484 
MR. W. K. PARKER OK THE DEVELOPMENT 
peculiarly simple foundation for all the aftergrowths. In Stage 1, in embryos 
1 Ogr m.m. long, nearly all the parts of the chondrocranium—including in this term 
the visceral arches—are present; the hinder arches become broken up, afterwards, 
but the two first, and largest, the mandible and hyoid, are already as much segmented 
as they will he in the adult. 
The skull-floor, only, is developed, as yet, and the rostral part, in front, is not 
chondrified, but its outlines can be traced, and the roof and walls of the skull are 
merely developments from the basal bands. 
Those bands in this type lend no support to the theory of the visceral (or ventral) 
nature of the 'pro- chordal tracts or trabeculce ; they are, manifestly, mere continuations 
of the undivided para- chordal cartilages, which expand and contract in relation to 
the parts around and over them. They diverge from the front third of the notochord, 
as though their relation to it was not intimate, and show—for a long while at least—• 
no tendency to grow up, with that axis, into the hollow of the mid brain. 
I see nothing in this lyriform basal skeleton of the skull but an undivided basi- 
neural structure comparable to, and a primary cephalic variation of, the tracts that 
form the paired rudiments of the neuro-central cartilages of the spine. The cessation 
at the end of the notochord (mesially), and close behind the oral opening, laterally, 
of the hypo-blastic layer, causes all the pre-oral and pre-pituitary parts to be, in a 
sense, imperfect; they are developed as porches and outworks to the full and complete 
structure further back, but this does not destroy their homology, nor break their 
continuity with the parts formed from their own embryonic layer, of which they are 
the direct ongrowths. 
Yet all parts growing out,—forwards, upwards, or downwards,—in front of the perfect 
axis, which ends close in front of the infundibulum, must be very cautiously named 
as “ serial homologues ” of the perfect base and its upper and lower arched growths ; 
they are probably mere outgroivths; at most they are only rudiments. 
The primary trabeculae are merely direct on-growths of the parachordals ; the 
cornua trabeculae are ow£-growths of the trabeculae. 
The intermediate element, or intertrabecula, is a fresh outbreak, so to speak, of 
the median mesoblast of the axis, which is tubular, behind, where it encloses the 
notochord, but, re-appearing in front, beyond it, it shoots forth as a solid process of 
the skeletal axis. 
Close to the fore end of the primary trabeculae there arises a similar but rather 
smaller bar, and the two parts are so close together that they chondrify continuously; 
these side bars are the palatine cartilages, evidently rudimentary structures. 
Here they are not distinct from the long spur (pterygoid cartilage), which shoots 
forwards from the dorsal element (suspensorium) of the mandibular arch; this is like 
what we see in the Tadpole, but unlike that which is found in Skates, Teleosteans, 
and Urodeles. Thus, with the palatine included, the suspensorium here is a palato- 
guadratc; in the Skate, Teleostean, and Urodele the suspensorium is a pterygo- 
