ZOOLOGY. 203 
of the frontal, and are chiefly to be recognized by their position 
as roofing the nares posteriorly, and their connection with the 
frontal. In one of our crania I observe that they are absent. 
The maxillary is much reduced, in connection with this position 
of the nares. It is separated from union with the prefrontal by 
the large lachrymal, which extends to both the frontal and the 
premaxillary. 
Such a determination of the bones of the roof of the cranium 
differs materially from that of Cuvier and Owen. The former 
(Ossemens Fossiles, Tab. 257, figs. 1-5, and p. 103, V, 2 plates) 
laid a wrong basis by assuming the bones (figs. 2, 3, Sq) to be the 
parietals: the parietals become then frontals, and the frontals are 
called nasals, the true nasals being entirely overlooked. Owen 
(Paleontology and Comp. Anat. Vertebrates) follows Cuvier in 
these points. Giinther falls into error as regards the squamosal 
branches in Sphenodon, uniting them with the true parietal as 
parietals. The frontals he names correctly. The parietal in Sphe- 
nodon is shown by Giinther’s figure to be a simple medial element, 
_ as in Ichthyosaurus. 
Having, however, observed a suture separating the squamosal 
from its supposed anterior plate in one young and one adult Ich- 
thyosaurus cranium, it has occurred to me that possibly the speci- 
men here described may have a coalescence of two elements really 
distinct. In that case the anterior bone will not be homologous 
with that in same position in Lystrosaurus, but may be, as usually 
stated, the parietal. The other bones in front of them would then 
retain their usual names, the supposed nasals (2) remaining with- 
out determination. 
Turning to the base of the cranium of Ichthyosaurus, we observe 
that the palatines and ectopterygoids are broad, flat bones, whose 
exterior margin is in contact with the maxillary and malar to 
opposite the posterior margin of the orbit, flooring the latter (fig. 
8, Ectp). The pterygoids, on the other hand, contract abruptly 
behind this point, and support the columella. They then expand 
to a degree unusual in the Reptilia, and extend over the whole 
space between the basioccipital and the quadrate, joining both 
closely, and projecting behind their posterior plane. Its margin is 
recurved as far as the stapes (Pt, figs. 1, 2, 3). 
The columella is very stout at its point of contact with the 
pterygoid, and above it; but higher it contracts much, and then 
expands anteriorly into the parietal branch of the ? squamosal with 
