BRITISH AVONIAN CONODONT FAUNAS 59 
Australia one of us (in Jones & Druce 1966) has recently discovered a series of 
abundant conodont faunas extending through some 3,000 ft. of Lower Carboniferous 
strata. There is a striking resemblance between the Australian and the Avonian 
conodont successions, and although Siphonodella is abundant in the Australian 
faunas, it does not extend up as high as the Spathognathodus costatus group (Fig. 
16). This could mean that all the Siphonodella zones of the Mississippi Valley lie 
below the first occurrence of S. costatus in the Avonian, and would therefore be of K 
age. 
The other implication of such a correlation would be that the North American and 
the German successions are more incomplete than has formerly been supposed, for it 
would be difficult to find any North American equivalent of the two overlying 
Avonian assemblage zones, if the correlation of the Avonian Polygnathus lacinatus 
Zone and the Mississippian Gnathodus semiglaber—Pseudopolygnathus multistriatus 
Zone (which seems to us to be very well established, p. 57) is accepted. This would 
suggest that the unconformity known to exist beneath the Sedalia Formation of the 
Mississippi Valley, is of greater magnitude than is generally assumed. 
There may not be such a break in the Oklahoma or parts of the Missouri successions 
(Cooper 1939 ; Branson & Mehl 1934A : 265). The Rockford Limestone fauna 
(Rexroad & Scott 1964) reveals an apparent transition in northern Indiana from the 
Siphonodella isosticha—S. cooperi Zone into the Gnathodus semiglaber—Pseudopoly- 
gnathus multistriatus Zone, however, and represents an anomaly if the present 
correlation is accepted, although most of the sections come from cores. There is, 
however, an unconformity in southern Indiana between the Rockford and the over- 
lying New Providence Shale. Even in areas of apparent transition of the two 
assemblage zones within the Rockford Limestone, the faunal transition is abrupt 
(Rexroad & Scott 1964, Table 1). Rexroad & Scott (1964 : 16-17) have written 
“ Although a number of species are common . . . this break between the Kinderhook 
and Osage Series is unusually well marked. The faunal break is sharp, but a number 
of species confined to the lower zone gradually decrease in number upward. Thus an 
unconformity within the Rockford at the Kinderhook—Osage boundary is not 
necessarily indicated. At the type section near Rockford (locality 10) the formation 
is exposed only in the bed of the East Fork of White River, and even at low water the 
evidence relating to a possible unconformity cannot be interpreted.” 
It may also be that the overlap of the Siphonodella and anchoralis faunas in 
Germany implies a stratigraphic break between them, in which case part of the 
Upper Cu IT « and Lower Cu II 8 could be unrepresented. One of the major problems 
concerning the anchoralis fauna is its patchy geographic distribution (p. 65). It is 
possible that its limited distribution in North America (it is recorded from the 
Pierson Formation of Missouri and the “ Sedalia’”’ of Illinois, above the uncon- 
formity) may mean that in places it is represented by the unconformity which we 
have postulated. 
This alternative correlation has stressed the inadequate numbers of siphonodellids 
in British Lower Carboniferous faunas, and its weakest point is the lack of a more 
detailed series of species of this genus in the Avonian. The first provisional correla- 
