190 Balanogilossus. [ZOE 
ticulars in which its structure and development resemble the typi- 
cal chordate: These are as follows: Of the first class, (1) the posi- 
tion and origin of the central nervous system; (2) the possession 
of anotochord; (3) the possession, method of origin and arrange- 
ment of gill-slits. Of the second class, (1) the origin of the meso- 
blast—the middle germ layer; (2) the asymmetry of the anterior 
parts; (3) the opercular fold; (4) the excretory funnels opening 
into the atrial cavity. It must serve our present purpose to con- 
sider the three primary features here enumerated; the secondary 
ones must be passed by with some general statements, merely. 
All vertebrates are characterized by the possession, at least in 
embryonic life, of a notochord arising from the dorsal portion of the 
-_ primitive digestive tract, and extending parallel with the long axis - 
_of the animal; by the possession of a cerebro-spinal nerve axis that 
arises from the ectoderm of the dorsal. portion of the embryo, and 
extends parallel with the notochord along its dorsal side; by the 
possession of paired respiratory organs that arise from the anterior 
portion of the digestive tube and communicate with the external 
world, either through the mouth or independently of it; and by the 
possession of a large median dorsal blood vessel situated between 
the digestive tract and the notochord, in which the blood flow from 
before backward. 
To all these fundamental features Balanoglossus certainly presents 
some remarkable resemblances. The nerve cord arises from the 
dorsal portion of the ectoderm of the embryo by a process that is 
quite. similar to that by which the same structure arises in many 
of the fishes, as the lamprey and the bony fishes. However, cer- 
tain important differences must not be disregarded. 
In all vertebrates the posterior end of the medullary plate—the 
nerve cord in its early stage—terminates at the blastopore, while in 
Balanoglossus, it does not extend so far back; in fact the portion of 
it that seems most nearly to resemble the vertebrate cord is ap- 
parently confined to the collar, while the anus is situated at the ex- 
treme posterior end of the animal. Furthermore it seems quite 
doubtful if the canal, or space that finally appears in the cord of 
Balanoglossus, is in any sense morphologically comparable to the 
vertebrate neural canal. 
Again, it is to be observed that in all vertebrates, even including 
Amphioxus, the nerve cord is encased in a connective-tissue or _ 
