9/1/1999

Australian Journal of Soil Research

By V. J. Barrett

Abstract

The relative abilities of 3 exotic lumbricid earthworms, the endogeic Aporrectodea caliginosa and A. trapezoides and the anecic A. longa, to bury surface-applied lime and help ameliorate soil acidity were measured in cages in 7 pasture soils in south-eastern Australia. All 3 species buried lime, mostly within the top 5 cm of the soil profile, but A. longa buried it deeper than A. caliginosa and A. trapezoides. A. longa significantly increased soil pH at 15-20 cm depth at some sites within 5 months (winter-spring, the earthworm `season' in the Mediterranean climate of south-eastern Australia). Lime burial varied markedly between sites. These site differences were explained, at least in part, by variations in rainfall. Lime burial increased with earthworm density. A minimum density of 214 A. longa/[m.sup.2] was needed to significantly enhance lime burial within one season. Higher densities were required for the other two species. However, per unit of biomass, A. caliginosa and A. trapezoides were generally more able to bury lime in the upper soil layers (2.5-10 cm depth) than A. longa. Agricultural soils in south-eastern Australia are dominated by shallow burrowing species such as A. caliginosa and A. trapezoides. Deeper burrowers such as A. longa are rare. Introduction of A. longa to soils in high-rainfall regions of south-eastern Australia, where it does not presently occur, should enhance lime burial and help reduce soil acidity.

Additional keywords: anecic, endogeic, lime incorporation, lumbricids, soil acidity.

Introduction

Soil acidity is a major problem in agricultural soils throughout the world, including Australia (Coventry 1985; Conyers and Scott 1989; Chartres et al. 1992; Von Uexkull and Mutert 1995). Soils acidify due to the use of ammonium-based nitrogenous fertilisers, growth of legumes, and removal of cations ([Ca.sup.2+], [Mg.sup.2+], [Na.sup.+], [K.sup.+]) (e.g. by harvesting, leaching, or erosion). Plant growth may be reduced in acid soils for a variety of reasons, most notably aluminium toxicity and phosphorus deficiency. Lime is commonly applied to the soil surface to ameliorate soil acidity and increase plant production, but it is often slow to move down the soil profile to the root-zone where it is needed (Helyar 1991). Tillage can obviously assist in the burial of lime, but this is not always feasible (e.g. in permanent pastures on hilly ground).

Earthworms can greatly enhance the burial of lime through their burrowing activities (Springett 1983, 1985; Baker et al. 1993c, 1995). Although the mechanism(s) for lime burial are poorly understood, it is generally accepted that rain water washes much of the lime down surface-venting burrows. However, the ability of earthworms to bury lime is likely to vary between species. Some earthworms are only active in the surface organic horizons of the soil (epigeic species), some feed predominantly within the mineral soil and rarely venture to the soil surface (endogeic species), and others feed and cast at the soil surface but burrow deeply (anecic species) (Bouche 1977). Anecic species are thus viewed theoretically as having the greatest potential for burying lime deep in the soil profile. Springett (1983) has indeed demonstrated in laboratory experiments that an anecic species, Aporrectodea longa (Ude) (Lumbricidae) (at an equivalent of 660 worms/[m.sup.2]), can bury surface-applied lime deeper than some epigeic and endogeic species (Lumbricus rubellus Hoff., A. caliginosa (Savigny), and Octolasion cyaneum (Say.) at 770, 1330, and 330 worms/[m.sup.2], respectively). Springett used the different densities of the various species in her experiments to simulate the relative abundances of these species that she had observed in the field. Springett (1983, 1985) also introduced A. longa into two pastures in New Zealand (at 170 worms/[m.sup.2]) where communities of A. caliginosa and L. rubellus already existed. Surface-applied lime moved deeper down the soil profile in the presence of A. longa in these pastures.

Accidentally introduced, endogeic lumbricids such as A. trapezoides (Duges), A. rosea (Savigny), and A. caliginosa are the most common earthworm species in soils used for pasture production in southern Australia (Baker 1998a; Baker et al. 1997). However, their distributions are patchy. Whilst some environmental variables (e.g. rainfall, soil type) help explain these patchy distributions, it seems likely that chance colonisation and poor dispersal ability have also been influential (Baker et al. 1997; Baker 1998a). These endogeic species are active in the top 10 cm of the soil for the coolest and wettest months of the year (winter-spring) (Baker et al. 1992, 1993a, 1993b). During summer, they aestivate deeper down the soil profile. In contrast to some other parts of the world (Lavelle 1983), anecic species, like A. longa, are generally rare in pasture soils in southern Australia, but A. longa is widespread in pastures in northern Tasmania (Kingston and Temple-Smith 1989; Baker et al. 1997). A. longa has only occasionally been found on mainland Australia but has the potential to colonise much of the temperate, higher rainfall region there (Baker 1998a; Baker et al. 1999a).

