Ammonites, extinct relatives of squids, octopuses, and cuttlefishes, are considered to be fossils “par excellence.” Their abundance in the fossil record and their long stratigraphic range, which covers more than 300 million years (from the Devonian up to the end of Cretaceous), make them one of the most studied fossils. Yet, very few remains of the animal living inside the shell are known and their paleobiology and paleoecology is still enigmatic. This lack of knowledge is a limiting factor in our understanding of their evolution and ecology. Many hypotheses have been proposed on ammonite feeding habits, but very few were based on detailed analyses of the available data. One of the very few elements that can be preserved of the animal inside the shell is the buccal mass, and it shows very distinctive features that could be related to trophic specialisation [1]. In order to promote new insights in this field, extremely well preserved ammonites belonging to the genus Baculites from the Upper Cretaceous of South Dakota (US) were scanned at beamline ID19 using propagation phase contrast microtomography (PPC-SRµCT) with a multiscale approach.

Fig. 139: a) Part of the shell of the straight-shelled ammonite Baculites. The green laser indicates the part where the radula was found. Below the green spot, one of the two ribbed halves of the lower jaws that was already exposed. b) Reconstruction of Baculites (A. Lethiers, UPMC).

The ammonites under examination already presented some evidence of mandible preservation (Figure 139), but ESRF phase contrast microtomography revealed the yet unknown upper jaw and in all the specimens an extremely rare structure still embedded in the matrix: the radula. The jaw consists of an upper and lower beak, and is typical of this group of ammonites (the aptychophorans [2]) in that the lower jaw is larger than the upper jaw and consists of two calcified halves separated along a midline. The radula (Figure 140) is found between the jaws in all the specimens. It is a toothed ribbon used by mollusks to feed and transport the food toward the oesophagus. The structure was preserved three dimensionally revealing the organisation of the teeth on the ribbon. Several teeth were still connected as they were in life and high resolution scans allowed us to see previously unknown details. The size of the radula was approximately 6 mm wide and 7 mm long and the tallest tooth cusp is 2 mm high. Teeth are very slender and show multiple cusps. The shape varies from saber to comb-like and differs from the tooth morphologies found in modern cephalopods.


Fig. 140: a) Reconstruction of the radula of Baculites, front view, b) Lateral view of the upper jaw, in grey, apex on the left, with the radular teeth preserved inside, c) One of the three pieces of crustacean (blue) found flanking the radular teeth (yellow) with a small gastropod larval shell (pink).

The detailed anatomical study of the buccal mass, as well as comparative studies with recent cephalopods, led to the hypothesis that ammonites with this kind of feeding apparatus (i.e., very large lower jaw, small and blunt upper jaw and delicate radula) fed on small organisms belonging to plankton. This hypothesis was reinforced by the discovery of a tiny snail and three small tiny crustaceans in one specimen, with one of the crustaceans having been cut into two parts (Figure 140c). Because these planktonic fossils are not found anywhere else on the specimen, the hypothesis was advanced that the specimen died while eating its last meal. Among Mesozoic ammonites, a large number of species share the same buccal mass morphology, therefore implying that all of these animals shared a similar position in the food web. One of the implications of this discovery is that the radiation of aptychophoran ammonites might be associated with the radiation of plankton during the Early Jurassic. In addition, plankton were severely hit at the Cretaceous-Tertiary boundary, and the loss of their food source probably contributed to the extinction of ammonites.

After this research, it is now clear that PPC-SRµCT is a powerful tool for the analysis of ammonite remains. This is a starting point for new developments on ammonite paleontological studies. In this research, we came close to realising the old dream of many paleontologists: knowing what an ammonite really looked like.


Principal publication and authors

I. Kruta (a,b), N. Landman (c), I. Rouget (d), F. Cecca (d) and P. Tafforeau (e), Science 331, 70-72 (2011).

(a) UMR-CNRS 7207, Département Histoire de la Terre, Muséum National d’Histoire Naturelle, Paris (France)

(b) Present address: Yale Department of Geology and Geophysics, New Haven (USA)

(c) Division of Paleontology (Invertebrates), American Museum of Natural History, New York (USA)

(d) UMR-CNRS 7207, Université Pierre et Marie Curie - Paris 6 (France)

(e) ESRF



[1] G.E.G. Westermann, in Topics in Geobiology, 607–707 (1996).

[2] T. Engeser and H. Keupp, Lethaia 24, 79–96 (2002).