UvA-DARE (Digital Academic Repository)
Macroevolution of Animal Body plans
Jenner, R.A.
Link to publication
Citation for published version (APA):
Jenner, R. A. (2002). Macroevolution of Animal Body plans Amsterdam: Grafische Producties, UvA
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),
other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating
your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask
the Library: http://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,
The Netherlands. You will be contacted as soon as possible.
UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)
Download date: 15 Jun 2017
Chapterr 4
Evolutionn of Animal Body Plans:
thee Role of Metazoan Phylogeny at
thee Interface between Pattern and
Process s
Ronaldd A. Jenner
Publishedd in Evolution & Development 2: 208-221 (2000)
75 5
76 6
Evolutionn of animal body plans: the role of metazoan phytogeny at the
interfacee between pattern and process
Ronaldd A. Jenner
Institutee for Biodiversity and Ecosystem Dynamics, University of Amsterdam, P.O. Box 94766,1090 GT Amsterdam,
Thee Netherlands
Conaspondenoss (emaB: jenmrObio.uva.nl)
SUMMARYY Comprehensive integrative studies are the haB- sicall view. First, I expose this tradrttonal view to be a sknpMed
markk of evolutionary developmental biology. A properly de- historicall abstraction that became textbook dogma. Second, I
finedd phytogenetic framework takes a central place in such discusss how tworecentimportant stuoles of animal body plan
analysess as the meeting groundtorobservation and inference. evolution,, examining the evolution of the platyhelminth body
Molecularr phytogenies take mis place in many current stuoles
planarKltr»evolutic<wrysignfflcaric^ ^
onn animal body plan evolution. In particular, 18S rRNA/DNA andd set-aside eels, have actively incorporated two problematic
sequencee analyses have yielded a new view of animal evolu- aspectecrftrwrawtyemerüjngrrwleculajvtew w
tiontion that is often contrasted with a presumed traditional orclaslution:: incomplete and unresolved phytogenies.
EVOLUT10NARYY DEVELOPMENTAL BIOLOGY
ANDD T H E CENTRALITY OF
PHYLOGENET1CC INFORMATION
Georgee Gaylord Simpson (1951, p. 51) wrote "Over and over
againn in the study of the history of life it appears that what can
happenn does happen." Particularly studies of animal developmentt and phylogeny have recently played key roles in elucidatingg the potentialities and actualities of the evolutionary
historyy of the animal kingdom, and it is in die field of evolutionaryy developmental biology (popularly called "evo-devo")
wheree such studies of development and phylogeny meet.
Evo-devoo seeks to explain the evolution of morphology and
developmentt and the underlying developmental and genetic
mechanismss by comparing different organisms. A properly
definedd phylogeny is the only meaningful framework for
comparisonn if the goal is the reconstruction of die nature and
directionn of evolutionary change, and the discrimination betweenn convergent similarity and homology (Raff 1996; Arthur
1997;; Hall 1999). Consequently, it is instructive to consider
inn detail the basis of some of die recent advances in our understandingg of the evolution of animal body plans. The use
off I8S rRNA/DNA phytogenies is rapidly gaining prominencee as the basis for building evolutionary scenarios. At
presentt there is no generally accepted morphology-based
phylogenyy of the Metazoa to compete with it, even with
thee widespread adoption of cladistic methods (Jenner and
Schramm 1999). Often, a traditional textbook phylogeny based
onn morphology is contrasted with the emerging view from
molecularr systematics.
Inn mis paper I address die following issuesfroma phylogeneticc perspective. First, 1 consider the traditional textbook
phylogenyy and explore its historical roots. Second, I criticallyy discuss die newly emerging view of animal evolution
basedd on 18S rRNA/DNA sequence data, which is sketched
inn the latest generation of review papers as a basis for advancingg scenarios of body plan evolution. The new view of
animall evolution is often presented in phytogenies characterizedd by two striking features: lack of resolution and incompleteness.. These data limitations have subsequently been activelyy incorporated as essential ingredients into various
importantt new hypotheses of animal body plan evolution.
Third,, I will re-evaluate two of these hypotheses that are beginningg to be considered major advances in our understandingg of animal evolution: platyhelminthes as derived coelomates,, and the evolutionary significance of set-aside cells
andd their rote in the Cambrian explosion.
METAZOANN MORPHOLOGY, UBBIE HYMAN,
ANDD THE "TRADITIONAL" VIEW OF
ANIMALL EVOLUTION
Sincee Ernst Haeckel codified our profession by coining its
currentt name and by establishing a coordinated research programm for the elucidation of evolutionary relationships be-
OO BLACKWEU. SCIENCE. MC.
77 7
tweenn organisms, the study of higher level animal phytogeny pliess structural resemblance or evolutionary propinquity.
hass yielded an expansive literature but relatively little de- Second,, Hyman published The Invertebrates over a time
tailedd consensus. It is therefore not surprising often to find spann of 27 years, creating enough room for change of opinmerelyy the shortest of possible summaries of morphological ion.. She considered views on animal relationships as ephemphylogeneticss in current papers on the subject Winnepen- erall products, necessarily changing with the accretion ofnew
ninckxx et al. (1998a, p. 888) epitomize tfiis approach in die dataa (Hyman 1940, p. 27; 1959, p. 697). Nevertheless, her
firstt sentence of their paper "Despite several decades of mor- distilledd views have been widely accepted as "traditional"
phologicall and anatomic research, numerous aspects of meta- consensuss by many biologists, and Hyman's viewpoint has
zoann relationships have remained uncertain." Their discus- beenn considered as so well established and typical mat it is
sionn then turns to die revisionary views from molecular noo longer necessary to refer to any particular of her works
systcmatics.. Aguinaldo and Lake (1998), Holland (1998), (seee Barth and Broshears 1982; Ruppert and Barnes 1994),
Balavoinee (1998), and Erwin (1999) adopt a similar approach andd that her view can stand as an epitome for the Anglobyy invariably presenting the Anglo-Saxon "traditional text- Saxonn view of animal relationships (Willmer and Holland
bookk view,'" die acoelomate-pseudocoelomate-coelomate 1991).. I will provide an alternative perspective on Hyman's
series,, as an icon for metazoan phytogeny based on morphol- viewss on animal phylogeny, which I regard as better supogy.. Balavoinc (1998), Balavoine and Adoutte (1998), Ad- portedd by her own statements than the traditional textbook
outtee et al. (1999), and Valentine et al. (1999) in fact claim treee for which her support is commonly claimed.
thatt such a phylogenetic series represents "... a glorious saga
Hymann indeed divided the Metazoa into diploblasts and
off progressive increase in complexity . . ." (Adoutte et al. triploblasts,, although her difficulties with die germ layer fheory
1999.. p. 105), arising from ". . . the intellectual bias for in- ledd her to regard die distinction between Radiata and Bilateria
creasingg complexity in evolution . . ." (Balavoine and Ad- ass more meaningful (Hyman 1940). Widiin the Bilateria, Hyouttee 1998, p. 397). It would be a serious problem indeed if mann did indeed distinguish between acoelomates, pseudocoeourr efforts to reconstruct animal phytogeny would produce lomates,, and coelomates stating: "Such a division stands firmly
nothingg more than manifestations of our intellectual disposi- onn a realistic anatomic basis and eschews all theoretical vationn for increasing complexity in evolution. However, I will porizings..."" (Hyman 1940, p. 35). She stated that "These
showw that these assessments are distorted and misleading.
groupingss [acoelomates, pseudocoelomates, coelomates] do
Inn point of fact, there exists no such thing as "the tradi- not,, however, entirely correspond to taxonomie relationships''
tionall textbook phytogeny." A diversity of different schemes (Hymann 195 la, p. 23). Hyman interpreted diese groups not as
cann be found, influenced by factors such as nationality, depth cladess or phylogenetic groups, but as organizational grades (Hyoff historical treatment, and a whole range of personal per- mann 1940, chapter II). Her consideration of priapulids strikspectivess of the textbook authors (e.g., Willmer 1990; Will- inglyy illustrates this distinction: priapulids are classified as
merr and Holland 1991). The most frequently mentioned schizocoell coelomates based on structural grade (Hyman
sourcee in Anglo-Saxon works is Hyman's influential series 1940,, p. 34) but phylogenetically united wim die pseudocoeTheThe Invertebrates (1940-1967) (e.g., Field et al. 1988; Rafflomatee aschelminths (Hyman 1951b, pp. 54,56, 196.197).
ett al. 1989; Schram 1991; Willmer and Holland 1991; Rup- Hymann stated that schizocoel coelomates are related to die
pertt and Barnes 1994; Balavoine 1998; Garey and Schmidt- acoelomatess and pseudocoelomates, together comprising die
Rhaesaa 1998; Adoutte et al. 1999). Hyman is often cited in Protostomiaa (Hyman 1940, p. 36; 1951a, p. 5). She envisioned
supportt of the traditional textbook phytogeny (e.g., Barth thee immediate splitting of die Bilateria from the last, acoetoandd Broshears 1982; Willmer 1990; Schram 1991; Willmer mate,, common ancestor into two lineages (Hyman 195 la, p. 4):
andd Holland 1991; Ruppert and Barnes 1994; Willmer I99S; onee leading to die Deuterostomia (enterocoelous coelomates)
Garcyy and Schmidt-Rhaesa 1998; Adoutte et al. 1999), andd one leading to the Protostomia, which includes the extant
whichh can be summarized as follows: after the divergence of acoelomates,, pseudocoelomates, and coelomate protostomes.
thee Radiata (Cnidaria and Ctenophora), the bilaterians are ar- Shee considered die Protostomia as".. .evidently relatedtoeach
rangedd in the acoelomate-pseudocoelomate-coelomate (pro- other"" (Hyman 1951a, p. 5). The only metazoan phylogeny from
tostomess and deuterostomes) series, which reflects succes- TheThe Invertebrates (Hyman 1940, Fig. 5) reflects diis grouping
sivee branchings of dwse mree clades. How well is mis Herr provisional acceptance oftiiisdiphyletic dieory ofbilaterianrian relationshipstiiereforecontradicts die monophyletic Eu
phylogeneticc scheme corroborated in Hyman's work?
