Basal euarthropod development a fossilbased perspective

nigel c. hughes, joachim t. haug and dieter waloszek

The morphological gap between Euarthropoda (the crown group that contains all extant arthropods) and living arthropod-like animals such as onychophorans, tardigrades and pentastomids is bridged by a number of fossils known primarily from rocks some 520 to 490 million years old (e.g. Fuxianhuia, Chengjiangocaris, Shankouia, see Figure 15.1; cf. Waloszek et al. 2005). These centimetre-scale fossil animals illuminate critical steps in early arthropod evolution (particularly head and limb development) but provide a limited amount of developmental information because of a lack of early ontogenetic stages. Small individuals that might represent pre-mature stages are scarce or absent, and the degree of allometry among the available individuals is generally modest. A limited number of early arthropod taxa do show more substantial ontogenetic information (Waloszek and Maas 2005). This chapter reviews the morphological development of early arthropods from two perspectives. The first is that provided by ontogenetic series based on the well-preserved biomineralised exoskeletons of trilo-bites, the best represented arthropod taxon in Palaeozoic rocks, but one whose development is seldom considered in broader comparative context. The second is that provided by 'Orsten'-type preserved faunas, in which the entire cuticle of numerous post-embryonic specimens of various species, mainly representatives of the crustacean evolutionary lineage (stem derivatives and Labrophora with phosphatocopines and members of the Eucrustacea; see Maas and Waloszek 2005) was replaced with spectacular fidelity by calcium phosphate in the absence of any compaction (Müller 1985). Such preservation has permitted detailed

Evolving Pathways: Key Themes in Evolutionary Developmental Biology, ed. Alessandro Minelli and Giuseppe Fusco. Published by Cambridge University Press. # Cambridge University Press 2008.

Figure 15.1 Phylogenetic relationships among major groups of living and fossil arthropod taxa (modified from Waloszek et al. 2005). We use the informal term 'allied taxa' to refer to a group of euarthropods with slim, long and feeler-like antennae and subsequent serial appendage pairs being biramous with lamellae-bearing exopods in cephalon and trunk, with a cephalic shield that covered basically three pairs of biramous limbs, and a trunk composed of largely homonomous exoskeletal segments commonly divided into an anterior region of freely articulating segments and a pygidium-like structure posteriorly.

Figure 15.1 Phylogenetic relationships among major groups of living and fossil arthropod taxa (modified from Waloszek et al. 2005). We use the informal term 'allied taxa' to refer to a group of euarthropods with slim, long and feeler-like antennae and subsequent serial appendage pairs being biramous with lamellae-bearing exopods in cephalon and trunk, with a cephalic shield that covered basically three pairs of biramous limbs, and a trunk composed of largely homonomous exoskeletal segments commonly divided into an anterior region of freely articulating segments and a pygidium-like structure posteriorly.

reconstruction of portions of the ontogeny of several late Cambrian euarthropods normally absent from the fossil record. Patterns of segment generation common to these different sources may provide insight into the developmental mode of basal Euarthropoda, and thus into the evolution of arthropod ontogeny.

trilobite structure and development

This discussion of trilobite segmentation focuses on the development of the segmentation of the biomineralised exoskeleton. Although appendages are known in some 20 species (see Hughes 2003a, Table 1, for a review) and there was a direct correlation between appendage pairs and dorsal exoskeletal segments in the non-terminal regions of the anterior-posterior (A-P) axis, ontogenetic information about trilobite development is almost entirely restricted to the development of the biomineralised exoskeleton. This was divided into two principal regions along the A-P axis, the cephalon and the trunk (Figure 15.2). Within

Subphylum Trilobites Structure

Figure 15.2 Basic dorsal morphology of two trilobites. A is anterior, P is posterior, Pyg. is the pygidium. Figure on the left, based on a generalised olenelloid trilobite, had a boundary between two distinct batches of segments located within the thorax, dividing the protrunk from the opisthotrunk (Opi.). Aulacopleura konincki, on the right, displayed the homonomous trunk condition in which all segments shared a similar morphology.

