Permian hypercarnivore suggests dental complexity among early amniotes

All vertebrates examined in this study and histologically sampled (Supplementary Table 1) exhibit polyphyodonty and dentine growth lines (Figs. 2–4 and Supplementary Figs. 2–9) that are morphologically consistent with the incremental lines of von Ebner of extant mammalian and crocodilian teeth: alternating opaque zones, line trajectories paralleling the pulp cavity, and widths ranging between 1 and 30 mm¹⁸. All functional teeth were continuously replaced through the development of the replacement tooth, lingual to the functional tooth, resulting in resorption of its base and shedding.

Replacement pattern in Mesenosaurus efremovi

Replacement in the gracile predator Mesenosaurus efremovi from the Richards Spur locality (Fig. 1) appears to occur as a wave in alternating tooth positions, with every other functional tooth in a sequence undergoing replacement during one event. Gaps in the tooth row represent stages in the replacement cycle when the old tooth has been shed, but the replacement tooth has not yet become functional and is not ankylosed to the jawbone. Frequently, these small replacement teeth are lost during fossilization, but in the case of the Dolese Mesenosaurus, preservation is so exquisite that these unattached replacement teeth are preserved, often in place (Fig. 1e). We found that numerous specimens of M. efremovi have tooth families containing a functional tooth and a single replacement tooth lingual to it, but one maxilla (ROMVP 85456) was observed to have a tooth family containing a functional tooth and two successive replacement teeth (Fig. 1c).

The replacement rate found in one tooth family within an M. efremovi dentary was 39 days (ROMVP 85502; Fig. 2), and 34 days for the left maxilla (ROMVP 85443; Supplementary Fig. 2). Replacement rates of three tooth families (mx10, mx12, and mx15) for ROMVP 85457 were estimated to be 46, 36, and 35 days. Thus, the replacement rate for M. efremovi does not appear to vary significantly in one specimen across tooth position, size, or ontogenetic age of tooth.

Replacement pattern in other synapsids

In contrast to the availability of many Mesenosaurus specimens for destructive sampling, other taxa are exceedingly rare, and few specimens were available for destructive analysis. Thus, only a single maxilla of the apex predator Dimetrodon with a replacement tooth in position was available (Fig. 3). The functional tooth had a total of 459 incremental lines, whereas the replacement tooth had a total of 354 lines, resulting in a replacement rate of 105 days. In contrast, the maxillary tooth for the basal sphenacodont Haptodus, was calculated to have functional tooth longevity of approximately 152 days and since neither a replacement tooth nor a resorption pit was present, the minimum replacement rate is 152 days.

Similarly, relatively little material was available for the larger varanopid predator Watongia meieri which is only known from the holotype material, with a resorption pit on one of the two teeth (mx19) on a maxillary fragment, but both teeth were missing the crown apex; thus, only a minimum age could be determined using the incremental line counts. The tooth with the resorption pit was determined to be a minimum of 81 days old, while the adjacent tooth not in the process of being replaced was approximately 68 days old. A second maxillary tooth with a resorption pit at mx18 was determined to be 145 days old. Additionally, one complete tooth with no resorption pit was longitudinally LL sectioned and estimated to be 108 days old.

One maxilla of the small, very rare herbivorous caseid Oromycter was available for destructive sampling (Supplementary Fig. 3). The tooth with a resorption pit in position mx07 was determined to have a total of 506 incremental lines, whereas the tooth without a resorption pit (mx09) had a total of 426 incremental lines. For the mx09 tooth family, the missing replacement tooth was estimated to have 115 incremental lines, resulting in an approximate replacement rate of 391 days.

The left dentary of the large herbivorous caseid Ennatosaurus, known only from five specimens, exhibited two posterior teeth with resorption pits on positions d08 and d07 (Supplementary Fig. 4). Tooth position d08 had a visibly larger and more developed resorption pit, with the functional tooth having a total of 628 incremental lines, whereas d07 had a smaller resorption pit and a total of 567 incremental lines. The missing replacement teeth for both d07 and d08 were estimated to have 136 and 169 incremental lines, resulting in a replacement rate of approximately 431 and 459 days, respectively.