Itt is not easy to represent accurately Hyman's opinion on coelomata,, comprising both coelomate protostomes and deumetazoann relationships for several reasons. First, she re- terostomes,, which is at die heart of die traditional textixwk
gardedd classification and phytogeny as two logically distinct phylogenyyforwhich Hyman's support is often claimed (Barm
devicess to represent the "relationships" between organisms andd Broshears 1982; Field etal. 1988; Raffetal. 1989; Willmer
(1940,, chapter 11). Consequently, when Hyman described 1990;; Schram 1991; Willmer and Holland 1991; Ruppert and
"affinities"" it may be difficult to determine whether this im- Barness 1994; Willmer 1995; Garey and Schmidt-Rhaesa 1998;
78 8
Adouttee et al. 1999). Seemingly reasonable arguments in support describess "higher" and "lower" groups, and specifies the oroff the traditional interpretation can be traced to two sources derr of treatment of the phyla in The Invertebrates according
inn Hyman: discussions of body cavities in 1940 (chapter II) too "... the general order of their structural complexity" (1940,
andd 195 la (chapter IX) and discussion of the phylogenetic p.. 38), she refers to a series of organizational grades, not a
affinitiess of lophophorates (1959, chapter XXII).
phylogeneticc series (e.g., see Willmer 1990, p. 5 for an illusDeducingg support for Eucoelomata as a phylogenetic unit trationn of this confusion). The only potential genuine phylofromm Hyman's discussions of body cavities is based on a geneticc series according to body cavity construction in Hyconfusionn of grades and clades. As mentioned above, Hyman man'ss work is die hypothesized series of ancestors leading to
regardedd the division of bilaterians into acoelomates, pseu- thee protostomes (1951 a, p. 17): planuloid ancestor—acoeloid
docoelomates,, and eucoelomates not as phylogenetic units, ancestorr (acoelomate grade)—trochozoon (pseudocoelomate
butt as organizational grades. Only by misinterpreting these grade).. But this is distinct from the progressive series leading
gradess as clades, by denying the argued phylogenetic unity too a monophyletic Eucoelomata depicted in most recent paoff the Protostomia, and by judiciously connecting the perss quoting Hyman's work.
Inn conclusion, citing Hyman as the principal source for the
branchess in ascending order from acoelomate to coelomate
cann one defend Hyman's discussion of body cavities to sup- "traditionall textbook tree" is either based on a confusion of
portt the traditional textbook tree. This reasoning may ex- gradess and clades or is at best a simplified abstraction of her
plainn the appeal to Hyman in support of the traditional phy- views.. Nevertheless, the Hymanian phylogeny has become
logcnyy in various papers (e.g., Schram 1991; Willmer 1995; dogmaa in subsequent works, and an icon for the Anglo-Saxon
Carranzaa et al. 1997; Garey and Schmidt-Rhaesa 1998; Ad- vieww on animal phylogeny. In contrast, I would argue that her
mostt cogently argued opinion on animal phylogeny is outouttee etal. 1999).
AA second potential source of support for the traditional in- linedd in the first volume of The Invertebrates (mainly chapter
terpretationn of Hyman may be her discussion of the lopho- II)) and is reinforced at various places in later volumes, and
phoratess (which she considered protostomes) as intermedi- paintss a different picture of animal phylogeny summarized in
atess between coelomate protostomes and deuterostomes. The Fig.. 5 of Hyman (1940). It emphasized die unity of the Promostt convincing statement is that"... deuterostomes show a a tostomia,, including acoelomates, pseudocoelomates, and cofurtherr development of characters beginning in an unclear elomatess (including lophophorates), and did not support a
wayy in lophophorates, and thus branch off from protostomes monophyleticc Eucoelomata and a linear, progressive phylobyy way of the latter" (1959, p. 605), although Hyman does not geneticc series according to body cavity organization. Neversupplyy any information as to how this link may be conceived. theless,, the 27 year time span between the first and the last
Thiss may be the only real support from Hyman for the tradi- volumee of The Invertebrates obviously provided enough
tionall textbook phylogeny. Strikingly, in the same volume roomm for a creative thinker such as Hyman to change her
Hymann argued that chaetognaths may have diverged early viewss and explore different alternatives. It may be difficult to
fromm die common ancestor of the Bilateria at die time that the interprett Hyman's phylogenetic views precisely since the
dipleurulaa ancestor of the deuterostomes became differenti- theoryy and practice of phylogenetics have changed so proatedd (1959, p. 66). This again points to the independence of foundlyy since her time. Representing Hyman's view in a clathee protostome and deuterostome lineages since their last dogramm (e.g., Eemisse et al. 1992; Adoutte et al. 1999) can
commonn acoelomate ancestor, and is in obvious contrast to thereforee never encompass the full subtlety of her perspective
becausee cladistics cannot easily accommodate die central pothee traditionally presented "Hymanian" phylogeny.
Howw well does Hyman's work support the traditional sitionn for direct ancestors such as die planuloid and acoeloid
acoelomate-pseudocoelomate-coelomatee series as a story of ancestors,, which would be needed for an accurate representalinearr progressive increase in complexity (Balavoine and tionn of historical information.
Acceptingg the traditional textbook tree as an impoverished
Adouttee 1998; Adoutte et al. 1999; Valentine et al. 1999)?
Hymann was convinced of the branching nature of evolution, historicall abstraction, and given the absence of a detailed
andd quite in contrast to our current penchant for relying on consensuss of animal phylogeny on the basis of morphology,
dichotomouslyy branching diagrams, she conceived of many manyy biologists have turned to molecular systematics in
groupss of bilaterians as arising at the same level, thus creat- searchh of a phylogenetic framework.
ingg polychotomies (1940, p. 39, Fig. 5). She emphasized the
bush-likee branching of the animal evolutionary tree, arguing
againstt a linear phylogenetic arrangement of organisms ac- MOLECULARR PHYLOGENY AND ANIMAL
cordingg to their structural complexity (1940, p. 39). Thus, in EVOLUTION:: THE "NEW VIEW"
Hyman'ss exegesis we cannot trace a conscious intellectual
biass to impose notions of necessary evolutionary progress in Molecularr phylogenetics is currently dominated by the study
complexityy on her views of animal phylogeny. When Hyman off 18S rRNA/DNA sequences. This research has yielded a
79 9
consensuss phytogeny, albeit of limited resolution (e.g., Aguinaldoo et al. 1997; Aguinaldo and Late 1998; Balavoine
1998;; Balavoine and Adoutte 1998; Maley and Marshall
1998;; McHugh 1998; Garey and Schmidt-Rhaesa 1998;
Halanychh 1998; Winnepenninckx et al. 1998a; Adoutte et al.
1999;; Knoll and Carroll 1999). This consensus proposes
threee major bilaterian clades. The first split divides the deuterostomess and protostomes, with protostomes consisting of
thee sister clades Lophotrochozoa and Ecdysozoa. The deuterostomess are the echinodenns, hemichordates, and chordates.. Lophotrochozoa comprises groups such as molluscs,
annelids,, platyhelmindis, rotifers, phoronids, bryozoans, and
brachiopods;; Ecdysozoa unites molting animals such as arthropods,, tardigrades, and nematodes. The Acoela may be
basall bilaterians (Ruiz-Trilloet al. 1999), although data from
elongationn factor I-a suggest they are lophotrochozoans
(Bemeyy et al. 1999). The relationships within the three main
cladess of the Bilateria are especially unresolved. Causes for
thee lack of resolution are discussed elsewhere (e.g., Lecointree et al. 1993; Phillipe et al. 1994; Abouheif et al. 1998;
McHughh 1998; Foster and Hickey 1999). Not all analyses
agreee in all aspects. A few analyses yield a paraphyletic Protostomiaa uniting deuterostomes either with ecdysozoans or
lophotrochozoanss (see Eemisse 1997; Winnepenninckx et
al.. 1998a; Zrzavy et al. 1998 for recent examples).
Thee icon of the new view of animal phytogeny depicts the
threee bilaterian clades (Lophotrochozoa, Ecdysozoa, Deuterostomia)) as polytomies of very limited membership, virtuallyy never depicting all the relevant taxa (Aguinaldo and
Lakee 1998; Balavoine 1998; Balavoine and Adoutte 1998;
Adouttee et al. 1999; Knoll and Carroll 1999). Of course, this
modee of representation should be regarded as a convenient
shorthand,, illustrating the uncertainty of the phylogenetic relationshipss in a cladogram of manageable size, whereas in
olderr studies these incomplete trees reflect limited availabilityy of sequence data. However, I will show that these incompletee and unresolved trees are being taken too literally as representationss of our new understanding of animal evolution in
aa significant proportion of these papers. Strikingly, these
limitationss now form integral parts of recently proposed scenarioss of animal body plan evolution, two of which will be
discussedd in detail: the evolution of the platyhelminth body
plan,, and the evolution of primary larvae and set-aside cells.
AA properly expanded molecular phytogeny including all pertinentt taxaformsthe basis of the following re-evaluation.
PLATYHELMINTHSS AS DERIVED COELOMATES
Balavoinee (1997,1998) investigated the phylogenetic positionn of the phylum Platyhelminthes on die basis of limited
18SS rRNA and Hox gene data. He proposed that these data
favorr the grouping of plaryhelminths with die coelomate spi-
raliann protostomes. The 18S rRNA data slightly favor a
groupingg of rhabditophoran platyhelmindis (acoelomorphs
aree excluded) witii lophotrochozoans as opposed to a positionn basal to either the Bilateria or all die protostomes. Balavoine'ss use of Hox gene data is innovative and potentially
providess independent support for a deep split between lophotrochozoanss (including platyhelmindis) and ecdysozoans
(seee also De Rosa et al. 1999; Kobayashi et aL 1999).
Balavoinee then considered the question of die origin of
thee platyhelminth body plan in light of this new phylogenetic
data.. Balavoine argued that me platyhelmindis tost die anus,
coelom,, and probably segmentation on the basis of what he
summarizedd in his Fig. 3 (reproduced here as Fig. la and lc).