Baker et al. (1999a, 1999b) conducted field experiments in 7 permanent pastures on various soil types in South Australia and Victoria to determine the short-term (5 months) survivals and influences on pasture production of A. caliginosa, A. longa, and A. trapezoides when inoculated into cages at varying densities (and biomass). This paper reports on the relative influences that these earthworms had on lime burial within the same cages. The research aimed to test the strength of arguments for extending the current distributions of these 3 earthworm species, in particular A. longa, in southern Australia.

Methods

Field trials were established at 3 sites in South Australia in 1994 (Penola, Kuitpo, and Woodside) and 4 sites in western Victoria in 1995 (Hamilton, Strathdownie, Balmoral, and Apsley) (see Baker et al. 1999a for locations of these sites). All sites were in permanent, dryland pastures. Common plant species included perennial ryegrass (Lolium perenne L.), subterraneum clover (Trifolium subterraneum L.), white clover (T. repens L.), cocksfoot (Dactylis glomerata L.), phalaris (Phalaris aquatica L.), and various broad-leaf and grassy weeds, such as cape weed (Arctotheca calendula L.), storksbill (Erodium spp.), onion grass (Romulea rosea L.), and barley grass (Hordeum leporinum Link). Soils varied between sites and included a siliceous sand, sandy loams, and clay loams, all with clay beneath (see Baker et al. 1999a for further descriptions). Average annual rainfalls ranged between approximately 580 and 850 mm. At all sites, summers are usually hot and dry and winters cool and wet. The abundance of local earthworms within the pastures at the 7 field sites varied between 31 and 286/[m.sup.2] during the winters when the experiments were conducted (Baker et al. 1999a).

Sections of PVC pipe (each 0.3 in diam., 0.25 in long, with 6-mm-thick walls) were driven 20 cm into the soil at each site during spring (usually September). The pipes, containing intact soil cores, were lifted out of the ground during the following summer (usually February). Fine nylon mesh was placed across the bottom edge of most of the soil cores and secured with plastic strapping. The soil cores were then replaced in the ground. At the start of the following winter (usually June), earthworms were added to the soil surface within the PVC `cages'. Three species of earthworm were used: A. caliginosa, A. longa, and A. trapezoides. A. longa were collected from a pasture at Woolnorth in north-western Tasmania, A. trapezoides from pastures near Adelaide, South Australia, and A. caliginosa from either Woolnorth or near Adelaide. Four densities of earthworms (8, 15, 30, and 45 worms/cage, which is equivalent to 114, 214, 429, and 643/[m.sup.2]) were inoculated for each species at each site. The different species were caged separately. Only large juveniles and adults were used for each species. The fresh weights of individual earthworms were approximately 1.8 g (A. longa) and 0.5 g (A. trapezoides and A. caliginosa).

Three treatments were included at each site in which no earthworms were added. The first of these (controls) had mesh strapped across the bases of the soil cores as described above. The second and third treatments (residents, with and without lime; see below) had no mesh across the bases of the cores. This enabled invasions of local earthworms within the pastures during the course of the experiments. Cages were arranged adjacent to each other in rows, with 1-m spacings between rows. The various treatments were allocated in a stratified random manner amongst the cages at each site. There were 7 replicate cages for all treatments at each site.

After the earthworms had been added in early winter and they had moved underground, lime (Yates Gro-Plus Garden Lime; equivalent Ca[CO.sub.3] = 80.0%, 95% [is less than] 850 [micro]m; effective neutralising value = 91) was added to the soil surface within all cages (28.6 g/cage, equivalent to 4 t/ha), except in 7 of the resident cages. Mesh bags were then draped over 30-cm-high wire frames and strapped to the tops of all the PVC pipes. These bags prevented escape/invasion of earthworms across the soil surface but allowed plant growth. The herbage within each cage was cut to within 1 cm of the soil surface at the time the earthworms were added. It was then cut a further 2 or 3 times during the course of each trial (see Baker et al. 1999b for findings on the relative abilities of the earthworms to increase pasture production).

