Ammonite shells have complex patterns of suture lines that vary across species. The lines are formed at the intersection of the outer shell wall and the septa. The wavy septa can form if the rear mantle of the ammonite, which functions as the template, has a complex shape. Previous hypotheses assumed that the rear mantle is like a flexible membrane that can be folded by some physical force. The elucidation of the mechanism of septa formation requires that the detailed shape of the septa should be known. We developed a new protocol of X-ray micro-computed tomography (CT) and obtained high-resolution three-dimensional (3D) images of the septa of the Upper Cretaceous ammonite Damesites cf. damesi. The obtained image suggested that the wavy and branched structures of the rear mantle grew autonomously. We found that some extant sea slugs have branched structures and showed similar shape and growth sequence as those in fossils, suggesting that the mantle of molluscs basically has the potential to form branched projections. Based on the characteristics of the obtained 3D structure, we explain how ammonites might have formed the complex suture patterns.
Ammonite conchs are separated into small chambers by septa, which might contribute to reinforce the shell wall1,2,3,4 and control buoyancy5,6,7. The wavy line of intersection between the shell and septum is called the suture line, which is highly diverged among ammonite species; it is a useful index for taxonomical identification8,9,10,11. However, very little is known about the mechanism of how this complex and diverged pattern was generated.
Living Nautilus also have septa; they form these septa by using their round rear mantle as a template12. Since ammonites are related to Nautilus, the ammonite septa should also have been formed by using the rear mantle as a template. However, the formation of the complex suture pattern requires extensive deformation of the rear mantle. In the two previously proposed hypotheses, the rear mantle of ammonites was assumed to be flexible and could be passively folded by some external force into complex folds.
The tie-point model, the best-known hypothesis, assumes that the rear mantle adheres to the shell wall at many small points known as tie-points13,14,15,16. If hydrostatic pressure was applied from the apical side, the mantle would be pressed to move adorally, but the tie-point region would remain at the same position and eventually form an arch-like shape.
The other model is the viscous fingering model17,18,19,20. According to this model, when two fluids with different viscosity come in contact, the Saffman–Taylor instability is induced at the interface, generating finger-like shapes that are similar to the ammonite suture patterns. This model assumes that the Saffman–Taylor instability occurs on the surface of the rear mantle and in the fluid of the shell chamber. Although these models explain the formation of the suture pattern via a simple method, both have a problem. These models are not based on the phenomena found in extant organisms. Therefore, validating these models is difficult.
Hammer presented another idea in 199921. His model does not involve any specific physical method, but simply assumes that the peripheral region of the rear mantle invaginates to form the folding of the septa.
Previous models mainly focused on the two-dimensional (2D) suture pattern and considered the behaviour of the marginal region of the ammonite mantle. However, because the suture lines are the intersection between the shell wall and septa, understanding how the three-dimensional (3D) structure of the septa is formed is necessary. Thus, we developed a method to obtain a super-high resolution 3D image of ammonite septa and then compared the previous models with the obtained image.
3D imaging of ammonite septa
To determine how ammonites formed the complex suture patterns, we performed detailed 3D analysis of the ammonite shells. Various types of ammonite fossils that vary in mineral diagenesis are available. In some types of ammonite fossils, the shell chambers were filled with calcite; diagenesis then dissolved and eliminated the shell wall and septa, generating an empty space at their place. We analysed such samples by using micro-computed tomography (CT) and obtained almost the complete sequence of septal structure with the highest resolution (Supplementary Figs 1–6). The 3D image for specimen NMA00802, Upper Cretaceous ammonite Damesites cf. damesi (Ammonitina, Desmoceratidae) from Hokkaido, is shown in Fig. 1. In addition to the septa, other structures such as septal necks, siphuncle and cameral sheets were visualised in high resolution (Supplementary Fig. 7). The septa appeared like repeated finger-like structures dividing at the edge (Fig. 1f,g). We named these finger-like structures as septal arms. To describe the detail of the septal structure, we referred these septal arms as external, lateral, umbilical and internal arms, following the conventional terminology of suture patterns. The external arm was the largest and the most extensively divided. The lateral and umbilical arms were smaller and had lesser branches. Therefore, the density of the septal tips was almost uniform, resulting in the fractal appearance of the entire structure.
