Substrate growth dynamics and biomineralization of an Ediacaran encrusting poriferan

1. Introduction

The appearance of biomineralization in the metazoan (animal) fossil record coincides with a rapid increase in their disparity and diversity, suggesting that this innovation was a major factor in the Cambrian explosion [1]. Understanding the origins of metazoan biomineralization requires study of both the biomineralizing proteome of phylogenetically early-branching metazoans such as poriferans (sponges), and the oldest known fossil record of skeletal metazoans from the Ediacaran.

In addition, the ability to encrust in order to secure and maintain growth to reproductive maturity on substrates was a key competitive innovation in benthic metazoans. Level bottom and reef communities grow not only where suitable substrate is available, but also where competitive overgrowth and predation can be overcome [2]. Communities dominated by organisms with short lifespans, high fecundity and rapid growth will differ from those with high longevity, low fecundity and heavy calcification. In particular, modular and clonal organisms often possess indeterminate growth with great flexibility of morphology and an ability to encrust, and so are superior competitors on stable, long-lived substrates [2].

Namapoikia rietoogensis is a skeletal macrofossil known from Driedoornvlagte reef complex in the Ediacaran–Cambrian Nama Group, Namibia [3]. Namapoikia is associated with coalesced thrombolite mounds that grew on the mid-ramp, high-energy platform margin of the final transgressive cycle of reef growth in the Upper Omkyk Member ([4]; Unit 3 m, OS2 of [5]). Uranium–lead chronology of an ash bed in the immediately overlying the Lower Hoogland Member yields an age of 547.32 ± 0.65 Ma [6,7].

Namapoikia is unique among calcified Ediacaran macrofossils in that it has a robust basal supportive skeleton with a modular, encrusting habit that reaches up to 1 m in diameter (figure 1a). Much remains unknown, however, as to its growth dynamics and possible affinity. Namapoikia has been suggested to resemble chaetetid sponges or simple colonial cnidarians (coralomorphs) [3], but a microbial or protozoan affinity has also been postulated [8]. Namapoikia had a preference for cryptic habitats, encrusting the walls of vertical synsedimentary fissures (neptunian dykes) within thrombolitic (clotted microbialite) reefs [3] which formed vertically perpendicular to reef front growth. This demonstrates that, as in modern reefs, Ediacaran reef metazoans developed preferences for either open surface or cryptic habitats, which may have hosted distinct communities. Namapoikia also actively interacted with other metazoans and microbialites, e.g. one individual is noted to have encrusted and fully enclosed a Namacalathus cup, but was subsequently further encrusted by several generations of laminated microbialites (stromatolites) (figure 1a; [3]).

• Download figure
• Open in new tab
• Download PowerPoint

Here we provide new observations on the internal organization and substrate relationships of Namapoikia, as well as its ecological interactions and growth dynamics with contemporary microbialite. This in turn enables inferences to be made as to biomineralization style and biological affinity.

2. Material and methods

Holotype material from the Geological Survey of Namibia, Windhoek (F 623) and paratype material (SMX 39260) from the Sedgwick Museum, University of Cambridge, were examined. In addition, large samples (200 × 200 × 100 mm) bearing Namapoikia were sliced with a diamond saw at intervals of approximately 20 mm, and cut surfaces polished and imaged wet at 4800 dpi on a flatbed document scanner. Highly polished thin sections (30 µm thickness), were examined under a Cathodoluminescence Cold Cathode CITL 8200 MK3A mounted on a Nikon optiphot microscope to identify differing luminescence and the degree of diagenetic alteration.

3. Morphology of Namapoikia

Intra-skeletal spaces within Namapoikia are commonly partially or fully filled with epitaxial, synsedimentary radiating fans of calcitic, neomorphosed aragonitic cement botryoids (figures 1 and 2a,b). Remaining pore space is filled with clotted thrombolite, microbialite, or white burial spar cement (figures 1c,d and 3d–f).

• Download figure
• Open in new tab
• Download PowerPoint

• Download figure
• Open in new tab
• Download PowerPoint

Namapoikia grew as nodular, encrusting, or domal individuals that either coalesce or extend laterally to form laminar growths (figure 1). The skeleton consists of multiple, incomplete, continuously conjoined vertical tubules that range from 1.5 to 5 mm in diameter, which in transverse section appear labyrinthine to occasionally polygonal (figure 2a,b).

