bory ^eoTives 



o 



r 



ana 



onia 



by Malcolm C. McKenna 



Paleontology is a combination of good 

 science and good luck. Most of the time, 

 we paleontologists work at determining 

 the meaning of what has already been 

 brought to a museum's storage cases. As a 

 new field season approaches, however, we 

 head for distant parts of the planet in hopes 

 of finding something important that will 

 improve our understanding of geological 

 history and biological evolution. Thus, in 

 June of 1991, ajoint Mongolian-American 

 expedition, including two Mongohan pale- 

 ontologists and six of us from the Ameri- 

 can Museum of Natural History, headed 

 for Naran Bulak ("Sunny Spring"), a re- 

 mote oasis in the otherwise sere Gobi 

 Desert of southwestern Mongolia. 



Due to Naran Bulak's coveted water — 

 and because many eighty-million-year-old 

 dinosaurs have been found in late Creta- 

 ceous sediments nearby — it has become 

 an important base for deeper paleontologi- 

 cal exploration of the Gobi. The area's 

 Mesozoic dinosaurs are about the same 

 age as those at Mongolia's famous Flam- 

 ing Cliffs — where, in the 1920s, the 

 American Museum's own Mongolian ex- 

 peditions first found dinosaur eggs, di- 

 nosaur skeletons such as those of Proto- 

 ceratops and Velociraptor (of cinematic 

 fame), and skulls and jaws of early mam- 

 mals such as Zalambdalestes {see "A 

 Pocketful of Fossils," page 40). 



Since the late 1940s, many expeditions 

 from scientific institutions in Russia, 

 Poland, and Mongoha, as well as our own 

 from New York, have used Naran Bulak as 

 a center from which to radiate in search of 

 extinct remains of late Cretaceous di- 

 nosaurs, lizards, birds, crocodiles, turtles, 

 and mammals. Our main interest in 1991, 

 like that of most of our predecessors, was 

 to explore the Cretaceous outcrops reach- 

 able from Naran Bulak. But equally im- 

 portant was the presence of younger fos- 



sils in the multicolored Cenozoic sands 

 and clays overlying the dinosaur beds, 

 where we unexpectedly had a stroke of 

 good luck, practically within shouting dis- 

 tance of camp, that set us off on a new path 

 of discovery. 



One day, research fellow James Clark 

 decided to prospect for fossils by follow- 

 ing a fifty-five-million-year-old Eocene 

 band of red early Cenozoic sediments that 

 extended southeast and east of camp. The 

 red layer rested on slightly older white- 

 colored sands, so we could easily trace the 

 boundary through the badland exposures. 

 Clark was almost immediately rewarded 

 by a hard nodule of stone that he found a 

 few feet above the base of the red layer. It 

 contained a complete skull and jaws of 

 what at first appeared to be some sort of 

 rodent. The fossil was mostly encased in- 

 side a limey nodule with only a couple of 

 front teeth protruding, but the nodule was 

 vaguely skull-shaped, which was what 

 had attracted Clark's attention. Being an 

 expert anatomist, he could almost see the 

 rest of the specimen through its coating of 

 limey silt. 



When a paleontologist finds and col- 

 lects a fossil, the next thing he or she does 

 (after wrapping it up and recording the de- 

 tails of its location on a map and its posi- 

 tion in the rocks) is to follow the same 

 layer of rock that produced it wherever the 

 layer can be seen. We often do this on our 

 hands and knees. Jim was not lucky again 

 that day, but in the ensuing days Priscilla 

 McKenna and I dihgently continued the 

 search several miles to the east. We knew 

 where to look: just a few feet above the 

 base of the Eocene red layer. We had only 

 to follow the red and white boundary 

 wherever it went and then prospect a few 

 feet above it. We also knew what to look 

 for: not so much for actual bones but 

 rather for limey rock nodules that looked 



vaguely skull- or bone-shaped, distributed 

 in the red beds Uke raisins in raisin bread. 

 After much effort in these areas both east 

 . and west of Naran Bulak, we eventually 

 found about fifty nodules with skulls or 

 other bones inside. Most of the specimens 

 were skulls and jaws of adult animals, but 

 juveniles with milk teeth were also pre- 

 sent. We even have a curled-up, articu- 

 lated partial skeleton that looks as if the 

 animal had died in a burrow. All these 

 specimens in their stony coatings had been 

 completely overlooked by our many pre- 

 decessors who had not had our expedi- 

 tion's brand of educated luck. 



Our first impression — that the speci- 

 mens belonged to some unknown member 

 of the Rodentia, the varied order that in- 

 cludes mice, rats, beavers, and porcu- 

 pines — was based on what we could see of 

 the front teeth, often the only part visible 

 at the surface of our rock nodules. They 

 looked like rodent incisors. 



When we returned to New York, our at- 

 tention was riveted by the dinosaurs, birds, 

 and lizards that we had found in the late 

 Cretaceous sediments far older than the 

 Eocene red beds. But finally, we found 

 time to begin removing some of the rock 

 from our red bed "rodents." We have been 

 able to dissolve some of the rock nodules 

 in weak acetic acid, leaving the specimens 

 inside more or less intact. We also have 

 used sharp needles and other tools to get at 

 the specimens. Gradually, some skulls and 

 jaws have emerged. However, behind the 

 gnawing front pair of theii" rodentlike in- 

 cisors were some surprises. 



A second pair of incisors bolstered the 

 front ones, not only in the upper teeth but 

 also in the lower jaw. Rabbits, hares, and 

 their short-eared relatives the pikas, col- 

 lectively known as lagomorphs, have a 

 second upper incisor pair, just behind the 

 main pair — but rodents do not. Modem 

 lagomorphs, as well as all known rodents, 

 have only one incisor in each lower jaw; 

 because our Naran Bulak specimens still 

 had two on each side, they are more prim- 

 itive. (Still more primitive mammals have 

 even more sets of incisors.) Thus, in the 

 early Eocene red beds at Naran Bulak, we 

 had found, not rodents, but early and prim- 

 itive Asian fossil lagomorph skulls and 

 bones, about twenty million years older 

 than any previously well-known lago- 

 morph skulls. 



Members of the mammalian order 

 Lagomorpha are well known from many 

 fine specimens, including complete skulls 

 and skeletons of Palaeolagus, dating back 

 to about thirty-five milhon years or so ago. 



56 Natural History 4/94 



