Microb. Ecol.7:229-234 (1981)
MICROBIAL CCOLOGV
M e a s u r e m e n t of the Microbial Biomass in Intact Cores of S o i l J. M. Lynch and Lynda M. Panting Agricultural ResearchCouncil,LetcombeLaboratory,Wantage,Oxon,OXI2 9JT, England
Abstract. The fumigation/respiration technique was used to estimate the size of the soil microbial biomass. Sieving decreased the biomass in winter but increased it in summer; we suggest that this was a consequence of the different substrates available and the different microbial populations during the year. The flush in respiration following fumigation correlated significantly with the CO2-C produced 10 days after fumigation (X), so that in the soils studied by us the biomass (B) can be calculated from Bk = 0.673X - 3.53, where k is the fraction of fumigated organisms mineralized to CO2, thus avoiding the need to measure CO2 production from unfumigated cores.
Introduction The technique in which a soil is fumigated with chloroform, inoculated with microorganisms, and a measurement made of the increase in carbon dioxide evolution from the soil over that from a soil which has not been fumigated is widely used for measuring biomass in soil [1, 6, 10, I 1, 13, 14]. It is less time consuming and probably more reliable than direct observation. Furthermore, the results can be more exactly interpreted than those from methods based on adenosine triphosphate (ATP) determinations, which can be misleading because measurements cannot be made instantaneously on sampling [5], and guanosine triphosphate can interfere [9], as also can cations and inhibitors. Not surprisingly, the ATP assay sometimes correlates only poorly with other methods [14]. Most users of the fumigation/respiration technique of Jenkinson and Powlson [6] have modified it, but all except ourselves [10, 11] have worked with sieved soil. Although sieving imparts greater homogeneity to the samples and can produce smaller apparent errors in measurement, we have shown [10] that it can reduce the size of the biomass and is thus best avoided when studying the effects of different cultivation treatments. Jenkinson and Powlson [7] have proposed an alternative interpretation of our results by suggesting that sieving the wet clay soil used in our experiments and recompressing it to its original bulk density closes the gas-filled pores, such that the fumigant penetrates less well and consequently a smaller biomass is measured. We have therefore re-examined the sieving effect. Biomass, B (mg C 100 g - l dry soil), is calculated from Bk = X-x; where X and x are the amounts of CO 2 produced by fumigated and unfumigated soils, respectively (mg C 100 g - l dry soil), and k is the proportion of fumigated organisms mineralized to CO2 (0.41). We have analyzed all our results from experiments with different soils for 0095-3628/81/0007-0229$01.20 9 1981Springer-VerlagNew YorkInc.
230
J.M. Lynchand L. M. Panting
possible relationships between biomass (B) and C O 2 production from fumigated (X) or unfumigated (x) soil. If a sufficiently close relationship exists, half the number of cores could be taken from any one treatment, allowing a greater number of treatments to be studied at any time.
Materials and Methods We have already publisheddetails of the methods [10, l 1]. Samples were taken from the soil surface with open-endedcans (7.6 cm diameter x 5 cm, 228 cm3). The microbialbiomass(B) of the surface 5 cm of soil was calculated from the relationshipBk = X-x, where X is the amount of CO2 produced in the 10 days followingfumigation,x is the amountof CO2 produced by non-fumigatedsoil in the period, and k (0.41) is the proportionof fumigatedorganismsmineralizedto CO2duringthe incubationat 220C. Where the effect of sieving was investigated,the soil was removed from the can, sieved (c 7 mm), and then eitherreplaced in the can and compressed to the same bulk densityas the soil in the fieldor left uncompressedin a glass beaker. To study temperaturechanges, thermistors (5 cm) were insertedsuch that the tip reached the centerof the core. The soils used were the clays fromthe Lawford series ofpH 6.5 at Northfield,Challow,Oxon(site 1), and of pH 6.4 at Oldfield, Faringdon, Oxon (site 2), a clay from the Denchworth series of pH 6.3 at Compton Beauchamp, Ashbury, Oxon, and a silt loam (Hambleseries) of pH 5.6 at Englefield,Reading, Berks. The classificationof the clays was UK, Stagnogley;USDA, Typic Haplaquept;FAO/UNESCO,EutricGleysol; and that the silt loam was UK, Argillic Brown Earth; USDA, Typic Hapladalf (in Alfisols); FAO/ UNESCO, Orthic Luvisol.