Figure 15.2 Basic dorsal morphology of two trilobites. A is anterior, P is posterior, Pyg. is the pygidium. Figure on the left, based on a generalised olenelloid trilobite, had a boundary between two distinct batches of segments located within the thorax, dividing the protrunk from the opisthotrunk (Opi.). Aulacopleura konincki, on the right, displayed the homonomous trunk condition in which all segments shared a similar morphology.

the cephalon exoskeletal segmentation was most evident in the axial region. The number of cephalic segments remained constant during the ontogenies of individual species, although in some cases adjacent segments differed markedly in shape. The number of segments within the cephalon (at least four) also appears to have been approximately constant throughout the Trilobita. It is not known whether these segments were specified simultaneously or sequentially. In striking contrast, the number of segments in the trunk region evidently varied both ontogen-etically within species and phylogenetically among them. Trunk segmentation was expressed in both axial and pleural (lateral) regions, and displayed several different attributes whose variation was partially independent of one another. Such attributes include: (1) the number of trunk segments in the mature, segment invariant (epimorphic) phase of postembryonic development; (2) the number and development of functional articulations within the trunk region; (3) the form of trunk segments.

the number and generation of trunk segments

The site at which new trunk segments were generated was a subterminal generative zone. This was located by studies of segments that were individualised from their first appearance, commonly by a unique axial or pleural spine (e.g. Stubblefield 1926, Chatterton 1994), which first appeared adjacent to the posterior end of the trunk, just as in the great majority of those arthropods in which segment expression is sequential (including those classified as having 'short germ-band' embryonic development). The development of trilobite trunk segmentation is also comparable to that of many other (eu)arthropods in that new trunk segments appeared sequentially through a series of postembryonic instars, a pattern that is termed anamorphic development.

In all species of trilobite in which ontogeny is well known, ana-morphic instars were succeeded by an ontogenetic phase during which moulting and size increase continued, but which was invariant in the number of segments expressed in the biomineralised exoskeleton. This biphasic accretive-invariant segmentation pattern seen in trilobites, some myriapods and some crustaceans is termed 'hemianamorphic' development (Enghoff et al. 1993). In trilobites, following the general usage of specialists on myriapod biology, the second, 'segment invariant' phase is known as the 'epimorphic' phase (Hughes 2003b, Hughes et al. 2006). Hemianamorphic development evidently characterised the vast majority of the over 15 000 trilobite species known. It is not clear whether those rare trilobites with over 100 trunk segments (Paterson and Edgecombe 2006) achieved an epimorphic phase, and it is possible that some trilobites continued to add segments throughout life. The range in mature trunk segment number among trilobites varied widely, from forms bearing fewer than 10 trunk segments to those with over 100. Such a range of variation was evident even amongst early Cambrian species.

The rate of production of segments during the anamorphic phase was variable both ontogenetically within species and phylogenetically among them, and phases in which successive instars maintained a stable number of segments intercalated within other phases of ana-morphic segment production have been reported (McNamara et al. 2003). Nevertheless, the number of segments expressed between successive instars was generally one or two segments, not the much larger numbers seen between instars in some derived myriapods (Fusco 2005) and crustaceans (Walossek 1993). Such a pattern of regular and gradual segment generation greatly aids in the reconstruction of trilobite ontogenies which are based on sclerites derived from multiple individuals and gives the impression that trilobite growth was 'tracklike' (Hughes and Chapman 1995), i.e. channelled along a trajectory of progressive, modest change. Nevertheless, more striking ontogenetic changes have been recognised between particular instars and have been labelled as 'metamorphosis' in trilobites (Evitt 1961, but see Hughes et al. 2006). There are also cases in which juveniles cannot be linked with any co-occurrent mature forms, perhaps suggesting morphological changes too extreme to permit such association in the absence of the sequential instars of any individual (Chatterton and Speyer 1997).

articulations between trunk segments

In later ontogenetic stages, the trilobite trunk was divided into the thorax, an anterior set of segments each with a functioning joint along its entire anterior and posterior margins, and the pygidium, a posterior set of conjoined segments that did not functionally articulate with one another. The similarity of form in thoracic and pygidial segments partially reflected their common site of origin at the subterminal growth zone. An important aspect of trilobite development is that at first appearance all trunk segments were conjoined to segments immediately anterior to them. During early post-embryonic development all segments, whether part of the cephalon or trunk, were dorsally conjoined. The appearance of a functional joint between the rear of the cephalon and the anterior of the trunk marked the transition from the protaspid to the meraspid ontogenetic stages (Figure 15.3). At this stage the dorsal exoskeleton was a two-part, hinge-like structure made up of two units each comprising conjoined segments: the cephalon and the trunk. New segments accreted anamorphically near the rear of the trunk. New articulations between segments resulted in the progressive construction of the thorax, via the sequential release into it of trunk segments previously conjoined in the meraspid pygidium. Hence the construction of the trilobite thorax was a gradual and prolonged process with the thorax recognisable as a distinct region several instars after the segments that came to constitute it first appeared near the rear of the trunk.

