INVESTIGADORES
PREVOSTI Francisco Juan
artículos
Título:
Reply to “Comment on “Isotopic insight on paleodiet of …” by Bocherens et al. (Gondwana Research, 48(1), 7–14)”
Autor/es:
BOCHERENS, HERVÉ; COTTE, MARTIN; BONINI, RICARDO A.; STRACCIA, PABLO; SCIAN, DANIEL; SOIBELZON, LEOPOLDO; PREVOSTI, FRANCISCO J.
Revista:
GONDWANA RESEARCH
Editorial:
ELSEVIER SCIENCE BV
Referencias:
Año: 2018 vol. 58 p. 243 - 245
ISSN:
1342-937X
Resumen:
Fariña and Varela (2018) consider that the observed pattern of differencebetween δ13C in the carbonate fraction of bioapatite and bonecollagen (Δδ13Ccarb-coll) presented for extinct giant Xenarthrans, especiallythe giant ground sloth Megatherium, do not demonstrate an herbivorousdiet, as concluded by Bocherens et al. (2017). The mainargument is that the influence of body mass on Δδ13Ccarb-coll valueswould not have been considered in Bocherens et al. (2017). However,contrary to the assertion by Fariña and Varela (2018), the possible influenceof bodymass on Δδ13Ccarb-collwas actually considered in thiswork.Indeed, small and large herbivorous as well as small and large carnivorousmammals were considered separately among the modern mammalsused as reference to compare the fossil specimens (Fig. 2 andTable 2 in Bocherens et al., 2017). Small and large carnivores on theone hand, and small and large herbivores on the other hand, presentedstatistically significant differences. Despite these differences linked tosize, small herbivores and small carnivores presented statistically significantdifferences (p b 0.0001), as did large carnivores and large herbivores(p = 0.008) (Table 2 in Bocherens et al., 2017). Therefore, thepossibility to infer dietary preferences from Δδ13Ccarb-coll in each sizeclasswas checked onmodern specimenswith known diets before inferringthe paleodiet of extinct Xenarthrans using the same approach.When considering a continuous variation in body mass as done byFariña and Varela (2018, Fig. 1B), it is noticeable that the correlationfor extant mammals, though statistically significant, has a r2 value ofonly 0.17 (Fariña and Varela, 2018), meaning that only 17% of the correlationis explained by body mass, and 83% by other factors. Moreover,the lack of any carnivorous species with a body mass higher than300 kg leads to a very large confidence interval that overlaps with thatof herbivores in the range of Log Body Mass above 3 (correspondingto bodymasses above 1000 kg). The fact that one value for Megatheriumfalls at the margin of this verywide confidence interval for carnivores isprobably a statistical artefact. To test if this is the case, we included intothe same type of graph the data from the late Pleistocene of Beringia(from Clementz et al., 2009) and combined them with data fromBocherens et al. (2017) for large modern mammals with different trophicecologies (herbivores, insectivores, omnivores and carnivores)and the data for the late Pleistocene Pampean mammals from thesame publication. The body masses were taken from Smith et al.(2003). The graph shows that all Δδ13Ccarb-coll values of carnivores, insectivoresand omnivores in the selected bodymass have average valueslower than those of herbivores in the same body mass range (Fig. 1).There is a slight overlap when considering the standard-deviations inGondwana Research 58 (2018) 243?245the range of Δδ13Ccarb-coll values between 5 and 6, but the overwhelmingmajority of large herbivores exhibit Δδ13Ccarb-coll values well abovethose of large carnivores (Fig. 1). As noted before, there is no large carnivoreor omnivore in a similar bodymass range as Megatherium, but interestingly,the woolly mammoth from Beringia exhibits an average aΔδ13Ccarb-coll value even lower than that of Megatherium, and there isno doubt that this proboscidean was purely herbivore (e.g. Bocherens,2003; Schwartz-Narbonne et al., 2015; Naito et al., 2016). Also,Macrauchenia, a pure herbivore with a body weight reaching 1 ton,has a Δδ13Ccarb-coll value similar to that of Megatherium, so not all verylarge late Pleistocene South American extinct herbivorous mammalshad Δδ13Ccarb-coll values above those of Megatherium. Thewhole rangeof Δδ13Ccarb-coll values should be considered in the discussion onMegatherium diet, not only the highest ones.Another argument used by Fariña and Varela (2018) to support acarnivorous diet for Megatherium is the similar range of δ15N values ofbone collagen between this ground sloth and the sabretooth catSmilodon as published in Bocherens et al. (2016) for the same fossilmaterial.However, in the same paper, Macrauchenia also falls in the samerange of δ15N values as Smilodon, therefore a carnivorous diet shouldbe also considered for this extinct species if we follow the reasoning ofFariña and Varela (2018). Such an option is absolutely not supportedby morphology and mechanical analysis (e.g. Varela and Fariña, 2015).This is not the only case of an herbivore with δ15N values of bone collagenoverlapping with those of some carnivores in the same context. Indeed,this is the case of the woolly mammoth in its distribution areafrom western Europe to north-western North America (Bocherens,2015). Such a pattern has also been documented for mammals frommodern ecosystems (e.g. Davie et al., 2014). The reasons for such a patternare a combination of dietary preference of these herbivores forplants with high δ15N values (such as graminoids and forbs) and theprey choice of the predator focused on herbivorous prey with ratherlow δ15N values, caused by the consumption of plants such as shrubsand tree leaves or legumes (e.g. Yeakel et al., 2009; Bocherens et al.,2015).In conclusion, we consider that the carbon and nitrogen isotopiccomposition of the collagen and carbonate fraction of the bones of theMegatherium specimens analyzed so far do not support the hypothesisthat they consumed animal resources.Whenmore isotopic data becomeavailable for Megatherium and coeval fauna, it will be possible to evaluatethe variability of the dietary ecology of this fascinating extinctspecies.