The tubules do not appear to expand with growth but grew by longitudinal fission. Vertical skeletal elements are highly variable, from 0.5 to 3.5 mm in diameter. Short (up to 2 mm) elements extend from vertical elements that resemble spines or incomplete tabulae, coral-like septa or pseudoseptae (figure 2e). Further transverse and complete, cross-linking structures form between longitudinal skeletal elements at millimetre-scale intervals (figure 3a). These transverse structures range in thickness from thin (less than 1 mm) to robust (3 mm). The resultant cross-linking and coalescence of skeletal walls results in a network of highly convoluted tubules, which changes growth direction and converges to form large open chambers within the skeleton (figure 3a). Thin, concave-shaped plates that resemble dissepiments also form between vertical skeletal elements (figure 4d,e).

• Download figure
• Open in new tab
• Download PowerPoint

Skeletal banding is present in many longitudinal skeletal elements, but preservation is highly variable (figure 4a). Skeletal bands form parallel to the main growth direction and have a spacing of 0.5–2.5 mm [3]. Bands range from 0.1 to 1.8 mm in thickness, with a mean thickness of 0.8 mm (n = 41). The banding density is fairly uniform between specimens, and is not present in the matrix. There are no ragged edges to the skeletal elements or incorporated sediment.

Other poorly-defined structures are noted sub-parallel to the walls within the vertical elements in thin-section (figure 4b). These appear as rod-shaped, straight or sinuous forms, up to 4 mm long and 0.2–0.4 mm wide.

4. Substrate relationships

Namapoikia was able to encrust two main substrate types: lithified thrombolite, and unlithified, presumably living, weakly-laminated microbialite and thrombolite. On lithified thrombolite, Namapoikia shows an encrusting and often domal morphology, and grew as sheet-like forms covering the substrate over centimetre-scale distances (figures 1 and 3a,c). Some substrate surfaces show evidence that they formed by fracturing, as they present sharp, sometimes stepped boundaries between substrate and encrusting Namapoikia skeletal material, or truncated features in the substrate (figure 3c). These surfaces are inferred to have been lithified, and sometimes fractured, fissure walls. Initial growth is via robust, extensively attached skeletal elements (figure 3a,c).

By contrast, we also note attachment to living microbial surfaces via an array of blunt holdfasts to weakly laminated microbialite substrates within dykes (figure 3b,d–f). The bulb-shaped holdfasts are approximately 2 mm diameter, and occupy shallow, millimetre-scale, depressions in the surface of laminated microbialite (figure 3b,e). Microbial laminae can also be interrupted by the base of the Namapoikia holdfasts (figure 3f). Namapoikia skeletal material interfingers with the laminated microbialite substrate, and in places is partially encased within it (figure 3e). Above the holdfasts, thicker skeletal elements form, sometimes with an inflated, conical form, from where skeletal walls divide and may cross-link the holdfasts, eventually forming transverse elements (figure 3d).

Namapoikia also shows a very dynamic relationship with actively growing microbialite. Often more than one generation of microbialite is present, here termed M1 (dark grey) and M2 (light grey), which may also alternate (figures 1d and 3b,d,f). Namapoikia shows repeated encrustation by, then subsequent encrustation of, multiple generations of microbialite, as well as thrombolite (figures 1c and 3d,f). Such a process may create enclosed secondary cavities, which become filled with synsedimentary botryoids and later burial spar cement (figure 3d).

We also note a direct relationship between the growth of thin, concave-shaped plates that resemble dissepiments and small domal growths of microbialite or thrombolite. The production of a dissepiment from a vertical skeletal element appears to occur as a response to the presence of adjacent microbialite growth, with the formation of the dissepiment effectively elevating the Namapoikia skeleton above the growing microbialite upper surface (figure 4c–e). In some cases, some skeletal elements appear to continue growth when other elements have been overgrowth by microbialite (figure 4f).

The presence of microbialite can cause apparent inter-fingering with the vertical skeletal elements (figure 4f), suggesting that Namapoikia and microbialite grew concurrently.

5. Insights into biomineralization

Skeletal preservation is as coarse spar mosaic-filled moulds (figure 2c), which show evidence for total recrystallization under cathodoluminescence (figure 2d). An originally aragonitic skeleton has been inferred on this basis [3,9].

The recrystallized form of Namapoikia skeleton reveals little as to its original microstructure, but evidence for rapid extension of the skeletal elements of Namapoikia suggest the presence of a pre-existing organic scaffold prior to calcification. Thin, concave-shaped plates relative to the growth direction that resemble dissepiments also form between vertical skeletal elements (figure 4d,e). These may sometimes represent the earliest stage of growth of the transverse elements, as they are thinnest in the youngest parts of the skeleton, and so subsequently thicken with age from the initial colonized substrate (figure 3d). Likewise, vertical skeletal elements also thicken with age (figure 3b,d). We also note that younger skeletal elements are both thinner and darker than older elements (those closer to the attached substrate) which are often pale grey and also show growth banding (figure 3d). It is likely then, that the organic matrix become increasingly calcified with time.

