“Proboscidean Agent” of Some Tertiary Megafaunal Extinctions

“Proboscidean agent” of some Tertiary megafaunal extinctions

P.V. Putshkov Schmalhausen Institute of Zoology, Kiev, Ukraine - [email protected]

SUMMARY: Proboscideans constantly and strongly modified Tertiary ecosystems. Their impact was mostly beneficial for herbivores, but not for the other elephant-sized mammals. So, proboscideans superseded arsinoitheres in Africa (Late Oligocene) and indricotheres in Asia (Early Miocene). Indricotheres efficiently browsed only at considerable height of standing trees. Gomphotheres, like modern elephants, were mixed browsers/grazers able to take food from different storeys owing to their long trunked snout and to tree-felling habits. They destroyed many trees. Still more trees were converted to a state inappropriate for the indricothere feeding. The Plio-Pleistocene contact of proboscideans with giant sloths did not result in either group extinction. Both were equally protected from predators and equally adapted for the efficient feeding on tree vegetation of different type.

1. INTRODUCTION 3. PROBOSCIDEANS, VEGETATION AND OTHER HERBIVORES The outstanding ecosystem impact of modern elephants and other pachyderms helped to Both modern elephants despite their differ- unravel the puzzle of Pleistocene ecosystem ences in dentition are mixed browsers/grazers functioning and that of the Wurm mass that feed in any storey using trunk, tusks and extinctions (e.g. Owen-Smith 1988; Putshkov felling trees (e.g. Kingdon 1979). In most cases 1989, 1994, 1997; Schule 1990). This impact the extinct proboscideans were ecologically may help to understand certain previous flexible and basically polyphagous too. Their events too. trophic and landscape preferences seem to be formed more by the competition and by the 2. PROBOSCIDEANS AND NON-HUMAN accessibility of one or another kind of food PREDATORS than by morphological differences (e.g. Putshkov 1989, 1997; Lambert 1992; Putshkov Various proboscideans evolved competing & Kulczicki 1995). with each other under the pressure of formida- Already early Miocene gomphotheres, true ble creodonts and, later, of large carnivorans mastodonts, and dinotheres ranging from rhino (mainly saber-toothed cats), killing calves, to Asian elephant size increased drastically the subadult and old individuals. In response, pro- nutrient recycling rates and vegetation cover boscideans not only increased in size, but mosaicity. They destroyed many trees both by became very ‘clever’ beasts too (Gabunia felling and by barking them. Also many dam- 1969). They withstood foes efficiently and their aged trees perished from diseases, fires, wood- population density stabilized just below their boring insects. Thus giants created clearings niche capacity level. Hence, proboscideans and then maintained them by the repeated could constantly exert a potent ecosystem- browsing over regrowth. Gomphotheres shaved shaping impact. the bark with their pointed upper and pointed or