Results
Effect of Sieving In winter samples taken in January 1978 and November 1979, sieving resulted in a smaller biomass, especially in the wettest soil (January 1978) (Table 1). In a spring sample taken in April 1980 sieving had no effect, whereas in the two samples taken in late spring/early summer (May 1978 and June 1980) it increased the biomass. The sieved samples compressed in the metal cans did not differ significantly in biomass from those left uncompacted in glass beakers. Although the soil moisture contents and air temperatures (11~ were similar when the winter (November 1979) and summer (May 1979, June 1980) samples were taken, the effects of sieving were in opposite senses. Samples taken in winter at I~ (January 1978) were not exposed to particularly sudden temperature shocks on sieving; the sieved cores reached ambient temperature in 1 h and unsieved cores took 3 h.
C02 Production by Fumigated and UnfumigatedCores The flush in CO2 production (X-x) correlated strongly (r, 63 degrees of freedom = + 0 . 8 8 2 5 * * * ) with CO 2 production from fumigated cores (X) (Fig. 1): X-x = 0.673 ( + 0.045) X - 3.53 (_+ 2.11). The correlation between the flush and C O 2 production from unfumigated cores (x) was significant but much weaker (Fig. 2) (r, 63 degrees of freedom = +0.2474*): X-x = 0.389 ( + 0.192)x + 19.15 ( + 3.72). Examination of the results from the different soils showed that no soil yielded results consistently on one side only of the regression line.
Microbial Biomass in Intact Soil Cores
231
Table 1. Effect of sieving on the soil biomass of the Denchworth series clay soil
Air ternperature (~
Date 16Jan 1978
1
25 May 1978
11
26 Nov 1979 b 11 28 Apt 1980
6
23 Jun 1980
11
Treatment
Moisture (% of max. Bulk water density holding (g c m - 3 ) capacity)
Microbial biomass and standard error a (mg C 100g - 1 dry soil)
Soil organic C Microbial in microbiomass bial bio0-5 cm (kg mass C ha - 1) (% w/w)
Sieved Undisturbed Sieved Undisturbed Sieved Undisturbed Sieved Undisturbed Sieved c Undisturbed c
0.94 0.94 0.99 0.98 1.03 0.99 0.97 1104 1.02 1.02
97 95 66 67 77 76 52 52 51 52
18 -I- 6.7 d 77 __+9.3 e 123 _ 1.4 d 69 ___5.3 e 20 -I- 1.2 d 38 -t- 4.9 e 91 ___7.8 d 92 __+2.9 d 88 -I- 7.8 d 89 -I- 2.9 d
85 361 610 337 102 187 439 478 447 453
0.4 1.5 2.3 1.3 0.8 1.6 1.7 1.7 1.7 1.7
Sieved Undisturbed Sieved c Undisturbed c
0.97 1.02 0.91 0.96
74 72 77 80
436 363 438 333
1.7 1.4 1.8 1.3
90 71 96 69
q- 5.9 d -I- 6.3 e + 4.6 d -t- 4.6 e
aOn each sampling date, results not followed by the same letter (d, e) are significantly different (P < 0.05), b Lawford series clay soil. c Samples not compressed in tins hut incubated in glass beakers.
6O
0
0
0
0
0
~-~
~~
9o
~
40
9
~
o
9
/~I
0
j
I
20
I
-(%o~ 9
I
40
~
"
9
-
o
~
o o
o o
I
I
I
0
I
80
I
I
100
X, COe produced by fumigated soil (mg COe-C 1000-1 dry soil)
Fig, 1. Relationship of the flush of CO 2 produced and the CO 2 produced by cores l0 days after fumigation. Each point represents the mean of at least 4 replicates, o, Denchworth series; m, Hamble series; e, Lawford series, site 1; n, Lawford series, site 2.