Traditionally the trilobite pygidium has been considered homologous to the abdomen of other (eu)arthropods (Burmeister 1846, Cisne et al. 1980), but this pattern of the exchange of segments between the 'abdomen' and thorax is unlike that in almost all other (eu)arthropods

Evolution Trilobites Segment

Figure 15.3 Generalised trilobite ontogeny showing the boundaries of ontogenetic stages based on three aspects of the development of segments: generation, articulation and morphology (see Hughes et al. 2006, fig. 3). 'Gn' refers to stages based on segment generation and contains a poorly known initial stage that may have had a constant set of cephalic segments, the anamorphic phase during which new segments appeared in the trunk, and the epimorphic phase after which the segment number was constant despite continued moulting. 'Form' refers to the morphology of newly generated trunk segments that in some trilobites are divided into discrete

Figure 15.3 Generalised trilobite ontogeny showing the boundaries of ontogenetic stages based on three aspects of the development of segments: generation, articulation and morphology (see Hughes et al. 2006, fig. 3). 'Gn' refers to stages based on segment generation and contains a poorly known initial stage that may have had a constant set of cephalic segments, the anamorphic phase during which new segments appeared in the trunk, and the epimorphic phase after which the segment number was constant despite continued moulting. 'Form' refers to the morphology of newly generated trunk segments that in some trilobites are divided into discrete

(see below, and for distinction between limb-bearing thorax and limbless abdomen, see Walossek and Müller 1998). All trilobites in which ontogenies are well known reached an instar in which segments ceased to be released into the thorax. Attainment of that state marked the transition from the meraspid stage to the holaspid stage. Like the anamorphic generation of trunk segments, the construction of the trilo-bite thorax was a gradual process, although it could depart from the regular release of one segment per instar. Cases in which more than one segment may have been released simultaneously, or two conjoined segments were released into the thorax together, are known but are rare. Other cases in which there may have been two or more instars for every segment released are also known but are uncommon.

the form of trunk segments

Recognition of distinct body regions in arthropods is generally based on the morphology of segments, rather than their pattern of articulation. What little is known of trilobite trunk appendages does not indicate major differences in appendage morphology along the trunk: changes, where they do occur, tend to be mainly size differences, or relatively minor morphological modifications (Hughes 2003a). Nevertheless, the trilobite trunk exoskeleton did show marked differences in segment morphology in some cases (Hughes 2005). Examples include individual segments marked by unique features such as axial or pleural spines, or the grouping of segments into batches, each of which had a broadly uniform segment morphology that was distinct from that of other batches. The latter case is more comparable to the traditional tagmatic

Fig. 15.3 (cont.) batches of anterior (protrunk) and posterior (opisthotrunk) segments. 'Art' refers to developmental stages based on dorsal sclerite articulation pattern and includes the stages previously applied in studies of trilobite ontogeny. Site of the appearance of new trunk segments is shown for the first trunk segment only. Solid grey triangle is the terminal piece, darker grey segments are conjoined and part of the pygidium. Lighter grey segments are thoracic. Individualised segments, such as those that bore unusually large axial or pleural spines (i.e. a 'macrospinous' condition), retained the same position relative to the cephalic margin following first appearance, indicating that the site of appearance of new segments was subterminal, and that the boundary between articulating and conjoined segments migrated posteriorly during the meraspid phase (Stubblefield 1926). The functional significance of this segment was considered by Hughes (2003a).

boundaries recognised along the anterior-posterior axis of other arthropods. Indeed, trilobites present an interesting case in which the evolution of a morphologically distinct posterior tagma apparently occurred independently several times within the group. In trilobites in which all trunk segments, whether thoracic or pygidial, are homono-mous, arguments for recognising the holaspid pygidium as a distinct tagma are weak (Minelli et al. 2003). However, where the trunk is heter-onomous, and divided into batches of distinct segments, it is more reasonable to consider the trunk to comprise two tagmata, particularly where the boundary in segment morphotype coincides with the mature thoracic/pygidial boundary (Hughes 2003a,b, 2005).

