306 REINERS 



these estimates are based on root biomass measurements from cores or 

 excavation pits. The simplest assumption is that the ratio of root production to 

 root biomass equals the ratio of shoot production to shoot biomass. 1 6 In some 

 cases, root material is sorted by diameter-size classes and turnover rates based on 

 separate intensive studies applied to these compartmentalized biomass measure- 

 ments. As in the above-mentioned cases, Kimura 1 8 assumed that root 

 contribution to detritus equaled root productivity, which was, in turn, simply 

 estimated as one-third that of stems and branches. Woodwell and Marples 1 9 

 estimated root contribution indirectly by subtracting aerial inputs from an 

 independent estimate of the A horizon annual decay rate. Kucera etal. 20-22 

 have conducted relatively detailed investigations on root turnover in a tall-grass 

 prairie. Their method essentially involved measurement of root biomass in soil 

 cores several times over a growing season and subtraction of minimum from 

 maximum values. Although executed very carefully, their method — along with 

 most other methods — could not measure rapid turnover of root organic matter 

 by root-feeding organisms, by root-hair dieback, or by root secretions. Therefore 

 their results are probably underestimates of energy and carbon flow from roots 

 to consumers. From the point of view of detritus-pool dynamics, however, such 

 losses may be more related to grazing pathways than detritus pathways since 

 direct and rapid consumption by heterotrophs is involved. Thus these data may 

 actually be reasonable estimates of the bulk of carbon involved in detritus-pool 

 storage and turnover. 



In summary, the method of estimating detritus-pool turnover with measures 

 of carbon inputs is sound for steady-state or near-steady-state systems. Its 

 principal difficulty lies in the inadequacy of methods for procuring root input 

 data. 



Direct CO .-Evolution-Measurement Method 



If we can assume that all but insignificant fractions of detrital carbon are lost 

 through the soil— air interface, the prospect of direct measurement of CO2 

 evolution is initially an attractive means of estimating detritus-pool turnover. 

 Theoretically, if the controlling environmental factors can be sufficiently 

 understood, an annual output may be estimated with information on rate 

 responses to environmental factors and data on these factors. An analysis of CO2 

 gas is usually performed with KOH absorption cups or by infrared gas analysis. 



Unfortunately this method is fraught with serious complications. The most 

 minor among these, at least for well-drained soils, are possible losses of other 

 volatile forms of carbon to the atmosphere or losses of dissolved CO2 or organic 

 substances to groundwater. This may be especially serious in circumneutral or 

 alkaline soils. Witkamp 2, illustrated a second problem — marked diurnal cycles 

 in CO2 evolution correlating with diurnal temperature changes. In some cases, 

 CO2 -evolution rates were inversely related to air temperature, presumably 

 because of the forced convection of warm soil air upward. Such cycles are 