232
J . M . Lynch and L. M. Panting
60
0 0
"(p o
A ,-r
0
0
0
..,.o
40
9
i
O
0
F/ o ~176 o - ~~
"~"
0
"" ~ 2 o L j
~ e--
oo
o
"~
_
o
o -
Fig. 2. Relationship o f the flush o f CO 2 produced and the CO 2 produced by unfumigated cores in 10 days. Each point represents the mean of at least 4 replicates, o, Denchworth series; n, Hamble series; o, Lawford series, site 1; a, Lawford series, site 2.
eo
@ee e II e eo
[J
0
I
o
I
2O
0
i
i
40
:r CO2 produced by unfumigated soil (mg CO2-C 100g" dry soil)
m
120
'~
o
o
80
o
~
4O
N
0
I
0
40
80
I
120
Biomass by regression (mg C 100g" dry soil) Fig. 3. Relationship between tim biomass reported by Jenkinson et al. (Bk = X - y) and that calculated for regression (Bk = 0.673 X - 3.53). The line indicates what would be expected for r [2];o, Jcnkinson and Powlson [6];D, Jenkinson ct al. [8]; a, Oades and Jenkinson [12].
o, Ayanaba et aL
Microbial Biomassin Intact Soil Cores
233
Table2. Relationshipsbetweenthe flush in CO2 production (X-y) and CO2 production from fumigatedcores (X) in the studies of Jenkinsonet al. Regressionequation
Reference
X-y = 0.787 (+0.036)X- 1.63(+0.63) X-y = 0.678 (-t-0.006)X- 0.09(-t-0.29) X-y = 0.569 (_0.075)x+ 0.99(5:4.19) X-y = 0.630 (+0.034)X- 2.91(+ 1.41)
[2] [6] [8] [12]
Discussion The variable effect of sieving on the biomass of the clay soils we studied is at first sight perplexing. Lack of penetration of the fumigant as suggested by Jenkinson and Powlson [7] does not seem to offer a satisfactory explanation because the variation was independent of the soil water content. Similarly, temperature at the time of sampling seemed not to cause the variation. The interpretation that we propose from our limited number of results concerns the different microbial populations of samples taken in winter and summer. In winter, root activity is minimal and the major source of energy to the biomass is from residues of the preceding crop plants. CeUulolytic fungi are probably the dominant colonists, and most are able to tolerate the long periods of poor aeration common in these heavy clay soils in winter. However, obligate and facultative cellulolytic anaerobic bacteria are also involved. The effect of sieving the soil on these organsism which have become adapted to low oxygen potentials is to remove both the substrates and the organisms on the sieve and to kill a proportion of them by exposure to lethally high oxygen potentials. This phenomenon is commonly observed in chemostat studies in which oxygen concentrations have to be changed slowly to allow time for enzyme adaptation [3]. The overall effect therefore is to increase the CO2 production from unfumigated cores after sieving, resulting in a smaller estimate of biomass because more dead organisms are present; this is indeed what we have observed. In spring and summer as roots grow we have proposed that the main components of the biomass that increase are the predominantly aerobic rhizosphere bacteria. When samples from such soils are sieved, the effect is to break up the roots and provide substrates in the form of root cell contents not usually available so rapidly to the microflora. At the ambient temperatures, microbial growth can be rapid during the I - 2 h that usually elapses between sieving and fumigation. Thus a greater biomass is measured as a result of sieving. Both the winter and summer effects of sieving give misleading estimates of the biomass in the field. We cannot be certain how widespread this is, for the problem may be peculiar to the surface 5 cm of the clay soils we have studied. However, under these circumstances, we consider that to measure biomass in the field with maximum reliability, the soils should not be sieved. The correlation of CO2 production from fumigated cores with the flush in CO2 production in our soils provides a means of estimating biomass without the need for
234
J. M. Lynch and L. M. Panting
measuring respiration from unfumigated cores. Equally, the results also show how inadequate is this last measurement alone as an index of biomass. We have calculated the regressions of (X-y) on X for the results obtained by Jenkinson et al. (Table 2). In each of their studies, the slopes do not differ significantly (P = 0.05) from ours, but in two of them [2, 6] the line is significantly displaced upward. The reasons for this displacement are unclear, but it may be caused by the differences in methods used. For example, they use the 10-20 day period to measure the CO2 produced by unfumigated cores (y), whereas we measure it in the 0-10-day period (x). However, when Jenkinson's results are recalculated using our regression line (Fig. 3), the relationship with the biomass calculated using the conventional method is quite close. We would not expect such a good relationship to be obtained if readily assimilable substrate had recently been added to the soil, and the fumigation/respiration method does not seem valid for acid woodland soil [12]. Acknowledgment. We thank Mr. B. O. Bartlett for statistical advice and Dr. D. S. Jenkinson for helpful comments.