Hee provided an evolutionary scenario based on heterochrony
(orogenesis)) to account for the origin of the platyhelminth
organization.. This is an interesting hypowesis, but is it well
supported?? Comprehensive use of available data necessitates
aa strikingly different alternative.
Thee cladograms presented in Balavoine (1997, 1998) are
cladogramss of bilaterian coetomates only. Applying die dubiouss principle "common is primitive" Balavoine states, "Most
off them are coetomates, which of course, can be best explained
byy die fact mat die ancestor was already a coelomate" (1998,
p.. 854). We might accept diis conclusion if there was a sufficientlyy broad sampling of ana, in particular Üiose taxa lacking
aa coelom. Unfortunately, Balavoine's (1998) cladogram is
highlyy pruned, lacking all aschelmindis, and quite unresolved.
Nevertheless,, Balavoine effectively used bom die lack of all
pertinentt taxa and die lack of resolution of the cladograms to
buildd his argument Using pruned, unresolved trees in this way
iss empirically inadequate and mediodotogically flawed. If we
wantt to make any sensible inferences about die evolution of
animall body plans, a first requisite is to take all pertinent taxa
intoo consideration in order to allow die proper estimation of
characterr distribution, i.e., to provide a comprehensive focus.
Notwithstanding,, too often audiors rely on incomplete trees
thatt do not allow any meaningful evolutionary hypodieses to
bee proposed (Jenner 1999). If we consider die fullness of die
metazoann phytogeny, eidier derived from morphology or molecules,, the picture could change drastically.
Lett us test Balavoine's conclusions (loss of anus, coelom,
segmentation)) in the light of a more complete consideration
off 18S rDNA phytogenies. The cladogram depicted in Figs,
lb,, Id, and 3c will serve as the basis for die discussions in
miss paper. It represents a conservative consensus estimate of
diee I8S rDNA phytogeny based chiefly on recent comprehensivee studies (Eemisse, 1997; Littlewood et al. 1998.
1999;; Winnepenninckx et al. 1998b; Zrzavy et aL, 1998;
Ruiz-Trilloo et al. 1999). For simplicity I consider the diploblastss (poriferans, placozoans, cnidarians, and ctenophores)
ass a sister clade to the Bilateria. The Acoela are not included
inn die cladogram. These simplifications do not affect die validityy of die reinterpretations.
80 0
Coelom m
I absent
present
t = ii equivocal
II
Segmentation n
I absent
present
^^33 equivocal
II
Fig.. 1. Cladogram from Balavoine (1998) (a. c) and a conservative consensus dadogram of metazoan relationships based on lfSSrDNA
sequencee data (b, d).with mapping of coelom (a. b), and segmentation (c,d). Consensus cladogram based on Eernisse (1997); Littlewood
ell al. (1998.1999): Winnepenninckx et al. (1998b); Zrzavy et al. (1998); Ruiz-Trillo et al. (1999). "Other ecdysozoans" comprises an unresolvedd assscmblage of nematodes, nematomorphs. priapulids. kinorhynchs, and "other lophotrochozoans" is an unresolved clade of
coelomatee protostomes including lophophoratcs, nemerteans. and entoprocts.
AA variety of molecular phylogenetic studies published beforee Balavoine's papers indicated that various aschelminth
orr pseudocoelomate phyla are likely to be closely associated
withh the lophotrochozoans and ecdysozoans (e.g., Winne-
penninckxx et al. 1995; Garey et al. 1996a, 1996b; Hanelt et
al.. 1996; Aguinaldo et al. 1997; Eemisse 1997). Recent studiess confirm this picture (e.g., Aleshin et al. 1998; Garey and
Schmidt-Rhaesaa 1998; Littlewood et al. 1998, 1999;Schmidt-
81 1
Rhaesaa et al. 1998; Winnepenninckx et al. 1998a; Zrzavy et iss the possible change in perspective concomitant with a
al.. 1998; Ruiz-Trillo et al. 1999). If we regard die anus of bi- broaderr sampling ofrelevanttaxa. I do not propose to solve
laterianss as homologous, we could conclude with some con- thee issue of die origin and evolution of coeloms in the Bilatfidencefidence that platyhebnindis have lost their anus, agreeing
eriaa in this paper by the sole criterion of parsimony. Detailed
withh Balavoine's conclusion. However, Balavoine's hypoth- considerationn of the morphology, ontogeny, and function of
esiss of the presence of a coelom in the ground pattern (the set coelomss is, of course, necessary for a proper understanding
off characters primitively present in a cladc) of the Bilateria off coelom evolution, but such treatment is outside the scope
andd subsequent loss of the coelom in platyhelminths is con- off this paper and largely irrelevant for the main message of
tradictedd by current information (Fig. lb). In plotting occur- thiss reinterpretation.
rencess of a coelom on his cladogram, Balavoine provided no
Furthermore,, Balavoine concluded that segmentation is
definitionn of a coelom. He discussed neither the divergent mostt probably ancestral for bilaterians. Balavoine gives no
morphologies,, nor the various developmental modes or func- definitionn for segmentation, although his tree (Fig. lc) sugtionss of coeloms in sufficient detail. Judging from Bala- gestss a very broad definition probably referring to anteroposvoine'ss mapping of coeloms in Fig. la, he adopted an indis- teriorlyy repeated mesodermal derivatives such as coeloms or
criminatee structural definition of a coelom, irrespective of musculature.. However, the distribution of segmentation in
ontogenyy or function (i.e., mesoderm lined body cavity, Niel- Balavoine'ss tree (Fig. lc) is misleading given current insenn 1995), although this does not explain why nemerteans sights.. There is mounting molecular (Black et al. 1997;
aree not scored, since their circulatory system (Turbeville 1986) Kojimaa 1998; McHugh 1997, 1999; but see Siddall et al.
andd rhynchocoel (Turbeville and Ruppert 1985) answer such 1998)) and morphological evidence (Bartolomaeus 1995,
aa definition of a coelom. The conservative 18S rDNA phy- 1998;; Meyer and Bartolomaeus 1996; Rouse and Fauchald
+ that pogonophores are derived polychaetes, equating
togenytogeny shows that Gastrotricha, Cycliophora, and Rotifera1997))
Acanthocephalaa are closely associated with rhabditophoran theirr segmentation with that of annelids. Segmentation of
platyhelminths,, and either the Gastrotricha alone orr the three molluscss is very dubious and the molluscan ground pattern
taxaa together form a sister clade to the remaining lophotro- doess certainly not include a situation comparable in structure e
chozoans.. The Gnathostomulida may also be associated with orr ontogeny to the segmented coelomic cavities found in anthesee three taxa, or with the ecdysozoans. Published compre- nelidss (e.g., Salvini-Plawen 1990; Haszprunar 1996). Seghensivee 18S rDNA analyses suggest independent evolution mentationn in deuterostomes is morphologically and ontogeoff the coelom in the deuterostomes, the lophotrochozoans, neticallyy distinct from protostome segmentation and it is
andd ecdysozoans as the most parsimonious solution (Fig. onlyy by virtue of the unresolved relationships in Balavoine's
11 b). Consequently, character distribution suggests that platy- deuterostomee clade that segmentation can be regarded as
helminthss never lost their coelom, nor did the other basal primitivee for deuterostomes. Nevertheless, adopting an equally
noncoelomates.. This interpretation corroborates the lack of indiscriminatee definition of segmentation as Balavoine and
evidencee for a coelom from platyhelminth embryology and provisionallyy accepting segmentation in the ground pattern
morphology.. Balavoine's conclusion is clearly dictated by off "other lophotrochozoans," Fig. Id indicates that (at least
consideringg highly incomplete trees, lacking virtually all the rhabditophoran)) platyhelminths are likely basal to annelids
importantt pseudocoelomate taxa. If loss and gain of a co- andd molluscs (grouped in "other lophotrochozoans"), indielomm are considered equally likely, convergent evolution of catingg the primitive absence of segmentation in platyhelaa coelom is the most parsimonious solution (Fig. lb). It minths. .
shouldd be noted that Fig. lb maximizes the probability of
Inn addition, Balavoine claimed support for the homology
findingfinding homologous coeloms. No distinction is made onoff
thesegmentation across all phyla from recent work on segbasiss of morphology, ontogeny or function, adopting a purely ment-polarityy genes and pair-rule genes. However, this argustructurall definition of a compartment between epidermis mentt is unconvincing. The demonstration of the role of these
andd gut, and lined widi a mesothelium (Ruppert 1991). If the geness in the elaboration of a segmented body plan in amnoncoelomatee ecdysozoans and lophotrochozoans do not phioxuss and Drosophila is no straightforward argument supformm sister taxa of the panarthropods and the other (mainly portingg homology of chordate and arthropod segmentation
coelomate)) lophotrochozoans, respectively, men the likeli- (e.g.,, Jenner 1999). Moreover, Balavoine does not supply
hoodd of convergence of coeloms increases. The coelomate anyy information on their presence or function in platyhelchaetognathss can at this moment not be positioned with minths,, while his argument hinges upon the one-to-one mapcertainty,, although the recent proposals of a relationship ei- pingg of the presence of these genes with die presence of mortherr with panarthropods (Zrzavy et al. 1998) or nematodes phologicall segmentation. There is mounting evidence from
(Halanychh 1996; Eernisse 1997) will not influence the map- developmentall genetics that this premise is very questionpingg of a coelom as depicted in Fig. lb. It needs to be empha- able,, and that the potency of using isolated gene expression
sized,, however, that the salient point of mis reinterpretab'on patternss to illuminate morphological homology can be seri-
82 2
ouslyy limited, especially in distantly related taxa (e.g., Davidsonn 1997; Scholtz et al. 1998; Holland and Holland 1999;
Janiess and DeSalle 1999).
Inn summary, Balavoine's conclusions that platyhelminths
aree derived coelomates that lost the coelom and segmentationn cannot be upheld. His pruned and unresolved phytogeny
bearss at best ambiguous testimony, while a proper considerationn of morphology in the context of molecular phytogeny
arguess against his hypothesis. Instead, multiple convergencess of coelom and segmentation are indicated.