In spring (either October or November), 2 soil cores (5 cm diam., 20 cm depth) were taken centrally within each cage with a metal corer. These soil cores were divided into sections of 0-2.5, 2.5-5, 5-10, 10-15, and 15-20 cm depth. The paired soil samples for each depth within a cage were bulked, air-dried, and then lightly ground ([is less than] 2 mm). Soil pH was measured in 0.01 in Ca[Cl.sub.2] (soil: solution ratio of 1: 5). After the soil samples were taken for pH analyses, the remaining soil within all cages at each site was hand-sorted for earthworms, which were then identified and counted.

Results

There was some contamination of the A. caliginosa, A. longa, and A. trapezoides cages by local earthworms, but the intended species and abundance biases were achieved for each treatment (Fig. 1). The survival of A. longa was greater when averaged across all sites (74.75) than that of A. caliginosa (63.9%) and A. trapezoides (50.4%). More detailed data for the survivals (and biomass) of the worms within the cages at each site are presented in Baker et al. (1999a).

[Figure 1 ILLUSTRATION OMITTED]

The numbers of earthworms that were recovered in the resident cages in spring varied between sites from 0.3/cage at Penola to 18.9/cage at Hamilton (overall means across all 7 sites = 6.5 and 7.0 worms/cage for cages with and without lime, respectively). This was slightly less than the average number of earthworms found by Baker et al. (1999a) in the pastures adjacent to each site in winter (equivalent of 2.2-20.0 worms/cage, mean = 10.4). The surface soil within the pastures had dried substantially by spring at several sites (compared with winter). Very probably, many of the local earthworms had retreated from the soils within the resident cages by the time they were hand-sorted. Indeed, cursory examination of the soil beneath the cages at several sites in spring yielded aestivating earthworms.

The soils at all 7 sites were acidic (Fig. 2). Soil pH was generally lowest at about 5 cm depth. Soil pH varied between treatments and to varying degrees at the 7 sites. Using Strathdownie as an example, soil pH varied between treatments at all depths except 5-10 cm (Fig. 3). Compared with the limed control treatment, soil pH was reduced by A. longa at 0-2.5 cm depth, but was increased by all 3 species at 2.5-5 cm depth, particularly at the higher densities. A. longa at the highest density also increased soil pH at 10-15 and 15-20 cm depth. Liming increased soil pH at 0-2.5 cm depth in resident cages (t = 22.45, P [is less than] 0.001) but had no effect further down the profile (Fig. 3).

[Figures 2-3 ILLUSTRATION OMITTED]

In contrast to Strathdownie, earthworm effects on soil pH at Kuitpo were more limited to the soil surface (Fig. 4). At this site, soil pH varied between treatments at only the 0-2.5 and 2.5-5 cm depths. Compared with the limed control treatment, soil pH was increased by A. caliginosa and A. trapezoides at 0-2.5 cm. Soil pH was also increased by A. longa at 2.5-5 cm depth, but was not influenced by the other species. Liming again increased soil pH at 0-2.5 cm depth in resident cages at this site (t = 9.57, P [is less than] 0.001) but had no effect further down the profile (Fig. 4).

[Figure 4 ILLUSTRATION OMITTED]

For the other 5 sites, the maximum depths at which earthworms influenced soil pH (across all species and densities, relative to limed controls) were 0-2.5 cm for Penola (F = 6.72, P [is less than] 0.001), 5-10cm at Woodside (F = 4.14, P [is less than] 0.001), 15-20 cm at Balmoral (F 1.86, P [is less than] 0.05), 5-10 cm at Hamilton (F =

6.82, P [is less than] 0.001), and 5-10 cm at Apsley (F = 2.27, P [is less than] 0.05). At each of these sites, liming increased soil pH at 0-2.5cm in the resident cages, but at only one site did it increase soil pH below that depth (at Hamilton, soil pH was increased in the resident cages down to 5-10 cm depth, t = 2.94, P [is less than] 0.01).

Fig. 5 illustrates the average soil pH, at the 5 soil depths across all 7 sites, as a function of the numbers of inoculated worms. Table 1 gives the results of analyses of variance for these data. Soil pH varied with site, species of earthworm, and initial density of earthworm. Most notably: (1) A. longa reduced soil pH more than the other species at 0-2.5 cm depth; (2) soil pH increased with density for all species at 2.5-5 cm depth; (3) soil pH again increased with earthworm density at 5-10 cm depth, but this was only the case for A. caliginosa and A. longa; (4) there was no effect of earthworm density at 10-15 and 15-20 cm depth; and (5) soil pH was higher in the A. longa treatments compared with those for the other 2 species at 10-15 and 15-20 cm depth.

Table 1. Results of analyses of variance for soil pH at each of the five soil depths at the seven sites

Data are only presented for interactions between main variables where they are significant

Variable    F           P           F           P  

              0-2.5 cm                 2.5-5cm  

 Site     25.58      <0.001     58.59        <0.001  

 Species  40.97      <0.001      1.8         > 0.05  

 Initial density3.23 <0.05      32.11        <0.001  

 Site x species                  2.76        <0.01  

 Species x density               2.74        <0.05  

               5-10 cm                10-15 cm  

 Site     193.78     <0.001     186.89       <0.001  

 Species    3.58     <0.05        4.69       <0.05  

 Initial density12.89<0.001       1.50       > 0.05  

 Site x species2.16  <0.05  

 Site x density2.80  <0.005  

 Species x density3.32<0.05  

               15-20 cm  

 Site   136.31        <0.001  

 Species 7.01         <0.005  

 Initial density1.19 > 0.05  

[Figure 5 ILLUSTRATION OMITTED]

Across all 7 sites, the minimum inoculation densities required to increase soil pH below 2.5 cm depth were 30 worms/cage for A. longa and A. caliginosa and 45 worms/cage for A. trapezoides (paired t-tests based on the average soil pH for control and earthworm treatments at each site: t = 3.64, P [is less than] 0.05 for A. caliginosa and t = 3.51, P [is less than] 0.05 for A. longa at 30 worms/cage; t = 2.99, P [is less than] 0.05 for A. trapezoides at 45 worms/cage). However, lower thresholds for influence on lime burial (i.e. increase in soil pH below 2.5 cm depth) were detected at individual sites (e.g. 2-sample t-test for A. caliginosa, 15 worms/cage at Hamilton, t = 3.00, P [is less than] 0.05; for A. longa, 15 worms/cage at Kuitpo, t = 3.16, P [is less than] 0.05; for A. trapezoides, 30 worms/cage at Strathdownie, t = 2.72, P [is less than] 0.05).

Fig. 6 illustrates the average soil pH, at the 5 soil depths across all 7 sites, as a function of the biomass of inoculated earthworms. Table 2 gives the results of analyses of covariance for these data. Soil pH again varied with site, species of earthworm, and initial biomass of earthworm. In particular: (1) A. longa reduced soil pH more than the other species at 0-2.5 cm depth; (2) soil pH increased with earthworm biomass at all depths below 2.5 cm depth, in particular for A. longa at the lower depths; and (3) A. caliginosa, and to a lesser extent A. trapezoides, increased soil pH more than A. longa at 2.5-10 cm depth, but below this depth no differences were detected between species.

Table 2. Results of analyses of covariance for soil pH at each of the five soil depths at the seven sites

No interactions Covariate is initial biomass of inoculated worms

Variable  F         P         F          P  

            0-2.5 cm             2.5-5 cm  

 Site   27.62      <0.001     37.82      <0.001  

 Species60.20      <0.001     20.16      <0.001  

 Initial biomass2.80> 0.05    51.69      <0.001  

               5-10 cm            10-15 cm  

 Site  115.25      <0.001    231.02      <0.001  

 Species 5.24      <0.01       0.07      > 0.05  

 Initial biomass24.23<0.001   10.30      <0.005  

               15-20 cm  

 Site  154.81      <0.001  

 Species 0.96     > 0.05  

 Initial biomass7.9<0.01  

 [Figure 6 ILLUSTRATION OMITTED]

Not all earthworms were recovered from the cages at the end of each trial (see above). Some of these `missing' earthworms may have escaped through the occasional tear in the mesh, some may have been overlooked during hand-sorting, and some presumably died during the course of the experiment. Analyses of covariance were computed for soil pH with the final numbers of worms per cage and the final biomass of worms per cage as the covariates. Conclusions based on such analyses were essentially the same as those indicated above for initial numbers and biomass. They are therefore not presented here.

There was no difference between the soil pH in the limed control and limed resident cages across all 7 sites (paired t-tests comparing average soil pH for limed control and resident treatments at each site; t ranged from 0.04 to 0.71 for each depth, in all cases P [is greater than] 0.05).