We infer this mode of biomineralization to have progressed via the initial formation of an organic scaffold, which thickened and also progressively calcified with time. The presence of dissepiments suggests rapid growth of this organic scaffold followed by potentially rapid calcification. The operation of incremental or episodic growth is supported by the presence of growth banding within skeletal elements. The absence of ragged edges to the skeletal elements or incorporated sediment suggests that the bands represent growth interruption surfaces. The orientation of the banding perpendicular to growth does not support a diagenetic origin, notwithstanding complete recrystallization of the originally aragonitic skeleton. We infer banding in Namapoikia is analogous to the density banding, often annual, found in ancient and modern corals and stromatoporoids [10].

The presence of an organic scaffold as a template for subsequent calcification is described from the living sphinctozoan-grade sponge Vaceletia, part of a lineage that first appears in the mid-Triassic and which has been suggested to show an ancient mode of biomineralization [11]. Here, the skeleton is secreted on a non-collagenous organic template formed by the basal pinacoderm. This template becomes substituted by crystalline aragonite deposited as tangled crystal bundles of aragonite [12]. The organic framework consists of proteins and polysaccharides rich in galactose, glucose and fucose, the latter suggesting that bacterial EPS (exopolymeric substances) may be involved in the biocalcification process [11]. In most cases, the basal parts of the skeleton, which is free from living tissue, is infilled by a micritic granular secondary deposit. Interestingly, not all the chambers of Vaceletia mineralize entirely, as inferred for Namapoikia.

The presence of such a pre-existing organic scaffold for biomineralization has been inferred for other Ediacaran macro-organisms including Cloudina and Sinotubulites, as well as Cambrian archaeocyath sponges [3,13].

The nature of the rod-shaped, straight or sinuous structures embedded within skeletal walls remains unclear, but it is possible that these may represent episodes of incremental calcification or even spicules.

6. Biological affinity

As in modern modular poriferans and cnidarians, the soft tissue of Namapoikia probably occupied only the uppermost, youngest skeletal parts, forming a relatively thin veneer of tissue. We note examples where the older parts of skeletal pore space in Namapoikia are infilled with cement botryoids or later calcite spar cement only, but by contrast the youngest skeleton, extending some 5–12 mm from the outer surface, is infilled with microbialite (figure 1a). We can infer that the older, abandoned parts of the skeleton which were progressively sectioned-off by skeletal structures created enclosed voids that became cemented, but that the skeleton filled with soft-tissue (penetrating approximately 12 mm from the upper surface) remained open and so was infilled by fouling microbialite after death.

The arrays of bulb-shaped holdfasts and free-standing skeletal elements are not consistent with the interpretation of Namapoikia as a calcified microbial colony or very large skeletal protozoan. The architecture of Namapoikia, with walls cross-linked by transverse skeletal elements, is reminiscent of the systems of tubules interrupted by tabulae in tabulate corals and in some hypercalcified poriferans, especially the polyphyletic chaetetid demosponges. These structures may indeed represent tabulae, as suggested by Wood et al. [3].

Living chaetetid-grade sponges such as Acanthochaetetes wellsi and A. seunesi, and the fossil forms Blastochaetetes dolomiticus, and Meandrioptera zardinii [14] show a similar internal structure of elements with transverse cross-linkages, containing a network of tubular voids. Acanthochaetetes and some fossil chaetetids, such as Chaetetes radians [12,14], also have pseudoseptae, similar to the elements that extend from longitudinal skeletal walls of Namapoikia that incompletely divide tubules in longitudinal sections (figure 2a). The living calcified demosponge Vaceletia shows inward facing short spines within ostia [12]. Vertical columns, transverse elements, pseudoseptae all form as part of the primary skeleton. Secondary thickening of all skeletal elements occurs later. Chaetetids may also have tubercles and astrorhizae on their upper surfaces, but these have not been observed in Namapoikia.

The tubules in Namapoikia both combine (fuse) and divide, and the spaces within the skeleton are also not always uniformly tubular but often labyrinthine. One differentiator between calcified sponges and tabulate corals is the style of tubule branching [15]. Tabulate coral tubules branch laterally from existing tubules, while calcareous sponges show longitudinal fission or intramural increase, where tubules subdivide or new tubules initiate within the walls of pre-existing ones. Namapoikia tubules show both lateral branching and longitudinal fission, and may also show intramural increase (figure 2a). However, the often highly irregular internal structure of Namapoikia does not unequivocally relate to any of these branching modes.