Finally,, Balavoine's interesting heterochronic scenario
forr the origin of platyhelminths is effectively proposing processs where pattern is lacking. Balavoine (1997) stated that
".... the Muller larva of polyclads is, in some traits, reminiscentt of the trochophore of annelids and molluscs, which may
indicatee that the possession of a trochophore-like larva is
ancestrall in a (platyhelminths + eutrochozoans) clade" (p.
92).. He also suggested homology of nemertean pilidium and
polycladd Müller's or Götte's larva, following the argument
off Nielsen (1995). Balavoine (1998) then proposed a progeneticc series beginning with an annelid-like ancestor with a
trochophoree larva, through a nemertean intermediate with
aa pilidium larva, and culminating in a platyhelminth with a
Müller'ss or Götte's larva. Balavoine (1998) stated that the
progeneticc origin of platyhelminths *'... always seemed to be
aa 'nice story'... but it is not readily testable" (p. 856). Indeed,, testing progenesis as a causal evolutionary process is
difficult,, but die first step can already be taken by rigorously
applyingg available phylogenetic information, which will necessitatee a reinterpretation on two levels.
First,, we need to ascertain that Müller's or Götte's larvae
andd pilidium larvae are homologous, and primitive for the
platyhelminthss and nemerteans, respectively. Conceding sufficientt special morphological similarities to warrant initial homologyy of polyclad and pilidium larvae (but see Salvini-Plawenn 1980; Rouse 1999), can we consider these larval types
primitivee for these taxa? There is general consensus that the
polycladd Müller's and Götte's larvae, and die nemertean pilidiumm larvae, are not present in die ground pattern of the two
phylaa (Haszprunar et al. 1995; Nielsen 1995,1998). Morphologicall and molecular phylogenetic studies support diis conclusionn in die platyhelmindis (e.g., Carranza et al. 1997; Camposs etal. 1998; Littlewoodetal. 1999; Ruiz-Trilloetal. 1999),
whereass a phytogeny for the Nemertea is not yet available, allowingg only a tentative conclusion (Henry and Martindale
1997).. However, pilidium larvae are restricted to some ancplann species only. Paradoxically, Nielsen (1995) united platyhelminthss and nemerteans as Parenchymia on die basis of proposedd similarities in polyclad and pilidium larvae, while
realizingg that these larval types are not likely to be ancestral
forr these phyla, and clearly his adherence to the trochaea theoryy is responsible for this conclusion. Support from Nielsen
(1995)) for Balavoine's hypothesis is therefore suspect.
Second,, the evolutionary branching order of the annelids,
nemerteans,, and platyhelminths needs to be consistent with
Balavoine'ss progenetic series if annelids and nemerteans are
too exemplify precursors in a series of evolutionary transitions.. It is only by virtue of die lack of resolution in his cladogramm that Balavoine can propose his speculative scenario.
Nonee of die current molecular and morphological studies
clearlyy indicate a phylogenetic position for annelids and
nemerteanss as steps in an evolutionary series culminating in
platyhelminthss (see Figs, lb and Id). It is more probable tint
diee trochophore larva evolved in a more restricted set of taxa
afterr platyhelminths split off, a hypothesis supported by
Rousee (1999), while pilidium and polyclad larvae are derivedrived within Nemertea and Platyhelminthes, respectively.
Somee 22 years ago, Gould warned about die speculative
invocationn of heterochronic mechanisms to explain the originn of higher taxa: "It is unfortunate that most literature on
paedomorphosiss is cast in the same mold diat bolstered recapitulationn during the previous half century—speculative
phytogenyy of higher taxa" (p. 277) (Gould 1977). Heterochronyy is a powerful concept to understand evolutionary
changess in animal form (e.g., McNamara 1995). Although
heterochronicc processes have been proposed to account for
thee origin of many higher level animal taxa, from mystacocaridd crustaceans to larvacean urochordates, a properly specifiedd phylogenetic framework is essential to determine die
ancestrall and descendant states, and therefore a properly resolvedd phytogeny can serve as an efficient test for a heterochronicc scenario.
SET-ASIDEE CELLS AS A MECHANISTIC
EXPLANATIONN OF THE CAMBRIAN EXPLOSION
Thee set-aside cell hypothesis developed by Davidson and
colleaguess (Davidson 1991; Davidson et al. 1995; Peterson
ett al. 1997; Cameron et al. 1998; Arcnas-Mena et al. 1998)
hass been regarded as an important advance in our current understandingg of animal body plan evolution (Hall 1999; Knoll
andd Carroll 1999; Olsson and Hall 1999). It addresses the
controversiall subject of die origin and evolution of animal
lifee cycles (e.g., Strathmann 1993; Strathmann and Eemisse
1994;; Haszprunar etal. 1995; Hart etal. 1997; Nielsen 1998;
Pechenikk 1999). The core argument of mis elegant scenario
iss an ingenious synthesis of descriptive and molecular embryology,, metazoan phytogeny, and the fossil record The
hypothesiss can be regarded as an explanation for the absence
off a pre-Ediacaran metazoan fossil record, almough it is
rootedd in negative evidence, and it may be a viable hypothe-siss explaining die origin and diversification of large bilateriann body plans (Vermeij 1996; Wray et al. 1996; Conway
Morriss 1997; Erwin 1999; Lieberman 1999; Smith 1999).
Thee central points of the scenario are: a biphasic life cycle
83 3
withh primary larvae and maximal indirect development (i.e.,
embryonicc [zygote to embryo/larva] and postembryonic
[larvaa to adult] stages of the life cycle are completely separablee both in terms of morphology [larval cells vs. set-aside
cells]] and developmental process [Type I embryogenesis vs.
patternn formation]) is primitive for Bilateria; the bilaterian
ancestorr is similar to a ciliated primary larva with a populationn of set-aside cells; and the rapid divergence of adult body
planss (Cambrian explosion) from homologous larvae with
homologouss set-aside cells.
Too put the following analysis into perspective, I will
brieflyy summarize pertinent discussions of the set-aside cell
hypothesiss in the literature. Critiques of the set-aside cell hypothesiss can be arranged along four lines dependent on the
chosenn perspective: convergence of adult body plans; adaptationistt reasoning; correlation of reproductive traits; and
distributionn of developmental types within the Metazoa.
Liebermann (1999) and Smith (1999) considered the unlikelihoodd of the synchronous, independent evolution of many
macroscopic,, adult body plans during the Cambrian explosion
ass the biggest stumbling block of the hypothesis. This appears
too be an important caveat, but at present we have no establishedd criteria forjudging the amount of expected convergence
inn animal evolution, although convergent evolution may be
prominentt in invertebrates (Moore and Willmer 1997).
Lacallii (1997) and Wolpert (1999) used adaptationist reasoningg as the basis for their critiques. Lacalli (1997) argued
thaii evolution should favor rapid development over slow development,, for example, to decrease predator pressure. The
needd to maximize the rate of development should favor the
evolutionn ofrigidmodes of developmentfrommore flexible
antecedents,, making it likely that the developmental processess forming the larva (Type I embryogenesis) have been
secondarilyy imposed upon pre-existing, more flexible patternn formation processes (forming the adult), although this
doess not explain why these primitive and supposedly slower
developmentall processes evolved in the first place. Wolpert
(1999)) argued that every evolutionary transition must be
graduall and adaptive. The evolutionary origin of set-aside
celtss would then be insurmountable, because what could
havee constituted the selective advantage of these cells before
thee adult stage that they would eventually give rise to had
evolved?? Although these criticisms may appear logical, they
aree difficult to test. I think Lacalli (1997) only shifts the
problem,, while Wolpert (1999) does not make the distinction
betweenn historical origin of a feature and its current utility, a
distinctionn that holds the key to some evolutionary riddles
andd that cannot be discounted in principle (Gould and
Lewontinn 1979; Gould and Vrba 1982). Although it may be
difficultt to reconstruct the evolutionary origin of set-aside
cellss with respect to their current function, set-aside cells
mightt turn out to be the most significant exaptation in the
historyy of the Metazoa.
Conwayy Morris (1998a, 1998c) argued against the primacyy of maximal indirect development on the basis of known
correlationss between reproductive traits in marine invertebrates.. Indirect development is virtually restricted to largebodiedd invertebrates (Olive 1985), and this would argue
againstt the likelihood of a small indirectly developing bilaterianrian ancestor. I do not consider this to be an effective argu
mentt since the set-aside cell hypothesis proposes a bilaterian
ancestorr reminiscent of a larva that itself represents die adult
stage,, thus exhibiting direct development (see discussion below).. The subsequent evolution of a biphasic life cycle could
bee linked with the origin of a large-bodied, new, adult stage.
Conwayy Morris (1998a, 1998b, 1998c) and Wolpert
(1999)) argued mat direct development may be primitive for
metazoanss based on the study of living taxa, in large measure
derivingg supportfromthe work of Haszprunar et aL (1995).
However,, the principal message in Haszprunar et al. (1995)
iss not that direct development is likely to be primitive for the
Bilateria,, but rather that a biphasic life cycle with planktotrophicc larvae is not likely to be plesiomorphic for the Bilateria.. Indirect development with Iecithotrophic larvae remainss a viable option for the original bilaterian life cycle.
Moreover,, since Haszprunar et al. (1995) do not provide a
phylogeneticc framework for their comparison of animal life
cycles,, I do not regard this study to be effective as an argumentt against the presence of indirect development in the
primitivee bilaterian life cycle. Similarly, Conway Morris'
(1998a,, 1998b, 1998c) and Wolpert's (1999) claim for supportt for an original metazoan life cycle with direct developmentt deduced from the proposed presence of various directly
developingg fossil metazoans in the Cambrian (Bengtson and
Yuee 1997) cannot be convincing since alternative modes of
development,, including direct and indirect development, are
widespreadd within the living phyla (e.g., Nielsen 1998). A
moree effective argument would have to rest on comparisons
withinn a phylogenetic framework of the Metazoa, allowing a
robustt assessment of plesiomorphic and apomorphic characterr states.