Discussion

All 3 exotic lumbricids (A. caliginosa, A. longa, and A. trapezoides) significantly increased lime burial in pastures in south-eastern Australia within 5 months of applying lime to the soil surface. These results suggest that the 3 species could have a significant role in helping to ameliorate soil acidity in the region. All 3 species can also greatly increase pasture production (Baker et al. 1996; Baker 1998a, 1999b). In addition, A. longa can significantly improve dung burial and markedly increase soil turnover (Baker et al. 1996; Baker 1998a, 1999b; Curry and Baker 1998). There are therefore strong supportive reasons for introducing these lumbricids to pastures that lack them. But which species: all three or just one? The optimal earthworm community would seem to be one which contains the greatest behavioural/functional diversity (Lee 1995). Since the endogeic A. caliginosa and A. trapezoides are already widely, although patchily, spread throughout south-eastern Australia, and the anecic A. longa is very restricted in distribution and current communities in the region lack anecic species, most interest is currently focussed on the merits of introducing A. longa (Baker 1998a). The short-term survival of A. longa when introduced to a range of soil types in South Australia and Victoria has generally been high (Baker et al. 1999a). Preliminary studies have suggested that competition between A. longa and exotic endogeic species in pasture soils and invasion by A. longa into native woodland soils is likely to be small (Dalby 1996; Dalby et al. 1998). Increases in pasture production attributable to A. longa and A. caliginosa are additive in mixed communities (Springett 1985; Temple-Smith et al. 1993; Baker 1998b). How mixed communities might affect the distribution and availability of lime through soil profiles in the field has yet to be broadly determined, but the results of Springett's (1983, 1985) two field studies in New Zealand which showed that introductions of A. longa to endogeic communities increased the depth of lime burial are encouraging. Springett (1983) also observed in laboratory studies that A. caliginosa was more proficient in moving lime laterally than A. longa, but this has yet to be confirmed in the field.

Is A. longa better at burying lime than A. caliginosa or A. trapezoides? Where soil pH was plotted against numbers of introduced earthworms, A. longa appeared superior to both A. caliginosa and A. trapezoides. A. longa reduced soil pH more at the limed soil surface and increased it further down the profile than the other two species. This result is consistent with Springett's (1983, 1985) findings. However, if biomass rather than numbers was used as the measure of earthworm `abundance', differences between species were not so clear. A. longa still reduced soil pH at the surface more than the other two species, but at 2.5-10 cm depth A. caliginosa (and to a lesser extent A. trapezoides) was superior to A. longa in increasing soil pH. Presumably, this result reflected the much greater numbers of individual earthworms per unit of biomass for the smaller species and consequently a greater number of burrows through which lime could be moved. An appropriate overview would seem to be that A. longa is more able to remove lime from the soil surface and influence soil pH at greater depth than A. caliginosa and A. trapezoides (i.e. at [is greater than] 10 cm depth), but that A. caliginosa and A. trapezoides have the potential to have more impact in the upper soil layers (2.5-10 cm). This conclusion matches well with the behavioural patterns attributed to these species (i.e. anecic and endogeic) and the concept of lime being washed down wide, deep burrows in the case of A. longa and narrow, shallow burrows in the case of A. caliginosa and A. trapezoides.

The reduction of soil pH at the surface (0-2.5 cm depth) by A. longa may have reflected greater leaching of the surface-applied lime, perhaps down surface-venting burrows. Alternatively (or in addition), the reduced pH at the soil surface may simply have reflected the burial of the lime under castings made by A. longa. A. longa cast prolifically upon the soil surface; much more than the other species. The soil component of these casts may have come from layers further down the soil profile where pH was lower than at the surface (e.g. approximately 5 cm depth at most sites).

In this and similar studies (Baker et al. 1993c, 1995), 30 worms/cage (approximately 400 worms/[m.sup.2]) commonly influenced lime burial within 5 months (i.e. one earthworm `season'). Such densities are relatively high for pastures in south-eastern Australia (Mele et al. 1996; Baker 1998a), but moderate in comparison with the densities cited by Springett (1983) for New Zealand. Threshold densities of 15 worms/cage were recorded at some sites in the present study. Presumably, the effects of low densities of earthworms would be cumulative over several seasons. Further studies are required to determine the threshold densities needed in the longer term and whether pastures in south-eastern Australia have the carrying capacity to sustain them.

The extent of lime burial varied markedly between sites. Most notably, lime burial was generally poorer at the 3 South Australian sites (e.g. Kuitpo) in 1994 compared with the 4 Victorian sites (e.g. Strathdownie) in 1995. Rainfall was much less than average in South Australia in 1994 (e.g. 345 mm annual rainfall in 1994 at Kuitpo compared with the long-term average of 848 mm). In 1995, rainfall was higher at the Victorian sites and closer to average (e.g. 702 mm near Strathdownie compared with the local average of 693 mm). Low rainfall would reduce earthworm activity and leaching of lime down the soil profile. The poor performance of the earthworms at the South Australian sites can therefore be explained, at least in part, by the prevailing weather.

