Cardabiodontidae Siverson, 1999

Siverson, Mikael & Lindgren, Johan, 2005, Late Cretaceous sharks Cretoxyrhina and Cardabiodon from Montana, USA, Acta Palaeontologica Polonica 50 (2), pp. 301-314 : 307

Cardabiodontidae Siverson, 1999

Discussion.—This family comprises an extinct group of large−sized lamniform sharks sharing a unique combination of dental characters (see Siverson 1999: 59). The enlargement of the anteriorly situated lower lateroposterior teeth (relative to the upper and lower anterior teeth) is probably a unique derived feature in cardabiodontids.

Siverson (1999) included Parotodus Cappetta, 1980 in the Cardabiodontidae , and noted the presence of cardabiodontids in the Eocene of Morocco and middle–upper Eocene of Mangyshlak (based on photographs of material in DJW’s collection). The Eocene records seemingly reduced the large gap in the fossil record of cardabiodontids between the middle Cenomanian Cardabiodon ricki Siverson, 1999 from the uppermost Gearle Siltstone in Western Australia and the oldest Parotodus from the lower Oligocene Boom Clay in Belgium, described and illustrated by Leriche (1910). However, in December 1999, one of us (MS) and David J. Ward had the opportunity to directly compare the teeth of the Cenomanian and Turonian Cardabiodon with those of the alleged cardabiodontids from the Eocene of Morocco and Kazakhstan.

The alleged (in Siverson 1999) cardabiodontid from the middle/upper Eocene Schorym Formation in Mangyshlak, was described formally by Kozlov in Zhelezko and Kozlov (1999) as Parotodus mangyshlakensis . It differs though from both the type species of Parotodus , P. benedeni (LeHon, 1871) and Cardabiodon in several aspects. In “ Parotodus ” mangyshlakensis , the lingual neck is broad in anterior teeth but much narrower in lateroposterior files, whereas there is no marked difference in the relative apicobasal width of the neck in neither Cardabiodon nor in Parotodus benedeni . Compare for example the narrow neck on an upper lateroposterior tooth of “ P ”. mangyshlakensis ( Zhelezko and Kozlov 1999: pl. 26: 2a) with the very broad neck in a upper lateroposterior tooth of C. ricki ( Siverson 1999: fig. 8.4A) and P. aff. P. benedeni ( Kemp 1991: pl. 32: M). In profile view, the lower anterior teeth of “ P ”. mangyshlakensis have an extremely prominent lingual protuberance of the root, a feature absent in Cardabiodon and Parotodus benedeni . The sample of teeth from the Rupelian Boom Clay, illustrated by Leriche (1910) as Oxyrhina benedeni , may be a heterogeneous mix of true Parotodus and “ Parotodus ” mangyshlakensis . One of the lower anterior teeth has an extremely large lingual protuberance of the root ( Leriche 1910: pl. 16: 10a) that is in stark contrast to the much lower protuberance in the other lower anterior teeth (e.g., Leriche 1910: pl. 16: 13a). We have not seen this feature on any Miocene–Pliocene P. benedeni , but it characterizes the lower anterior teeth of “ P ”. mangyshlakensis . Two of the lateroposterior teeth illustrated by Leriche (1910: pl. 16: 5, 6) as P. benedeni differ from the other illustrated lateroposteriors in being larger, having more angular lobes of the root, and in having lateral cusplets. They conform extremely well to two of the lateroposteriors of “ P ”. mangyshlakensis illustrated by Zhelezko and Kozlov (1999: pl. 26: 2, 3). Purdy et al. (2001) pointed out that the former two teeth (i.e., those illustrated by Leriche 1910: pl. 16: 5, 6) are not Parotodus , but erroneously referred them to “ Lamna ” rupeliensis Le Hon, 1871 (see Ward and Bonavia 2001).

The undescribed Eocene Morrocan teeth referred to the Cardabiodontidae by Siverson (1999) differ from those of Cardabiodon and Parotodus in having a considerably narrower lingual neck, rendering their assignment to the Cardabiodontidae questionable. Moreover, these specimens (DJW collection) were purchased from commercial dealers and do not have absolutely reliable collecting data. As it is, the gap in the fossil record between Cardabiodon venator sp. nov. from the Fairport Member of the Carlile Shale and the earliest known unquestionable Parotodus from the upper Rupelian Boom Clay amounts to no less than approximately 60 Ma.

The absence of cardabiodontids in the upper Turonian– lower Rupelian strata may be artificial, as few selachian faunas have been recorded from this time interval at higher palaeolatitudes. The appearance of Parotodus in the fossil record (Boom Clay in Belgium) followed the onset of Antarctic glaciation (see e.g., Zachos et al. 2001). If Parotodus was primarily a cool−water shark during the Paleogene, its appearance in the lower Oligocene of western Europe could be interpreted as a temperature induced migration from high palaeolatitudes (which are poorly sampled for selachians) to mid−palaeolatitudes.

The current lack of confirmed occurrences of cardabiodontids in upper Turonian–lower Rupelian strata, does however weaken the rationale for Siverson’s (1999) assignment of Parotodus to the Cardabiodontidae .