Rickards, Sherwin and Williamson (2001) have described a small graptolite fauna from New South Wales which they interpret to be Gisbornian (Caradoc), and provide evolutionary schemes for several lineages (see also Sherwin and Rickards 2000) in which the precise age is crucial. My study of more diverse Late Ordovician graptolite faunas in more complete successions indicates that the age of their fauna is considerably younger, most likely Bolindian Bo1. The evolutionary lineages that Rickards et al. propose are therefore implausible. Rickards et al. (2001) consider the possibility that the fauna is 'bimodal', containing both late Gisbornian and mid-to late Eastonian, verging on Bolindian, elements, but opt for a Gisbornian age because the younger forms are poorly documented. The Gisbornian age is based on identification of two forms: Climacograptus bicornis, which does not occur in beds younger than Gisbornian in Australia (VandenBerg 1990;VandenBerg and Cooper 1992) and disappears at the same level elsewhere, and Orthograptus calcaratus calcaratus, with probably a similar range.
Climacograptus bicornis is one of three 'climacograptids' in the Upper Ordovician that bear curved basal spines-Climacograptus bicornis (J. Hall), Diplacanthograptus spiniferus (Ruedemann) and Appendispinograptus longispinus (T.S. Hall). All three are very useful because their ranges do not overlap, they are worldwide, and all are quite abundant. However, they are often misidentified, and all three are listed as Climacograptus bicornis in almost all of the Australian literature prior to 1981. Publication of detailed studies of all three by Riva (1974Riva ( , 1976) ) has shown the clear stratigraphic and morphological distinction between them. Reference to Riva's work indicates that the specimens identified as Climacograptus bicornis by Rickards et al. (2001) in fact belong to Appendispinograptus longispinus longispinus. Points of difference between the two are as follows. The rate of widening in the proximal region is different, resulting in a steadily tapering rhabdosome in C. bicornis, but a spindle-shaped rhabdosome in A. longispinus (well shown in their figure 5G). Thecal apertures are always very clearly visible in C. bicornis but smaller and often difficult to observe in A. longispinus (e.g. figures 5G, H, I, J, L, 8D). The proximal spines are, in comparable growth stages, much smaller in C. bicornis and secondary thickening shows quite a different pattern (Riva 1976).
The form identified as Orthograptus calcaratus calcaratus shows several characters that suggest it is not a member of the O. calcaratus group. Its proximal end is very slender, and the rhabdosome widens more gradually than in O. calcaratus (see Elles and Wood 1907). These characters indicate that the Cheesemans Creek specimens belong to Orthograptus thorsteinssoni (Melchin 1987). An added point of distinction between the two is that all members