Thee hypothesis claims that a biphasic life cycle with maximall indirect development is primitive for Bilateria. This necessitatess the evolution from essential direct development
(forr the bilaterian ancestor resembled a reproductively mature
larva-likee organism) to maximal indirect development with a
separatee larva and adult bearing no morphological resemblancee at the base of the Bilateria. This hypothesis is only
testablee by a phylogenetic study when the out-groups suggest
directt development as primitive at the out-group node, and
thee in-group suggests indirect development as primitive for
thee in-group node, thereby enabling an evolutionary change
att the intemode basal to the Bilateria (Fig. 2). Note that we are
thenn dealing with two ancestors. This would better accommodatee the evolution of set-aside cells, because we can circumventt contradictory characters in one ancestor (at the same
84 4
Outgroupf») )
nott the reverse," because this direction of evolution is frequentlyy observed within phyla. So they write about"... the
\\
\
Bilateria
premisee that [maximal indirect development] represents the
ancestrall mode by which adult body plans are ontogeneticallyy produced." Cameron et aL (1998, p. 615) conclude that,
// ^ » maximal indirect "Thee important point is that the process of indirect developdevelopment
//
menttfroma ciliated larva, which itself bears little or no relationn in structure to the adult body plan to which it will give
rise,rise, is a character shared by the majority of animal phyla.
^directt
Therefore,, we postulate that the latest common ancestor of
development t
bilaterianss minimally had a larval stage similar to what is
foundd in modern indirectly developing marine organisms.**
Iff higher level phylogenetic research is to have any indeFig.. Z. Cladogram illustrating the phylogenetic «interpretation of
directt and maximal indirect development in the ground pattern of pendentt status as a potential source of insight into evolutionthee Bilateria. Note that there are two distinct ancestors, represent- aryy processes, we have to berigorousin analyzing pattern. In
edd by two intemodes. See text for discussion.
thiss context, a consideration of die acoelomate and pseudocoelomatee bilaterians is necessary. Aschelminths, for example,, have direct development, completely lacking primary
timee the larva is the adult and there is maximal indirect devel- ciliatedd larvae and set-aside cells that give rise to the adult
stage.. As discussed above, 18S rDNA data indicate basal poopmentt with a maximally different larva and adult).
Lett us examine the central points of this hypothesis in de- sitionss in the bilaterian clade of various directly developing
tail.. First, we must determine the phylogenetic level at which groupss (Fig. 3c). Therefore, it would appear that primary larthee set-aside cell scenario is said to be effective. Davidson ct vaee are not homologous in the Bilateria, and the evolution of
al.. (1995) and Peterson et al. (1997) proposed to explain the set-asidee cells is convergent It should be noted that the phydiversificationn of the Bilateria, although their cladograms logeneticc positions of the directly developing Acoela and the
excludee most of the non-coelomate directly developing problematicc ctenophores and placozoans will be crucial for a
aschelminthss (Figs. 3a and 3b). It should be noted that with- properr resolution.
outt sufficient attention to noncoclomate bilaterians there is a
Davidsonn et al. (1995) and Peterson et al. (1997) seal their
problemm to pinpoint the exact clade under discussion. (For argumentt by discussing the arthropods and chordates as diinstance,, Conway Morris [1998a, 1998b], Hall [1999], and rectlyy developing exceptions to the general dominance of
Olssonn and Hall [1999] regard the hypothesis as explaining maximall indirect development Both groups are characterizedd by the loss of larval characters present in closely related
thee origin of the Metazoa.)
Thee scenario hinges on die assumption that "maximal in- taxa.. Molecular systematics, however, suggests an interestdirectt development" is primitive for Bilateria, and monopha- ingg alternative. According to 18S rDNA information the
sicc life cycles with direct development (absence of a primary Chordataa (Urochordata, Cephalochordata, and Vertebrata)
larva)) are thought to be derived within the Bilateria. To but- mostt likely are the sister group to the Enteropneusta 4- Pterotresss this conclusion, Davidson et al. (1995) and Peterson et branchiaa + Echinodermata (Turbeville et al. 1994; Halanych
al.. (1997) provided two phytogenies, Figs. 3a and 3b, with a a 1995;; Eemisse 1997; Littlewood et al. 1998; Wada 1998;
mappingg of maximal indirect development. Neither of these Winnepenninckxx et al. 1998a; Zrzavy et al. 1998; Bromham
cladogramss allow the primitive developmental mode of the andd Degnan 1999). A A dipleurula type larva then is an autapoBilateriaa to be deduced with certainty. Moreover, the likeli- morphyy of the hemichordate/echmoderm clade, indicating
hoodd for maximal indirect development in the bilaterian thatt chordates and deuterostomes may in fact be primitively
groundd pattern decreases when the previously discussed ca- directt developers. When confirmed, this may have important
veatss for indirect development in platyhelminths and nem- consequencess for recent hypotheses deriving the chordate
erteanss is considered. In addition, the provided phytogenies bodyy plan with reference to enteropneust anatomy (e.g.,
aree heavily pruned, leaving out all direct developing "minor" NObler-Jungg and Arendt 1999; Ruppert et al. 1999) or larval
bilateriann phyla, except Nematoda (Fig. 3a), or Rotifera (Fig. morphologyy (Lacalli 1994; Salvini-Plawen 1998; Nielsen
1999). .
3b).. Understanding the phylogenetic positions of these excludedd taxa is crucial if the goal is the reconstruction of the
ancestrall mode of bilaterian development Rather dian givingg a detailed justification for their premise, Davidson et al.
(1995,, p. 1320) state: "We are convinced that the direction
off the transition is from indirect to direct development and
Arthropodss fall within the ecdysozoan clade. All ecdysozoanss lack classic primary larvae and indirect development
Studyingg die distribution of indirect development on the consensuss 18S rDNA phytogeny (Fig. 3c) reveals the possibility
thatt lack of larvae in chordates and arthropods (as well as the
85 5
Maximall indirect d e v e l o p m e n t
II I absent
present
F ^^ equivocal
Fig.. 3. Cladogram from Davidson el at. (1995) (a); Peterson et al. (1997) (b); 18S rDNA consensus cladogram (c), with mapping of maximall indirect development.
basall ecdysozoans and lophotrochozoans) is primitive. In this maximall indirect development, and the independent evolution
schemee indirect development and set-aside cells would have off set-aside cells in arthropods. These conclusions corroborate
evolvedd multiple times in parallel, e.g., in various lophotro- andd extend the suggestion by Valentine et al. (1999) that setchozoanss such as cycliophorans, platyhelminths, nemerteans, asidee cells may not be homologous across the Bilateria. Appliandd the coelomate lophotrochozoans, and in hemichordates cation,, therefore, of set-aside cells as a mechanism to explain
andd echinoderms. Presence of set-aside cells in the directly de- thee Cambrian explosion should be regarded with caution. The
velopingg arthropods (e.g., imaginal discs, teloblasts, and de- basall position of directly developing taxa without set-aside
rivatives)rivatives) indicates decoupled evolution of set-aside cellscellss
and in the Ecdysozoa and Lophotrochozoa indicates that in-
86 6
dependencyy evolved set-aside cells could have contributed to anss and lophotrochozoans. However, with current phylogethee radiation of distinct clades of macroscopic metazoans. The neticc information we cannot conclusively decide whether a
radiationn of the various small-bodied basal taxa such as gas- biphasicc life cycle is primitive for the Bilateria. At least introtrichs,, rotifers, loriciferans, and Idnorhynchs probably oc- dependentt evolution of set-aside cells and larvae in deuterostomess and protostomes is indicated (Fig. 3c). The combinacurredd independently of set-aside cells.
Doo primary larvae actually exist? Historically, the defini- tionn of Type I embryogenesis and later pattern-formation
tionn of primary larvae was developed as a phylogenetic rather processess were probably already in place in the bilaterian anthann morphological concept Diagnosing a primary larva then cestor.. The modem basal noncoelomate bilaterians did not
iss inferential, not direct observation. Francis Maitland Balfour loosee maximal indirect development and set-aside cells. Notwass perhaps the most devoted British recapitulationist of the withstanding,, the set-aside cell hypothesis has great heuristic
19thh century, who sought to explain everything about the na- valuee for evolutionary developmental biology when placed
turee of larval forms by exclusive use of evolutionary argu- inn a proper phylogenetic framework. The study of the develments.. Accordingly, Balfour (1880, p. 383) defined primary opmentt of larvae and adults in terms of molecular developlarvaee as **... more or less modified ancestral forms, which mentall processes in various indirect developing deuterohavee continued uninterruptedly to develop as free larvae from stomess has already yielded very important insights into body
thee time when rhey constituted the adult form of the species." plann formation (e.g., Arenas-Mena et al. 1998; Peterson et al.
Moree recently, Gösta Jigersten (1972) in his influential book. 1999a,, 1999b). Nevertheless, detailed study of indirectly developingg
EvolutionEvolution of the Metazoan Life Cycle: A Comprehensive
The- protostomes and the relatively unknown directly
ory,ory, similarly defined primary larvae as a phylogenetic con- developingg noncoelomates is much needed if our goal is to
cept,, although not in die same recapitulationist mold as Bal- understandd the evolutionary origin and significance of setfour.. Jagersten (1972, p. 4) states that, "The qualification asidee cells across the Bilateria.
'primary'' thus neither refers to any special characters in the
morphologyy of the larvae... nor to the degree ofcomplication
off structure, but to the fact that the pelagic larval type has persistedd in ontogeny without interruption since its first appearance."" Therefore not surprisingly, Jagersten designates both
crustaceann nauplii and sponge larvae as primary larvae despite
veryy different morphologies. Without explicit morphological
justification,, these definitions are linked with a particular view
off metazoan phylogeny, and thus represent phylogenetic concepts.. Primary larvae can then only be recognized with referencee to a particular phylogeny. The specter of these phylogeneticc definitions is still present in the literature. In discussing
thee phylogenetic position of the most recently discovered animall phylum, the Cycliophora, Fundi and Kristensen (1997)
citee Jagersten (1972) to support their notion that the cycliophorann chordoid larva is a primary larva. Cameron et al.
