason for this is pretty simple—lack of

adequate character and taxon sampling—but that is rapidly

changing.

Our understanding of mammalian interrelationships

has made great strides in recent years because of the addition

of very large morphological and molecular data sets.

Yet, at the same time, as discussed by O’Leary and colleagues

(ch. 28 in this vol.), there exists a great deal of conflict

among data sets, and over their interpretation. All agree

that monotremes are the sister group of the marsupials and

placentals, but within the latter group there is considerable

debate about how the traditional orders are related. The

increasingly large molecular data sets appear to be converging

on an answer, but morphological (and paleontological)

data often conflict.

Finally, the symposium included a discussion of our current

picture of hominid phylogenetics (Wood and Constantino,

ch. 29 in this vol.)—for after all, it is a subject that

generates great scientific and public interest and controversy.

In contrast to other contributors, Wood and Constantino

focus attention on the basal taxa of Homo—what systematists

generally call species—because it is difficult to understand

human evolution without delimiting those units. These authors

come down on the “many species” side of the debate,

as opposed to “just a few,” and they argue that deciphering

relationships among these taxa is challenging because so

much of the fossil material is fragmentary and difficult to

compare. They also demonstrate that debates over human

origins—in the sense of which species is related to which—

are likely to continue for quite some time.

Perspectives on the Tree of Life

A volume like this was not possible a decade or so ago, as a

comparison with The Hierarchy of Life (Fernholm et al. 1989)

makes clear. New analytical methods and new and more

abundant data have transformed the field. But there has also

been a sea change in biology’s attitude toward systematics

and TOL research. Our interest in life on Earth has accelerated,

not only because it is rapidly disappearing, or is in our

self-interest to find new ways to make money from it, or

because increased understanding will contribute to the wellbeing

of humanity, or it is intrinsically interesting. It is all

these reasons. In his perspective, E. O. Wilson (ch. 30) makes

the case for “a complete account of Earth’s biodiversity, pole

to pole, bacteria to whales, at every level of organization from

genome to ecosystem, yielding as complete as possible a

cause-and-effect explanation of the biosphere, and a correct

and verifiable family tree for all the millions of species—in

short a unified biology.” Amen to that. Indeed, discovering

and describing biodiversity and understanding the TOL go

hand in hand, and both are increasingly seen as a foundation

for all of biology. Importantly, TOL research has moved

into mainstream experimental and molecular biology.

560 Perspectives on the Tree of Life

Growth in TOL research over the past decade, as David

Wake (ch. 31) and David Hillis (ch. 32) observe, is readily

apparent. Hillis also makes the important point that as TOL

research expands, so do its applications to science and

society. We are certainly on a roll, but how we might

measure progress is not so straightforward, as Michael

Donoghue notes (ch. 33). His tentative conclusion, discussed

more below, is that it is the recognition and abandonment

of paraphyletic groups that is perhaps the best

measure of progress. Although it seems that some new paraphyletic

groups will inevitably be created as more taxa are

investigated, a successful war against paraphyly is the surest

measure of success.

The Tree of Life: Progress Against Paraphyly

A survey of previous literature leads one to the conclusion

that assembling the TOL must be an exceedingly complex

problem because very few have attempted to resolve the

whole tree (one of the few attempts has been in the popular

literature; Tudge 2000). The present volume signals that we

have entered a new era of research in phylogenetics. If we

look back more than a decade ago, the overall state of knowledge

discussed at the 1988 Nobel symposium might appear

disappointing, and in today’s terms, it was. If we compare,

for example, the “summary tree” from that symposium

(fig. 34.2) with the one discussed here (fig. 34.1), the contrast

is striking. As noted above, it reflects a change not only

in data and data analysis but also in attitude toward what we

now know we can accomplish. The latter is not to be dismissed:

a decade ago, not everyone was convinced a universal

tree was at hand, or possible (even for relatively small

chunks of the tree). Today, the attitude of systematists has

changed. We will have a universal tree, and the operative

questions are when, how well supported it will be, and how

we are going to create a new field of phyloinformatics to tap

the tree’s benefits.

Concepts of monophyly, paraphyly, and polyphyly are

not really associated with phylogeny per se but how relationships

map to classification. When we say that the goal

of TOL research is to discover and eliminate paraphyly, we

mean eliminate named groups that are not natural groups.

