Response to comment on Ward et al.’s ‘Insights into the procurement and distribution of fossiliferous chert artefacts across southern Australia from the archival record’

We appreciate the opportunity to respond to the arguments put forward by Bird et al. against the premise of a long-distance source and trade of bryozoan fossiliferous chert across southern Australia. Given the long-standing enigma of fossiliferous chert artefacts and their apparent offshore source, it is appropriate for there to be some debate when this enigma is challenged. However, it is difficult to understand why Bird et al. ignore the geological evidence that indicates unequivocally that the source of fossiliferous chert cannot be from the Perth Basin (O’Leary et al. 2017), and offer no alternative source. Bird et al. themselves seem to acknowledge, with reference to Glover (1975a), that ‘no local sources [of fossiliferous chert] are known, but it most closely resembles chert from the Eucla area’. As noted in O’Leary et al. (2017:37), the idea of a transport pathway of Eocene-age fossiliferous chert along the south coast (from Eucla) was first proposed by Glover and Cockbain (1971). Only after petroleum exploration wells were drilled on the Rottnest Shelf, which contained bands of fossiliferous chert, did Glover (1975a, 1975b) and Quilty (1978) opt for an offshore source. This change in thinking was considered to account for the apparent westward increase in frequency of chert artefacts, absence of a suitable local onshore chert source, and absence of chert artefacts in strata younger than 4.5 ka. The latter was attributed to an elimination of source following post-glacial flooding of the continental shelf. Yet an offshore source in the Perth Basin remains unlikely given that the well data show chert bands in Eocene to Miocene age formations (a similar age to the chert deposits on the Nullarbor) at depths of 50 400m below lowest sea levels at the Last Glacial Maximum (LGM). Critically, chert can only form under pressure from burial. There are extensive supporting datasets including geological (e.g. borehole), geochronological (Glover and Cockbain 1971), geophysical (e.g. shallow seismic) and neotectonic evidence that show the surficial and shallow subsurface sediments of the Rottnest Shelf consist of Pleistocene marine calcarenites. It is a geological impossibility for in situ Eocene chert deposits to exist at or just below the seabed on the Rottnest Shelf where it could be accessed as a resource. The main part of Bird et al.’s argument revolves around the distance-from-material-source concept, namely that raw material distribution declines with increasing distance from source. While this decline may exist for local Plantagenet chert, this trend (effect) and the various processes that are involved in making it (cause) may not hold up when considering material sources over distances of hundreds of kilometres where research and preservation bias are significant factors. Even within the Perth region, Bird et al.’s figure highlights the distribution of sites with fossiliferous chert in a broad arc around the Perth floodplain. This mirrors the distribution of archaeological sites generally, with the vast majority, whether a result of research bias or preservation, associated with the Bassendean sand (Bowdler et al. 1991). The archaeological record is largely based on material remains being created, preserved and found. Thus, an easterly decline in archaeological material may reflect one or more of the following aspects: