26 
Records of the Australian Museum (2009) Vol. 61 
tooth-like denticles (Fig. 13) along their anterior and lateral 
margins, and by blunt node-like denticles along the posterior 
margins (Fig. 14D). Underneath each specimen is a wide, 
smooth, distally arched basal face (Fig. 12A, G), which is 
crescentic in outline and divided anteroposteriorly into two 
lobes (lateral processes; Fig. 13 A-E, K) by a broad, median 
groove. Typically a small ring-like node underneath the cusp 
represents the initial stage of the element (Figs 13N, 14E), 
without basal cavity. 
Natural breakage surfaces on some of the specimens show 
these laminar layers to be tightly compacted (Fig. 12G, H) 
with partitions generally observed only near the edge of 
each layer (Fig. 12E-H). In most cases, no microstructure 
is observable on the surface of the breakage. However, some 
specimens show fine laminations (growth lamellae) formed 
by flattened crystallites (Fig. 121). 
Serratognathus extensus Yang in An et al., 1983 differs 
from S. diversus in having a more robust cusp and a long, 
laterally extended process on each side. This species is 
similar to S. diversus and S. bilobatus in that the anterior 
face bears 2-3 rows of denticles that extend continuously 
from the end of one lateral process to the other, and in lack 
of a basal cavity. Yang’s type material included 14 specimens 
from the Liangjiashan Formation in Hebei Province (An et 
al., 1983, tables 6-7). An et al. (1983, p. 26) suggested that 
S. extensus might be directly evolved from S. bilobatus, and 
recognized the S. extensus Zone succeeding the S. bilobatus 
Zone in the Liangjiashan Formation. 
Tropodus Kennedy, 1980 
Chionoconus Smith, 1991: 22. 
Type species. Tropodus comptus (Branson & Mehl, 1933). 
Remarks. Tropodus Kennedy, 1980 is treated herein as 
a valid genus, having a rather different species apparatus 
as that of Acodus, particularly its S elements that exhibit 
a much wider variation characterized by the occurrence 
of multi-costate elements, and the P elements typically 
with weaker development of a lateral costa that may be 
represented by a broad carina. Tropodus was proposed to 
consist of a bimembrate apparatus (comprising comptiform 
and pseudoquadratiform elements) with Paltodus comptus 
Branson & Mehl, 1933 as the type species (Kennedy, 1980). 
Although the genus was originally defined as consisting of 
a symmetry transition series of elements bearing “three or 
more, prominent keel-like costae” (Kennedy, 1980, p. 65), 
both elements of his revised T. comptus have five costae, 
likely representing only part of a species apparatus. The 
comptiform element (= Sc herein) was represented by the 
form species P. comptus, an asymmetrical element with 
costate anterior and posterior margins, and with two costae 
on the outer face and one on the inner face (Kennedy, 1980, 
pi. 2, figs 21-24), and the pseudoquadratiform element (= 
Sd herein) by the form species Scolopodus pseudoquadra- 
tus Branson & Mehl, 1933, a symmetrical quinquicostate 
element with a broad anterior face and a costate posterior 
margin (Kennedy, 1980, pi. 2, figs 25-27). 
The other species originally included in Tropodus was 
Walliserodus australis which was defined as consisting of a 
transitional series from tricostate to multicostate elements 
(Serpagli, 1974, p. 89). Kennedy (1980) admitted that “the 
two types of elements in T. comptus are very similar to 
two of the many morphologies of elements in T. australis 
(Serpagli, 1974)” although the latter presented “a multitude 
of variably costate forms.” Serpagli (1974) originally defined 
a quinquimembrate apparatus for T. australis including 
asymmetrical tricostate (= Sbl herein), asymmetrical quadri- 
costate (= Sb2 herein), strongly asymmetrical, laterally com¬ 
pressed multicostate (= Sc herein) and a nearly symmetrical 
quinquicostate (= Sd) elements; Sa, P and M elements were 
originally not recognized in either species. 
Based on material from the Cow Head Group of 
Newfoundland, Bagnoli et al. (1988) and Stouge & Bagnoli 
(1988) included P and M elements in the species apparatus 
of Tropodus. Stouge & Bagnoli (1988) also revised both 
T. comptus and T. sweeti (Serpagli, 1974), and suggested 
that W. australis proposed by Serpagli (1974) comprised 
elements belonging to both T. comptus and T. sweeti. As 
Smith (1991) correctly pointed out, distinguishing between 
S elements belonging to these species is rather uncertain, 
as Stouge & Bagnoli (1988) had included the holotype 
(tricostate element) of W. australis in the synonymy lists 
of both T. comptus and T. sweeti. However, the P elements 
they defined for T. comptus and T. sweeti show remarkable 
differences. The P elements of T. comptus illustrated from 
the Cow Head Group (Stouge & Bagnoli, 1988, pi. 16, fig. 
2) and also from Utah (Ethington & Clark, 1982, text-fig. 
22, pi. 11, figs 6, 7) are more or less scandodiform bearing 
a smooth, convex outer face and a concave inner face with a 
broad, prominent mid carina, while the P element of T. sweeti 
is a typical acodiform element with a prominent costa on one 
side (Serpagli, 1974, pi. 14, figs 13, 14, pi. 24, figs 8-10). 
Smith (1991) restricted Tropodus to Kennedy’s original 
definition by including only costate elements in the species 
apparatus and proposed Chionoconus to accommodate those 
elements which Stouge & Bagnoli (1988) defined as the P 
elements of T. comptus. 
Zhen et al. (2004) reported the occurrence of T. Australis 
(assigned to T. comptus ) and T. sweeti from two samples 
within the Early Ordovician Hensleigh Siltstone in central 
New South Wales. The P and M elements of both these species 
are comparable with those described by Serpagli (1974) from 
the San Juan Formation of Precordilleran Argentina and by 
Stouge & Bagnoli (1988) from Newfoundland. However, 
Zhen et al. (2004), following Ji & Barnes (1994), applied 
a rather different concept for the S elements by including 
a symmetrical Sa element in the apparatuses. These studies 
also raised uncertainties about the definition of the constitu¬ 
ent species of Tropodus. Firstly, if P elements of Tropodus are 
confined to “scandodiform” elements as Stouge & Bagnoli 
(1988) originally proposed, T. sweeti should be excluded 
from it. Secondly, although Kennedy (1980) suggested that 
Triangulodus consisted of hyaline elements, if Tropodus is 
considered as bearing “scandodiform” P elements, it would 
be very similar to Triangulodus, if it is not considered as a 
junior synonym of the latter. Among the specimens from the 
Hensleigh Siltstone, two types of P elements were recognized 
for T. australis (Zhen et al., 2004). Both Pa and Pb elements 