The practical manifestation of the chapters in this book is

to rid systematic biology of nonmonophyletic groups, but

this activity will be resisted by some. TOL research is caught

to some extent in the language of the past, in which groups

are ranked on the basis of distinctness. In the past, it was morphological

distinctness, but today “genetic distinctness”—

however that might be measured objectively—is increasingly

an important criterion. The notion that distinctness should

enter into hierarchical classifications through ranking has

created paraphyletic groups in its wake and hindered phylogenetic

progress. The plethora of high taxonomic ranks,

such as domains, kingdoms, phyla, and the like, does nothing

to clarify the phylogenetic history of life.

Figure 34.2. A “summary” tree

(hierarchy) of life of selected major

groups of organisms in which those

taxa underlined were the subject of

discussion at the 1988 Nobel

symposium (Fernholm et al. 1989).

Assembling the Tree of Life 561

Although we can be “immeasurably” optimistic that progress

on the TOL will continue unabated, those involved in

research know the task is a difficult one. A theme of the 1988

Nobel symposium was molecules versus morphology. In the

early years of molecular systematics, there was an abundance

of exuberance that molecules were going to sweep away morphology

in reconstructing the TOL. That has not exactly happened,

if one is to judge by the myriad molecular data sets

that conflict with one another. Indeed, as this volume attests,

more and more workers are seeking to combine molecular

and morphological data, and there is a growing realization that

if we are truly to have a TOL, extinct life—at least 90% of all

of it—must be included. The view here is that most of the

conflicts we see among different data sets are more a matter

of the selection of data, method of analysis, and lack of sufficient

data than they are anything substantially “wrong” with

a particular kind of data. Evidence is evidence, and we should,

as scientists, bring all that is relevant to bear on a problem

that we can. This view echoes Colin Patterson’s (1989) closing

remarks for the 1988 Nobel symposium: molecules allow

us to gather large amounts of data quickly, but morphological

data give us access to other dimensions of life—ontological,

paleontological, temporal, and of form and function.

Systematics needs all this. Biology needs all this.

Acknowledgments

All the participants and coauthors of the chapters in this book,

especially Sandie Baldauf, Mark Siddall, Ward Wheeler, Pam

Soltis, Kathleen Pryer, Timothy Rowe, Douglas Eernisse, Tim

Littlewood, Max Telford, and John Taylor, provided expert

advice and help in constructing figure 34.1.

Literature Cited

Benton, M. (ed.). 1988. The phylogeny and classification of the

tetrapods, vols. 1 and 2. Clarendon Press, Oxford.

Cracraft, J. 2002. The seven great questions of systematic

biology: an essential foundation for conservation and the

sustainable use of biodiversity. Ann. Missouri Bot. Garden.

89:127–144.

Fernholm, B., K. Bremer, and H. Jцrnvall (eds.). 1989. The

hierarchy of life: molecules and morphology in phylogenetic

analysis. Elsevier, Amsterdam.

Fortey, R. A., and R. H. Thomas. 1998. Arthropod relationships.

Chapman and Hall, London.

Hennig, W. 1950. Grundzьge einer Theorie des phylogenetischen

Systematik. Deutscher Zentraverlag, Berlin.

Hennig, W. 1966. Phylogenetic systematics. University of

Illinois Press, Urbana.

Judd, W. S., C. S. Campbell, E. A. Kellogg, P. F. Stevens, and

M. J. Donoghue. 2002. Plant systematics: a phylogenetic

approach. 2nd ed. Sinauer, Sunderland, MA.

Kenrick, P., and P. R. Crane. 1997. The origin and early

diversification of land plants: a cladistic study. Smithsonian

Institution Press, Washington, DC.

Littlewood, D. T. J., and R. A. Bray (eds.). 2001. Interrelationships

of the platyhelminthes. Taylor and Francis, London.

Patterson, C. 1989. Phylogenetic relations of major groups:

conclusions and prospects. Pp. 471–488 in The hierarchy of

life: molecules and morphology in phylogenetic analysis

(B. Fernholm, K. Bremer, and H. Jцrnvall, eds.). Elsevier,

Amsterdam.

Stiassny, M. L. J., L. R. Parenti, and G. D. Johnson (eds.).

1996. Interrelationships of fishes. Academic Press, San Diego.

Tudge, C. 2000. The variety of life. Oxford University Press,

Oxford.

This page intentionally left blank

563

Навигация