152 
ERIK A : SON STENSIO 
may, according to our present knowledge, venture to estimate its length in large spe¬ 
cimens at between i and 2 m. 
The primordial neurocranium and its ossifications. 
The best and most completely preserved neurocranium (P. 165; text fig. 61; PI. 20, 
fig. 6; PI. 22, figs, i, 3 ; PI. 24, fig. 2) was probably about 11 —12 cm wide at its widest 
part. Its length in its present state of preservation is only about 11 cm, but has un¬ 
doubtedly been much greater, as both the occipital and the ethmoidal are lacking. When 
nothing to the contrary is stated the following description is based on specimen P. 165. 
The greatest width of the neurocranium was about between the well developed 
postorbital processes or was at least not far behind these (text fig. 61). The conditions 
of the anterior orbital boundary are impossible to state, but it is evident, however, that the 
orbits had a rather considerable size. The interorbital wall is in the wohle thick and strong. 
The labyrinth region is wide in its anterior part; the posterior part grows fairly rapidly 
narrower and in this way passes, very much narrowed, into the occipital region. The 
greatest height of the neurocranium seems to be relatively slight, only 3.5 cm., which is 
probably partly due to the state of preservation of the cranium in question, as this is 
somewhat compressed in a dorso-ventral direction. The maximum height (text fig. 62) 
is situated in the posterior part of the labyrinth region; then there follows a somewhat 
lower part at about the boundary between the labyrinth and the orbitotemporal regions, 
and then once more a somewhat higher portion in the middle part of the last-mentioned 
region, after which the height continuously decreases anteriorly. The dorsal side of the 
neurocranium slopes gently and uniformly forward and the variations in height present 
here are due, as text fig. 62 and PI. 20, fig. 6 show, to sinuations in the basal surface. 
The lateral surfaces of the labyrinth region face strongly downwards in the anterior 
part. Backwards, on the other hand, this state of affairs gradually changes, so that they 
first turn directly laterally and then, towards the boundary of the occipital region, 
somewhat backwards as well owing to the rapid narrowing that, as already pointed out, 
takes place here. The lateral surfaces of the orbito-temporal region are directed in the 
same way as the anterior part of those of the labyrinth region. 
The primordial neurocranium is to an unexpectedly great extent occupied by 
substitution bones. These are often considerably thick and massive and in several places 
they have undoubtedly corresponded to the entire thickness of the cranial wall. The 
smooth periostal lamellae that characterize the exposed surfaces on the substitution bones 
in the majority of Teleosts are here in B. mougeoti (P. 165) very thin and usually wea¬ 
thered away, so that the other, cancelleous substance appears and gives the bone a 
characteristic granulated surface. The centres of ossification, which in most cases are 
very distinct in the substitution bones of the Teleosts cannot be traced in this species. 
The macroscopic appearance of the bone substance might possibly give the impression 
of calcified cartilage. A close investigation shows, however, immediately that this cannot 
be the case, but that we are concerned here with real bone substance. This bone sub¬ 
stance is formed of thin lamellae, connected with each other and surrounding small cavities, 
which are filled with calcite, infiltrated during the process of fossilization. Afterwards, 
when the lamellae between these cavities are weathered, and the contents, which have 
