I would like to continue the discussion of the mantle carbon flux featured in the January 3, 1989, issue of Eos, between B. Marty and D. Des Marais. Although the controversy over carbon recycling by subduction is not new, the current discussion suggests that the time has come to organize a special session for the AGU 1990 Spring Meeting to address this important issue. I will be taking the initiative to organize this session and would appreciate input on the appropriate format and focus.
My first of three points is that all of the isotopically light carbon (−20 to −30 ‰) liberated below 500°C in stepped-heating carbon isotope measurements of basalts and mantle xenoliths is probably not contamination in most cases, as has been argued by Des Marais and others [Des Marais and Moore, 1984; Mattey et at., 1984]. Exley et al. , in their study of basalts from Loihi seamount presented in their reply to Craig , showed that a low-temperature iso opically light component is observed even when special precautions are taken to avoid contamination. Mantle xenoliths and basalt are known to contain condensed carbonaceous films on crack surfaces and grain boundaries [Mathez and Delaney, 1981; Mathez, 1987]. It has been suggested that the isotopically light carbon is derived from these carbonaceous films [Nadeau et al, 1987] and that the films may be partly organic [Mathez, 1987; S. Nadeau et al., unpublished manuscript, 1989]. This hypothesis has been confirmed in our laboratories by thermal desorption combined with photoionization time-of-flight mass spectrometry of mantle-derived olivine single crystals, xenoliths, and host basalt from San Carlos, Ariz., and Hualalai Volcano, Hawaii. A brief report of our results for San Carlos olivine crystals is given here.