Dr. A. Smith Woodward — TJie Bone-beds of Pikermi. 483 



and eroded by the present stream before being covered with the 

 three or four metres of superficial gravel which now preserves them. 

 The bones are also broken by the penetrating rootlets of trees. 

 The lower horizon is at a depth varying from one to two metres 

 below the upper horizon, and thus secure from destruction by surface 

 agencies. Like each of the two upper bone-beds, it is rarely more 

 than 30 cm. in thickness ; while the marl above and below it is 

 almost destitute of bones, rarely yielding more than rotten fragments, 

 but quite prolific in scattered land and fresh-water shells. The 

 deepest excavations beneath the lower bone-bed descended for about 

 three and a half metres, and furnished the bone-fragments and shells 

 throughout. No traces of vegetable matter were observed in any layer. 



So far as can be judged at present from the new excavations, the 

 three bone-beds of Pikermi are all of the same nature and contain 

 the same mammalian remains. The bones are massed together in 

 inextricable confusion, and are often mixed with a few pebbles. 

 Large and small bones, whole specimens and splintered fragments, 

 all occur together ; but the small bones are usually most numerous 

 at the bottom of the layer. Several specimens of approximately the 

 same shape and size are often met with in groups, as if they had 

 been sorted by water in motion. On one occasion, for example, the 

 scattered remains of many gazelles were found together ; in another 

 spot there were several skulls of Tragoceras in one mass ; in other 

 cases nearly all the bones belonged to limbs of Eipparion ; while 

 one area was specially chai'acterized by pieces of vertebral columns 

 of Euminants and Eipparion. The elongated bones and elongated 

 gi'oups, however, were never observed to trend in one definite 

 direction, but were always disposed quite irregularly ; thus 

 indicating that in the region where the bones eventually 

 accumulated, the water by which they had been transported either 

 became still or moved only in gentle eddies. 



Very few nearly complete skeletons occur, and even when chains 

 of vertebrae are preserved most of the ribs are lacking. The only 

 approximately complete skeletons observed during the recent 

 excavations were those of some Garni vora (Ictitherium, Metarctos, and 

 Machcerodus). It is, however, obvious that many of the bones were 

 still held together by ligaments at the time when they were buried ; 

 for numerous complete feet and nearly complete limbs are found 

 with all the bones in their natural position. It is also to be noted 

 that in most cases these limbs are sharply bent so that the two or 

 three segments are almost parallel, as if they had retained the 

 contraction assumed at death. Some decomposition of the soft parts 

 had already taken place even in these instances ; for a few of the 

 phalanges of the hipparions and ruminants are often wanting when 

 the other bones of the limb are still in their natural association, 

 while the phalanges of the rhinoceros feet seem to be always lost, 

 though the three associated metapodials are quite common. 

 Similarly, the loosely articulated mandible of the Ungulata is nearly 

 always removed from the skull ; it is only commonly preserved in 

 place in the Carnivora and Quadrumana. 