Baker et al. (1993c, 1995) reported that A. trapezoides buried lime to 10-15 cm depth beneath both a permanent pasture and a wheat crop. Increases in soil pH of almost I unit were measured at 4-6 cm depth. This degree of lime burial far exceeded those reported here. Again, soil moisture may help explain the discrepancy. Baker et al. (1993c, 1995) conducted their work during 1991 (pasture) and 1992 (wheat crop). These years were relatively wet. In 1991, 791 mm annual rainfall fell at the pasture site compared with a local average of 800 mm, and in 1992, 826 mm fell at the site with the wheat crop compared with an average of 496 mm. Lime burial was very poor beneath another wheat crop in 1994, which was a dry year (320 mm) (sown in the same field as used in 1992) (G. Baker, unpubl, data).

Other potential reasons for the variability in lime burial between the 7 sites studied in 1994 and 1995 include soil type and earthworm survival. However, there were no obvious associations between any of the soil characteristics measured at each site (e.g. clay content, soil C, N, P) (Baker et al. 1999a) and the degree of lime burial (taken as the maximum depth at which soil pH in one or more earthworm treatments significantly exceeded that of the control treatment), nor were there any obvious associations between earthworm survival and lime burial. For example, both Strathdownie and Penola had very sandy soils (Baker et al. 1999a), but lime burial was much better at Strathdownie than at Penola. A. longa buried lime to 15-20 cm depth at both Strathdownie and Balmoral, yet the soils were very different (much coarser textured and more acidic at Strathdownie). A. longa's survival was relatively poor at Strathdownie compared with Kuitpo (Baker et al. 1999a), yet its influence on lime burial was much greater at the former site. The large variability in lime burial recorded between sites provides a useful reminder of the need to replicate field trials at a number of contrasting sites in such studies.

The increases in soil pH noted at 10-20 cm depth in the presence of A. longa, whilst significant, were not large (e.g. average increases across all 7 sites of approximately 0.1 unit). Larger increases may of course have been achieved over a longer time interval than 4-5 months. Collections of larger volumes of soil for pH analyses would also probably have reduced sampling variance and increased the chance of detecting earthworm influences at some of the sites. The corers used to collect the soil only sampled 6% of the soil volume in each cage. Assuming that most of the lime that was incorporated deep into the soil was washed down burrows, it seems likely that the horizontal distribution of the lime in the deeper soil layers would have been quite patchy. The lime used for the experiments was not the finest quality available for ameliorating soil acidity in southern Australia (Merry et al. 1995). Fine lime may be leached through the soil profile more rapidly than coarse lime (Hodge and Lewis 1994). Perhaps better lime burial may have been obtained in this study if finer lime had been used.

Lime burial was generally poor in the resident cages at the 7 pasture sites in South Australia and Victoria (rarely any burial below 2.5 cm depth). The exception to this was at Hamilton, where lime was incorporated to 5-10 cm depth. Notably, Hamilton had the densest population of resident earthworms (186/[m.sup.2]).

Many field trials have been conducted in Australia and elsewhere in the world to evaluate pasture and crop responses to the liming of acid soils (Rowe and Johnson 1988; Cregan et al. 1989; Merry et al. 1990; Ridley and Coventry 1992; Scott and Cullis 1992; Date et al. 1995; Peoples et al. 1995). Lime movements down the soil profiles and plant growth responses have been very variable, both between years and between sites. Very rarely have the abundance, diversity, and activity of the soil fauna, in particular earthworms, been considered when evaluating these trials. The work described here strongly suggests that account should be taken of the soil fauna and its potential effects on lime burial when evaluating lime response trials in future.

Acknowledgments

Many people contributed to the running of the trials at particular sites. We especially thank Adam Brown, Paul Dalby, Li-Mei Loo, Chris Obst, Jason Woods, and Wendy Whitby for their assistance in the field and laboratory. We also thank the local farmers for access to their land, to collect worms and conduct the trials, particularly David Ford at Woolnorth. The research was funded in part by the Land and Water Research and Development Corporation.

References

Baker, G. H. (1998a). The ecology, management and benefits of earthworms in agricultural soils, with particular reference to southern Australia. In `Earthworm Ecology'. (Ed. C. A. Edwards.) pp. 229-57. (Soil & Water Conservation Service: Ankeny.)

Baker, G. H. (1998b). Recognising and responding to the influences of agriculture and other land-use practices on soil fauna in Australia. Applied Soil Ecology 9, 303-10.

