v THE SKELETON 341 



scheme of a vertebral segment in addition to the central element. 

 But a difficulty is at orfce raised by the not infrequent appearance 

 (as in Amia and many Elasmobranchs) of two centra within the 

 limits of a segment. To get over this it has been suggested that 

 primitively there were actually present two complete vertebrae, each 

 with centrum, neural arch and haemal arch, within the limits of a 

 single segment (dispondylous or diplospoudylous condition). Physio- 

 logical considerations however support the probability of there having 

 been primitively a single vertebral centrum, extending from about 

 the middle of one pair of myotomes to about the middle of the next 

 pair. If it is borne in mind that the material of the anterior and 

 posterior halves of the centrum is derived from two independent 

 sources successive pairs of arch-elements or of sclerotomes it will 

 seem reasonable to explain the occasional duplicity as due primarily 

 to absence of the complete fusion which normally comes about 

 between the halves of the centrum derived from the two sources, the 

 two halves proceeding with their development independently. Con- 

 versely a more complete fusion, extending over a number of these 

 potential half- vertebrae instead of merely two of them, would lead 

 to cases of elongated definitive vertebrae carrying a number of 

 arches. 



BONY SKULL. The caution expressed on p. 335 is especially 

 necessary in connexion with the bones of the skull. Here we have 

 a department of morphology which took shape in the early days of 

 that science. The efforts of the older anatomists were devoted to 

 the working out of homologies between the bones of different groups 

 of Vertebrates and individual bones were given the same name they 

 were decided to be homologous mainly on the basis of similarity of 

 relations in the adult animal, with only the most slender basis of 

 either palaeontological or embryological knowledge. The consequent 

 uncertainty as to the precise homology of similarly named bones in 

 different groups of Vertebrates makes it in the author's opinion 

 impossible to write a satisfactory account without treating each of 

 the main groups in detail by itself. As to do this would require 

 more space than is available he would refer readers who desire such 

 information to Gaupp's volume and the literature there cited and 

 will confine himself here to a very brief sketch. 



In the lowest Gnathostomata, as represented at the present day 

 by the Elasmobranch fishes, the skull retains throughout life its 

 cartilaginous character, the bony tissue being confined to the placoid 

 elements of the skin. In the other groups of Gnathostomata the 

 purely cartilaginous condition is temporary, the cartilaginous skull 

 becoming strengthened and in places, though never entirely, replaced 

 by bone. 



In the floor of the chondrocranium there make their appearance 

 a mid-ventral row of replacement bones the basi-occipital, the basi- 

 sphenoid, and the presphenoid. Laterally to each of these elements 

 the cartilage becomes replaced by a pair of bones known respectively 



