85 
Observations .—From the foregoing measurements and the accompanying 
figures, it is seen that the spines show some variation in length, curvature 
and proportionate breadth ; but the general character leads one to infer that 
they all belong to one species. The ornament, also, differs to some extent, 
the longitudinal ridges in certain specimens being very pronounced owing 
to slight decortication, whilst in others the ridges are thicker. All the 
examples, however, show the distinctive character of the knotting of the 
ridges at short intervals. 
The present genus was instituted by McCoy to include some small to 
moderate-sized spines found in the Lower Carboniferous limestone of Armagh, 
Ireland. The present species, P. micracanthus, shows certain related 
characters with P. arcuatus , McCqy 1 . The spines of P. micracanthus are 
curved, as in P. arcuatus , but are not nearly so large, nor are they so pro¬ 
portionately broad at the base. 
In Dre'payiacanilius (—Physonemus ), Newbery and Worthen 2 have noticed 
characters which they regard as evidence for the forward direction of the 
spines of this type ; but judging from the general arrangement of dermal 
processes on fishes, which are always conducive to their facile move¬ 
ment, this seems somewhat improbable. 
Another genus of ichthyodorulites or fish-spines closely approaching the 
specimens now described, is Asteroptychins, McCoy (Agassiz MS.) 3 . The 
spines of' this type, however, are furnished with marginal denticles and 
striated intercostal surface. 
The ichthyodorulites are summarized by A. S. Woodward 4 5 in five classes, 
Physonemus being placed in the third group defined as “ Paired spines, of 
which some may have been placed in front of fins, but of which many are 
broad, with insignificant base of insertion, and must have been arranged 
as independent dermal armour.” 
The exact morphological relationship of these spines from the Grampian 
Sandstones, like other species of the genus Physonemus , is doubtful and 
must await further evidence. It is a distinctly Lower Carboniferous type 
of fish, which by its association in England, Ireland and North America, 
with teeth of genera belonging to the Cochliodontidce , may yet turn out to 
be more closely related to that family than we yet know. Physonemus 
has indeed already been placed in the Cochliodont family by Jaekel ; whilst 
Smith Woodward regards that family as nearly related to the CestraciontidfS 
(example, Port Jackson sharks). In view of the importance of this 
occurrence in Victoria, and in order to secure further evidence of this in¬ 
teresting fauna, a renewed search should be made ,in this locality at the 
earliest opportunity. 
Locality. —The holotype (No. 16076) comes from Mount Rosea, the para- 
type (No. 12833) from near Hall’s Gap. 
Physonemus attenuatus, Davis . 6 
Plate. V. Fig. 4. 
Observations. —Although the present specimen is much smaller than that 
from the Lower Carboniferous limestone of Armagh, described by J. W. 
Davis, yet in the broad form of the base and comparative straightness of 
the exsert portion, it approaches more nearly to this than to any other 
described species. Davis’s specimen is 6 inches in length, whereas the 
Victorian example is not quite one and a quarter inches long. The Vic¬ 
torian specimen is distinguished from the previously described P. micra¬ 
canthus, both in outward form and build, and also in internal structure. 
1. Ann. Mag. Nat. Hist., ser. 2, vol. IT., 1848, p. 117. Brit. Pal. Foss. 1855, p. 638, pi. III., I., Fig. 29 ; 
pi. III., K., Fig. 25. Davis, J. W., Trans. B. Dubl. Soc., ser. 2, vol. I., 1883, p. 367, pi. XLVII.,Figs. 8, 8a. 
2. Pal. Illinois, vol. II., 1866, p. 121. 
3. Ann. Mag. Nat. Hist., ser. 2, vol. II., 1848, p. 118. 
4. Cat. Foss. Fishes. Brit. Mus., pt. II., 1891, p. 93. 
5. Physonemus attenuatus, Davis, 1883, Trans. B. Dublin Soc., Ber. 2, vol. I., p. 369, pi. XLVII. 
Fig. 10. 
