Reply

Arp et al. bring a geomicrobiological dimension to bear on ancient microbially mediated calcium carbonate precipitation. The basic bone of contention is whether this precipitation was triggered by the metabolism—photosynthesis—of the cyanobacteria presumed to be involved, or took place in dead material during degradation. I have been struggling with this issue since finding calcified filaments in Holocene stromatolitic intraclasts (Pratt, 1979). The notion of postmortem calcification that I entertained for Girvanella (see Pratt, 1984, 1995, for Tubiphytes and the Epiphyton-Renalcis plexus) was based purely on circumstantial, petrographic evidence.

Arp et al. submit that the polysaccharide-dominated composition of cyanobacterial sheaths is not conducive to CaCO₃ precipitation during postmortem aerobic degradation by bacteria because released CO₂ would lower CaCO₃ supersaturation. By contrast, von Knorre and Krumbein (2000) suggested that this CO₂ buffers pH instead by forming and any surplus escapes to the atmosphere. The instructive examples documented by Défarge et al. (1996) and Sprachta et al. (2001) show that precipitation in fact does occur in decaying sheaths, in the latter case with the potential to replicate them.

Arp et al. claim that Chafetz and Buczinski's (1992) experiments using seawater enriched in peptone do not reproduce the appropriate chemistry needed to trigger CaCO₃ precipitation in exopolysaccharides. While this may be true, similar experiments with acetate-enriched seawater performed by von Knorre and Krumbein (2000) also resulted in precipitation. It should be remembered as well that although carbohydrates dominate the hydrated polymers comprising cyanobacterial sheaths, uronic acids and other organic constituents, plus adsorbed organic compounds, are also present. Furthermore, the decay products of cyanobacterial cell contents and the heterotrophic bacteria involved complicate the biofilm microenvironment and increase the number of reactive carboxyl and hydroxyl functional groups that are the binding ligands for cations (e.g., Decho, 1990).

Arp et al. may well be correct that Girvanella formed by in vivo calcification of sheaths, and I can accept this for tubular specimens exhibiting finely calcified walls. However, Bartley (1996) showed that cyanobacterial sheaths remain reasonably intact for quite some time after death (i.e., 1–2 weeks). In addition, the excess Ca²⁺ and/or Mg²⁺ in seawater may help maintain the structural integrity of dead or abandoned sheaths for longer than might be expected because these ions form cross-links between adjacent sugar polymers (Decho, 1990). It is also conceivable that while the living filaments likely possessed inhibitors to spontaneous precipitation (Westbroek et al., 1994), their active photosynthesis might have increased [] enough to promote precipitation in closely associated dead material. In any case, my original interpretation that “permineralization…generally took place postmortem” (Pratt, 2001, p. 764) does not exclude a role for in vivo calcification. Furthermore, both could have operated in some cases, because crystal nucleation and crystal growth can be two distinct stages (see also Sprachta et al., 2001). The fact is that Girvanella is commonly poorly defined, grading into micrite threads and clots, and this range of preservation is difficult to explain by photosynthetically induced precipitation in living sheaths.

Arp et al. reiterate the case they made while my paper was in press (see Arp et al., 2001) in which they calculated that impregnation of cyanobacterial sheaths by CaCO₃ would occur in seawater only if it were more than tenfold supersaturated with respect to calcite, and when photosynthesis further increases [] within the sheath microenvironment. Based on a higher atmospheric pCO₂ believed to have characterized the Paleozoic, seawater needed to be considerably enriched in Ca²⁺ before cyanobacterial photosynthesis would have induced precipitation. According to existing models, it was. Thus Arp et al. (2001) considered Ca enrichment to be responsible for the seemingly greater abundance of cyanobacterial microfossils and other calcified microbial structures such as thrombolites and stromatolites in the Paleozoic. Nonetheless, these constitute a wide variety of features over a long span of geological time, and I am loath to ascribe them to a single overarching factor.

Certainly there is much more observational and experimental work to be accomplished before a clear understanding of marine microbial CaCO₃ precipitation will emerge, as Arp et al. signal. And these processes have to be considered in the context of probably differing chemistries in ancient seas (Arp et al., 2001; see also Holmden et al., 1998).