References 1. Anderson, J. P. E., and K. H. Domsch: A physiological method for the quantitative measurement of microbial biomass in soil. Soil Biol. Biochem. 10, 215-221 (1978) 2. Ayanaba, A., S. B. Tuckwell, and D. S. Jenkinson: The effects of clearing and cropping on the organic reserves and biomass of tropical forest soils. Soil Biol. Biochem. 8, 519-525 (1976) 3. Carter, B. L. A., and A. T. Bull: The effect of oxygen tension in the medium on the morphology and growth kinetics ofAspergillus nidulans, J. Gen. Microbiol. 65,265-273 (1971) 4. Eiland, F., and B. S. Nielsen: Influence of cation content on adenosine triphosphate determinations in soil. Microb. Ecol. 5,129-137 (1979) 5. Ellwood, D. C., J. N. Hedger, M. J. Latham, J. M. Lynch, and J. H. Slater (eds.): Contemporary Microbial Ecology, p. xiv. Academic Press, London (1980) 6. Jenkinson, D. S., and D. S. Powlson: The effects of biocidal treatments on metabolism in soil. V. A method for measuring soil biomass. Soil Biol. Biochem. 8, 209-213 (1976) 7. Jenkinson, D. S., and D. S. Powlson: Measurement of microbial biomass in intact soil cores and in sieved soil. Soil Biol. Biochem. 12,579-581 (1980) 8. Jenkinson, D. S., S. A. Davidson, and D. S. Powlson: Adenosine triphosphate and microbial hiomass in soil. Soil Biol. Biochem. 11,521-527 (1979) 9. Karl, D. M.: Adenosine triphosphate and guanosine triphosphate determinations in intertidal sediments. In C. D. Litchtieid and P. L. Seyfried (eds.): Methodology for Biomass Determinations and Microbial Activities in Sediments, ASTM STP 673, pp. 5-20. American Society for Testing Materials, Philadelphia (1979) 10, Lynch, J. M., and L. M. Panting: Cultivation and the soil biomass. Soil Biol. Biochem. 12, 29-33 (1980) 11. Lynch, J. M., and L. M. Panting: Variations in the size of the soil biomass. Soil Biol. Biochem. 12, 547-550 (1980) 12. Oades, J. M., and D. S. Jenkinson: Adenosine triphosphate content of the soil microbial biomass. Soil Biol. Biochem. 11,201-204 (1979) 13. Paul, E. A., and R. P. Voroney: Nutrient and energy flows through soil microbial biomass. In D. C. Ellwood, J. N. Hedger, M. J. Latham, J. M. Lynch, and J. H. Salter (eds.): Contemporary Microbial Ecology, pp. 215-237. Academic Press, London (1980) 14. Ross, D. J., K. R. Tate, A. Cairns, and E. A. Ponier: Microbial biomass estimations in soils from tussock grasslands by three biochemical procedures. Soil Biol. Biochem. 12, 375-383 (1980)
MICROBIAL CCOLOGV
M e a s u r e m e n t of the Microbial Biomass in Intact Cores of S o i l J. M. Lynch and Lynda M. Panting Agricultural ResearchCouncil,LetcombeLaboratory,Wantage,Oxon,OXI2 9JT, England
Abstract. The fumigation/respiration technique was used to estimate the size of the soil microbial biomass. Sieving decreased the biomass in winter but increased it in summer; we suggest that this was a consequence of the different substrates available and the different microbial populations during the year. The flush in respiration following fumigation correlated significantly with the CO2-C produced 10 days after fumigation (X), so that in the soils studied by us the biomass (B) can be calculated from Bk = 0.673X - 3.53, where k is the fraction of fumigated organisms mineralized to CO2, thus avoiding the need to measure CO2 production from unfumigated cores.