133 The World of Elephants - International Congress, Rome 2001

chisel-like lower tusks aided with trunk possible that gomphotheres with a shoulder (Lambert 1992; Putshkov & Kulczicki 1995). height of 1.7-2.5 m and a weight of 1.5-3.5 tons The long trunked snout permitted them to outcompeted the indricotheres with a shoulder browse efficiently on low, medium and quite height of 4-6 m, head height of 6-9 m and a tall trees (these were often uprooted), bushes, weight of 3-15 tons? The unravel of the puzzle herbs and grasses (Lambert 1992; Putshkov seems to be just there. & Kulczicki 1995). Due to such mixed ‘all- Indricotheres ate twigs and stripped the bark storeys’ feeding habits these strong and ‘clever’ upwards mostly at height near 4.0-8.5 m with gregarious creatures not only lived in savannas, their strong lower incisors (Putshkov & but created them too in seasonally dry regions, Kulczicki 1995). Earlier Indricotherium also like modern elephants do. used large upper incisors for the bending of The still large later proboscideans exerted high branches. Later Paraceratherium had even more strong impact on Old and New small or no upper incisors while the lower ones World ecosystems. This impact usually was became even larger. Gromova (1959) supposed beneficial for smaller megafauna (up to the that the former was more leaf-eating and less modern black rhino size) (e.g. Owen-Smith bark-eating than the latter. Yet, these changes 1988; Putshkov 1989, 1997, Schule 1990) but reflect perhaps the giants-driven changes of not for the elephant-sized mammals. Only the tree-form rather than trophic changes last aspect is considered here. (Putshkov & Kulczicki 1995). Due to the upward barking with lower incisors some trees 4. INDRICOTHERE ‘DEMISE’: DROUGHT OR perished, but other died back only above the GOMPHOTHERES ‘ACCESSION’? leisured zone. Beneath it, trees reacted by abundant trunk-shooting. Giants, attracted by The sudden final Early Miocene extinction of the shoots, browsed upon them repeatedly. Due impressive indricotheres (baluchitheres) to the compensatory growth the dense ‘brush’ (Indricotheriidae) has been ascribed to the fact of shoots appeared. Some trees, whose terminal that they were simply too big. Hence, they buds were repeatedly browsed could never could not exist for a long time due to their slow reach their normal height. They formed dense generation turnover (Janis & Carrano 1991). distorted crowns at the height convenient for Yet, indricotheres were taller but not heavier the indricothere feeding. Such distorted trees than large elephants (e.g. Gromova 1959) and made up the considerable part of all trees in the thrived for more than 15 million years (e.g. ‘indricothere’ savannas and parklands. The Savage & Russell 1983). beasts found enough of leaved twigs at 3.5-8 m Other version blames the increase in aridity height (Putshkov & Kulczicki 1995). It became caused by rain shadow effect of the uplift of unnecessary to bend the branches of the mountain ranges from Himalayas to Dinarids remaining higher trees that were, however, (e.g. Gromova 1959; Gabunia 1969). However, more convenient for the upward barking than open steppes have occupied most of the Inner trees distorted as yet. These indricothere-driven Eurasia more than 10 million years after the changes of tree-form became the selective indricothere extinction. At the time of extinc- agent causing the mentioned changes in the tion mountains were low and caused only mod- indricothere dentition. Huge indricotheres erate-scaled increase in aridity. As a result, the damaged trees more than the modern giraffes, indricothere life zone (parklands and savannas but far less than the elephants do. Unlike pro- of warm-temperate, subtropical and tropical boscideans, indricotheres could not inhabit types) expanded while closed forests contract- both temperate and tropical forests. Like ed in range (e.g. Yassamanov 1985). giraffes, they remained exclusively parkland Indricotheres disappeared from the geologi- and savanna-dwellers (Putshkov & Kulczicki cal record soon after the gomphothere invasion 1995). from Africa (e.g. Savage & Russell 1983). Is it Gomphotheres came to Eurasia through the 134 ‘Proboscidean agent’ of some Tertiary megafaunal extinctions new-established land bridge. They felled trees constantly remained ‘too numerous’ and went or barked them upwards and downwards at the on converting many trees to the state inappro- far lower level than indricotheres did. Neither priate for the indricothere feeding. The ‘ill-fed’ plants nor carnivores of Asian ecosystems were indricotheres brought feeble calves and often adapted to gomphotheres. It is probable that at could not protect them from predators. The first the latter increased in number explosively small residual indricothere populations were (as recently elephants in Tsavo Park: e.g. Beard doomed to the extinction by stochastic events, 1989) and occupied the indricothere range rap- although they could linger in some regions till idly not only geologically but ecologically more proboscidean species appeared due to speaking as well (Putshkov & Kulczicki 1995). migration and speciation processes (Putshkov They soon destroyed many trees (both undis- & Kulczicki 1995). torted and distorted by indricotheres) in savannas, parklands and riparian forests or converted 5. PROBOSCIDEANS AND OTHER ELEPHANT SIZED them into the non-climax state of low regrowth, MAMMALS bushes, scrub, dense stool-shoots, etc. Such changes were adverse neither for gomphotheres Probably the small-brained arsinoitheres of with their mixed browsing-grazing ‘all-storeys’ African Oligocene were superseded by ‘clever’ feeding nor for the most of smaller herbivores proboscideans that defended their calves (Putshkov & Kulczicki 1995). For indri- against creodonts more efficiently. On the other cotheres the changes were disastrous. Similar hand, the gomphotheres (and partially to modern giraffes that can not cover their ener- mastodonts and mammoths as well) coexisted gy losses while being forced to browse upon with elephant sized sloths (eremotheres and vegetation lower than 2 m, giraffe-like indri- megatheres) in America during late Pliocene cotheres could not find enough food in and Quaternary times. These sloths were diffi- tree-stands transformed by gomphotheres. cult prey for any predator due to their mon- Indricotheres as hindgut fermenters needed strous claws combined with exceptional more food than the foregut fermenting and strength and thick skin. Also, unlike indri- much smaller giraffes need. Probably, only cotheres, they could live in both savannas and small indricothere herds near forest edges did heavily wooded regions and feed efficiently on not starve to death (Putshkov & Kulczicki trees broken or fallen by proboscideans or by 1995). sloths themselves. After several generations gomphothere num- bers should have arrived to the dynamic equi- 6. CONCLUSIONS librium with their environment. The interaction between their natality rates and mortality fac- Since late Oligocene proboscideans were tors may have resulted in fluctuations of gom- both giant and ‘clever’ creatures. Predators and phothere density just below the level permitted other adverse agents could only confine their by their trophic resources. Besides other mor- population density just below their niche tality agents (diseases, periodic attrition, etc.), capacity level. Hence, they strongly modified predators that previously killed indricothere ecosystems. These changes were usually useful calves had to switch to subadult gomphotheres. for smaller megafauna. For the other elephant- Certain formidable foes, as probably sized giants they may be harmful, especially if Hyainailouros, may have arrived from Africa these giants were less able to withstand preda- following on their proboscidean prey ‘heels’. tors. Therefore, proboscideans superseded arsi- Yet, indricothere populations could not recover. noitheres in Africa (Late Oligocene) and indri- Gomphotheres with their far more developed cotheres in Asia (Early Miocene). Indricotheres brain (Gabunia 1969) and more flexible feed- efficiently browsed only on trees that were tall ing strategy withstood the hostile factors more enough. Trees distorted by their repeated efficiently than indricotheres did. Hence, they browsing had abundant shoots and/or ramifica- 135 The World of Elephants - International Congress, Rome 2001

