140 ME. E. PHELPS ALLIS, JUN., ON 



enter tliis layer, the latero-sensory organs and terminal bucls all 

 lying external to it, as do apparently also the corresponding 

 tissues of the ear and eye. 



Conclusions. 



In the chondrocranium of the adult Chmicera the trabeculae 

 arise from the ventral surface of the parachordals at a con- 

 siderable distance posterior to their anterior ends, and they 

 project antero-ventrally at a marked angle to the parachordals. 

 This shows that there was not only marked cranial liexure at the 

 time the trabeculse began to chondrify, but also that this flexure 

 was not, as in the Plagiostomi, later greatly reduced ; for that 

 there has here been reversion from the conditions found in the 

 alult Plagiostomi, the cranial flexure being first reduced and 

 then later reacquired, seems wholly improbable. The Holocephali 

 must accordingly be descended from some form in which con- 

 ditions existed similar to those that are now found in embryos 

 of the Plagiostomi, and as these conditions were_ probably not 

 found in any adult form, the Holocephali must be descended 

 directly from embiyos of the Plagiostomi by conservation and 

 modification of the conditions there found. 



The chondrocranium is generally considered to have been 

 formed as a cast which lies between the brain, as a core, and the 

 external epidermis, the form of the cast depending primarily 

 upon the form of the brain. Tliere must accordingly have been 

 some potent influence determining, in these fishes, the retention, 

 by the forebrain, of the position imposed upon it by the marked 

 cranial flexure of embryos. The precocious development of the 

 eyeballs, and their shifting forward and mesially dorsal to the 

 forebrain, where they are actually found in the adult, would 

 doubtless have furnished such an influence; but, unfortunately 

 for this assumption, Schauinsland's descriptions of Ccdlorhynchus 

 show that the eyeballs are still widely separated fiom each other 

 in early embryonic stages, and that the midbrain lies between 

 them. It would accordingly seem as if the determining influence, 

 whatever it may have been, must have been related either to 

 a. precocious development of the olfactoiy oigans, or to such a. 

 development of the functional mouth. 



In all embryos of the Plagiostomi the mouth is at first directed 

 ventrally, and if it became functional before the cranial flexure 

 was reduced, it would evidently tend to remain ventral, this 

 possibly explaining why it is found in this position in the adults 

 of most of these fishes. If in such an embryo the cranial flexure 

 were relatively late in being reduced, or, what is the same thing, 

 if the mouth became precociously functional, the perpetuation of 

 the conditions then existing might give vise to the relations of 

 trabeculfe to parachordals actually found in the adult Holocephali. 

 The eyeballs would then na.tui-ally tend to shift forward, and 

 when the head was later compressed' latero-mesially they would 

 lie dorsal to the forebrain. The other peculiarities of the 