Baker, G. H., Barrett, V. J., Carter, P. J., Buckerfield, J. C., Williams, P. M. L., and Kilpin, G. P. (1995). Abundance of earthworms in soils used for cereal production in south-eastern Australia and their role in reducing soil acidity. In `Plant Soil Interactions at Low pH'. (Eds R. A. Date, N. J. Grundon, G. E. Rayment, and M. E. Probert.) pp. 213-18. (Kluwer Academic Publishers: Dordrecht.)

Baker, G. H., Barrett, V. J., Carter, P. J., Williams, P. M. L., and Buckerfield, J. C. (1993a). Seasonal changes in the abundance of earthworms (Annelida: Lumbricidae and Acanthodrilidae) in soils used for cereal and lucerne production in South Australia. Australian Journal of Agricultural Research 44, 1291-301.

Baker, G. H., Barrett, V. J., Carter, P. J., and Woods, J. P. (1996). Method for caging earthworms for use in field experiments. Soil Biology and Biochemistry 28, 331-9.

Baker, G. H., Barrett, V. J., Grey-Gardner, R., and Buckerfield, J. C. (1992). The life history and abundance of the introduced earthworms Aporrectodea trapezoides and A. caliginosa (Annelida: Lumbricidae) in pasture soils in the Mount Lofty Ranges, South Australia. Australian Journal of Ecology 17, 177-88.

Baker, G. H., Barrett, V. J., Grey-Gardner, R., and Buckerfield, J. C. (1993b). Abundance and life history of native and introduced earthworms (Annelida: Megascolecidae and Lumbricidae) in pasture soils in the Mount Lofty Ranges, South Australia. Transactions of the Royal Society of South Australia 117, 47-53.

Baker, G. H., Carter, P. J., and Barrett, V. J. (1999a). Survival and biomass of exotic earthworms, Aporrectodea spp. (Lumbricidae), when introduced to pastures in south-eastern Australia. Australian Journal of Agricultural Research 50 (7). (In press.)

Baker, G. H., Carter, P. J., and Barrett, V. J. (1999b). Influence of exotic earthworms, Aporrectodea spp. (Lumbricidae), on pasture production in south-eastern Australia. Australian Journal of Agricultural Research 50 (7). (In press.)

Baker, G. H., Thumlert, T. A., Meisel, L. S., Carter, P. J., and Kilpin, G. P. (1997). `Earthworms Downunder'. A survey of the earthworm fauna of urban and agricultural soils in Australia. Soil Biology and Biochemistry 29, 589-97.

Baker, G. H., Williams, P. M. L., Barrett, V. J., and Carter, P. J. (1993c). Burial of surface-applied lime by earthworms in South Australian pastures. Proceedings of the XVII International Grassland Congress 1, 939-41.

Bouche, M. B. (1977). Strategies lombriciennes. In `Soil Organisms as Components of Ecosystems'. (Eds U. Lohm and T. Persson.) Proceedings of the 6th International Colloquium on Soil Zoology. Swedish Natural Science Research Council Ecological Bulletin No. 25, pp. 122-33.

Chartres, C. J., Helyar, K. R., Fitzpatrick, R. W., and Williams, J. (1992). Land degradation as a result of European settlement of Australia and its influence on soil properties. In `Australia's Renewable Resources: Sustainability and Global Change. Bureau of Rural Resources Proceedings No. 14'. (Eds R. M. Gifford and M. M. Barson.) pp. 3-33. (Australian Government Publishing Service: Canberra.)

Conyers, M. K., and Scott, B. J. (1989). The influence of surface incorporated lime on subsurface soil acidity. Australian Journal of Experimental Agriculture 29, 201-7.

Coventry, D. R. (1985). Changes in agricultural systems on acid soils in southern Australia. In `Proceedings III Australian Agronomy Conference'. (Ed. J. J. Yeates.) pp. 126-45. (Australian Society of Agronomy: Melbourne.)

Cregan, P. D., Hirth, J. R., and Conyers, M. K. (1989). Amelioration of soil acidity by liming and other amendments. In `Soil Acidity and Plant Growth'. (Ed. A. D. Robson.) pp. 205-64. (Academic Press: Sydney.)

Curry, J. P., and Baker, G. H. (1998). Cast production and soil turnover by earthworms in soil cores from South Australian pastures. Pedobiologia 42, 283-7.

Dalby, P. R. (1996). Competition between earthworms in high rainfall pastures in the Mount Lofty Ranges, South Australia. PhD Thesis, University of Adelaide.

Dalby, P. R., Baker, G. H., and Smith, S. E. (1998). Potential impact of an introduced lumbricid on a native woodland in South Australia. Applied Soil Ecology 9, 351-4.