Tubular structures in ammonite septa
The margin of the arms occasionally joined with the previous septum as well as with the shell wall (Fig. 2). The number of septum-to-septum joints was low in the outer (ventral) region and high in the inner (dorsal) region. In the inner regions, the arm sections often formed a complete circle, indicating that they did not have any margin, but had a tubular structure, which is known as septal lobe in lytoceratid ammonites22 (Fig. 2). We believe that the formation of such a tubular structure might not be possible by the previous models for the following reasons. Since the viscous fingering effect occurred only in the quasi-2D field, such 3D structure should be beyond expectation of the model. Formation of the septal lobes by the tie-point mechanism requires an irregular condition. That is, tie-points need to occur at the inside of the rear mantle, but not at the edge. In addition, the tips of the tubular structures divide into many fine branches. The gaps of these branches are alternately filled with the fine branches of the adjacent arm (white circle in Fig. 2a). Formation of such a structure requires that the branches are formed before they touch the old septum.
To determine whether such a structure is common among ammonite species, we examined one more species, Hauericeras sp., belonging to the same family as that of Damesites cf. damesi and two other species from different suborders, Tetragonites sp. (Litoceratina, Tetragonitidae) and Hypophylloceras sp. (Phylloceratina, Phylloceratidae), which confirmed the generality of this structure (Supplementary Fig. 8).
Similarities between ammonite septa and sea slug cerata
Branched projections are not a rare structure, but rather commonly found in various animals; in particular, gastropod molluscs of the superfamily Dendronotoidea (a group of sea slugs) have a wide variety of branched projections known as cerata; their patterns resemble the suture lines (and naturally also the septa) of ammonoids (Supplementary Fig. 9). Because the cerata of sea slugs develop from the mantle, a similar mechanism can be assumed in ammonites.
Although living ammonites are not available, the micro-CT imaging allowed the observation of a series of septa that led to the growth of the mantle projection. When the growth sequence of sea slug cerata and ammonite septa were compared (Figs 3 and 4), the similarity between the two organs could be estimated.
The growth sequence of a lateral arm of Damesites cf. damesi is shown in Fig. 3b–e. When the organisms become older and larger, the septal folding becomes more complex by the repetitive branching. A part of the septal arm (Fig. 3f) is very similar to the entire structure of the younger arm (Fig. 3d), suggesting that the arm branching is a fractal structure. We could grow the sea slug Tritoniopsis elegans in a sea water tank and record how the shape of the cerata changed (Fig. 3h–j). A single ceras repetitively divided and formed a branched structure in the similar manner as that in the ammonite (Fig. 3h–j). A part of a large ceras (Fig. 3l) is similar to the entire ceras of a young sea slug (Fig. 3j), suggesting that the structure is fractal.
During the enlargement of the shell, small new arms appeared between the old and big arms. New septal arms appeared first at the most ventral region where the interval between the arms was the widest and then gradually appeared in more lateral regions (Fig. 4b–e). This mode of arm insertion implies that the timing of the new arm insertion was dependent on the space without the arms (Hammer also pointed out this tendency and suggested the involvement of the lateral-inhibition mechanism21). A similar tendency was observed in the sea slug. The wider the spacing between the old cerata, the earlier was the appearance of the new ceras (Fig. 4g–j). Therefore, the growth process of the ammonite septa and sea slug cerata is very similar and both might be controlled by similar mechanisms.
Our observations suggest that the rear mantle of ammonites might have self-grown projections whose shape and growth sequence resembled the cerata of sea slugs. The projections could form without the external physical method because sea slugs can form the dorsal projections autonomously in the absence of shells. This ‘cerata-septa model’ is close to the idea presented by Hammer (1999) and has some advantages compared to the other models. First, the cerata-septa model is based on existing biological phenomenon, which is more reasonable than postulating the phenomena not found in extant organisms. Branching is one of the common phenomena in organisms, especially in molluscs. Therefore, ammonites could have formed projections in a similar manner.
Second, the model can account for the variety of suture patterns among ammonite species and can be applicable to ammonoid species because the branching pattern of cerata varies extensively among the species (Supplementary Fig. 9). Comparing various ammonoid septa and sea slug cerata is necessary for regarding the cerata-septa model as a unified theory.