While stromatoporoid and chaetetid-grade sponges show a similar domal or encrusting habit and internal structure of elements and spaces to Namapoikia, stromatoporoid skeletal elements tend to be far smaller, more densely spaced, and with a more pronounced and well-organized lamination than found in Namapoikia.

Although inferred for probable extinct poriferans, in living taxa the presence of an organic scaffold as a template for subsequent calcification is described in living sponges such as the sphinctozoan Vaceletia and chaetetid Acanthochaetetes [12].

In sum, the highly irregular, non-deterministic skeletal organization, open tubular structure showing lateral branching and longitudinal fission, the presence of transverse elements and potential pseudoseptae, and the inferred style of biomineralization of Namapoikia places probable affinity within total-group poriferans.

We can then reconstruct the relationship between Namapoikia and substrates, as well as soft-tissue and skeletal elements of Namapoikia as an inferred poriferan (figure 5). Vertical elements form on lithified surfaces; bulb-shaped holdfasts develop when attached to living substrates. The basal pinacoderm may secrete the organic matrix. The organic matrix forms first as vertical elements and perhaps spines, in which calcification subsequently takes place by substitution of crystalline aragonite. We infer that the tabulae and pseudosepta (spines) are also precipitated as part of the primary calcareous skeleton: there is no secondary skeleton but secondary thickening occurs with age. Tabulae serve to separate off abandoned parts of the skeleton. Plate-like dissepiments form in response to microbial overgrowth, which may become thickened to form tabulae. Secondary thickening of the pillars may occur via additional formation of organic matrix, and then calcification. We have not noted any traces of the excurrent (exhalent) canal system in the form of astrorhizae, but the excurrent canal may inhibit skeletal formation if in contact with the area of skeletogenesis. Skeletal voids away from the growth of living microbialite become cemented, but voids close to living microbialite surfaces may become infilled with microbialite.

• Download figure
• Open in new tab
• Download PowerPoint

7. Discussion

Although it is not known at what periodicity the growth bands occurred, their presence, together with the large size of Namapoikia, is suggestive of a notable longevity for Namapoikia. Such incremental growth over extended timescales is in contrast to the inferred opportunistic life habit of other Ediacaran skeletal metazoans such as Namacalathus, which also occurs abundantly at Driedoornvlagte [16,17]. Namapoikia was a relatively long-lived, persistent member of reef communities, suggesting that its cryptic habitat was also persistently oxic and stable over relatively long timescales, in contrast to the transiently hospitable environments often colonized by Cloudina and Namacalathus [17,18].

Namapoikia shows a diverse range of encrusting to domal morphologies. Such intraspecific morphological plasticity is common in extant and ancient poriferans and cnidarians, where it is often a response to different hydrodynamic conditions, substrates, or rates of sediment supply [19].

The ecology of Namapoikia suggests a strategy of rapid and effective substrate coverage, which may have been facilitated by the initial growth of an organic skeleton that subsequently calcified. Namapoikia is capable of two distinct attachment strategies: encrustation on lithified thrombolitic or fractured substrates, or the production of arrays of bulb-shaped holdfasts on unlithified microbial substrates. The interruption of microbial laminae by holdfasts that become embedded in microbialite suggests that Namapoikia colonized living microbial substrates. We therefore conclude that Namapoikia could attach to both lithified and living surfaces. This suggests a level of indeterminant growth and plasticity in attachment strategy unusual among marine invertebrates.

Some extant sponges can attach to both lithified and soft, living substrates via holdfasts, which are often small and conical on soft substrates, though such individuals tend to be small in size [20]. Indeed some extant sponges, cnidarians and other benthic marine invertebrates preferentially colonize substrates coated with living bacterial films, possibly in response to specific chemical cues [21]. The ability to adopt different attachment strategies in response to variation in the mechanical properties of available substrates implies a high level of sensory ability and regulatory control in Namapoikia.

Namapoikia shows a highly dynamic behaviour of intergrowth with microbialites. This enabled a successful response to potential overgrowth, and may also have enabled re-growth from partial mortality, as shown by continued growth of some skeletal elements when other elements have been overgrowth by microbialite (figure 4f). This suggests that like modern reefs, substrate space was at a premium, and competition for substrate was a major determinant of ecology. Such a modular and clonal construction greatly enhances competitive fitness on stable substrates [2].

