TUE SKULL OF CniM.TDUA. 135 



undergoing cliondritication. The presence of these three ligu- 

 - ments connecting the three rostral processes of Cldmcera, and 

 their possible chondrification, thus suggest that they and the 

 processes together represent a rostral basket similar to that 

 found in the Oarchariidpe and Scylliidae, but everted. In 

 certain specimens of Mastelus I find similar ligamentous or 

 fibrous strands connecting the median and lateral limbs of the 

 rostral basket, and if such a basket were to be everted, by 

 the lateral limbs passing antero-ventrally external to the median 

 limb, the conditions found in Chhnmra would apparently arise ; 

 the lateral processes of Chimcer-a thus representing the basal 

 portions of the lateral limbs of the basket of the Selachii, those 

 portions of those limbs being attached to the nasal capsules as are 

 the processes in my specimens of Chvmcera. 



Garman says that, in Chwutira monstrosa^ the median rostral 

 process " presents the appearance of having oi-iginally been 

 attached near the nasal capsule, as in CallorJiynchus, and of 

 having the basal portion, for a short distance, brought back 

 against and fused with the frontal region of the skull." This is 

 certainly also the appearance presented by the process in Chimcera 

 colliei ; but it is probable that this apparent folding backward of 

 this process is due to the marked antei'o-ventral growth of the 

 anterior portion of the naso-ethmoidal cartilage, the rostral stalk 

 being carried forward bej'ond the base of the median rostral 

 process. Comparison of Schauinsland's figure 125 of Callorhynchus, 

 giving a median vertical sectional view of an embryo of Callo- 

 rhynchus, with the accompanying figure 2 (PI. I.), giving a similar 

 view of the adult Chimcera colliei, will show that this must be so. 

 Comparison with Sewertzoff's figures of embryos of Fristiurus, 

 and with Dean's figures of embryos of Chimcera colliei, then 

 further shows that, because of the antei-o-ventral growth of this 

 part of the chondrocranium, the antero-ventral half of the 

 fenestra pi^aecerebralis has been carried forward on either side 

 of the median rostral process, that the edges of the fenestra 

 anterior to the process have then fused with each other, and that 

 posterior to the rostral process the fenestra has been closed by 

 lateral compression of the cranium, the ethmoidal canals of oppo- 

 site sides coalescing in the median line and their floors forming 

 the roof of the cranial cavity and their roofs a secondary roof to 

 the ci-anium. The fenestra prjecerebralis is thus completely closed 

 in the adult Chimcera, but in embryos of Callorhynchus a part of 

 it still persists as the opening t of Schauinsland's figures. 



A stout ligament arises from the dorsal sur-face of the chondro- 

 cranium at the posterior edge of the nasal capsule, there Ij'ing 

 postero-lateral to the corresponding lateral rostral process. This 

 ligament lies at first at right angles to the chondrocranium, and 

 as it is stout and stiff", it looks like a short ligamentous horn 

 which extends to the intei'nal svirface of the fibrous layer of the 

 corium above referred to as being related to the tubules of the 

 ampuUary organs and to be described immediately below. There 



