Ghost in the shell

Paleontologists always seek the most complete reconstruction of their beloved study objects, e.g., dinosaurs, mammals, or extinct invertebrates like the ammonites. The latter are representatives of the cephalopods that populated the oceans for more than 340 My. Among modern cephalopods only Nautilus shares an external shell with ammonites, while the majority of modern cephalopods evolved an internal and often reduced shell. All shell-forming mollusks—such as cephalopods and gastropods, among others—share the same mechanism of shell formation, continuously adding material to the growth front by a specialized tissue called the mantle (1). This means that their complete ontogeny can be recorded in the shell. Unfortunately, the fossil record of mollusks is mostly limited to shells and much is unknown about their soft-tissue organization. Chirat et al. (1) present a physical model to look at mollusk soft-body organization and growth as manifested in shell form, which also informs us about the soft-body symmetry of the extinct ammonites.

At first glance, the reader may wonder why this subject is of general interest. Mollusks belong to the Bilateria and are organized along an anteroposterior and a dorsoventral axis both defining the plane of symmetry, case closed. A look at gastropod anatomy, however, shows that the majority of gastropods form helicospiral shells and their soft body is anything but bilaterally symmetric. This is also true for non-shell-forming gastropods (2). Due to these peculiarities, gastropods became model organisms for studying the potential bases for observed left–right asymmetry. Gastropods and externally shelled cephalopods may seem intuitively comparable due to their spirally coiled shells. Both groups share the same basic anatomy of the mantle edge secreting system. However, does that mean that cephalopods forming a helicospiral shell, e.g., ammonites and nautiloids, also evolved an asymmetric soft body?

Only 1% of all known (fossil and recent) externally shelled cephalopods develop a nonplanispiral shell (helicospiral or meandering), but helicospiral shells evolved independently four times during ammonite evolution (?Late Devonian, Late Triassic, Late Jurassic, and Late Cretaceous) (3). Although these forms, which do not form a monophyletic group, are derived from bilaterally symmetric ancestors, and phylogenetic bracketing would support their bilateral symmetry, the question arises whether or not these helicospiral ammonites developed an asymmetrical soft body. Chirat et al.’s (1) model shows how a bilaterally symmetric ammonite body plan can produce an asymmetric shell through mechanically induced twisting.

The model of Chirat et al. (1) includes developmental mechanisms and physical constraints. It is composed of two elastic rods (dorsal-internal/ventral-external) representing the mollusk soft body. The ventral rod grows faster compared to the dorsal rod and under stress-free conditions, meaning a perfect match of soft body and shell growth, a planar logarithmic spiral results. A small mismatch between soft body growth rates and shell secretion rates results in mechanical stress and will deform the mollusk soft body. When secretion outpaces soft-body growth, the body becomes stretched and will be in tension. In such cases a planispiral shell always results. However, where body growth rates exceed shell secretion rates the soft body will be in compression. Here, according to the model of Chirat et al. (1), the soft body performs a twist within the shell to reduce this compressive stress. The dorsoventral axis including growth and secretion gradients becomes rotated—thus the shell shape will change into a nonplanar form. This change in shell shape in turn creates new stress on the growing animal. A constant mechanical stress rate (i.e., constant twist rate) will result in helicospiral shells, and an oscillatory twist rate will result in a meandering shell. One of the major outcomes of this model is to explain how bilaterally symmetric animals like the ammonites can form asymmetrical shells when their body is under compression. Second, gastropods with an intrinsic asymmetrical body form helicospiral shells without compression.

One of the main arguments for a bilaterally symmetrical body in ammonites with a helicospiral shell is the 50:50 ratio of left (sinistral) and right (dextral) coiled shells (Fig. 1A), while in gastropods coiling is typically dextral (90%). The equal distribution of sinistral and dextral shells suggests a random determination, supporting the physical model in which twisting is equally likely to occur in either direction. Such twisting within the shell, resulting in an irregularly coiled shell, has also been reported for gastropods, demonstrating the general possibility of such scenarios (4), and a twist of the ammonite soft body is recorded by the angular offset between parts of the ventral siphuncle (the buoyancy-regulating tube that runs through shell chambers) in a sinistral Turrilites (5). Further, hatchlings of ammonoids always start with a bilaterally symmetric shell. Moreover, shell openings in heteromorph ammonites generally remain bilaterally symmetrical, whereas gastropod shell openings are asymmetrical (6). Of major importance is the genetically fixed cell cleavage pattern which, as in the majority of mollusks, is spiral and typically dextral in gastropods. That cleavage pattern can partially explain shell morphogenesis in gastropods, but not all mollusks produce helicospiral shells (e.g., bivalves). For example, limpet gastropods have a dextral soft-body anatomy but produce bilaterally symmetrical shells. Obviously, the development of the shell–muscle system (asymmetric in helicospiral gastropods and symmetric in limpets and cephalopods) as a mechanical factor plays an important role. Cephalopods with their large yolky eggs show a highly modified cleavage pattern, with bilaterally symmetrical cell division occurring in a disk-like area followed by direct development (2). Rare ammonoid soft-tissue records further support the bilateral symmetry of ammonoids: bilaterally symmetric radulae for ammonoids and nautiloids (7), and jaws of Turrilites (8) and Didymoceras (9) are symmetrical as well. Recently, a rare case of soft-body preservation with paired organs (upper and lower jaw and gills) for a Late Jurassic planispiral ammonite was reported (10).

