First Report of Enamel Hypoplasia in Extinct Tragulids: A Marker Bearing on Habitat Change

Enamel hypoplasia (EH) is a type of tooth anomaly characterized by thinning of enamel and disruption of crystallite deposition. Ameloblasts (enamel forming cells) are highly sensitive so their formation can be depleted by the stresses faced by an animal during its tooth development. The enamel is the hardest tissue of the body with no remolding in it so marks of EH once formed on tooth enamel will persist forever even if the affected tooth had been fossilized (Goodman and Rose, 1990; Franz-Odendaal et al., 2004). Keeping in view all these facts EH has been used by different researchers as a credible stress marker for different extinct mammalian groups i.e. Pleistocene hominids (Molnar and Molnar, 1985), Pleistocene Neanderthals (Trinkaus, 2018), early Miocene primates (Lukacs, 2001), early Pliocene grazers and browsers (Franz-Odendaal et al., 2003), Pliocene giraffids (Franz-Odendaal et al., 2004), Miocene rhinocerotids (Mead, 1999; Böhmer and Rössner, 2018), Siwalik rhinocerotids (Roohi et al., 2015), Siwalik giraffids (Ahmad et al., 2018). There is no report about occurrence of EH in any extinct tragulid species from the Neogene or Quaternary deposits globally even though these are ecologically significant ungulates. This study provides the baseline data on occurrence of enamel hypoplasia in extinct tragulids.

There are three broader categories of EH according to FDI (Federation Dentaire International, 1982) that are Pits, Grooves and Areas missing enamel. Areas missing enamel can further be categorized into semicircular enamel hypoplasia (SEH) and linear enamel hypoplasia (LEH). There can be different etiologies for these types of EH i.e. birth trauma, metabolic and nutritional disorders, infections, exposures to toxic chemicals (Seow, 1991), rickets (Nikiforuk and Fraser, 1981), low birth weight (Slayton et al., 2001), poor health status of the mother (Armelagos et al., 2009), nutritional conditions of an area (El Najjar et al., 1978; Ogilvie et al., 1989), general physiological stress (Guatelli-Steinberg et al., 2004), weaning stress (Mead, 1999), post weaning stress (Moggi-Cecchi et al., 1994), nutritional and/or environmental stress (Franz-Odendaal et al., 2004). In the types of enamel hypoplasia, LEH is the most common type and its marks can macroscopically be observed on the enamel of living as well as fossilized dental remains. The physiological, nutritional and/or environmental stresses are probable precursors for LEH (Franz-Odendaal et al., 2004). According to Franz-Odendaal et al. (2004), LEH can provide a unique perspective into vegetational and environmental conditions present during growing year of an extinct animal’s life. This shows that LEH can be a credible proxy for ecological studies of extinct tragulids so the aim of this study is to assess habitat stability for the Neogene Siwalik tragulids by using LEH as a proxy. The dental remains used for this analysis were recovered from the Siwalik outcrops of Pakistan.

The multi-storied interbedded layers of sandstone characterize the Siwalik group of Potwar Plateau. The tragulid material used in this LEH analysis has been recovered from the Siwalik deposits having a chronological range of ca. 13.5-4.0 Ma (Fig. 1 and 2). The Chinji Formation marks the middle Miocene of the Siwaliks of Pakistan. The boundary between Chinji and Nagri formations is identified at the upper part of Chron 5r, dated around 11 Ma (Barndt et al., 1978; updated in Barry et al., 2013). When we move from North to South in the Siwaliks, we find that the Nagri lithofacies at the lower boundary are time-transgressive. The upper boundary of the Nagri Formation is time-transgressive and gradational with the overlying Dhok Pathan Formation. The upper boundary of the Dhok Pathan Formation may be assigned an age of 5.1 Myr on the basis of stratotypes studied by Opdyke et al. (1979), Johnson et al. (1982) and Barry et al. (2002). Pilgrim (1913) identified the Dhok Pathan type locality on the basis of lithological characters and associated the fauna of the locality with the Dhok Pathan Formation. Tauxe and Opdyke (1982) identified that this section dates to around 8 Ma, but at some other study sections Dhok Pathan rocks (e.g. in Bhandar section) may be as younger as 5.8-5.3 Ma. These estimations are based on the main stratotype and certain reference sections as described by Opdyke et al. (1979); Tauxe and Opdyke (1982) and Johnson et al. (1982).

