490 Dr. R. H. Traquair—On the “ Dendrodont’”? Fishes. 
and following a general scheme for a Magnetic Survey of the whole 
globe, and united efforts might considerably reduce the labour 
involved in so great an undertaking. It is very desirable that such 
a systematic and complete survey should be taken in hand, instead of 
giving particular attention to limited districts, and I solicit the help 
of those interested in the subject, in starting an undertaking in which 
we may be sure that all nations will gladly join. 
(Lo be concluded in our next Number.) 
IV.—On rue Systematic Posrrion oF THE “DENDRODONT” FISHES. 
By Dr. R. H. Traquatr, F.R.S., F.G.S. : 
N a short paper on the nomenclature of the Old Red Sandstone 
Fishes published in this Magazine for November, 1888, I 
expressed the opinion ‘that the scattered teeth and fragments of 
jaws known as Dendrodus and Lamnodus belong to fishes at 
present known to us by their scales as species of Holoptychius and 
Glyptolepis.” The family terms “ Holoptychiide” and ‘ Dendro- 
dontides” I consider absolutely synonymous. On the other hand, 
the Rhizodontide (Gyroptychius, Tristichopterus, Rhizodus, ete.) 
present a somewhat different form of tooth-structure, and one which 
is, in the main, identical with that which, in so many Stegocephalous 
Amphibia, is called “ labyrinthodont.” 
The reasons in support of this opinion are something considerably 
beyond the region of conjecture, —they amount to positive proof. 
As far back as 1849, Hugh Miller? figured portions of jaws with 
teeth, as well as microscopic sections of teeth from Thurso, which 
are undoubtedly referable to Owen’s genus Dendrodus, both as regards 
external configuration and internal structure. Of this Miller was 
aware, though he refers them to Asterolepis, Hichwald, along with 
the large cranial shields and other remains of a great coccostean fish, 
Homosteus of Asmuss. Now nothing can be more certain than that 
these dendrodont teeth from Thurso belong to a large species of 
Glyptolepis, in more than one head of which, belonging to the 
Edinburgh Museum, they may be seen in situ. This Glyptolepis is 
also identical with Agassiz’s “ Platygnathus” paucidens. 
As regards the very closely allied Upper Old Red genus Holopiychius, 
I have not obtained any microscopic sections of teeth found in situ; 
but, to judge from the external characters of these teeth, as seen 
through a good hand-lens, it is impossible to doubt their dendrodont 
nature. And finally the portions of jaws which occur displaying 
teeth of “* Dendrodus” are undoubtedly Holoptychian in their con- 
figuration. 
Now, the occurrence in the Holoptychiide: of pectoral fins dis- 
playing the ‘“archipterygeal” configuration clearly enough shows 
that there must be some genetic connection between this family 
and the Dipnoi. I was therefore interested to read in the introduction 
to a recent paper on this subject? by Dr. J. V. Rohon that the 
1 Footprints of the Creator, first edition, 1849, figs. 30, 31, 32 and 33. 
* Die Dendrodonts des devonischen Systems in Russland, Mem. Acad. Imp. 
Sc. St. Petersbourg (vii.) vol. xxxvi. No. 14, 1889. 