One maxilla of the herbivorous edaphosaurid Edaphosaurus had a resorption pit at tooth position mx09 (Fig. 4) and was estimated to have a total of 506 incremental lines. The adjacent tooth at position mx10 had no resorption pit and was determined to have a total of 429 lines. For the mx09 tooth family, the missing replacement tooth was estimated to have 131 incremental lines, resulting in a replacement rate of 381 days.

Replacement pattern in early and extant reptiles

For the insectivorous parareptile Delorhynchus the functional tooth had a total of 147 incremental lines, while the replacement tooth had 43 lines (Supplementary Fig. 5), resulting in a replacement rate of 104 days. For the other parareptile Colobomycter the premaxillary functional tooth had a total of 157 incremental lines, whereas the replacement tooth had a total of 59 lines, resulting in a replacement rate of 98 days (Supplementary Fig. 6). For the omnivorous eureptile Captorhinus, the functional tooth was 146 days, and the replacement tooth was 69 days, resulting in a replacement rate of approximately 77 days. For the other eureptile, the highly specialized insectivore Opisthodontosaurus, the maximum tooth age for positions d04 to d07 was 151, 155, 206, and 258, respectively (Supplementary Fig. 7). Although no replacement teeth were present, it was possible to use the resorption pit heights to estimate the replacement rates of 182 and 193 days for d06 and d07, respectively. These rates, although different from Captorhinus are not unexpected since this small, close relative of Captorhinus has a very odd, unusual dentition, specialized for feeding on harder shelled invertebrates.

In addition to the above Paleozoic amniotes, two skulls were examined for the extant varanid lizards, Varanus bengalensis and Varanus komodoensis, as well as shed teeth of the latter were also available for study and comparison. The maxillary bone of Varanus bengalensis carried dentition showing six replacement events, but only the mx04 tooth position was sectioned. The functional tooth was determined to have 188 incremental lines, and since a continuous record for the replacement tooth’s incremental lines was not visible, the replacement rate was estimated based on its entire dentine area divided by the functional tooth’s mean line width. The estimated replacement rate for V. bengalensis was approximately 110 days. Unlike M. efremovi, the base of the teeth is characterized by plicidentine, and neither tooth serrations (ziphodonty; Supplementary Fig. 8) nor resorption pits were observed for V. bengalensis.

Similar to Mesenosaurus, Varanus komodoensis, a highly endangered varanid lizard, exhibits ziphodonty on both the mesial and distal tooth surfaces and provides a valuable comparison with the fossil taxon. Two isolated teeth of an adult individual that were in the process of attachment, but not yet ankylosed with the jawbone, were sectioned. The age of the first tooth was determined to have 106 lines, and the second tooth had approximately 135 lines. A third isolated shed tooth (due to resorption from replacement tooth or from the processing of food)²⁹ provided by the Toronto Zoo was determined to have approximately 227 incremental lines. Thus, from the age of initial tooth attachment to the age of shedding, a tooth appears to be functional for an average of 107 days. Additionally, as in Mesenosaurus, the adult skull of V. komodoensis (ROM R7565) showed that each tooth position exhibited multiple replacement teeth for both the dentary and the maxilla, also confirmed by the data from Auffenberg³⁰.

Replacement pattern in a stem amniote

For the representative carnivorous stem amniote Seymouria (Supplementary Fig. 9) the functional tooth was determined to have a maximum of 171 incremental lines, while the missing replacement tooth was estimated to have had approximately 36 lines. Thus, the estimated replacement rate for Seymouria was calculated to be 135 days.

Replacement rate and body mass

There seems to be no significant relationship between replacement rate and body mass (kg) for the taxa examined (Supplementary Fig. 10). Although the largest body sized taxon Ennatosaurus had the longest replacement rate, but the other large species had varying rates, while the smallest taxa (Captorhinus, Delorhynchus, Colobomycter, and Opisthodontosaurus) all have varying replacement rates. Instead, replacement rates appear to be related to feeding behaviour since the herbivorous synapsids all exhibited long replacement rates and great tooth longevities (Fig. 5).

Source: Ecology - nature.com