Introduction The technique in which a soil is fumigated with chloroform, inoculated with microorganisms, and a measurement made of the increase in carbon dioxide evolution from the soil over that from a soil which has not been fumigated is widely used for measuring biomass in soil [1, 6, 10, I 1, 13, 14]. It is less time consuming and probably more reliable than direct observation. Furthermore, the results can be more exactly interpreted than those from methods based on adenosine triphosphate (ATP) determinations, which can be misleading because measurements cannot be made instantaneously on sampling [5], and guanosine triphosphate can interfere [9], as also can cations and inhibitors. Not surprisingly, the ATP assay sometimes correlates only poorly with other methods [14]. Most users of the fumigation/respiration technique of Jenkinson and Powlson [6] have modified it, but all except ourselves [10, 11] have worked with sieved soil. Although sieving imparts greater homogeneity to the samples and can produce smaller apparent errors in measurement, we have shown [10] that it can reduce the size of the biomass and is thus best avoided when studying the effects of different cultivation treatments. Jenkinson and Powlson [7] have proposed an alternative interpretation of our results by suggesting that sieving the wet clay soil used in our experiments and recompressing it to its original bulk density closes the gas-filled pores, such that the fumigant penetrates less well and consequently a smaller biomass is measured. We have therefore re-examined the sieving effect. Biomass, B (mg C 100 g - l dry soil), is calculated from Bk = X-x; where X and x are the amounts of CO 2 produced by fumigated and unfumigated soils, respectively (mg C 100 g - l dry soil), and k is the proportion of fumigated organisms mineralized to CO2 (0.41). We have analyzed all our results from experiments with different soils for 0095-3628/81/0007-0229$01.20 9 1981Springer-VerlagNew YorkInc.
230
J.M. Lynchand L. M. Panting
possible relationships between biomass (B) and C O 2 production from fumigated (X) or unfumigated (x) soil. If a sufficiently close relationship exists, half the number of cores could be taken from any one treatment, allowing a greater number of treatments to be studied at any time.
Materials and Methods We have already publisheddetails of the methods [10, l 1]. Samples were taken from the soil surface with open-endedcans (7.6 cm diameter x 5 cm, 228 cm3). The microbialbiomass(B) of the surface 5 cm of soil was calculated from the relationshipBk = X-x, where X is the amount of CO2 produced in the 10 days followingfumigation,x is the amountof CO2 produced by non-fumigatedsoil in the period, and k (0.41) is the proportionof fumigatedorganismsmineralizedto CO2duringthe incubationat 220C. Where the effect of sieving was investigated,the soil was removed from the can, sieved (c 7 mm), and then eitherreplaced in the can and compressed to the same bulk densityas the soil in the fieldor left uncompressedin a glass beaker. To study temperaturechanges, thermistors (5 cm) were insertedsuch that the tip reached the centerof the core. The soils used were the clays fromthe Lawford series ofpH 6.5 at Northfield,Challow,Oxon(site 1), and of pH 6.4 at Oldfield, Faringdon, Oxon (site 2), a clay from the Denchworth series of pH 6.3 at Compton Beauchamp, Ashbury, Oxon, and a silt loam (Hambleseries) of pH 5.6 at Englefield,Reading, Berks. The classificationof the clays was UK, Stagnogley;USDA, Typic Haplaquept;FAO/UNESCO,EutricGleysol; and that the silt loam was UK, Argillic Brown Earth; USDA, Typic Hapladalf (in Alfisols); FAO/ UNESCO, Orthic Luvisol.