The freedom to vary segment morphology within the trunk was strongly constrained in the trilobite thorax because in that region segments were required to articulate with one another. This constraint also apparently applied to the segments in the juvenile trunk that would eventually become thoracic: ultimately they had to achieve a form that permitted articulation. Trilobite trunk segments did not apparently change radically in shape as they were released from the pygidium into the trunk. This is evident in the early ontogenies of those trilobites in which segments that would ultimately become thoracic and those ultimately part of the mature pygidium are differentiated within the meraspid pygidium (Chatterton 1971, Hughes 2003a). Interestingly, peripheral features, such as marginal spines, seem to vary onto-genetically independently of their ultimate identity as thoracic or pygidial segments. However, variation in axial features, such as the courses of furrows marking segment boundaries, and in the relationship between axial and pleural segmentation, could be markedly different in the two regions, with tight covariation in these features in those segments that would become thoracic, and more independent variation in those that would remain pygidial throughout life.

diversity in the segmentation process among trilobites

Each of the three attributes of trunk segmentation discussed above reached a mature phase after which, although growth via moulting continued, each state remained invariant (Figure 15.3) (Hughes et al. 2006). In the case of segment generation this was the onset of a stable number of segments (epimorphic phase); for articulation, it was the attainment of a constant number of thoracic segments (holaspid stage); and with respect to segment form, it was the onset of production of the distinct set of posterior trunk segments (the opisthotrunk) in those trilobites with a heteronomous, two-batch trunk (Figures 15.2 and 15.3). Given the rather regular and progressive development of the trunk segmentation, we might expect that transitions to the mature phases of each of these attributes would be strongly coordinated. In some cases this was so, and particularly in those trilobites in which the mature thorax and pygidium represent distinctly different structures, each presumably with a distinctly different function (Figure 15.4). Nevertheless, trilobites showed a surprising diversity in the relative developmental timing of the transitions to the mature phases of these attributes. Cases in which transition to a stable number of segments coincided with onset of the holas-pid phase are termed synarthromeric, those in which segment generation was completed prior to the termination of trunk articulation are protomeric, and those in which articulation was completed prior to onset of a stable number of segments are protarthrous (Hughes et al. 2006).

A review of such cases shows a variety of patterns (Hughes et al. 2006). For example, the numbers of trunk segments in the Silurian aula-copleurid trilobite Aulacopleura konincki varied, apparently intraspecifi-cally, over a range of five segments, but each of the five morphs apparently showed synarthromeric growth (Fusco et al. 2004). On the other hand, two putative intraspecific morphs of the early Cambrian eodiscid trilobite Neocobboldia chinlinica ultimately apparently achieved the same total number of trunk segments. A suggested explanation for this is that one may have developed synarthromerically, the other proto-merically (Hughes et al. 2006). A comprehensive review of trunk trilobite development has yet to be attempted, but it appears that variation in developmental mode commonly occurred at low taxonomic levels, although some taxa may be characterized by a particular mode. One case is known in which the epimorphic stage was achieved prior to the onset of trunk articulation: the late Cambrian Schmalenseeia fusilis never developed joints in the trunk and thus remained a permanent pro-taspid (Peng et al. 2005, Hughes et al. 2006).

The principal value of the developmental record of trilobites is the ability to explore how these different aspects of the segmentation process map onto the phylogeny of the group, and to consider the evolutionary trade-offs between the flexibility to vary aspects of the trunk independently, and the advantages of increasingly integrated covariation. Key to exploring this is the delay during development between the appearance of segments and the attainment of their final functional role in the thorax. Thorough exploration of the developmental record of

Figure 15.4 The 'two-batch' heteronomous trunk condition in which the mature thoracic and pygidial segments had very different morphologies occurred in a variety of major trilobite taxa, as illustrated by the large specimens in each of the three figures. All specimens are from the AM Limestone of the Devonian of Morocco. A, Ceratarges sp. (Order Lichida), length 8.4 cm. B, The larger specimen is the lichiid trilobite Acanthopyge (Acanthopyge) cf. haueri (Order Lichida), length 8.4 cm. The smaller trilobite (Order Phacopida, species undescribed) shows the homonomous trunk condition. C, Kolihapeltis sp. (Order Corynexochida), length about 5 cm. Note that the boundary in segment morphotype coincides with the mature thoracic-pygidial boundary in each of these cases. Modified from Hughes (2007).

