helium. The gases were concentrated on Porapak Q cooled with 

 liquid nitrogen, and then injected into a gas chromatograph 

 equipped with a flame ionization detector. Calibrations were 

 made by injecting known volumes of a gas standard (23 ppm CH4 in 

 helium). Precision of the analysis was +5%. 



Dissolved inorganic carbon measurements were made on 3.5 ml 

 samples taken from the same bottles as the methane samples. The 

 samples were analyzed using a gas chromatograph following the gas 

 stripping method described by Weiss and Craig (1973). Samples 

 were calibrated against standards prepared gravimetrically from 

 Na2CC>3 solutions as described by Wong (1970). The precision of 

 replicate analyses was +0.5%. 



After sampling, a gas phase formed in several of the sealed 

 bottles. Partitioning of methane and carbon dioxide into this 

 gas phase can alter their dissolved concentrations. This was 

 especially noticeable in the warmest waters sampled (i.e., those 

 with dissolved Si approximately twice that of ambient bottom 

 seawater). Consequently, we restrict our statistical analysis of 

 the relationships between dissolved gas concentrations and 

 dissolved silica to those samples containing <160 ym Si. 

 Concentrations of methane and total carbon dioxide in the warmer 

 vent fluids are estimated by extrapolation of the linear trends 

 observed in the CH4 or ECO2 versus Si relationships to the Si 

 concentrations measured in the 30-35°C vent samples (as follows). 



Dissolved nutrients 



Individual vent water samples and control (non-vent) deep 

 (1000 m) seawaters were processed immediately upon their arrival 

 on the support vessel ( R/V Atlantis II for Alvin dives or R/V 

 Kila for Pisces V dives). For the Alvin dive series, subsamples 

 (125 ml) for dissolved nutrients ( [N0 3 ~ + N0 2 "], NH 4 + , P0 4 ~ 3 , 

 Si [OH] 4) were filtered through pre-combusted (450°C, 3 hr ) 

 Whatman glass fiber filters (GF/F type) and placed into acid- 

 washed polyethylene bottles and stored frozen (-20°C). Inorganic 

 nutrients were measured by the Hawaii Biogeochemical Analytical 

 Facility using standard Technicon II autoanalyzer techniques for 

 the analysis of seawater. We also compared the total nutrient 

 content of filtered versus unfiltered vent water samples and 

 concluded that filtration (GF/F filter) had no significant effect 

 on the nutrient determinations at Pele's Vent. 



Technically speaking, the spectrophotometric methods used 

 in this study measure soluble reactive phosphorus ( SRP ) rather 

 than total dissolved phosphate and reactive silica rather 

 than total dissolved silica. The latter may be of special 

 interest in the analysis of samples collected from deep-sea vents 

 due to the possible presence of Si, either as polymerized Si or 

 as micro-particulate (<0.4 ym diameter) Si which would escape 

 detection by our autoanalyzer methods. For this reason we also 



193 