Results
Effect of Sieving In winter samples taken in January 1978 and November 1979, sieving resulted in a smaller biomass, especially in the wettest soil (January 1978) (Table 1). In a spring sample taken in April 1980 sieving had no effect, whereas in the two samples taken in late spring/early summer (May 1978 and June 1980) it increased the biomass. The sieved samples compressed in the metal cans did not differ significantly in biomass from those left uncompacted in glass beakers. Although the soil moisture contents and air temperatures (11~ were similar when the winter (November 1979) and summer (May 1979, June 1980) samples were taken, the effects of sieving were in opposite senses. Samples taken in winter at I~ (January 1978) were not exposed to particularly sudden temperature shocks on sieving; the sieved cores reached ambient temperature in 1 h and unsieved cores took 3 h.
C02 Production by Fumigated and UnfumigatedCores The flush in CO2 production (X-x) correlated strongly (r, 63 degrees of freedom = + 0 . 8 8 2 5 * * * ) with CO 2 production from fumigated cores (X) (Fig. 1): X-x = 0.673 ( + 0.045) X - 3.53 (_+ 2.11). The correlation between the flush and C O 2 production from unfumigated cores (x) was significant but much weaker (Fig. 2) (r, 63 degrees of freedom = +0.2474*): X-x = 0.389 ( + 0.192)x + 19.15 ( + 3.72). Examination of the results from the different soils showed that no soil yielded results consistently on one side only of the regression line.
Microbial Biomass in Intact Soil Cores
231
Table 1. Effect of sieving on the soil biomass of the Denchworth series clay soil
Air ternperature (~
Date 16Jan 1978
1
25 May 1978
11
26 Nov 1979 b 11 28 Apt 1980
6
23 Jun 1980
11
Treatment
Moisture (% of max. Bulk water density holding (g c m - 3 ) capacity)
Microbial biomass and standard error a (mg C 100g - 1 dry soil)
Soil organic C Microbial in microbiomass bial bio0-5 cm (kg mass C ha - 1) (% w/w)
Sieved Undisturbed Sieved Undisturbed Sieved Undisturbed Sieved Undisturbed Sieved c Undisturbed c
0.94 0.94 0.99 0.98 1.03 0.99 0.97 1104 1.02 1.02
97 95 66 67 77 76 52 52 51 52
18 -I- 6.7 d 77 __+9.3 e 123 _ 1.4 d 69 ___5.3 e 20 -I- 1.2 d 38 -t- 4.9 e 91 ___7.8 d 92 __+2.9 d 88 -I- 7.8 d 89 -I- 2.9 d
85 361 610 337 102 187 439 478 447 453
0.4 1.5 2.3 1.3 0.8 1.6 1.7 1.7 1.7 1.7
Sieved Undisturbed Sieved c Undisturbed c
0.97 1.02 0.91 0.96
74 72 77 80
436 363 438 333
1.7 1.4 1.8 1.3
90 71 96 69
q- 5.9 d -I- 6.3 e + 4.6 d -t- 4.6 e
aOn each sampling date, results not followed by the same letter (d, e) are significantly different (P < 0.05), b Lawford series clay soil. c Samples not compressed in tins hut incubated in glass beakers.
6O
0
0
0
0
0
~-~
~~
9o
~
40
9
~
o
9
/~I
0
j
I
20
I
-(%o~ 9
I
40
~
"
9
-
o
~
o o
o o
I
I
I
0
I
80
I
I
100
X, COe produced by fumigated soil (mg COe-C 1000-1 dry soil)
Fig, 1. Relationship of the flush of CO 2 produced and the CO 2 produced by cores l0 days after fumigation. Each point represents the mean of at least 4 replicates, o, Denchworth series; m, Hamble series; e, Lawford series, site 1; n, Lawford series, site 2.