Date, R. A., Grundon, N. J., Rayment, G. E., and Probert, M. E. (Eds) (1995). `Plant-Soil Interactions at Low pH.' (Kluwer Academic Publishers: Dordrecht, The Netherlands.)

Helyar, K. R. (1991). The management of acid soils. In `Plant-Soil Interactions at Low pH'. (Eds R. J. Wright, V. C. Baligar, and R. P. Murrmann.) pp. 365-82. (Kluwer Academic Publishers: Dordrecht, The Netherlands.)

Hodge, T. J. V., and Lewis, D. C. (1994). Influence of lime rate and particle size on soil pH and vegetative and seed yields of subterranean clover in the south east of South Australia. Australian Journal of Experimental Agriculture 34, 367-71.

Kingston, T. J., and Temple-Smith, M. G. (1989). Earthworm populations under Tasmanian pastureland. In `Proceedings of the V Australasian Conference on Grassland Invertebrate Ecology'. (Ed. P. P. Stahle.) pp. 192-8. (D&D Printing: Vic.)

Lavelle, P. (1983). The structure of earthworm communities. In `Earthworm Ecology'. (Ed. J. E. Satchell.) pp. 449-66. (Chapman & Hall: London.)

Lee, K. E. (1995). Earthworms and sustainable land use. In `Earthworm Ecology & Biogeography in North America'. (Ed. P. F. Hendrix.) pp. 215-34. (Lewis Publishers: Boca Raton, FL.)

Mele, P. M., Baker, G. H., and Blakemore, R. (1996). Introduced species: populations and distribution in agricultural land. In `The Role of Earthworms in Agriculture and Land Management'. (Eds M. Temple-Smith and T. Pinkard.) pp. 39-46. (Department of Primary Industries and Fisheries: Launceston, Tas.)

Merry, R. H., Hodge, T. J. V., Lewis, D. C., and Jacka, J. (1995). Evaluation of liming materials used in South Australia. In `Plant Soil Interactions at Low pH'. (Eds R. A. Date, N. J. Grundon, G. E. Rayment, and M. E. Probert.) pp. 497-503. (Kluwer Academic Publishers: Dordrecht, The Netherlands.)

Merry, R. H., Tiller, K. G., and Richards, A. F. (1990). Variability in characteristics of some acidic pasture soils in South Australia and implications for lime application. Australian Journal of Soil Research 28, 27-38.

Peoples, M. B., Lilley, D. M., Burnett, V. F., Ridley, A. M., and Garden, D. L. (1995). Effects of surface application of lime and superphosphate to acid soils on growth and [N.sub.2] fixation by subterranean clover in mixed pasture swards. Soil Biology and Biochemistry 27, 663-71.

Ridley, A. M., and Coventry, D. R. (1992). Yield responses to lime of phalaris, cocksfoot, and annual pastures in north-eastern Victoria. Australian Journal of Experimental Agriculture 32, 1061-8.

Rowe, B. A., and Johnson, D. E. (1988). Residual effects of limestone on pasture yields, soil pH and soil aluminium in a krasnozem in north-western Tasmania. Australian Journal of Experimental Agriculture 28, 571-6.

Scott, B. J., and Cullis, B. R. (1992). Subterranean clover pasture responses to lime application on the acid soils of southern New South Wales. Australian Journal of Experimental Agriculture 32, 1051-9.

Springett, J. A. (1983). Effect of five species of earthworm on some soil properties. Journal of Applied Ecology 20, 865-72.

Springett, J. A. (1985). Effect of introducing Allolobophora longa Ude on root distribution and some soil properties in New Zealand pastures. In `Ecological Interactions in Soil'. (Ed. A. H. Fitter.) pp. 399-405. (Blackwell: Oxford.)

Temple-Smith, M. G., Kingston, T. J., Furlonge, T. L., and Garnsey, R. B. (1993). The effect of introduction of the earthworms Aporrectodea caliginosa and Aporrectodea longa on pasture production in Tasmania. In Proceedings VII Australian Agronomy Conference'. (Eds G. K. McDonald and W. D. Bellotti.) p. 373. (Australian Society of Agronomy: Melbourne.)

Von Uexkull, H. R., and Mutert, E. (1995). Global extent, development and economic impact of acid soils. In `Plant-Soil Interactions at Low pH'. (Eds R. A. Date, N. J. Grundon, G. E. Rayment, and M. E. Probert.) pp. 5-19. (Kluwer Academic Publishers: Dordrecht, The Netherlands.)