1896. ] NEW YORK ACADEMY OF SCIENCES. 103 
MARSIPOBRANCHIAN CHARACTERS OF PALAOSPONDYLUS. 
Positive. Negative. 
Tentacles Similar to those of the naso- Not necessarily marsipo- 
of head. mouth region of Bdellostoma, branchian; mouth surround- 
/or even to the buccal cirrhi of ing tentacles evolved independ- 
_Amphioxus Apparently as ently in some of the larger 
_many as eleven or twelve were groups of fishes, Sharks, and 
_ present. many Teleostomes, e. g., Silu- 
| roids, Pogonias, Hemitripterus. 
| Moreover it is possible that the 
‘ventral cirrhi’’ are displaced 
structures from the cranial re- 
gion, as one of the specimens 
examined by the present writer 
seems to indicate. 
Jaw parts. Unknown. Unknown. 
She Ta Sf Chondrified, massive, filling 
the entire head, this with the 
huge auditory capsules, is far too 
large in size for any known 
Marsipobranch. 
Vertebral —= Highly specialized, massive 
column. centra, with differentiated 
neural arches, entirely un-cy- 
_ clostomian. 
Cauda | Essentially marsipobranchian. But its diphycercal (or per- 
fin. | haps heterocercal) condition 
also common to many groups, 
Shark, Lung-fish, Teleostome. 
Paired ——— | Decidedly non-marsipobran- 
fins. _chian: also the ‘‘ post-occipital 
plates’ which might well rep- 
resent basalia of pectoral fins. 
| 
It will thus be seen that the only character which Palzospon- 
dylus retains allying it with the Cyclostomes is the presence of 
tentacles in the anterior head region, a strong ground of kinship 
be it admitted. But in view of the many un-marsipobranchian 
features, especially if the presence of paired fins be added, can 
the head-tentacles be taken to be in any way a crucial test of 
kinship? Or, on the other hand, would it not be more reason- 
able to believe that these structures, if they be a ring of ten- 
tacles, arose independently within the group to which Palzeospon- 
dylus belongs ? 
The position of the fossil, if not to be regarded as marsipo- 
branchian, is certainly undefinable. One is loth to accept the hy- 
