S24: AXSALS XEW YORK ACADEMY OF SCIEXCEs 



I incline to the opinion that Dollo's view is the correct one. The skull 

 pattern of Scaumenacia, as figured by Hussakof (1912, p. 136), is suffi- 

 ciently close to that of Dipterus (Pander, ISqS, Tab. 3) :o enable one to 

 homologize the principal bony elements in these two genera (Fig. 1). 

 The Scaumenacia skull, I think, points the way to that of PhaneropJeuron 

 (Goodrich, 1909, p. 239) and this in turn brings us within reach of the 

 Ceratodus skull-pattern. In short it appears probable that the Ceratodus 

 skull-pattern has been derived from a Dipterus-like type through the 

 following changes : (a) the small front als and parietals have become 

 enlarged, so as to cover the median, pre-parietal element; (h) the oppo- 

 site frontals have united to form the wrongly called '•'ethmoid'' of Ceroy- 

 todiis: (c) the opposite parietals have united with each other and with 

 the median supraoccipital or nuchal plate to form the large median 

 ^'occipitar' : {d) the longitudinal series of small elements running from 

 above the orbits to the back of the skull have coalesced and broadened, 

 to form the large pair on either side of the median plate ; {e) the small 

 plates behind the orbit and above the operculum have coalesced and 

 broadened into the large plate marked pterotic in Goodrich's Fig. 206 

 (1909, p. 23 T) : (/) the remaining small elements immediately behind 

 the orbit are part of the circumorbital series: {g) the exposure of the 

 chondrocranium above the occiput and above the front part of the skull 

 I regard as quite secondary and as a relatively modem specialization: 

 the massive character of the chondrocranium may be due to the fact that 

 the stout tritoral plates on the roof of the mouth and the heav\' mandible 

 require a massive firm skull for their support : it may also be related with 

 the fact that the scales and dermbones of the skull are losing their dense 

 osseous character and becoming horny. The exposure of the chondro- 

 cranium is carried to an extreme in Lepidosiren. and it cannot be claimed 

 that this degenerate, eel-like form is more primitive than the numerous 

 Palaeozoic Dipnoi, Crossopterygii and Actinopterygii that have a complete 

 skull-roof. 



In view of the foregoing discussion it is hardly necessary to state that 

 I am unable to accept Dr. Eastman's comparison (190", p. 132) of the 

 Ceratodus skull with that of Dinichfliys at its face value, and, with Dr. 

 Dean (1907), I must ascribe these resemblances to convergence, referring 

 again to the arthrodiran appearance of the skull-top of one of the modern 

 catfishes {vide supra, p. 320). 



Eetuming to Dipterus I would also be inclined to disagree with Good- 

 rich's statement (1909, p. 256) that "the arrangement of the cranial 

 covering bones points rather to the dipterids V)eing a highly specialized 

 offshoot from the base of the dipnoan stem.'' As already stated, the 



