THE MAMMALIA OF THE UINTA FOUMATIOX. 567 



number of forms, it cannot be due to mere coincidence, but must be the result of cer- 

 tain modifications of the pressure and strain transmitted to the carpus in i)araxonic 

 types. The loss of the Literal toes in the highly reduced Artiodactyla. in which the 

 pressure is most jjerfectly divided between mts. iii and iv, from Eocene to recent times, 

 has not been accompanied by any alterations of the relations of the lunar to the mag- 

 num and unciform (see Anoplotherium and Cervus), apparently because there is no 

 alteration in the axis of the metai)odium. But in the Perissodactyla the reduction of 

 mc. V, as can be readily observ'ed in the rhinoceroses (see De Blainville, Osteographie, 

 R. tetradactylus of Auvei-gne, of Sansans, and li. sumafrensis, utiicorms, bicornh), is 

 accompanied step by step by the displacement of the lunar from the magnum. Lateral 

 compression of the carpus has nothing to do with this displacement, because the 

 tapir, which is even now in the transition stage between Types TV and Y, has a 

 much narrower carpus than Aphelops. The tapir manus is now in the stage of evo- 

 lution which was passed by the rhinoceros in the Miocene period. 



It follows from these cases that I'eduetion per se does not influence displacement, 

 but only in so far as (in the Perissodactyla) by releasing the carpus from the pressure 

 of the V digit it brings the major axis directly througli the centre of the third digit, as 

 seen in Type Y. This type thus embraces a large number of Ibiins, in which reduc- 

 tion of the fifth digit has been very rapid. The writer can find no exceptions to the 

 law that mesaxonic tridactyl types of all the Tei'tiai-y pei-iods show the extreme of 

 diplarthrism. * 



Dr. Scott has called the writer's attention to a third correlation of the lunar and 

 reduction, as found by a comparison of the lunar, sca|)hoid and unciform in tiie 

 adaptive and inadaptive Artiodactyla. In the inadaptive types (Euteledon, Ano- 

 plotherium, etc., see Kowalevsky) the lunar is enlarged at the expense of the sai[)h- 

 oid ; the latter resting upon the trapezoid, which in turn suppoi-ts only the rudi- 

 mentary second digit, the third digit foi-miug no trapezoidal union. In the adajHive 

 types, on the other hand, the third digit acquiring a trapezoidal facet, the scaphoid is 

 somewhat larger and the lunar less enlarged. 



TAiisrjs. 



The above observations api)ly chielly to the carpu>;, bill |)i('ciscly the same |)riu- 

 ciples operate in the tarsal displacements. 



•Tlie manus o( PaUiolheriitm medium, us figured liy KowulevnUy (Piiln-onloRrniiliica, Hd. XXI[, Tnf. VII). prt-!tcnl» 

 an exception. But reference to Do Rlainvllle's flgure anil to a cast of the original foot recently presented to Princeton 

 by llic I'ari'i Museum, shows that Kowalevsky's flgure is incorrect. There i.s no ejtiensivp lunar niagniiin facet, !«ucli 

 as he represents. 