Fig. 1.

Helicospiral cephalopod shells. (A) Turonian Hyphantoceras reussianum from Halle (Northrhine-Westfalia, Germany). Both specimens are about 6.5 cm high; the upper specimen sinistral, the lower specimen dextral. (Scale bar, 1 cm.) (B) Virtual model of Campanian Didymoceras stevensoni from the Western Interior Seaway with a planispiral shell of the hatchling (not shown), straight to slightly bent juvenile shell parts, turning into a helicospiral shell that transitions into a meandering and finally planispiral shell part (both green). (C) Internal helicospiral shell of the mesopelagic coleoid cephalopod Spirula spirula. (Scale bar, 2 mm.)

Chirat et al. present a physical model to look at mollusk soft body organization and growth, which also informs us about the soft body symmetry of the extinct ammonites.

The model put forward by Chirat et al. (1) is a provocative morphogenetic explanation of a mechanical origin for the occurrence of helicospiral shells in bilaterally symmetric cephalopods, albeit incomplete and difficult to verify. It is important to note that development of dextral or sinistral shells is not fully disconnected from genetics according to the model. Accordingly, the path followed by the shell edge and the resulting form is partly governed by the mechanics of the body inside the shell. The mechanical twist is growth-dependent and undoubtedly was governed by growth-regulatory genes in ammonites. Reports on equally frequent left and right asymmetry in scale-eating cichlid fish due to a combination of genetic and environmental factors (11) support the notion that no single factor alone can explain the phenomenon. Asymmetry seems to be intrinsic for these fish because already juveniles, feeding exclusively on zooplankton, show this morphological trait, and left-symmetric pairs produce more, but not exclusively, left-symmetric offspring and right-symmetric pairs more right-symmetric offspring, indicating that a genetic bias need not always result in a directional asymmetry. A genetic factor controlling helicospiral shells would also explain their consistent morphology allowing for a taxonomic treatment at the species level. In the framework of Seilacher's concept of “Konstruktionsmorphologie” (12), the observed sinistral and dextral helicospiral shells should be discussed taking phylogenetic, functional, biomineralization, and environmental aspects into account. A well-balanced mixture of different processes was regarded as more suitable to explain heteromorph shells of ammonites and their morphological uniformity in general (6). Other aspects like hydrostatics (see ref. 13 for Didymoceras and ref. 14 for Nipponites) and hydrodynamics (15), including orientation and stability, were suggested to influence the mode of shell coiling as well. The recurring patterns among distantly related heteromorph ammonites supports the idea of an adaptive value for these shell shapes. These findings may not only hold true for ammonites but also helicospirally coiled nautiloids like Oncocerida, Discocerida, and Barrandeocerida (16). When it comes to the environment, the notion that soft-body growth and secretion rates may vary during ontogeny based on either environmental stimuli or internal cues such as stress is important. According to the presented model, these individual changes would result in a chaotic distribution of heteromorph shell shapes. This is not the case. A planispiral ammonite (Aplococeras) formed a helicospiral shell in response to infestation by epizoans, which caused the shell to tilt during life due to changes in hydrostatic conditions (17).

Also interesting to explore in future work will be the relaxation of the assumption that growth rates are isometric and mostly static, due to the difficulty of mathematical handling of allometric growth. As allometric growth occurs in all cephalopods, with isometric growth rather exceptional among animals in general (18), nonstatic growth rates can be assumed in cephalopods; they have been observed for Nautilus (19).

Finally, a few questions remain in light of this mechanical model. Why did pathological heteromorphs (Aegocrioceras, Pictetia, Proaustraliceras, and Ptychoceras) with nontouching whorls (figure 5 of ref. 20) not develop meandering or helicospiral shells, assuming a slight mismatch between growth rates was induced by traumatic events? Why does Spirula have a slightly helicospiral internal shell (Fig. 1C)? The internal position of the shell should exclude a mismatch between the two growth rates resulting in compression of the soft body. A completely straight body would create very high stresses according to the model, and the energy-minimizing shell forms would be quite unrealistic. It is concluded by Chirat et al. (1) that these shells are unlikely to exist, but they do: juvenile Didymoceras, Baculites, Sciponoceras, Bochianites, Bactrites, many Paleozoic nautiloids, and scaphopods (Fig. 1B). If the occurrence of helicospiral shells is largely related to a mismatch between two growth rates, why does it occur only in 1% of all externally shelled cephalopods and within these forms so consistently? This model is an important step that will stimulate many interesting questions regarding the occurrences of asymmetry in mollusk and asymmetry in general, providing a new innovative computer-based approach.