The Miocene Evolutionary diversification

At the dawn of the Miocene, which spanned 24-5 My, a warming trend set in again, this time accompanied by an even drier climate. Shifts in the continental plates closed the Tethys Sea (creating the Mediterranean), while a passage was opened between Antarctica and South America. These changes, combined with other tectonic events, such as the uplifts of mountains (including the Himalayas, the Andes, and the Cordilleras), brought about changes in oceanic and atmospheric circulation to set up new global climatic patterns. The climatic contrasts of heat and cold increased between higher and lower latitudes. The warming in the middle latitudes seems to have caused some expansion of tropical and subtropical forests here, but the arid trend also encouraged thorn scrub and grasslands. Importantly, the new land bridges formed by lower sea levels and continental tectonics paved the way for vast migrations across continents by the new radiation of mammals.

The Miocene witnessed the greatest proliferation of the proboscidean tree, giving rise to the mammutids and the gomphotheres, the latter leading to the progenitor of the true elephants before the close of the epoch. Before a closer look at the evolutionary adaptations and radiations of these two groups, another bizarre proboscidean must be introduced—Deinotherium (the "terrible beast"), whose fossils first appear in middle Miocene strata. The deinotheres differentiated earlier, possibly during the late Eocene, and enjoyed phenomenal evolutionary success with little change over a period spanning the Oligocene, Miocene, Pliocene, and early Pleistocene. Deinothere fossils were first discovered in Europe, where they were wrongly classified with rhinos and tapirs. Later, their remains were found in North and East Africa and in Asia (as far east as India), but none have been recovered in the Americas. One genus, Deinotherium, alone lived through 20 My, a testimony to its great adaptability to a changing world. Over this time, it remained remarkably stable in form, save for an increase in body size (Deinotherium giganteum reached 4 m in height, much larger than any modern elephant).

The main peculiarity of Deinotherium was the absence of tusks in the upper jaw, but it possessed a pair of inward-curving tusks from a down-recurving lower jaw, not seen in any other proboscidean group. (The story goes that the Rev. Dr. William Buckland, Dean of Westminster, hypothesized that the deinothere was an aquatic creature that used its downward-pointing tusks to anchor itself to the riverbank while sleeping.) The tusks were possibly used for digging up plant roots and vegetation. At the same time, their role as merely a sexual display cannot be discounted (see chapter 3).

The deinotheres had a mixture of general and specialized proboscidean characters. The first molar was trilophodont, while the other two molars were bilophodont. Cheek teeth were replaced in a vertical fashion as in primitive proboscideans, rather than horizontally as in advanced forms. To prolong the life of the teeth, the enamel on deinothere teeth was much thicker (5-8 mm) than in the elephantids (1-5 mm). The shearing teeth were ideal for processing soft vegetation. The elevated position of the external nostrils suggests the presence of a trunk, perhaps only a bit shorter than that of the elephants. One of the earlier forms Prodeinotherium originated during the early Miocene in eastern Africa and possibly radiated into Asia (Deinotherium indicum is one species), Europe (D. giganteum), and other African localities (D. bozasi).

How did the deinotheres live through a changing world over millions of years with not much adaptive change, but only an increase in size? One possibility is that the deinotheres adapted to increasing aridity and poorer quality forage by increasing body size to process higher quantities of food. Jeheskel

Shoshani feels that the mixture of general and specialized characters in the deinotheres helped in adapting to changing environments across continents through time.

The early Miocene also saw the emergence of mammutids (family Mam-mutidae). The mammutids possibly differentiated during the Oligocene, but the earliest known fossils of a mammutid, the species Eozygodon morotenesis, are associated with the early Miocene of Uganda. The mammutids radiated from Africa into Eurasia (where Zygolophodon turicensis from the middle Miocene is a type species) and North America (Mammut americanum or the American mastodon from the late Miocene onward is one of the best known of extinct proboscideans). The mammutids were also a very successful proboscidean group, surviving to the end of the Pleistocene.

A mammutid was as different from a mammoth or an elephant as "a dog is from a cat" (to quote Shoshani 1992a, p. 24). Although it had a shortened lower jaw compared to many of the early gomphotheres, a mammutid had a long jaw compared to the mammoth. While some earlier mammutids had a short pair of tusks in the lower jaw, this was completely lost in later forms. The upper jaw had a pair of well-developed tusks. The molars were low crowned with little bonding material, but with thick enamel. Pairs of domed cusps lined the length of the teeth (hence the term mastodont, meaning nipple-tooth in Greek). Among mammutids, there are the beginnings of horizontal replacement of the cheek teeth. Only two teeth were in wear at any given time. Mammutids such as the American mastodon were clearly adapted to a browsing mode of feeding.

