4O THE CLASSIFICATION OF BIRDS 



family which have an ambiens muscle.* They appear to have 

 been isolated from the Coraciiformes at a very early date, before 

 the power to re-develop the ambiens muscle was entirely lost. As 

 soon as they began to be more terrestrial in their habits it may be 

 assumed that it was advantageous to them to arrest the decline of 

 the ambiens muscle and restore it to its former use. Some species 

 belonging to the Coraciiformes (some of the Bucerotidce, to wit) are 

 perhaps more terrestrial than the CcUhartidat, but they seem to 

 have become so at a much later date, when the ambiens was lost 

 beyond recall. 



* There are very few aquincubital birds that have no ambiens muscle, but 

 there are no quincubital families that contain both birds with and birds without 

 it. It is so very much easier for characters to be independently lost than to be 

 independently acquired, that we must assume the possession of an ambiens 

 muscle to have been a primitive avian character. 



Of the llalliformes the Podici/iidce. have no ambiens muscle. 



Of the Charadriiformes the Alcidce and some of the (Mumbidce have lost it. 



Of the Ciconiiformes the Ardeidce, the Scopidce, and the Pelecanidce have 

 lost it. 



Of the Falconiformes the Striges and some of the Psittaci, and of the Ralli- 

 formes several genera of the Tubinares have the ambiens muscle either absent or 

 in a vestigial condition. 



It is not known that any of the Sphsenisciformes have lost the ambiens muscle. 



Of the Struthioniformes the Casuarii are the only Order which have no 

 ambiens. 



It is quite clear that amongst aquincubital birds it is a normal condition to 

 possess an ambiens muscle. Amongst quincubital birds the contrary appears to 

 be the rule. 



It is not known that any of the Passeriformes have an ambiens muscle, but, 

 so far as is known, none of the Galliformes are without it, whilst amongst the 

 Cuculiformes the Upupse have lost it and the Cuculi have retained it. 



The presence or absence of the ambiens muscle was a character to which 

 Garrod and Forbes attached primary value. In Garrod's Classification the Class 

 Aves is divided into two subclasses the Homalogonatse (consisting of birds which 

 have an ambiens muscle, or which were supposed by Garrod to have only recently 

 lost it) and the Anomalogonatas (consisting of birds which were assumed by 

 Garrod to have lost it long ago). Garrod regarded the Passeres and their allies 

 as the most highly developed birds that had adopted the latest improvements of 

 the avian structure, of which the suppression of the ambiens muscle was one. 

 The Gallinse, on the other hand, were supposed by Garrod to be more archaic in 

 their structure, and he regarded the retention of the Ambiens muscle as an 

 archaic character. It is quite possible, however, that exactly the opposite is the 

 case. If we assume that the ancestral birds lived in trees, and acquired the 

 power of flight because they lived in trees, it seems probable that the ancestral 

 birds had no ambiens muscle, and that those birds which now possess it have 

 acquired it in order to adapt themselves to a life on the ground. It may or may 

 not have been a retrograde step, but it was doubtless made in order to find a 

 modus vivendi, in the struggle for existence, for a variety of little aberrant groups 

 of birds, which were thus enabled to fill up some of the holes and corners left 

 by the great mass of their relatives who maintained the ancestral habit of perch- 