Figure 15.4 The 'two-batch' heteronomous trunk condition in which the mature thoracic and pygidial segments had very different morphologies occurred in a variety of major trilobite taxa, as illustrated by the large specimens in each of the three figures. All specimens are from the AM Limestone of the Devonian of Morocco. A, Ceratarges sp. (Order Lichida), length 8.4 cm. B, The larger specimen is the lichiid trilobite Acanthopyge (Acanthopyge) cf. haueri (Order Lichida), length 8.4 cm. The smaller trilobite (Order Phacopida, species undescribed) shows the homonomous trunk condition. C, Kolihapeltis sp. (Order Corynexochida), length about 5 cm. Note that the boundary in segment morphotype coincides with the mature thoracic-pygidial boundary in each of these cases. Modified from Hughes (2007).

trilobites will require a combination of careful phylogenetic and functional analyses and detailed individual studies of growth mechanics, and firm conclusions are not yet warranted. Nevertheless, those trilobites that varied trunk segment numbers at low taxonomic levels do appear to be those in which the trunk consisted of homonomous segments, while those with a highly tagmatised trunk were invariant in segment numbers (Hughes 2005). The fact that the tendency toward a more highly integrated, segment invariant trunk occurred independently among several trilobite taxa (and possibly also some allied taxa often referred to as 'trilobitomorphs') offers promise for exploring the functional context of such a transition.

developmental patterns among ' orsten' arthropods

Several 'Orsten' (eu)arthropods are preserved with multiple developmental stages, the ontogeny of the eucrustacean Rehbachiella kinnekul-lensis being by far the best known (Walossek 1993) (Figure 15.5B). The earliest larva of this possible stem branchiopod was a true nau-plius, a short-headed larva that is autapomorphic for Eucrustacea. This larva bore only three pairs of appendages, one pair of unira-mous antennulae and two pairs of biramous limbs, the so-called antenna and mandible. Altogether, four 'naupliar' and 26 'post-nau-pliar' stages have been distinguished in R. kinnekullensis. During this phase, the trunk segments were added progressively at the rate of one segment for every two moults. This sequence led to a trunk with 13 limb-bearing segments, while the posteriormost trunk

Figure 15.5 Three crustaceans from the 'Orsten' fauna. Top row shows early larvae, bottom row latest known stages. A, Bredocaris admirabilis (lower picture composed of different specimens). B, Rehbachiella kinnekullensis. C, Martinssonia elongata. All specimens illustrate the exceptional three-dimensional preservation.

Figure 15.5 Three crustaceans from the 'Orsten' fauna. Top row shows early larvae, bottom row latest known stages. A, Bredocaris admirabilis (lower picture composed of different specimens). B, Rehbachiella kinnekullensis. C, Martinssonia elongata. All specimens illustrate the exceptional three-dimensional preservation.

limbs remained of larval shape and the hind body remained unseg-mented (Walossek 1993).

The extant brine shrimp Artemia salina adds one trunk segment per moult, and after that phase the abdominal segments are added successively, while the appendages are modified into adult shape. Comparison with Rehbachiella indicates that even the latest stages of the fossil (size-and developmentally correlated with that of the latest larva of the first phase of Artemia) were still immature. Likewise, Rehbachiella would have required additional moults to develop the segmentation of the abdomen. Isolated limbs and body fragments also hint at older developmental stages, some being twice the size of the largest complete specimen. The 30-stage sequence of Rehbachiella is, therefore, far from being complete.

The strictly anamorphic developmental pattern exhibited by Rehbachiella has been used as a reference for the developmental patterns of other Crustacea (Figure 15.6A, B), serving to highlight deviations from such a regular system. It can even enable us to predict the size of a larva of a taxon based on its segmental stage via the correlation between the number of clearly expressed segments and overall size. This applies between moults that show large developmental 'jumps' between stages, with addition of several segments in one step (Figure 15.6C for penaeid decapod malacostracans).

As compared to the regular developmental pattern in R. kinnekullen-sis, the 'Orsten' eucrustacean, Bredocaris admirabilis (Figure 15.5A), exhibited a derived mode of development. The first larva was already a

0 0

Post a comment