(1998)) cite Jagersten (1972) for arguing a priorii that direct developmentt is secondary, failing to recognize the nature of
Jagersten'ss phylogenetic concepts. Zrzavy et al. (1998) use the
presencee of a primary larva to sort animal relationships withoutt providing the necessary morphology-based definition, thus
enteringg a phylogenetic concept as primary data into die reconstructionn of a phylogeny. It is therefore noteworthy mat
Cameronn et al. (1998, p. 617) conclude in the Balfourian mode
that,, "The larval forms of modern indirect developing marine
speciess indicates to us the probable nature of these ancestral
organismss [the ancestral bilatcrians], except mat they then
constitutedd the terminal or adult stage of development
Inn summary, the bilaterian ancestor did probably not resemblee a ciliated primary larva. More likely, this ancestor (at
leastt the protostome ancestor) was at the organizational
gradee of a noncoelomatc as suggested by the basal ecdysozo-
Thesee two studies are not unique in evolutionary developmentall biology in that they can benefit from greater attention to
thee phylogenetic basis of their conclusions. Problematic hypothesess of animal body plan evolution can be identified in a
rangee of other recent studies, suffering from biased taxon samplingg and/or equivocal phylogenetic deductions. For example,
wee can diagnose ambiguous inferences of ground pattern attributess in the bilaterian ancestor such as presence of a coekwn
(e.g.,, in Balavoine and Adoutte 1998; Martindale and Henry
1998;; Adoutte et al. 1999; Knoll and Carroll 1999), die presence
offenterocody based comisrepiesenting me conclusions ofValentinee (1997) (e.g., in Martindale and Henry 1998; Knoll and
Carrolll 1999), the presence of "anterior tentacular appendages
formedd by coelomic outpouching" (Knoll and Carroll 1999),
diee presence of morphological seriation or metamerism (e.g., in
Balavoinee 1997,1998; Adoutte et al. 1999; Holland and Hollandd 1999; Knoll and Carroll 1999), and the presence of antenniformm appendages in the protostome ancestor (e.g., in Panganibann et al. 1997; Tabin et al. 1999). Rigorous application of
phylogeneticc information has yielded remarkable insights into
animall evolution at lower taxonomie levels for diverse taxa and
characterss (e.g., Heeg 1995; Sturmbauer et aL 1996; Lee and
Shinee 1998; McHugh and Rouse 1998; Werdelin and Nilsonne
1999).. If studies on animal body plan evolution aim at a similar
effectiveness,, phylogenetic information of higher level animal
taxaa should be approached with appropriate care.
CONCLUSIONS S
Evolutionaryy developmental biology attempts to understand
thee genesis and evolution of organismal design by compara-
87 7
tivee research, integrating information from disparate fields
off biological and paleontological research. A proper phytogenyy is necessary because the interpretiveframeworkin many
studiess of evolutionary developmental biology. The emergencee of a consensus on higher level metazoan relationships
onn the basis of 18S rRNA/DNA sequence data has provided
aa new framework for studying animal body plan evolution.
Frequently,, papers contrast this new molecular view of animall evolution wim a traditional, morphology based perspective.. I explored the nature and historical basis of mis widely
citedd traditional textbook tree, and provided an amended reinterpretationn that better reflects historical information.
Nextt I discussed the newly emerging molecular view on
higherr level animal relationships as a basis for studies of animall body plan evolution. Two recent studies, on the evolutionn of the platyhelminth body plan and on the evolutionary
significancee of indirect development and set-aside cells, are
thenn used to illustrate the danger of overlooking two aspects
off the newly emerging molecular metazoan phylogeny depictedd in many recent papers: incompleteness and lack of
resolution.. In addition, the need for rigorous phylogenetic
reasoningg is stressed. Both case studies rely on phylogenies
thatt exhibit a strong taxon selection bias for coelomate, indirectlyy developing bilaterians. In both instances, by ignoring
pseudocoelomates,, the probability of recovering "linear"
transformationn series and homology is maximized by maintainingg a maximum of connections between features in differentt taxa. Pseudocoelomates have been the classical locus
off some of the most enduring disputes in comparative zoology,, namely the origin and evolution of body cavities and
larvae.. The properly expanded new molecular view of animall evolution forcefully draws attention to the necessity to
betterr incorporate the poorly understood bilaterians into our
hypothesess of animal body plan evolution.
methodologiess o f "telling die tree" remain standing, as illustrated
byy the continued reference to their earlier papers for support o f their
phylogeneticc interpretations.
REFERENCES S
Abouhcif,, E., Zardoya. R, and Meyer, A. 1998. Limitations of metazoan
18SS rRNA sequence data: implications for reconstructing a phylogeny
off the animal kingdom and inferring die reality of die Cambrian explosion.. J. Mol. Evol. 47:394-405.
Adoutte,, A , Balavoine. G., Laniltot, N.. and De Rosa. R. 1999. Animal
evolution,, die end of intermediate taxa? TIG 15:104-106.
Aguinaldo,, A. M. A., and Lake, J. A. 1998. Evolution of die multicellular
animals.. Amer. Zool. 38:878-887.
Aguinaldo,, A. M. A., et al. 1997. Evidence for a clade of nematodes, arthropodss and other moulting animals. Noturt 387:489-493.
Aleshin,, V. V., Mityutina. I. A., Kedrova, O. $., Vladychenskaya, N. S„ and
Petrov,, N. B. 1998. Phytogeny of Nematoda and Cephalorhyacha derivedd from 1SS rDNA. / . Mot. Evol. 47:597-605.
Arenas-Mena.. C Martinez, P., Cameron, R. A , and Davidson, E H. 1998.
Expressionn of the Hot gene complex in the indirect development of a
seaa urchin. Proc. Natl. Acad. ScL USA 95:13062-13067.
Arthur,, W. 1997. The Origin of Animal Body Plans: A Study in Evolutionaryary Dex-elopmental Biologv. Cambridge University Press. Cambridge,
pp.. 1-336.
Balavoine,, G. 1997. The early emergence of Platyhclminthes is contradictedd by the agreement between 18S rRNA and Hox genes data. C R. Acad.
Sci.Sci. Paris 320:83-94.
Balavoine,, G. 1998. Are Platyhclminthes coelomates without a coelotn?
Ann argument based on the evolution of Hox genes. Amer. Zool. 38:
843-858. .
Balavoine,, G.. and Adoutte, A. 1998. One or three Cambrian radiations?
ScienceScience 280:397-398.
Balfour,, F. M. 1880. Larval forms: their nature, origin, and affinities. Q. J.
Micr.Micr. Sci. 20:381-407.
Berth,, R. H.,and Broshears, R. E. 1982.77» Invertebrate World Saunders
Collegee Publishing, Philadelphia, pp. 1-446.
Bartolomaeus.. T. 1995. Structure and formation of die uncini in Pectinaria
koreni,koreni, Pectinaria auricoma (TerebeOida) and Sptrorbls spirorbtt (Sabdlida):: implications for annelid phylogeny and the position of the
Pogonophora.. Zoomorphology 115:161-177.
Bartolomaeus,, T. 1998. Chaetogenesis in polychaetous Annelida—significancee for annelid systematic! and the position of the Pogonophora.
Zooi.Zooi. Anal. Comp. Syst. 10ft 348-364.
Bengtson,, S., and Yue, Z. 1997. Fossilized metazoan embryos from the earAcknowledgments s
liestt Cambrian. Science 2TI: 1645-1648.
II thank Professor Frederick Schram for his many valuable c o m Berney,, C . Pawlowski, J., and Zaninetti, L1999. Protein sequences do not
mentss on the manuscript. My work is supported by grant 805supportt an early divergence of die Acoela. Programme and Abstracts of
33.4311 -P from the Earth and Life Sciences Foundation (A. L. W.),
InterrelationshipsInterrelationships of the Ptatyhetmlnthes, p. 9.
o ff the Netherlands Organization for Scientific Research (N. W. O.).
Black.. M. B.. et al. 1997. Molecular systematic: of vestimentiferan tubewarmss from hydrodwrmal vents and cold-water seeps. Mar. Blot. 130:
141-149. .
NoteNote added in proof. While this article was in press, a paper (PeterBromham,,
L D.. and Degnan, B. M. 1999. Hemichordates and deuterosonn et al. 2000) was published that contains new relevant informastomee evolution: robust molecular phylogenetic support for a hemitionn for the discussion on the evolutionary significance o f set-aside
chordatee + echinoderm dade. Evol. Dev. 1:166-171.
cells.. Exhibiting a striking series o f parallel lines o f thought, PeterCameron.. R. A., Peterson, K. J , and Davidson. E. H. 1998. Developmental
sonn ct al. (2000) independently resolved some o f the specific critigenee regulation and the evolution of large animal body plans. Amer.
Zooi.Zooi. 38:609-620.
cismss aired in this paper. Their adjusted perspective now also inCampos..
A., Cummings, M. P., Reyes, J. L , and Lacktte. J. P. 1998. Phycludess a more balanced sampling of bilaterian taxa; a new phylogenetic
logeaeticc relationships of Ptatytielminthes based on 18S ribosomal gene
frameworkk chiefly based on 18S rRNA/DNA sequence data, a consequences.. Mot. PhyL Evoi 10:1-10.
siderationn o f a sequence o f succeeding bilaterian ancestors rather
Carranza,, S„ BaguBa. J., and Riutort, M. 1997. Are the Platyhebnindies a
thann a single ancestor to better accommodate the evolution from dimonopfiyieticc primitive group? An assessment using IBS rDNA sequences.. Mot. Biol. Evol. 14:485-497.
rectt to indirect development at the base o f the Bilateria; and the inConwayy Morris, S. 1997. Molecular docks: Defusing the Cambrian "exploterpretationn o f the directly developing bilaterian ancestor as represion?"" C«rr. Biol. 7: R71-R74.
sentingg the organizational grade o f modem primary larvae, rather
Conwayy Morris, S. 1998a. Early metazoan evolution: reconciling paleonthann possessing specific larval structures. However, our different
tologyy and molecular biology. i4m«r. Zooi 38:867-877.
88 8
Conwayy Morris, S. 1998b. Metazoan phytogenies: falling into place or fallingg into pieces? A palacontologkal perspective. CUT. Opin. Genet.
O n .. 8: «62-667.
Conwavv Morris, S. 1998c. Eggs and embryos from the Cambrian. BioEssaysEssays 20:676-682.
