8 o 
ERIK A : SON STENSIO 
As is shown by PI. 9, fig. 2, the neural and haemal arches are of the usual Coe- 
lacanthid type. The neural and haemal spines are long and narrow, and are attached to 
the lepidotrichia of the caudal fin in the usual way (Schmalhaussen, igi3a, pp. 47 — 59X 
While in the anterior caudal region these spines form, both dorsally and ventrally, a 
somewhat big angle with the chorda, they are so inclined backwards that this angle 
gradually becomes more acute. The most posterior of them consequently take up a 
position almost parallel to the longitudinal axis of the fish. Their distal free segment 
about corresponds in length to the proximal one, which was continuously connected 
with the arch. Both in the neighbourhood of and at the place where the two segments 
are connected with each other they are somewhat thickened. The bone substance in 
them only forms a thin perichondral layer. The interior undoubtedly consisted of cartilage 
and both the distal end of the proximal segment and the two ends of the distal segment 
were furnished with a cartilaginous epiphysis. 
The caudal fin is not completely preserved at its posterior part and it is thus 
uncertain if the so-called supplementary fin («Pinselflosse») has been developed. In the 
other parts the lepidotrichia are long and fairly robust but not expanded distally. Their 
number is about the same dorsally and ventrally — as far as one can judge about 
twenty. On one side — it is not possible to say if it is the dorsal or the ventral'one — 
they are distinctly more robust than on the other, and, except for the three or four 
most anterior ones, they are jointed, although the proximal joints are of considerable 
length. J ) Even the distal joints are fairly long, which is also the case in the lepidotrichia 
on the opposite side, but the latter lepidotrichia, on the other hand, have long unjointed 
proximal parts. With the exception of a piece of the proximal part all the lepidotrichia 
of the caudal fin have on their lateral sides a row of fine, short striae, or, if the 
expression is preferred, a row of somewhat lengthened tubercles. 
The scales (PI. 9, fig. 3 ) seem most closely to resemble those of W. sinuosa in their 
sculpture. As in this species, the sculpture consists of a rather large number of fine, 
but not specially long, striae. A few of the middle striae on each scale run, as in 
W. sinuosa, fairly horizontally in the longitudinal direction of the scale, the rest, situated 
dorsally and ventrally of these, show a certain concentric arrangement along the posterior 
margin and consequently their posterior ends are directed towards the longitudinal axis 
of the scale. — Nothing is known in detail as to the shape of the scales. 
* * 
Remarks: With regard to the sculpture of the scales the two Goela'canthid remains 
just described resemble, as has already been pointed out, W. sinuosa more closely than 
any of the other Spitzbergen species so far known. At the same time the deviations 
are very obvious, and as in other respects we are without any points of departure for 
a comparison, it is impossible at present to decide whether these two fragments really 
belong to W. sinuosa or the genus Wimania at all. Under these circumstances I have 
only hesitatingly included them in this genus. 
Geological occurrence and localities. — The fish horizon at Mt Milne Edwards 
and Mt Trident. 
z ) The material at hand does not show whether these rays too have been longer than those on the 
opposite side. 
