7 o PRIMAR Y FA CTORS OF OR GANIC E VOL UTION. 



the tarsus, and, what is not stated in the above table, 

 of the carpus also. Finally there is the remarkable 

 development of horns on the head. When we come 

 to examine the phylogeny of this order we will find 

 how completely these characters are the result of the 

 fixation of variations which have appeared in past geo- 

 logic ages, and how various are the combinations and 

 modifications presented by the extinct types. 



Few natural groups permit of representation of 

 their subdivisions in linear series. The only correct 

 representation is in the form of a branching tree, and 

 this cannot be well done in flat projection on the pages 

 of a book. Each branch taken by itself, however, 

 yields itself for a longer or shorter space to linear 

 treatment. 



For an example of such linear series in higher 

 groups I turn again to the Batrachia Salientia. Here 

 the two suborders of the Arcifera and Firmisternia pre- 

 sent the following interesting parallels : 



ARCIFERA. FIRMISTERNIA 



I. Without teeth. 

 a. With sacral diapophyses dilated. 



Brevicipitidae. 

 Bufonidae J Engystomidse. 



[ Phryniscidae. 

 aa. Sacral diapophyses cylindric. 



Dendrophryniscidae. Dendrobatidae 



II. With premaxillary and maxillary teeth only. 

 a. With sacral diapophyses dilated. 



Pelod >" i<te I f Dyscophid*. 



Suf) 153? 



aa. With sacral diapophyses cylindric. 



Cystignathidae... . f Colostethidae. 



I Ranidae. 



