Chemistry 105 



by taking a core and thoroughly inoculating it with K''NO.!. After 

 incubation for 16 days in the original core-hole, the part of the core that 

 had been injected was extruded into a core squeezer and the interstitial 

 water was collected. Gas was stripped from the water (Goering and 

 Dugdale 1966) and analyzed for ''N on an AEl MS-20 mass 

 spectrometer. The two measurements gave 0.17 and 0.19 Mg N liter ' 

 day ' or an average of 32 Mg N m ^ day ' . Because of these low rates of 

 denitrification, the mean mass ratio of 28 : 29 changed very little; there was 

 a larger change in the mass ratios of 30 : 28 and so this ratio was used as the 

 indicator of denitrification. 



Hauck et al. (1958) have established that molecular nitrogen 

 produced through denitrification has mass 28, 29, 30 distribution 

 determined by the nitrogen source and that isotopic equilibration with the 

 preexistent N2 pool does not occur. With this, and the further assumption 

 that the nitrogen gas produced came only from the interstitial water 

 nitrate pool, the amount of nitrogen gas produced per liter of water was 

 calculated: ng N2 liter"' day ' = excess '^N2 (at %)xN2 (^g 

 liter ')x(100)(at % "NO.3)''. The N2 concentration in the interstitial 

 water was not measured but was assumed to be 2.032x10^ Mg liter ' 

 (saturation value at 5°C, Weiss 1970). The initial atoms percent ''NO3 

 used in the equation above was calculated from the amount of tracer 

 originally added to the core, the nitrate concentrations, and the water 

 content. 



We did not include N2O in our measurements. Under certain 

 circumstances such oxides of nitrogen may be produced, and for most 

 denitrifying organisms N2O is a precursor of N2 (Alexander 1971). 

 However, Cady and Bartholomew (1960) found complete reduction of 

 N2O to nitrogen even in their experiment in acid soil, although previously 

 N2O reducti.on had been found strongly inhibited below pH 7 (Wijler and 

 Delwiche 1954). Any error in our results due to production of N2O would 

 be in the direction of an underestimate, but for the reasons discussed 

 above, we feel that such error is likely to be small. 



Another possible source of error is the effect of the added nitrate on 

 the process. Hart et al. (1965) and Clasby (personal communication) 

 concluded that low levels of nitrate do not limit the rate of denitrification 

 as long as measurable nitrate is present so we believe that the added ' 'NO3 

 had no effect (levels were kept low, however). In fact, rather than the level 

 of nitrate limiting the rate, it is more likely that the denitrification process 

 is limited by other nutrients. For example, when tundra soils were tested 

 with added glucose and phosphorus, the denitrification rate rose 4-fold. 



Because the water above the sediment always contains abundant 

 oxygen, denitrification never occurs except in the sediments. This is not 

 true of shallow subarctic lakes where anoxic conditions occur beneath the 

 ice in late winter. Goering and Dugdale (1967) measured rates of 

 denitrification as high as 15 Mg N 'iter ^ ' day ' in the water column and 

 even higher rates in the presence of lake sediment. 



