346 EMBRYOLOGY OF THE LOWER VERTEBRATES CH. 



Chapter II., according to which new nerve -paths may arise in 

 response to the short circuiting of nerve-impulses. 



A tympanic cavity with membrane and columella occurs in many 

 of the Anura as for example the ordinary Frogs and Toads (Rana, 

 Bufo) while in others such as the genera Boinbinatbr and Atelopus 

 (" Phryniscus ") it is absent. In the Urodele Amphibians the 

 tympanic cavity and membrane have not yet made their appearance. 

 The inner end of the columella is however represented by a movable 

 plate of cartilage fitted into the fenestra and in various genera the 

 extra-columellar portion is represented by a rod-like outgrowth 

 from this. The former apparently develops from the auditory 

 capsule while regarding the latter there is much difference of 

 opinion as to the extent of its relation to the cartilage of the hyoid 

 arch. The disagreement between different observers probably means 

 that there are actual differences between different genera of 

 Amphibia. This is quite what is to be expected, for whenever we 

 find a single organ of which part is derived from one embryonic 

 source and part from another the proportion contributed by the two 

 sources is liable to vary, so that in one case it may be the portion 

 derived from the one source which is conspicuous and in another 

 case that derived from the other. 



The tympanic ring within which the tympanic membrane is 

 stretched arises in the form of an outgrowth from the rudiment of 

 the quadrate, i.e. from the upper portion of the skeleton of the 

 mandibular arch. This outgrowth separates off and grows round the 

 outer end of the hyomandibular pouch in the form of a crescent the 

 two horns of which eventually meet to form a complete ring. 



SKELETON OF THE MEDIAN OR UNPAIRED FINS. The median 

 fin, thin and membranous as it is in its most highly evolved condi- 

 tion, is supported by characteristic skeletal arrangements. Into 

 these two distinct elements enter, one mesial represented by rays of 

 cartilage or bone (radials), the other superficial and of dermal origin. 



The mesial fin -rays are frequently in close relation to the 

 neural and haemal arches and it is reasonable to suppose that in the 

 process of evolution, as the hind end of the body became extended 

 in a dorsal and ventral direction, so as to attain to the flattened 

 form conducive to efficiency in propelling the body, the neural and 

 haemal spines underwent a corresponding extension for the purpnsi s 

 of support. This view is corroborated by the existing Dipnoi in 

 which the mesial fin remains a comparatively slightly differentiated 

 extension of the body dorsally and ventrally and in which the 

 mesial supporting elements are simply the prolonged neural and 

 : lines, each secondarily subdivided into three segments. 

 The same is the case in Fishes generally so far as the ventral port nm 

 of the caudal fin IS concerned in whieli the mesial supports develop 

 also for the most part as typieal haemal spines. 



The mesial support^ of the dorsal portion of the median tin >n 

 I!M contrary do not ii, I nerally nlmw this relation to the 



