Manchester Memoirs, Vol Ix. (1916), No. 2. 43 



by our series are shown to be homologous with the 

 "squamosal" of Ceratodus are not at first sight repre- 

 sented in the roof of the head of Osteolepis. Our finest 

 skull of Dipterus platycephalus shows that these plates 

 cover the upper end of the quadrate, in this agreeing with 

 the '■ squamosal " of Ceratodus. 



It will be noted that they carry part of the broad belt 

 of pits for the reception of lateral line sensory organs. 

 Their connection with the quadrate at once suggests that 

 they are represented by the squamosal and bones asso- 

 ciated with it in Rhipidistia, as we know that these are 

 actually applied to the lateral border of the quadrate. It 

 will be remembered that these cheek plates in Rhipidistia 

 do actually support a lateral line canal although, as we 

 have been unable to trace anything more than a very 

 broad parallel between the general distribution of lateral 

 line canals in Osteolepis and Dipterus, such evidence is not 

 of any great value. The fact that up to now no repre- 

 sentatives of the cheek plates which form so large an 

 element of the Rhipidistian skull have been found in 

 Dipnoi,, has always been troublesome to those who 

 believed in their close connection. It remains to be 

 shown that it is possible to explain the methods by which 

 this reduction of the cheek plates to two small bones on 

 the top of the head has been brought about, and the 

 reasons for this reduction. It is obvious that the Rhipi- 

 distia, taken as a whole, are a group of predacious fish 

 living mainly on other fishes. In many cases the massing 

 of large fins at the end of the body shows that they were 

 active rapid-swimming fish built for pursuing others. 

 The Dipnoi, on the other hand, have always been slow 

 swimmers, probably becoming slower as time went on. 

 Their flat crushing tooth plates are often very badly worn, 

 suggesting that their owners either lived on molluscs, the 