The new tragulid remains reported in this study were collected from the fossiliferous localities of different strata of the Siwaliks of Pakistan (Figs. 1 and 2). The tragulid remains already present at Dr. Abu Bakr Fossil Display and Research Center, University of the Punjab, Lahore, Pakistan are also included in the study. The outcrops from where this already available tragulid material is recovered are given in Figures 1 and 2. The specimen under study were thoroughly washed and cleaned for linear enamel hypoplasia analysis. The species identification of the included tragulid remains is based on the dental morphometric features of the Siwalik tragulids given in Farooq et al. (2007a, 2007b, 2007c, 2008).

Table I. Occurrence of enamel hypoplasia in the studied samples of Dorcabune anthracotherioides from the Siwaliks of Pakistan.

In the six reported valid Siwalik tragulid species the three species; Dorcatherium minimus, Dorcatherium nagrii and Dorcabune nagrii that had the chronological ages of 14.2-11.2 Ma, 14.2-10.0 Ma and 11.2-3.3 Ma in the Siwaliks respectively (West, 1980; Gaur, 1992; Khan et al., 2012) were excluded from the study. In order to extract a more reliable and precise LEH based information about the habitat preference of Neogene tragulids only three tragulid species named as Dorcabune anthracotherioides, Dorcatherium majus and Dorcatherium minus (chronological age 14.2-3.3 Ma according to Khan et al., 2012) that had their existence throughout the middle Miocene-early Pliocene epoch in the Siwaliks; were selected for the study.

Table IV. Statistical results of Chi square test for difference in occurrence of enamel hypoplasia between the middle Miocene and the late Miocene-early Pliocene Siwalik tragulids.

*Numerator is indicating the molars having LEH and denominator is indicating the total analyzed molars, **The chronological ages given in this table are based on the overall ages of the middle Miocene and the late Miocene-early Pliocene Siwalik fossiliferous localities included in this study and are given in Fig. 1.

Nearly 102 fossils of the three selected Siwalik tragulid species were scrutinized for LEH analysis based on criteria given by Mead (1999) and Franz-Odendaal et al. (2004). The highly fragmented, weathered, cemented and worn down to the cervix teeth were excluded and 72 well preserved tragulid molars were selected for LEH analysis. The numbers of studied molars are given in Table IV. The development of molars in mammals begins prenatally and the first molars are affected by the quality of mother’s milk. First molars may show weaning EH. Nutrition during later development depends on foraging and has a direct interaction with the vegetation and other ecological conditions of an ecosystem. Therefore, molar hypoplasia, especially second and third molars, may reflect nutritional stress. Keeping in view this fact, only molars were selected for this ecological study of extinct tragulids.

The tragulid remains were observed carefully for LEH analysis under 10 X hand lens. If a mark of LEH observed; its presence was confirmed by observing the same mark at different orientations of the specimen. Further the existence of LEH mark/s was checked under both artificial light of 50-watts lamp as well as sunlight. There were two to three repetitions of this procedure to increase the reliability of the results. The measurements for height of LEH on the tooth crown from the root crown junction were taken in mm by using student manual Vernier Caliper having a least count of 0.1 mm. The classification and methodology for current LEH analysis is based on the work of Franz-Odendaal et al. (2004) and Roohi et al. (2015).

The photographic results of EH given in different publications i.e. Mead (1999), Franz-Odendaal et al. (2004), Lacruz et al. (2005), Teegen and Kyselý (2016) and Lyman (2018) were used to differentiate the observed marks of LEH on tragulid remains from the other possible observable irregularities on the surface of a tooth enamel.

It is an observation based analysis so to ensure the authentication of the results the LEH analysis was carried out by two independent raters. The Cohen’s (1960) kappa statistical test was applied to access the level of agreement between the two raters (Table V). The Cohen’s kappa test is available on line, https://idostatistics.com/cohen-kappa-free-calculator.

The inference for trend of ecological stress faced by the Siwalik tragulids was drawn by applying Chi square test on LEH results with the help of IBM SPSS Statistics-20 software. The level of significance was taken as 0.05. The Chi square test is selected in this study as it is already in use for interpretation of EH results in different publications i.e. Goodman et al. (1980), Guatelli-Steinberg and Skinner (2000), Slayton et al. (2001), but these references are for living or archeological specimens. The current statistical analysis of LEH in extinct tragulids is an innovative approach in the field of paleontology. The validity of the statistical results was checked online using the link http://turner.faculty.swau.edu/mathematics/math241/materials/contablecalc/

The results illustrate that like many other mammals the extinct tragulids had suffered with EH during their life histories (Fig. 3). In the analyzed dental remains of tragulids 13% of the specimens of the middle Miocene epoch have occurrence of LEH whereas for the tragulid fossils of late Miocene-early Pliocene age this value is 48% (Fig. 4, Table IV). The results for occurrence of LEH in the included tragulid species are given in Tables I, II and III. The comparative trend for LEH between the Siwalik tragulids of middle Miocene and late Miocene-early Pliocene epoch are presented in Table IV.

