NATURAL SCIENCES OF PHILADELPHIA. 149 



which aided in the production of the burrs, is also probable. We 

 may thus believe it to have been dermal lilie that of the giraffe or 

 the Aittilocajira at the period of immaturity of its horny sheath. 



It may be concluded then tliat the genus Cosoryx represents 

 the ancestral type of the Cervidae, and explains the origin of the 

 remarkable type of horns of that family as follows : Ruminants 

 with fixed horns of structure more dense and brittle than others 

 of the same type, in their annual combats at the rutting season, 

 ver}^ frequently broke the beams off not far above the base. The 

 usual location of nutrition followed, which being annually re- 

 peated, became as periodical in its return as the activity of nutri- 

 tion of the reproductive system. This activity ceasing, the horn 

 being dense lost its vitality, the more so as the normal covering 

 would have already perished in its distal portions. The natural 

 consequence, the separation of the dead from the living bone by 

 suppuration, would follow. This process would, however, probably 

 require a longer time for the establishment of its periodical return 

 than the fracture and attachment of the existing horn. 



This. appears to be the only explanation of the origin of the 

 phenomena exhibited by the horns of the Gervidse, and is sug- 

 gested by the specimens of Cosoryx to be described. 



Char. Specif. This species is larger than the C.furcatus^ Leidy, 

 and exhibits two antlers instead of one, of which the first is given 

 off at a point much further from the base than in that species. 



The beam near the base is curved a little inwards, and is semi- 

 circular in section, the outer face being slightl}^ concave, the inner 

 very convex. The base is situated a short distance within the 

 free superciliary border. The beam becomes more cylindric, and 

 then expanding in a fore and aft direction, gives off an antler at 

 right angles, nearly paraljel to the cranial axis. At a distance 

 little over half the elevation of the first antler, the beam gives off 

 a second, in a plane transverse to the axis of the skull. The 

 terminal portion of the beam is cjdindric, curved, and acute at 

 the apex. 



Mandibles with teeth of two species of this genus were found, 

 the smaller of which, occurring with the other portions of C./ur- 

 catus. belong to it. The larger differs in the elevation of the 

 intercrescentic column of the first molar, which is worn into a 

 loop at ordinary' maturity ; this may however be but an individual 

 variation. The diastema is long, and the ramus of that point 

 quite slender. 