Third, the cerata-septa model also considers the soft part of the ammonites (Fig. 5; Supplementary Fig. 10), which remains almost unknown23,24. If the rear mantle of ammonites undergoes suturing via a similar mechanism as that in sea slugs, experimentally inferring the nature of the ammonite soft part by using extant sea slug might become possible.
Our observation of Damesites cf. damesi is based on the fossil specimen NMA00802 reposited at the Nakagawa Museum of Natural History, Hokkaido, Japan. This specimen was excavated from an outcrop (44.229°N, 142.005°E) exposed along a small tributary of the Chimeizawa River in Haboro district, Hokkaido, Japan. The fossil was preserved in a calcareous sandy mudstone nodule of the Haborogawa Formation, Yezo Group, with fine woody fragments. According to Hayakawa et al.25, the fossil-bearing horizon can be correlated with Santonian. In this study, we followed the classification of Damesites established by Nishimura et al.26 and NMA00802 can be referred to as Damesites cf. damesi, S-group morphotype. The following three additional ammonites are shown in Supplementary Fig. 2: Tetragonites sp. is based on NMA00803 occurring with NMA00802; Hypophylloceras sp. is based on NMA00811 from the Yezo Group (Upper Cretaceous) of Tomamae district, Hokkaido; Hauericeras sp. is based on NMA00812 from the Yezo Group (Upper Cretaceous) of Hokkaido; all these specimens are reposited at the Nakagawa Museum of Natural History. These specimens are selected by sight on the basis of the condition that the chambers are filled with calcite and the septa are completely dissolved.
Fossils were scanned using ScanXmate-D150-S270 (Comscantecno) to obtain projection data and the data, including 3D reconstruction data, were processed and visualised using Molcer Plus software (White Rabbit). The tube voltage was optimised to 151 kV. The reconstructed 3D images consisted of cubic voxels and their resolution was 30.71, 29.14, 22.37 and 39.92 μm for Damesites cf. damesi, Tetragonites sp., Hypophylloceras sp. and Hauericeras sp., respectively. A continuous domain of the shell and septum was extracted for analyses by manually fixing minor discontinuities between voxels that were probably caused by tiny cracks or noise. A domain of the outer shell was also manually selected and masked in order to observe clearly the structure of the septum (Figs 3 and 4).
Observation of a sea slug
Tritoniopsis elegans was collected from Hachijo-jima, Tokyo, Japan. It was maintained in artificial seawater at 23–24 °C and fed soft coral, Cladiella digitulata. Subsequently, it was weakly anesthetised with artificial seawater containing 100 mM MgCl2 and covered with a 50 × 70-mm cover glass (Matsunami Glass); it was viewed and imaged using a Leica MZ16 FA stereomicroscope equipped with a DFC300FXR2 digital camera (Leica) every three days.
How to cite this article: Inoue, S. and Kondo, S. Suture pattern formation in ammonites and the unknown rear mantle structure. Sci. Rep. 6, 33689; doi: 10.1038/srep33689 (2016).
Buckland, W. 1836. Geology and Mineralogy Considered with Reference to Natural Theology (W. Pickering, London, 1836).
Westermann, G. E. G. The significance of septa and sutures in Jurassic ammonite systematics. Geological Magazine 95, 441–455 (1958).
Hassan, M. A., Westermann, G. E. G., Hewitt, R. A. & Dokainish, M. A. Finite-element analysis of simulated ammonoid septa (extinct Cephalopoda): septal and sutural complexities do not reduce strength. Paleobiology 28, 113–126 (2002).
De Blasio, F. V. The role of suture complexity in diminishing strain and stress in ammonoid phragmocones. Lethaia 41, 15–24 (2008).
Weitschat, W. & Bandel, K. Organic components in phragmocones of Boreal Triassic ammonoids: implications for ammonoid biology. Paläontologische Zeitschrift 65, 269–303 (1991).
Seilacher, A. & LaBarbera, M. Ammonites as Cartesian divers. Palaios 10, 493–506 (1995).
Daniel, T. L., Helmuth, B. S., Saunders, W. B. & Ward, P. D. Septal complexity in ammonoid cephalopods increased mechanical risk and limited depth. Paleobiology 23, 470–481 (1997).