The current LEH analysis indicates that overall 26% Siwalik tragulid molars out of the total analyzed material has occurrence of EH (Table IV). The result shows that the Siwalik tragulids of the middle Miocene habitat had significantly lower occurrence of EH (p<0.05) as compared to tragulids that were present in the late Miocene-early Pliocene habitats (Table IV). The limitations of food resources to some extent and interspecific competition might be the stressors responsible for observed 13% occurrence of EH in tragulid molars of the middle Miocene time period (Fig. 4, Table IV), even then the middle Miocene habitat was comparatively much favorable for the Siwalik tragulids (based on their lower EH occurrence) as compared to the tragulids of the late Miocene-early Pliocene habitats as the observed occurrence of EH in tragulid molars of this time period is 48% (Fig. 4, Table IV).

The middle Miocene Siwalik palaeoclimate is characterized by warm and humid conditions with dominance of dense forests (Heissig, 2003). Such conditions supported the browsing habit as deduced from the dental brachydonty of the faunal elements of middle Miocene time spans. The Siwalik tragulids had brachydont dentition that indicates a browsing nature for these animals thus one main reason of tragulid success in middle Miocene habitat might be the vegetation pattern of the middle Miocene Siwaliks. As the climate became less humid and seasonality played its role in the late Miocene with emergence of C4 grasslands, the grasslands started expanding at the expense of forests and woodlands which ultimately stressed the fauna of latest Miocene to early Pliocene time spans, resulting in the greatest faunal turnover in the latest Miocene (Barry et al., 1982; Cerling et al., 1993; Barry et al., 2002; Nelson, 2005; Bibi, 2007; Flynn et al., 2016). Possibly due to late Miocene-early Pliocene arid climatic conditions under which C4 grasslands expanded, the tragulids of these habitats had faced comparatively high degree of stress due to climatic and dietary changes.

The global warming event of the early middle Miocene period caused an increase in temperature (You et al., 2009) on the other hand late Miocene-Pliocene was the time of lower temperature and seasonal precipitation. Quade et al. (1992) reported cooling temperature in the Siwaliks during Plio-Pleistocene epoch. So the warm climatic conditions might be another possible reason for the preference of middle Miocene habitat by the Siwalik tragulids. Open vegetation ecosystem increased after 7 Ma (Meijaard and Groves, 2004), which increased predatory stress for tragulids. This might add up another reason for preference of the middle Miocene on the late Miocene-early Pliocene habitats by the Siwalik tragulids.

Along with the above given probable reasons population density of middle Miocene tragulids is another evidence for compatibility of the Siwalik tragulids to the middle Miocene habitat. In the Siwaliks 30-60% of the small ruminant species were tragulids before 9 Ma but this value declined to only 04% at the end of late Miocene (Barry, 2013).

The current LEH results are unable to trace out the exact reason of the ecological preferences for extinct tragulids but these results clearly indicate that for the Siwalik tragulids the middle Miocene ecological conditions were more favorable as compared to late Miocene-early Pliocene. Rössner (2007) has described that the living tragulids also prefer to live in tropical ecosystem. In our view the vegetational pattern might be the key precursor for the middle Miocene habitat preference by the Siwalik tragulids.

The current study is the first ever report on EH in extinct tragulids so it adds up new information in the literature of palaeontology and paleopathology. The results indicate that the middle Miocene habitats were preferred for the extinct tragulids but the exact reason for this preference could not be traced out at the moment. Further studies by using other ecological proxies may resolve this mystery. The study can provide important data for conservation of living tragulids. This study can be a road map for future research on mammals as no significant data is available on EH occurrence in living tragulids nor are data known on the LEH based ecological comparison of extinct tragulids with other members of their community i.e., bovids.

The authors of this article are very much thankful to John Luckas (Professor Emeritus, University of Oregon, USA) and R. Lee Lyman (Professor, University of Missouri, Columbia, Missouri) whose comments on earlier version of this article proved to be highly significant in improving this article. Furthermore, this study is also selected after peer review by the experts and editors for an oral presentation in 2nd Conference of the Arabian Journal of Geosciences, Springer 2019. The authors are also very much thankful to the anonymous reviewers and editors of the conference for their remarkable suggestions. The financial support for this study has been provided by HEC under the indigenous scholarship for Amtur Rafeh, PIN NO: 112-33857-2BM1-187.

We declare no conflict of interest in this study.