The gomphotheres are a group of Miocene proboscideans considered to represent the second major radiation of the order. There has been much confusion over the use of the term gomphothere. Indeed, Jeheskel Shoshani feels that the gomphotheres have been a "wastebasket" of the proboscideans into which taxa of uncertain position or affinities have been dumped at various times. This is ironic given the importance of the gomphotheres as the stem group of the true elephants. It is beyond the scope of this volume to go into gomphothere taxonomy; I use the term gomphothere in the broader sense to include an assortment of Miocene proboscideans with the exception of the mammutids, stegodontids, and the true elephants. Given the taxonomic and phylogenetic uncertainties, I only briefly trace the radiations of some of the better-known gomphotheres.

As with most of the ancestral proboscidean groups, the gomphotheres originated in Africa, where fossils (of the genus Gomphotherium) have been traced to early Miocene deposits in Egypt and Kenya. Depending on how one defines evolutionary success of a taxon, the gomphotheres were the most successful originally, migrating to all continents save Australia and Antarctica (no other group went to these either), diversifying into a large number of forms and living from the early Miocene to the Pleistocene. Unlike the mammutids, the gomphotheres were "long-jawed" proboscideans.

The genus Gomphotherium is the most morphologically conservative of this group, with few diagnostic characters to distinguish it from the more advanced gomphotheres. It had tusks in both the upper and lower jaws plus a well-developed, though short, trunk. The molars were typically trilophodont (three transverse ridges), low crowned (brachyodont), with thick enamel. A sequence of progressively larger teeth indicates both horizontal and vertical replacement. Gomphotherium was as large as a modern Asian elephant.

From its eastern African origins, Gomphotherium migrated into Eurasia, where fossils are known from Arabia, Pakistan, France, Portugal, and even Japan and into North America through the Bering land bridge during the Miocene. Several species are described for the Eurasian sites, but their relationships are still unclear. One early Miocene proboscidean that caught everyone's imagination is Platybelodon, which differed from Gomphotherium in having reduced upper tusks (which were vestigial in females) but possessing a pair of enormously long and flattened lower tusks juxtaposed to form a shovel. The specialized nature of this character seems to point definitely to a creature that lived in marshy habitats and used its shovel to dig up soft vegetation and roots. Clive Spinage, however, argues that this large gomphothere could not have obtained sufficient nutrition from low-quality aquatic plants (which have large air spaces for buoyancy). Rather, the shovel could have been used to crop land plants just as hippos do with their broad, horny lips. Perhaps, the "shovel-tuskers" adapted to an increasingly arid Miocene by switching from marshy plants to drier ground herbs. A later form, Gnathabelodon, lost its lower tusks, and its protruding bony jaw was presumably covered with tough skin. The related genus Amebelodon, also widespread across Eurasia and North America, had narrower, more elongated shovels with differences in dentine structure. Amebelodon is well known from North American sites, where they may have occupied similar moist habitat niches.

By the middle Miocene, the globe had reached its peak warming since the beginning of the Oligocene.

A cooling trend accompanied by further desiccation set in during the late Miocene, about 10 My ago. Mountain uplift, especially in the tectonically active Himalayan-Tibetan plateau region, during the early-to-middle Miocene had exposed more rock to chemical weathering. In the process, carbon dioxide was removed from the atmosphere and transported to the oceanic sink. This "reverse greenhouse" effect resulted in greater loss of heat from the atmosphere. The increasing aridity promoted the spread of grasslands in the two American continents. Drier vegetation types also spread in the other continents, although true grasslands were not as yet expansive.

The reduced atmospheric carbon dioxide also brought about another ecological change that could have had important consequences for the evolution of herbivores, including the elephants. This was a shift toward grasses with the C4 pathway of carbon fixation during photosynthesis (see chapter 5). Plants with the more common C3 photosynthetic pathway are very inefficient at fixing carbon when CO2 levels fall below a certain threshold (this is partly temperature dependent, but may be in the range of 160-180 ppm). Under such conditions, the C4 plants (primarily tropical kinds of grasses and sedges) successfully

Figure 1.4

Phylogenetic relationships within the Elephantidae. (Updated from Maglio 1973, figure 40, p. 105, and Todd and Roth 1996. Reproduced with permission granted by the American Philosophical Society.)

Figure 1.4

Phylogenetic relationships within the Elephantidae. (Updated from Maglio 1973, figure 40, p. 105, and Todd and Roth 1996. Reproduced with permission granted by the American Philosophical Society.)

out-compete their C3 counterparts. The work of Thure Cerling, Jay Quade, and others on fossil soils and herbivore tooth enamel using carbon isotopic analysis showed that C4 grasslands underwent a major global expansion during the late Miocene, anywhere between 8 and 6 My ago (see chapter 5 for details of the stable carbon isotopic method and its use in dietary inferences). The evidence for expansion of the C4 grassland (used here in the broader sense to include sedges) comes from sites as far apart as North America, South America, East Africa, and South Asia.

Grasses had been around for a long time, probably evolving during the early Paleocene. Many herbivorous mammals undoubtedly included grasses as a part of their diets. The grasses initially played a relatively minor ecological role. Their increased success during the late Miocene, however, introduced a new dimension to the equation between herbivores and vegetation. Unlike most other plants, the grasses grow from basal tissues that are not destroyed by herbivory. Moderate levels of herbivory actually stimulate their productivity. Animals adapted to feed on grass ensured themselves a constant supply of a plant that spread across large landscapes. This is precisely what a variety of the later Miocene mammals did; equids, bovids, rodents, and proboscideans are some of the groups that evolved newer structures to take advantage of this resource.