The major competitor of Namapoikia for substrate appears to have been microbial growth, not other skeletal metazoans. But as in other Ediacaran macro-organisms, both skeletal and soft-bodied, Namapoikia had a strong ecological relationship with the pervasive microbial mats of its habitat. While skeletal metazoans in the late Ediacaran have microbial mat-associated lifestyles [17,22], this is the first documentation of direct competitive growth between metazoans and microbial mat communities. This is notable as first, it counters the assertion that the late Ediacaran was a generally low-competition environment [13,17,23], and secondly, such microbial mats must have presented a potential fouling risk to early metazoans.

Namapoikia provides evidence that the ability to compete effectively on stable and living substrates had evolved by the terminal Ediacaran. Ediacaran skeletal metazoan communities had therefore differentiated into those dominated by organisms with inferred relatively short life histories, high fecundity and rapid growth such as Namacalathus and Cloudina [16,17], and those with relatively high longevity, inferred low fecundity but clonal behaviour, and heavy calcification. The restriction of Namapoikia, however, to cryptic niches, might suggest that Namapoikia was unable to compete successfully for substrate on open surfaces, which may have been dominated by faster-growing microbial communities or metazoans adapted to well-lit and/or higher energy conditions.

The similar style of Namapoikia biomineralization to the living demosponge Vaceletia is noteworthy. Poriferans are widely regarded to be the earliest branching metazoans, and are among the first multi-cellular animals to display the ability to biomineralize in the fossil record. Forty proteins from both the head and stalk regions of Vaceletia have been identified, including known biomineralization compounds such as carbonic anhydrase, spherulin, extracellular matrix proteins and very acidic proteins [11]. The presence of deeply conserved biomineralization proteins such as α-CA demonstrates that the Last Common Ancestral Metazoan (LCAM) contributed several key enzymes and matrix proteins to descendants that in turn supported the metazoan ability to biocalcify [11]. Indeed, although a further calcified demosponge Astrosclera willeyana shows very different skeletal morphologies and skeletal microstructure to Vaceletia sp., these two taxa apparently share at least two important biomineralization proteins, astrosclerin and spherulin [11]. This suggests therefore that there may be notable conservation in the molecular details of skeleton formation across the Metazoa despite divergent skeletal morphologies.

8. Conclusion

Namapoikia was a large, robustly calcified clonal metazoan with a complex, though irregular skeletal structure consistent with poriferan-grade organization. Biomineralization probably proceeded via calcification of a rapidly growing organic matrix, which also facilitated rapid and successful substrate colonization and enabled recovery from partial mortality. This is likely an ancient mode of biomineralization with similarities to the living calcified demosponge Vaceletia.

In cryptic habitats, Namapoikia shows evidence of substrate-specific attachment strategies and colonization of both lithified and living microbial mats. Namapoikia also shows a high degree of morphological plasticity, with notably indeterminate growth that was adapted to compete effectively for substrate space with, and combat overgrowth by, living microbial surfaces. The large size and possible original growth banding of Namapoikia suggests a long-lived life strategy. This is the first documentation of direct competitive growth between Ediacaran metazoans and microbial mats, and is notable as it provides a counter-example to the assertion that the late Ediacaran was a generally low-competition environment, and demonstrates that microbial mats must have presented a potential fouling risk to some early skeletal metazoans.

Namapoikia provides evidence that the ability to compete effectively on stable and living substrates had evolved by the terminal Ediacaran. Ediacaran skeletal metazoan communities had therefore differentiated into those dominated by organisms with inferred relatively short life histories, high fecundity and rapid growth such as Namacalathus and Cloudina [16,17], and those with relatively high longevity, inferred low fecundity but clonal growth, and heavy calcification.

In sum, such an interpretation confirms the presence of poriferans, with calcareous skeletons, in the terminal Ediacaran, and also that sophisticated adaptations for effective substrate competition had evolved in the cryptic benthos where microbial mats presented a fouling risk to early skeletal metazoans.

Ethics

This work has not involved any living subjects and conforms to the Ethics guidelines of the School of GeoSciences, University of Edinburgh.

Data accessibility

All data used in this study are presented in the text.

Authors' contributions

R.W. and A.P. designed the study; R.W. collected the material; A.P. carried out the morphological description; R.W. and A.P. drafted the manuscript; all authors gave final approval for publication.

Competing interests

We have no competing interests.

Funding

R.W. and A.P. acknowledge support from the International Centre for Carbonate Reservoirs, and A.P. from the University of Edinburgh and the Daniel Pidgeon fund of the Geological Society of London.