Davidson.. E. H. 1991. Spatial mechanisms of gene regulation in melazoan
embryos.. Development 113:1-26.
Davidson.. E. H. 1997. Insights from the ccainoderms. Nature 389:679-680.
Davidson.. E. H.. Peterson, K. J., and Cameron. R. A. 1995. Origin of btlateriann body plans: evolution of developmental regulatory mechanism.
ScienceScience 270.1319-13».
Dee Rosa. R-, et al. 1999. Hox genes in brachiopods and priapuiids and protostomee evolution. Nature 399:772-776.
Ecrnisse.. D. J. 1997. Arthropod and annelid relationships re-examined. In
R.. A. Fortey and R. K. Thomas (eds.). Arthropod Relationships. Chapmann and Hall, London. The Systematic* Association Special Volume
Seriess 55. pp. 43-56.
Ecrnisse.. D. J . Albert. J. S„ ami Anderson, F. E. 1992. Annelida and Arthropodaa are not sister taxa: a phylogenetk analysts of spiralian meuzoann morphology. Syst. Biol. 41:305-330.
Erwin.. D. H. 1999. The origin of body plans. Amer. Zool. 39:617-629.
Field.. K. C . et al. 1988. Molecular phytogeny of the animal kingdom. Sci««239:748-753. .
Foster.. P. C . and Hkkey. D. A. 1999. Compositional bias may affect both
DNA-basedd and protein-based phylogenetk reconstructions. / Mol.
Evol.Evol. 48:284-290.
Funch.. P.. and Kristcmen, R. M. 1997. Cycliophora. In F. W. Harrison and
R.. M. Woollacoil (eds.) Microscopic Anatomy of invertebrates. Vol. IX
Lopltoplioraies,Lopltoplioraies, Entoprocta, and Cycliophora. Wiley-Liss,
pp.. 409-474.
Garcy.. J. R.. and Schmidt-Rhaesa. A. 1998. The essential role of "minor"
phylaa in molecular studies of animal evolution. Amer. Zool. 38:907-917.
Carey.. J. R.. Kroujc. M., Nelson. D. R.,and Brooks. J. 1996a. Molecular analysiss supports a tardigrade-arthropod association. Inv. BioL 115:79-88.
Garey.. J. R.. Near. T. J., Nonnemacher. M. R.. and Nadler. S. A. 1996b.
Molecularr evidence for Acanthocephala as a subtaxon of Rotifer». J. Mol.
Evol.Evol. 43:287-292.
Gould.. S. J. 1977. Ontogeny and Pltytogeny. Belknap. Harvard University
Press.. Cambridge, pp. 1-501.
Gould.. S J., and Lewontin. R. C. 1979. The spandrels of San Marco and the
Panglossiann paradigm: a critique of the adaptationiit programme. Proc.
R.R. Sac. Load. B 205:581-598.
Gould.. S. J., and Vrba, E. S.. 1982. Exaptation—a missing term in the sciencee of form. Paleobiology 8:4-15.
Halanych.. K. M. 1995. The phylogenetk position of the pterobranch hemichordatess based on I8S rDNA sequence data. Mol. Phyl. Evol. 4:72-76.
Halanych,, K. M. 1996. Testing hypotheses of caaetognath origins: long
branchess revealed by 18S ribosomal DNA. Syst. Biol. 45:223-246.
Halanych.. K. M. 1998. Considerations for reconstructing metazoan history:
signal,, resolution, and hypothesis testing. Amer. Zool. 38:929-941.
Hall.. B. K. 1999. Evolutionary Developmental Biology. Second edition.
Kluwcrr Academic Publishers. Dordrecht, the Netherlands, pp. 1-491.
Hanelt,, B.. Van Schyndel. D„ Adema, C M., Lewis, L. A., and Loker, E. S.
1996.. 'I"he phylogenetk position of Rhopatura ophtocomat (Orthonectida)) based on 18S ribosomal DNA sequence analysis. Mol. Biol. Evol.
13:1187-1191. .
Hart.. M. W„ Byrne. M.. and Smith, M. J. 1997. Molecular phylogenetk
analysiss of life-history evolution in asterinid starfish. Evolution 51:
1848-1861. .
Haszprunar.. G. 1996. The Molluscs: coelomate turbelUrians or mesenchymatee annelids? In J. Taylor (ed). Origin and Evolutionary Radiation of
thethe Moilmca. Oxford University Press. Oxford, pp. 1-28.
Haszpninar.C.Salvini-Plawen.. L V„artd Rieger, R. M. 1995. Larval planktotrophy—aa primitive trait in the Bilateria? Acta Zool. 76:141-154.
Henry.. J., and Manindak. M. Q. 1997. Nemerteans. the ribbon worms. In
S.. F. Gilbert and A. M. Raunio (eds.). Embryology: Constructing the
Organism.Organism. Sinauer Associates, Sunderland, MA, pp. 151-166.
Hoeg.. J. T. 1995. Sex and the single cirripede: a phylogenetk perspective.
Inn F. R. Schram and J. T. Heeg (eds.). Ne*' Frontiers in Barnacle Evolution.lution. A. A. Balkema. Rotterdam, the Netherlands, pp. 195-207.
Holland,, P. W. H. 1998. Major transitions in animal evolution: a developmentall genetk perspective. Amer. Zoo/. 38:829-842.
Holland,, N. D„ and Holland, L. Z. 1999. Amphioxus and the utility of molecularr genetk data for hypothesizing body part homologies between
distantlyy related animals. Amer. Zool. 39:630-640.
Hyman,, L H. 1940. The Invertebrates, vol. I. Protozoa through Ctenophora.phora. McGraw-Hill Book Company, New York. pp. 1-726.
Hyman.. L. H. 1951a. The Invertebrates, vot II. Platyhelmlnthes and Rhynchocoeta.chocoeta. McGraw-Hill Book Company, New York, pp 1-550
Hyman,, L H. 1951b. The Invertebrates, vol III. Acanthocephala, Aschetmiathesmiathes and Enioproaa. McGraw-Hill Book Company. New York, pp.
1-572. .
Hyman,, L H. 1959. The Invertebrates, vol. V. Smaller Coelomate Groups.
McGraw-Hilll Book Company. New York. pp. 1-783.
Jagersten.. G. 1971 Evolution of the Metazoan Life Cycle: A Comprehensivesive Theory. Academk Press. New York. pp. 1-282.
Janies.. D., and DeSehe. R. 1999. Development, evolution, and corroboration.. The Anal. Rec 257:6-14.
Jenner,, R. A. 1999. Metazoan phytogeny as a tool in evolutionary biology:
currentt problems and discrepancies in application. Belg. J. Zool. 129:
245-262. .
Jenner,, R. A., and Schram, F. R. 1999. The grand game of metazoan phytogeny:: rules and strategies. Biol. Rev. 74:121-142.
Knoll,, A. H.. and Carroll. S. B. 1999. Early animal evolution: emerging
viewss from comparative biology and geology. Science 284:2129-2137.
Kobayashi.. M., Furuya. H.. and Holland, P. W. H. 1999. Dicyemids are
higherr animals. ,V<jwr»401:762.
Kojima.. S. 1998. Paraphyletk status of Polychaeta suggested by phylogeNewnetkk
York,analysis based on the amino acid sequences of elongation factor-1
alpha.. Mol. Phyl. Evol 9:255-261.
Lacalli.. T. C. 1994. Apical organs, epithelial domains, and the origin of the
chordatee central nervous system. Amer. Zool. 34:533-541.
Lacalli.. T. C. 1997. The nature and origin of deuterostomes: some unresolvedd issues. Invert. Biol. 116:363-370.
Lecointre.. G.. Philippe. H., Van Le. H. L . and Le Guyader, H. 1993. Speciess sampling has a major impact on phylogenetk inference. Mol. Phyl.
Evol.Evol. 2:205-224.
Lee.. M. S. Y., and Shine, R. 1998. Reptilian viviparity and Dollo's law.
EvolutionEvolution 52:1441-1450.
Lieberman.. B. S. 1999. Testing the Darwinian legacy of the Cambrian radiationn using trilobite phytogeny and biogeography./ Paleon. 73:176-181.
Litikwood,, D. T. J., Telford, M. J- dough, K. A., and Robde. K. 1998.
Gnaihostomulida—ann enigmatic metazoan phylum from both morphologicall and molecular perspectives. Mol. Phyl. Evol 9:72-79.
Littlewood.. D. T. J , Robde, K.. and dough. K. A. 1999. The interrelationshipss of all major groups of Platyhelmimbes: phylogenetk evidence
fromm morphology and molecules. Biol. J. Una. Soe. 66:75-114.
Maley.. L. E., and Marshall, C R. 1998. The coming of age of molecular systematic.. Science 279:505-506,
Martindale,, M. Q., and Henry, J. Q. 1998. The development of radial and
biradiall symmetry, the evolution of bilaterality. Amer. Zool. 38:672-684.
McHugh,, D. 1997. Molecular evidence that echiurans and pogonophorans
aree derived annelids. Proc. Nad. Acad. Sci. USA 94:8006-8009.
McHugh,, D. 1998. Deciphering metazoan phytogeny: the need for additionall molecular data. Amer. Zool 38:859-866.
McHugh.. D. 1999. Phytogeny of the Annelida. Siddall et al. (1998) rebutted.. Cladistia 15:85-89.
McHugh,, D™ and Rouse, G. W. 1998. Life history evolution of marine invertebrates:: new views from phylogenetk systematica. TREE 13:182-186.
McNamara,, K. J. 1995. Evolutionary Change and Heterochrony. John
Wileyy & Sons. Chkhester. pp. 1-286.
Meyer,, K-and Bartotomaeus, T. 1996. Ultrastructure and formation of the
hookedd setae in Owenia fusiformis delle Chiaje. 1842: implications for
annelidd phytogeny. Can. J. Zool 74:2143-2153.
Moore.. J., and Wiflmer. P. 1997. Convergent evolution in invertebrates.
Biol.Biol. Rev TZ: 1-60.
Nielsen,, C. 1995. Animal evolution. Interrelationships of the living phyla.
Oxfordd University Press. Oxford, pp. 1-467.