Grasses, especially the perennial siliceous grasses of the C4 type, have the disadvantage (to herbivores) of being highly abrasive to teeth. The adaptive response of grazing mammals, therefore, is to develop teeth with high crowns (hypsodonty), complex surface patterns, and stronger reinforcement of cement. This is most striking in the evolution of horses, for instance. The proboscideans, too, show dental adaptations to the late Miocene vegetational shift.

The late Miocene saw the third major radiation of the Proboscidea—the emergence of the true elephants (family Elephantidae) (fig. 1.4), which diversified through the Plio-Pleistocene. An "advanced" gomphothere was most likely the ancestor of the elephants. But, before riding the elephant wave, we must briefly attend a sideshow in the great proboscidean drama.

The stegodontids, derived from an advanced gomphothere, were once believed to be ancestors of elephants, but are now considered a sister group of the elephantids. They evolved in Asia around the same time that the true elephants arose in Africa. Actually, one genus with less-specialized characters, Stegolophodon, may have arisen in earlier Miocene times in Southeast Asia (Indochina) and stagnated in molar development, before a late Miocene radiation to other parts of Asia. Stegodontid fossils are known from China, Japan, Southeast Asian islands, and the Indian subcontinent, as well as Europe and Africa. The genus Stegodon diversified more rapidly during the Pliocene.

A particularly impressive form was Stegodon ganesa (named after the Asian deity with an elephantine form; see chapter 2), which stood 3.5 m at the shoulder and sported a formidable pair of tusks. Emerging from the upper jaw, these thick tusks almost touched the ground, curving sideways and upward at the ends. Only vestigial tusks remained in the shortened lower jaw. Molar replacement in the stegodons was horizontal, with only two teeth being in wear at a given time in each jaw half. The molars had up to 14 lophs (transverse ridges), with each loph composed of a row of thickly enameled cusps. The low-crowned teeth, however, suggest that stegodons were adapted to a browsing diet of forest leaves and bamboo shoots.

We must, therefore, seek for the ancestral elephant among the subfamily Stegotetrabelodontinae (fig. 1.4). The two recorded genera, Stegotetrabelodon and Stegodibelodon, are known from fragmentary remains in Africa. The prefix stego is confusing as it implies that these were stegodontids, as earlier believed, but they are now considered true elephantids, or at least intermediate between the gomphotheres and the elephantids, based on their dental characters. They had low-crowned teeth like the gomphotheres, but with a few plates characteristic of the elephantids. The ridged surface, with cusps in the more primitive forms, had now evolved into plates, each plate consisting of a flattened loop or lamella with enamel on the outside and dentine filling inside. The space between the enamel loops was U or V shaped as in the elephants. While Stegotetrabelodon had short incisors in the lower jaw, these are absent in the more progressive Stegodibelodon, which also had a shorter mandible.

One of these early African elephantids is believed to have given rise to Primelephas just prior to the end of the Miocene (5 My ago). Vincent Maglio considered Primelephas to be the progenitor of the genera Loxodonta, Mammu-thus, and Elephas, while M. Beden thought that Loxodonta could be derived directly from the stegotetrabelodonts. The former view seems the more favored today. A more recent discovery of an unnamed species of Loxodonta from late Miocene deposits in Uganda and from Kenya seems to imply that Primelephas, although an elephantid and related to Elephas and Mammuthus, may not be a direct ancestor. The molars of Primelephas gomphotheroides show some structural advances over those of the stegotetrabelodonts. Although still low crowned, the molar ridges are more like those of later elephants. A pair of small incisors was still retained in the short symphysis of the lower jaw.

Thus, 6 My ago, or close to that time, there were many proboscidean players, belonging to an amazing 22 of the 38 or so recognized genera, marching across all continents except Australia and Antarctica. They included the primitive deinotheres, the mammutids, the gomphotheres, the stegodontids, and the true elephants. There is no precise figure for the number of species that were found at a given point in time, but this was undoubtedly large.

How could such a variety of large-bodied proboscideans coexist? One explanation obviously is that the various genera and species were geographically spread over the different continents. Even where more than one species of proboscidean were found together in the same locality, there could have been ecological separation in use of habitats and plant types consumed. The climatic changes occurring during the late Miocene were also giving rise to newer vege-tational compositions, structures, and mosaics, providing the stimulus to evolutionary adaptations among the mammals. Thus, gradation in use of semi-aquatic or soft vegetation, browse from moist forests or drier woodlands, and grass from savannas or true grasslands may have contributed to a certain de gree of niche separation and coexistence among proboscideans. By the close of the Miocene, about 5 My ago, the emergence of the true elephants ensured that this amazing run of proboscidean diversity, although in decline, was not going to end in a hurry.

0 0

Post a comment