Arkell, W. J. Introduction to Mesozoic Ammonoidea. Treatise On Invertebrate Paleontology, Part L, 81–129 (Geological Society of America, Boulder, & University of Kansas Press, Lawrence, 1957).
Miller, A. K., Furnish, W. M. & Schidewolf, O. H. Paleozoic Ammonoidea. Treatise On Invertebrate Paleontology, Part L, 11–79 (Geological Society of America, Boulder, & University of Kansas Press, Lawrence, 1957).
Kullmann, J. & Wiedmann, J. Significance of sutures in phylogeny of ammonoidea. Paleontological Contributions 47, 1–32 (1970).
Korn, D., Ebbighausen, V., Bockwinkel, J. & Klug, C. The A-mode sutural ontogeny in prolecanitid ammonoids. Lethaia 46, 1123–1132 (2003).
Ward, P. D., Greenwald, L. & Magnier, Y. The chamber formation cycle in Nautilus macromphalus. Paleobiology 7, 481–493 (1981).
Seilacher, A. Fabricational noise in adaptive morphology. Systematic Zoology 22, 451–465 (1973).
Westermann, G. E. G. Model for origin, function and fabrication of fluted cephalopod septa. Paläontologische Zeitschrift 49, 235–253 (1975).
Seilacher, A. Mechanische Simulation und funktionelle Evolution des Ammoniten-Septums. Paläontologische Zeitschrift 49, 268–286 (1975).
Seilacher, A. Why are nautiloid and ammonite sutures so different? Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen 177, 41–69 (1988).
García-Ruiz, J. M., Checa, A. & Rivas, P. On the origin of ammonite sutures. Paleobiology 16, 349–354 (1990).
García-Ruiz, J. M. & Checa, A. A model for the morphogenesis of ammonoid septal sutures. Geobios 26, 157–162 (1993).
Checa, A. G. & García-Ruiz, J. M. Morphogenesis of the septum in ammonoids. Ammonoid Paleobiology, 253–296 (Plenum Press, New York, 1996).
Checa, A. G. Fabrication and function of ammonite septa—Comment on Lewy. Journal of Paleontology 77, 790–791 (2003).
Hammer, Ø. The development of ammonite septa: an epithelial invagination process controlled by morphogens? Historical Biology 13, 153–171 (1999).
Nagao, T. & Saito, R. Peculiar septal features observed in ammonites of certain lytoceratid genera. Proceedings of the Imperial Academy 10, 357–360 (1934).
Tanabe, K., Mapes, R. H., Sasaki, T. & Landman, N. H. Soft-part anatomy of the siphuncle in Permian prolecanitid ammonoids. Lethaia 33, 83–91 (2000).
Klug, C., Riegraf, W. & Lehmann, J. Soft–part preservation in heteromorph ammonites from the Cenomanian–Turonian Boundary Event (OAE 2) in north–west Germany. Palaeontology 55, 1307–1331, doi: 10.1111/j.1475-4983.2012.01196.x (2012).
Hayakawa, T. & Hirano, H. A revised inoceramid biozonation for the Upper Cretaceous based on high-resolution carbon isotope stratigraphy in northwestern Hokkaido, Japan. Acta Paleontologica Polonica 63, 239–263 (2013).
Nishimura, T., Maeda, H., Tanaka, G. & Ohno, T. Taxonomic Evaluation of Various Morphological Characters in the Late Cretaceous Desmoceratine Polyphyletic Genus “Damesites” from the Yezo Group in Hokkaido and Sakhalin. Paleontological Research 14, 33–55 (2010).
We are grateful to T. Okamoto for advice on the life orientation of Damesites cf. damesi; R. Nakano, A. Omori and K. Tanaka for identifying the habitat of Tritoniopsis elegans and collecting it; and S. Mochizuki for helpful advice. We also thank T. Iruma for collecting the specimen of Damesites cf. damesi. This study was supported by the JST, CREST to S.K.
The authors declare no competing financial interests.
Electronic supplementary material
About this article
Cite this article
Inoue, S., Kondo, S. Suture pattern formation in ammonites and the unknown rear mantle structure. Sci Rep 6, 33689 (2016). https://doi.org/10.1038/srep33689
This article is cited by
Scientific Reports (2021)
Scientific Reports (2021)
Scientific Reports (2020)
Swiss Journal of Palaeontology (2019)