Nielsen.. C. 1998. Origin and evolution of animal life cycles. Biol. Rev. 73:
125-155. .
89 9
Nielsen.. C. 1999. Origin of tbc chordate central nervous system-and the
Schottz,, G., Mittmann, B., and Gerberding. M. 1998. The pattern of Distaloriginn of chordate*. Dev. Cents Evot. 209:198-205.
lessless expression in the mouthpartsof crustaceans, myriapods and insects
NUMerJung.. K„ and Arende D. 1999. Dorsoventral axis inversion: enneww evidence for a gnathobasic mandible ami the common origin of
teropneustt anatomy links invertebrates to chordates turned upside
Mandibulata.. Int. J. Dev. Biol. 42:801-810.
down.. / Zool. Syst. Evot. Res. 37:93-100.
Schram.. F. R. 1991. Cladtsticanarysis of metazoan phyia and the placement
Olive,, P. J. W. 1985. Covariability of reproductive traits in marine inverteoff fossil problematka. In A. Simonetta and S. Conway Morris (eds.).
brates:: implications for the phytogeny of the lower invertebrates. In S.
TheThe Early Evolution of Metazoa and the Significance of Problemat
Conwayy Morris. J. D. George, R. Gibson, and H. M. Plan (eds.). The
Taxa.Taxa. Cambridge University Press. Cambridge, pp. 35-46.
OriginsOrigins mid Relationships of Lower Invertebrates. Clarendon Press.
Ox- M. E., Fitzhugh, K.. and Coates, K. A. 1998. Problems determining
Siddall,,
ford,ford, pp- 42-59.
thee phytogenetic position of echniraits and pogonophorans with limited
Olsson.. L . and Halt, B. K. 1999. Introduction to the symposium: developdataa Ciadistics 14:401-410.
mentall and evolutionary perspectives on major transformations in body Simpson,, G. G. 1951. The Meaning of Evolution. A special and abridged
organization.. Amer. Zool. 39:612-616.
edition.. The New American Library. New York, pp. 1-192.
Panganiban.. C . et al. 1997. The origin and evolution of animal appendSmith,, A. B. 1999. Dating the origin of metazoan body plans. EvoL Dev. I:
ages.. Proc. Natl. Acad. Sd. USA 94:5162-5166.
138-142. .
Pechenik.. J. A. 1999. On the advantages and disadvantages of larval stages
Strathmann.. R. R. 1993. Hypotheses on the origins of marine larvae. Annu.
inn benthic marine invertebrate life cycles. Mar. Ecot. Prog. Ser. 177:
Rev.Rev. Ecot. Syst. 24:89-117.
269-297. .
Strathmann,, R. R., and Eemisse. D. J. 1994. What molecular phylogenies
Peterson.. K. J.. Cameron. R. A„ and Davidson. E. H. 1997. Set-aside cells
telll us about the evolution of larval forms. Amer. Zool. 34:502-512.
inn maximal indirect development: evolutionary and developmental inSiurmbauer.. C , Levinton, J. S.. and Christy, J. 1996. Molecular phytosights.. BioEssays 19:623-631.
genyy analysis of fiddler crabs: test of the hypothesis of increasing behavPeterson.. K. J.. Cameron. R. A , Tagawa. K.. Satoh,, N.. and Davidson. E. H.
iorall complexity in evolution. Proc. Natl. Acad. Sd. USA 93: 108551999a.. A comparative molecular approach to mesodermal patterning in
10857. .
bas.ill deuterostomes: the expression pattern of Brachyury in the enTahin,, C. J.. Carroll. S. B.. and Panganiban. G. 1999. Out on a limb: paralleropneustt hemichordatc Plychodera flaw. Dcwtopmeni 126:85-95.
lelss in vertebrate and invertebrate limb patterning and the origin of apPeterson.. K. J.. Harada. Y.. Cameron. R. A., and Davidson. E. H. 1999b.
pendages.. Amer. Zool. 39:650-663.
Expressionn pattern of Brachyury and Not in the sea urchin: comparative
Turbeville..
J. M. 1986. An ultrastructural analysis of coelomogenesis in the
implicationss for the origins of mesoderm in the basal deuierusiomes.
hoplonemertincc Prosorhochmusamericanus and the polychaete MagetDewDew Biol. 207:419-431.
onaona sp. J. Morph. 187:51-60.
Peterson.. K. J.. Cameron. R. A., and Davidson. E. H. 2000. Bilaterian origins:: significance of new experimental observations. Dev. Biol. 219: Turbeville,, J. M.. and Ruppert E. E. 1985. Comparative ultrastucture and
thee
evolution of nemertines. Am. Zool. 25:53-71.
1-17. .
Turbeville.. J. M.. Schulz. J. R.. and Raff, R. A. 1994. Deuterostome phyPhillipe.. H.. Chenuil. A., and Adoutte, A. 1994. Can the Cambrian explotogenyy and the sister group of the chordates: evidence from molecules
sionn be inferred through molecular phytogeny? Development (Suppleandd morphology. Mot. Biol. EvoL 11:648-655.
ment):: 15-15.
Valentine.. J. W. 1997. Cleavage patterns and the topology of the metazoan
Raff.. R. A. 1996. Tlte Shape of Life: Genes, Development, and the Evolution
treee of life. Proc Natl. Acad. ScLUSA 94:8001-8005.
ofof Animal form. The University of Chicago Press, Chicago, pp. 1-520.
Raff.. R. A., et al. 1989. Metazoan phytogeny based on analysis of 18S ribo- Valentine,, J. W.. Jablonski. D.. and Erwin. D. H. 1999. Fossils, molecules
somall RNA. In B. Femholm, K. Bremer, and H. Jomvalt (eds.). The Hiandd embryos: new perspectives on the Cambrian explosion. Developerarchyerarchy of Ufe. Elsevier Science Publishers, Amsterdam, pp. 247-260.memmem 126851-859.
Rouse.. C. W. 1999. Trochophore concepts: ciliary bands and the evolution Vermcij.. G. J. 1996. Animal origins. Science 274:525-526.
Wada.H.. 1998. Evolutionary history of free-swimming and sessile lifestyles
off larvae in spiralian Metazoa. Biol. J. Unix. Soc. 66:411-464.
inn urochordates as deduced from 18S rDNA molecular phytogeny. Mot.
Rouse,, d. W.. and FauchaU. K. 1997. Ciadistics and potychaetes. Zool. Set.
Biol.Biol. Evot. 15:1189-1194.
26:: 139-204.
Werdelin,, L., and Nibonne, A. 1999. The evolution of the scrotum and testicRuiz-Trilio.. I.. Riutort, M.. Litllewood, D. T. J.. Herniou. E. A., and Baularr descent in mammals: a phytogenetic view./ Theor. Biol. 196:61-72.
guna,, J. 1999. Acoel flatworms: earliest extant bilaterian metazoans, not Willmer,, P. 1990. Invertebrate Relationships: Patterns m Animal Evolution.
memberss of Plalyhelminthes. Science 283:1919-1923.
Cambridgee University Press, Cambridge, pp. 1-400.
Ruppert.. E. E. 1991. Introduction to the aschelminth phyla: a considerWillmer,, P. G. 1995. Modern approaches to the phytogeny of body cavities.
ationn of mesoderm, body cavities, and cuticle. In F. W. Harrison and
Inn G. Lanzavecchia, R. Valvassori, and M. D. Candia (eds.). Body CavE.. E, Ruppert (eds.). Microscopic Anatomy of Invertebrates. Vol. 4. ities:ities: Function and Phytogeny. Mucchi, Modena. pp. 23-39.
Aschelminthes.Aschelminthes. Wiley-Liss. New York. pp. 1-17.
Willmer.. P., and Holland. P. W. H. 1991. Modern approaches to metazoan
Ruppert.. E. E.. and Barnes. R. D. 1994. Invertebrate Zoology. Saunders
relationships.. J. Zool., Land. 224:689-694.
Collegee Publishing. New York. pp. 1-1056.
Winnepenninckx.. B.et al. 1995.18S rRNA data indicate that aschclminths
Ruppert.. E. E.. Cameron. C. B., and Frick. J. E. 1999. Endostyle-tike feaaree poryphyletic in origin and consist of at least three distinct dades.
turess of the dorsal epibranchial ridge of an enteropneust and the hyMot.Mot. Biol Evot. 12:1132-1137.
pothesiss of dorso-ventral axis inversion in chordates. Invert. Biol. 118: Winnepenninckx.. B. M. H.. Van De Peer. Y.. and Backeljau. T. 1998a.
202-212. .
Metazoann relationships on the basis of I8S rRNA sequences: a few
Salvini-Plawen.. L. v. I960. Was ist eine Trochophora? Eine Analyse der
yearss later. Amer. Zool. 38:888-906.
Larventypenn mariner Protostomier. Zool. Sb. Anat. 103:389-423.
Winnepenninckx.. B. M. H.. Backeljau. T , and Kristensen. R. M. 1998b.
Salvini-Plawen.. L v. 1990. Origin, phytogeny and classification of the phyRelationss of the new phylum Cycliophora. Nature 393:636-638,
lumm Mollusca. Ibena^r. 1-33.
Wolpertt L. 1999. From egg to adult to larva. Evot. Dev. 1:3-4.
Salvini-Plawen,, L. v. 1998. The urochordate larva and archkhordate orgaWray.. G. A.. Levinton. J. S. and Shapiro, L. H. 1996. Molecular evidence
nization:: chordate origin and anagenesis revisited. /. Zool. Syst. EvoL
forr deep precambrian divergences among metazoan phyla. Science 274:
ResearchResearch 36: 129-145.
568-573. .
Schmidt-Rhaesa,, A.. Banolomaeus, T , Lemburg, C Ehlers. U.. and
Zrzavy,, J.. Mihulka, S.. Kepka, P, Bezdek, A., and Tietz, D. 1998. PhytogGarey,, J.R. 1998. The position of the Anhropoda in the phylogenetk
enyy of the Metazoa based on morphological and I8SribosomalDNA
svstem.// Morph. 238: 263-285.
evidence.. Ciadistics 14:249-285.
90 0