232
J . M . Lynch and L. M. Panting
60
0 0
"(p o
A ,-r
0
0
0
..,.o
40
9
i
O
0
F/ o ~176 o - ~~
"~"
0
"" ~ 2 o L j
~ e--
oo
o
"~
_
o
o -
Fig. 2. Relationship o f the flush o f CO 2 produced and the CO 2 produced by unfumigated cores in 10 days. Each point represents the mean of at least 4 replicates, o, Denchworth series; n, Hamble series; o, Lawford series, site 1; a, Lawford series, site 2.
eo
@ee e II e eo
[J
0
I
o
I
2O
0
i
i
40
:r CO2 produced by unfumigated soil (mg CO2-C 100g" dry soil)
m
120
'~
o
o
80
o
~
4O
N
0
I
0
40
80
I
120
Biomass by regression (mg C 100g" dry soil) Fig. 3. Relationship between tim biomass reported by Jenkinson et al. (Bk = X - y) and that calculated for regression (Bk = 0.673 X - 3.53). The line indicates what would be expected for r [2];o, Jcnkinson and Powlson [6];D, Jenkinson ct al. [8]; a, Oades and Jenkinson [12].
o, Ayanaba et aL
Microbial Biomassin Intact Soil Cores
233
Table2. Relationshipsbetweenthe flush in CO2 production (X-y) and CO2 production from fumigatedcores (X) in the studies of Jenkinsonet al. Regressionequation
Reference
X-y = 0.787 (+0.036)X- 1.63(+0.63) X-y = 0.678 (-t-0.006)X- 0.09(-t-0.29) X-y = 0.569 (_0.075)x+ 0.99(5:4.19) X-y = 0.630 (+0.034)X- 2.91(+ 1.41)
[2] [6] [8] [12]
Discussion The variable effect of sieving on the biomass of the clay soils we studied is at first sight perplexing. Lack of penetration of the fumigant as suggested by Jenkinson and Powlson [7] does not seem to offer a satisfactory explanation because the variation was independent of the soil water content. Similarly, temperature at the time of sampling seemed not to cause the variation. The interpretation that we propose from our limited number of results concerns the different microbial populations of samples taken in winter and summer. In winter, root activity is minimal and the major source of energy to the biomass is from residues of the preceding crop plants. CeUulolytic fungi are probably the dominant colonists, and most are able to tolerate the long periods of poor aeration common in these heavy clay soils in winter. However, obligate and facultative cellulolytic anaerobic bacteria are also involved. The effect of sieving the soil on these organsism which have become adapted to low oxygen potentials is to remove both the substrates and the organisms on the sieve and to kill a proportion of them by exposure to lethally high oxygen potentials. This phenomenon is commonly observed in chemostat studies in which oxygen concentrations have to be changed slowly to allow time for enzyme adaptation [3]. The overall effect therefore is to increase the CO2 production from unfumigated cores after sieving, resulting in a smaller estimate of biomass because more dead organisms are present; this is indeed what we have observed. In spring and summer as roots grow we have proposed that the main components of the biomass that increase are the predominantly aerobic rhizosphere bacteria. When samples from such soils are sieved, the effect is to break up the roots and provide substrates in the form of root cell contents not usually available so rapidly to the microflora. At the ambient temperatures, microbial growth can be rapid during the I - 2 h that usually elapses between sieving and fumigation. Thus a greater biomass is measured as a result of sieving. Both the winter and summer effects of sieving give misleading estimates of the biomass in the field. We cannot be certain how widespread this is, for the problem may be peculiar to the surface 5 cm of the clay soils we have studied. However, under these circumstances, we consider that to measure biomass in the field with maximum reliability, the soils should not be sieved. The correlation of CO2 production from fumigated cores with the flush in CO2 production in our soils provides a means of estimating biomass without the need for
234
J. M. Lynch and L. M. Panting
measuring respiration from unfumigated cores. Equally, the results also show how inadequate is this last measurement alone as an index of biomass. We have calculated the regressions of (X-y) on X for the results obtained by Jenkinson et al. (Table 2). In each of their studies, the slopes do not differ significantly (P = 0.05) from ours, but in two of them [2, 6] the line is significantly displaced upward. The reasons for this displacement are unclear, but it may be caused by the differences in methods used. For example, they use the 10-20 day period to measure the CO2 produced by unfumigated cores (y), whereas we measure it in the 0-10-day period (x). However, when Jenkinson's results are recalculated using our regression line (Fig. 3), the relationship with the biomass calculated using the conventional method is quite close. We would not expect such a good relationship to be obtained if readily assimilable substrate had recently been added to the soil, and the fumigation/respiration method does not seem valid for acid woodland soil [12]. Acknowledgment. We thank Mr. B. O. Bartlett for statistical advice and Dr. D. S. Jenkinson for helpful comments.
