DEVELOPMENT OF THE SKULL IN THE BATEACHIA. 
649 
on both sides, not on the left only, as in common Tadpoles ; here, with limbs such as 
usually are seen when the tail is lessening, the operculum (Plate 60. fig. 1, op.) is a 
small free flap, growing back from the angle of the gaping mouth. This operculum is 
like that seen in the “ first stage ” of the Salmon (“ Salmon’s Skull,” plate i. figs. 3 & 9, 
op.). The gills are well developed in Dactylethra ; but I had no stage in that type to 
show the external free branchiae, as in the Frog (“ Frog’s Skull,” plate iii. fig. 10). In 
Pipa , as in Salmo, there are no signs of such filaments (this embryo is immature enough 
to show them if they existed), and there are no branchiae under the operculum, although 
the arches (Plate 60. fig. 4) are all present, and the vascular arches are here also. The 
normal free branchiae exist in the unborn young of Salamandra maculosa , and are just 
aborted at birth ; here, neither kinds have been found. As in the first stage of the 
Salmon, with which these embryos are in structure and in stage so closely comparable, 
the skull is largely membranous ; it is only floored with cartilage ; the brain is quite 
visible through the delicate gelatinous roof; and the eyeballs lie loosely, as it were, 
in their sockets, as though they had been planted in them, and were not a part of the 
primary structure (Plate 60. figs. 1, 2, and “Salmon’s Skull,” plate i.). The ear-sacs, 
with the hind brain, take up half the head, so large are they, much larger than in the 
Salmon’s embryo, or in their counterpart in the Toad and Tadpole (Plate 55. fig. 3, au.). 
Moreover, in many respects, Tadpoles of the Common Frog and Toad only half the 
length of these embryos of Pipa are more advanced, and are free active individuals at 
one third their length. But in Pipa the whole metamorphosis is perfect before they 
awake ; for they lie wrapped up like unborn Lizards, not in a proper primary uterus or 
oviduct, but in a secondary dorsal marsupium. 
Notwithstanding the slight advance in the parts relating to organic life, the animal 
organs are fast advancing ; the skull (Plate 60. fig. 3), in this recently differentiated band 
of blastoderm, is well nigh as far advanced as that of the youngest Dactylethra (Plate 56. 
figs. 4, 5). 
These two “ chondrocrania ” come much nearer to each other than to those of the 
common kinds, and they mutually illustrate each other ; yet, for all that, they are very 
diverse from each other. Here, as in Dactylethra , only t\\e floor is, in this stage, solidified 
into cartilage, and in this type the basal view gives most of the morphology. In both the 
pituitary floor is chondrified earlier than in the Frog, and in Pipa it must have rapidly 
filled in. In few embryos have I seen such a long cephalic notochord (fig. 3, nc.) : it 
is half the length of the chondrocranium, in the Tadpole of Dactylethra it is one third, — 
that is, in the latter the trabecular region is twice as long as the parachordal, and in 
the embryo of Pipa they are equal. This is partly a peculiarity of the type ; but it 
shows that these embryos whose chondrocranium can be compared with that of the 
active Tadpoles of Dactylethra are very young, and that they develop with great rapidity. 
The “ first stage ” of the Pig’s skull shows a similar state of things (“ Pig’s Skull,” 
plate xxviii. fig. 8). The long parachordal tracts (Plate 60. fig. 3, iv.) are much more 
distinct from the auditory capsules than in the corresponding skull of Dactylethra ; 
mdccclxxvi. 4 x 
