Total organic carbon content in Leg 184 sediments was generally low (<0.1-1.1 wt%) but typically ranged 0.2-0.5 wt% OM. The organic matter concentration in sediments is important for many reasons; one is that it controls the sediment oxidation-reduction state, which can be determined by observation of the bulk color of the sediment:
Organic matter in sediments can have mixed marine and land sources. The land-sourced OM signal is related to climate and erosion and transport of bulk land minerals to the sea. The relative amount of land-sourced OM can record this change with time. The relative amounts of land- and marine-sourced OM are often determined using C/N ratios; unfortunately, the low concentrations of OM in the Leg 184 sediment samples made this determination difficult. Because TOC content is most often determined by difference (total carbon [CHN analysis/Carlo Erba combustion] minus carbonate carbon [wet chemical/titration]) (Emeis and Kvenvolden, 1986) and because the sample was >30 wt% carbonate carbon, the analytical uncertainty was at least 0.1% (or 0.2%), which means the TOC value often had an uncertainty >20%. These "by difference" values were often as much as 30% lower than the Rock-Eval TOC values.
With the effort and sharp attention of the shipboard sedimentologists, seven centimeter-size samples of organic carbon-rich material, such as fragments of woody material, were noted and collected by the author from cores from the five sites. Five samples were later identified as lignite and two as "plant fragments." In order to determine the land-plant carbon isotopic end-member value, these seven samples were analyzed.
On board ship, some samples were air-dried at room temperature and others were freeze-dried and stored in glass screw-top vials. Headspace samples were stored in crimp-top glass containers. The shore-based laboratory procedure was as follows. Samples were dried in air or under vacuum at <70°C. Sediment samples were pulverized/ground to <100 µm. A 1- to 5-g portion of the sample was treated successively with cold 1-N HCl and cold 3-N HCl and then warmed to 60°C in a test tube (in a hot water bath) and allowed to react for 24 hr to dissolve calcite. The sample was then centrifuged, and the acid solution was decanted. Samples were washed with deionized water (usually three or four times) and centrifuged until neutrality (pH = 7), and the solution was decanted. Finally, samples were vacuum dried at <60°C prior to isotope analyses.
Woody and OM-rich concretion samples were treated somewhat differently. Portions of three of these samples were treated with 0.1-N NaOH to extract humic material (indigenous terrestrial and/or adsorbed or absorbed marine). Soluble humic materials give a brown or tan color to the solution. The solution was decanted (and saved), and new NaOH solution was added to further extract humic material. The sample was centrifuged, and the humic material was decanted and combined with the first extract. This humic extract was acidified dropwise to pH = 1 with 1-N HCl, which caused the humic acid fraction to precipitate. The solution was decanted, and the humic acid was dried under vacuum at <60°C. Both the humic acid and the extracted woody material were analyzed for stable carbon isotopes. The humic acid extracts usually represented 4%-9% of the original OM and were sometimes only 5-mg samples. Fortunately, the new mass spectrometer systems are quite sensitive and are able to measure carbon isotopes on such small samples. The other four samples were much smaller and therefore not large enough to undergo extraction; these were analyzed without pretreatment.
Splits of the woody samples collected shipboard were given to Co-Chief Scientist Prof. Pinxian Wang for genus/species identification at the Chinese National Laboratory of Dendrology. Unfortunately, the samples were not suitable or too small(?) for identification. My colleague, Dr. Neely Bostick (U.S. Geological Survey [USGS] Emeritus Scientist), produced the identifications used in this report. He positively identified some samples as peat or lignite, but two samples do not have the characteristic "lignite" woody texture and may represent the leaf or other soft parts of plants (apparently in a clay matrix).
The carbonate-free residue was combusted to CO2 and purified using an elemental analyzer coupled to a Micromass Optima isotope ratio mass spectrometer in continuous flow mode (USGS laboratory in Denver, Colorado). The 
13C isotope value of this CO2 was measured relative to Vienna Peedee belemnite (VPDB) and working standards, where a of 1
is 0.1% different from the 13C/12C ratio of the VPDB standard. Several replicate preparations were run (due to unknown organic carbon values), and, therefore, multiple analyses of many samples were conducted to adjust the amount of CO2 into the range for optimum mass spectrometer determination. The results were generally ±0.2 for replicate analyses of these low-TOC samples.
A total of 15 headspace samples (from shipboard determinations and in their original crimp-top glass containers) from the high-methane zones at Site 1146 were analyzed after extraction with a 10-mL gas syringe and injection into a sample loop connected to a combustion furnace coupled to a Micromass Optima mass spectrometer. Unfortunately, the headspace gas contained essentially no methane (<100 ppm), probably due to air leakage (despite a silicone-sealed septum) or microbial oxidation on the sediment surface (after storage at room temperature for >2 months). The lesson to be learned from this is that headspace gas containing methane must be separated from its sediment host if not analyzed within 24 hr of sampling. The headspace gas should be transferred to a vacutainer-type container (as was done for samples collected for Dr. Matsumoto of Tokyo University, Japan).
A total of 24 organic matter samples were measured to determine 13C stable isotope values. Seven centimeter-sized organic matter samples (wood fragments) were measured for bulk 13C values. Humic acid was extracted from several samples and the 13C value determined as well as, in one case, the 13C value of the residue after extraction (Table T1).