References 1. Anderson, J. P. E., and K. H. Domsch: A physiological method for the quantitative measurement of microbial biomass in soil. Soil Biol. Biochem. 10, 215-221 (1978) 2. Ayanaba, A., S. B. Tuckwell, and D. S. Jenkinson: The effects of clearing and cropping on the organic reserves and biomass of tropical forest soils. Soil Biol. Biochem. 8, 519-525 (1976) 3. Carter, B. L. A., and A. T. Bull: The effect of oxygen tension in the medium on the morphology and growth kinetics ofAspergillus nidulans, J. Gen. Microbiol. 65,265-273 (1971) 4. Eiland, F., and B. S. Nielsen: Influence of cation content on adenosine triphosphate determinations in soil. Microb. Ecol. 5,129-137 (1979) 5. Ellwood, D. C., J. N. Hedger, M. J. Latham, J. M. Lynch, and J. H. Slater (eds.): Contemporary Microbial Ecology, p. xiv. Academic Press, London (1980) 6. Jenkinson, D. S., and D. S. Powlson: The effects of biocidal treatments on metabolism in soil. V. A method for measuring soil biomass. Soil Biol. Biochem. 8, 209-213 (1976) 7. Jenkinson, D. S., and D. S. Powlson: Measurement of microbial biomass in intact soil cores and in sieved soil. Soil Biol. Biochem. 12,579-581 (1980) 8. Jenkinson, D. S., S. A. Davidson, and D. S. Powlson: Adenosine triphosphate and microbial hiomass in soil. Soil Biol. Biochem. 11,521-527 (1979) 9. Karl, D. M.: Adenosine triphosphate and guanosine triphosphate determinations in intertidal sediments. In C. D. Litchtieid and P. L. Seyfried (eds.): Methodology for Biomass Determinations and Microbial Activities in Sediments, ASTM STP 673, pp. 5-20. American Society for Testing Materials, Philadelphia (1979) 10, Lynch, J. M., and L. M. Panting: Cultivation and the soil biomass. Soil Biol. Biochem. 12, 29-33 (1980) 11. Lynch, J. M., and L. M. Panting: Variations in the size of the soil biomass. Soil Biol. Biochem. 12, 547-550 (1980) 12. Oades, J. M., and D. S. Jenkinson: Adenosine triphosphate content of the soil microbial biomass. Soil Biol. Biochem. 11,201-204 (1979) 13. Paul, E. A., and R. P. Voroney: Nutrient and energy flows through soil microbial biomass. In D. C. Ellwood, J. N. Hedger, M. J. Latham, J. M. Lynch, and J. H. Salter (eds.): Contemporary Microbial Ecology, pp. 215-237. Academic Press, London (1980) 14. Ross, D. J., K. R. Tate, A. Cairns, and E. A. Ponier: Microbial biomass estimations in soils from tussock grasslands by three biochemical procedures. Soil Biol. Biochem. 12, 375-383 (1980)
