434 ANNUAL, EEPOET SMITHSONIAN INSTITUTION, 1918. 



ber of terms intermediate between the extremes (Dollo, 5, p. 97 1 ). 

 Just such a series we possess in the paleontological series of the 

 Dipnoi : Dipterus valenciennesi, Dipterus macropterus, Scaumen- 

 acia, Phaneropleuron, Uronemus, Ctenodus, Ceratodus, Protopterus, 

 Lepidosiren (1. c., p. 88). Dollo shows that the structure of the 

 tail as well as that of the top of the head, the squamation, the pugu- 

 lar plates, the opercular apparatus, the ganoin, and the ossification 

 of the mandible, the suborbital band — all this proves that the course 

 of evolution has been in the direction from Dipterus to Ceratodus. 

 and not the opposite (1. c, p. 89-97). Ife is especially by the struc- 

 ture of the tail that the concept of irreversibility is illustrated. In 

 a long and thorough treatment of the subject Dollo shows (1. c, 

 pp. 89-97) that the diphy cereal tail of the Dipnoi (and of the other 

 known ancient and modern fishes) is a secondary diphycercal tail 

 whose morphological value in the Dipnoi (the second dorsal fin, the 

 second anal fin) is not equivalent to the morphological value of the 

 primitive diphycercal tail (caudal fin). In this secondary diphy- 

 cercal condition there is therefore no return to the primitive 

 structure. 



The most important other cases of phylogeny which Dollo has 

 considered are the phylogeny of the sirenians (Dollo 3, p. 119), the 

 phylogeny of the Leather-backed turtle (Dollo, 7, p. 9), and the 

 phylogeny of the Holocephali (Dollo, 13, p. 107). 



One of the most important cases with regard to the ethological 

 application of the law of irreversibility is found in the memoir on 

 the Dipnoi. If it be assumed that Dipterus comes from Ceratodus, 

 as the latter is an adaptation to life in turbid water, it would be 

 necessary to suppose either that Dipterus represents an adaptation 

 to life in mud (excessively turbid water), or else that it represents 

 a return to life in clear water. The first alternative being that of 

 Lepidosiren, the second is the only one which remains open for dis- 

 cussion (Dollo, 5, p. 99). But, putting aside paleontological and 

 purely ethological reasons, the law of irreversibility is sharply op- 

 posed to such a view. 



"Would the lost ganoin return? Would the cephalic shield resolve itself 

 into its ancestral elements? Would the suborbital band with its ossicles in 

 varying number become once more a solid arch? Would the opercular appa- 

 ratus resume its original dimensions? Would the vanished jugular plates re- 

 appear?" As all of these structures are reduced in Ceratodus (1. c, p. 100), 

 Dipterus can only represent a primary adaptation to life in clear water, that 

 is to say it is purely a fish (" the most pisciform of Dipnoi," 1. c, p. 101). 



'Discussing the subject of the phylogeny of the Holocephali (Dollo, 13, pp. 107-108), 

 Dollo says : " The idea of the irreversibility of evolution, which has led me to the results 

 that I have just demonstrated, has once more shown its usefulness. Without it one would 

 be led to assert that specialized organisms could become primitive again and then once 

 more specialize themselves in the same or another direction. Such an assumption, unless 

 supported by absolutely complete paleontological series — which we are far from possess- 

 ing — would destroy all possibility of discovering phylogeny, the main object of mor- 

 phology." 