tions at the height of 3.5-8.0 m. Gomphotheres shovel-tusked gomphotheres: evidence were mixed browsers/grazers taking food from from tusk wear patterns. Paleobiology 18: different storeys owing to their long trunked 133-147. snout and to tree-felling habits. They destroyed Owen-Smith, N.R., 1988. Megaherbivores. many trees and converted still more trees to the The influence of very large body size on state inappropriate for the indricothere feeding. ecology. London; Cambridge Univ. Press. The Plio-Pleistocene contact of proboscideans Putshkov, P.V. 1989. Uncompensated extinc- with giant sloths in Americas did not result in tions in Pleistocene: the supposed mecha- either group extinction. Both were equally well nism of the crisis. Kiev: Inst. Zool. protected from predators and equally well Putshkov, P.V. 1994. Uncompensated Wurm ‘equipped’ for the efficient feeding on tree veg- extinctions. 6. Final communication: the etation of different type; both were mighty ‘tree crisis outside the Old World. Vestnik fellers’. Zoologii 1994 (2): 65-73. Putshkov, P.V. 1997. Were the mammoths killed 7. REFERENCES by the warming? (Testing of the climatic versions of Wurm extinctions). Kyiv: Beard, P.H. 1989. La fin d’un Monde. Le Inst.Zool. [Suppl. n.4 to Vestnik Zoologii] dernier echo du Paradis. Paris: Chene Putshkov, P.V. & Kulczicki, A.H. 1995. The Gabunia, L.K. 1969. Extinctions of ancient rep- Early Miocene Drama: Mastodonts against tiles and mammals. Tbilisi: Metsniereba. Indricotheres. Vestnik Zoologii 1995 (6): Gromova, V.I. 1959. Giant rhinoceroses 54-62. (Indricotheriidae). Moscow: Ac. Sci. Savage, D.E. & Russell, D.E. 1983. USSR. Mammalian palaeofaunas of the world. Janis, C.M. & Carrano, M. 1991. Scaling of London, Addison-Wesley. reproductive turnover in archosaurs and Schule, W. 1990. Landscapes and climate in mammals: why are large terrestrial mam- prehistory: interaction of wildlife, man and mals so rare? Ann. Zool. Fenn. 28: 201-206. fire. In J.G. Goldammer (ed.), Fire in the Kingdon, J. 1979. East African mammals. An tropical biota: 273-318. Berlin, Springer atlas of evolution in Africa. (Large mam- Yassamanov, N.Ya. 1985. Ancient climates. mals). London, Acad. Press. Leningrad: Hydrometeoizdat. Lambert, D. 1992. The feeding habits of the

136