Characterisation of the three-dimensional structure of earthworm burrow systems using image analysis and mathematical morphology

The aim of this work was to exemplify the specific contribution of both two- and three-dimensional (3D) X-ray computed tomography to characterise earthworm burrow systems. To achieve this purpose we used 3D mathematical morphology operators to characterise burrow systems resulting from the activity of an anecic ( Aporrectodea nocturna), and an endogeic species ( Allolobophora chlorotica), when both species were introduced either separately or together into artificial soil cores. Images of these soil cores were obtained using a medical X-ray tomography scanner. Three-dimensional reconstructions of burrow systems were obtained using a specifically developed segmentation algorithm. To study the differences between burrow systems, a set of classical tools of mathematical morphology (granulometries) were used. So-called granulometries based on different structuring elements clearly separated the different burrow systems. They enabled us to show that burrows made by the anecic species were fatter, longer, more vertical, more continuous but less sinuous than burrows of the endogeic species. The granulometry transform of the soil matrix showed that burrows made by A. nocturna were more evenly distributed than those of A. chlorotica. Although a good discrimination was possible when only one species was introduced into the soil cores, it was not possible to separate burrows of the two species from each other in cases where species were introduced into the same soil core. This limitation, partly due to the insufficient spatial resolution of the medical scanner, precluded the use of the morphological operators to study putative interactions between the two species     

Structural parameters and functional patterns of simulated earthworm burrow systems

Summary Earthworm burrow systems have been simulated by modelling, from field observations, five main characteristics of burrow units: density (number of burrows per unit volume), mean length, angular orientation, middle coordinates and diameter. The functional properties of the burrow systems have been analysed using structural parameters, i. e. number of full perforations, mean available burrow surface and minimal burrow spacing, and by estimating the shortest pathway through the burrow system. The simulations are used to study the effects of each of the main characteristics on structural parameters and the shortest pathway. This work reveals that the specific role of density is to lead to a threshold by determining an optimal horizontal distance, and that the mean length governs the vertical critical dimension of the system. These simulations may be applied to explain the nature of field burrow systems.     

Factors affecting surface casting behaviour in several species of earthworm

Summary Surface cast production was investigated for several species of earthworms, including some not normally considered to produce surface casts. In single-species culture, the amount of cast material deposited on the surface by different species varied, with Allolobophora chlorotica < Lumbricus rubellus < Aporrectodea caliginosa < Aporrectodea longa. In field communities, results indicated that a single species dominated surface casting activity. The order of species dominance was consistent with the above findings from single-species cultures, if Lumbricus terrestris was considered to be equivalent to Ap. longa.Surface casting is an essential function within earthworm communities which maintains their living space. However, it involves an energy cost and carries a risk of predation. Therefore, there are disadvantages associated with this activity. All of the species studied produced surface casts unless other, usually larger, species were present.     

Effects of metal pollution on soil macroinvertebrate burrow systems

A reduction in the numbers of macroinvertebrates present in soil may have a negative effect on soil structure, infiltration rates, and gas exchanges. Soil pollution by metal is known to have a detrimental effect on soil macrofauna. The aim of the present study was to evaluate (1) direct and indirect effects of soil pollution on soil macroinvertebrate bioturbation and (2) effects of the two macroinvertebrate communities found in a polluted and a nonpolluted area (one supposed sensitive, the other tolerant to metals) on burrow systems parameters. Macroinvertebrate porosity was studied using X-ray tomography. Three-dimensional reconstructions and characterisation of the burrow system were obtained using image analysis. Results showed that metal pollution principally affected the spatial distribution of macropores (more macropores were found near the soil surface) and the shape of the burrow system (branching rate was higher in the polluted soil), whereas soil macroinvertebrate composition principally affects burrow density parameters (the number of burrows was higher for the sensitive macroinvertebrate community).     

Assessment of the burrow system of Lumbricus terrestris, Aporrectodea giardi, and Aporrectodea caliginosa using X-ray computed tomography

Artificially packed soil columns were inoculated with individuals of different earthworm species (Lumbricus terrestris, Aporrectodea giardi or Aporrectodea caliginosa) and placed under controlled conditions in the laboratory. At the end of the incubation period, which lasted 8 months, three-dimensional reconstructions of the burrow system of each species were obtained using X-ray computed tomography. The particular features of the three burrow systems and the differences between them are discussed in terms of density, orientation and distribution of the burrows and the complexity of the network in relation to variations in soil depth and soil density. Received: 5 February 1997     

Volume density of earthworm burrows in compacted cores of soil as estimated by direct and indirect methods

After earthworms of the species Aporrectodea caliginosa and A. rosea had burrowed in compacted cores of soil for 68 days the cores were sectioned horizontally. The upper surface of each sectioned layer of soil was photographed before it was dissected and the dimensions of all burrows within the layer measured. Volume densities calculated from the direct measurement of burrows were compared with the values calculated by stereology from data obtained from two image analysis methods, computerised image analysis and point counting with a systematic lattice. The assumption implicit in all stereology calculations was satisfied for this experiment in that the burrows of both species showed no preferred orientation in the compacted soil. Computerised image analysis could not measure the density of all burrows in the photographs because of the lack of contrast between cast-filled burrows and the soil and the complex shapes of the burrows. Although the volume densities of A. caliginosa burrows calculated from point counts were correlated with the values calculated from direct measurement, point counting over-estimated volume densities by two to three times. In the experiment, A. rosea produced too few airfilled burrows for the lattice to detect. The relative ratios of air-filled to cast-filled burrows calculated from the point counts suggested that approximately two-thirds and eight-ninths of the burrows of A. caliginosa and A. rosea, respectively, were filled with casts.     

Assessment of earthworm burrowing efficiency in compacted soil with a combination of morphological and soil physical measurements

Summary Column experiments were carried out to quantify the effect of earthworms on compacted soil. The earthworms (Lumbricus terrestris) were able to burrow into soil which was artificially compacted to a pore volume as low as 40%; they may also penetrate an artificial plough pan deep in the soil. The effect of the burrowing activity of Lumbricus terrestris was quantified by measuring hydraulic conductivities and infiltration rates through the whole soil column (19 cm wide, 40 cm long). Morphological parameters, mainly the vertically projected burrow depth, were correlated with the saturated hydraulic conductivity. The amount of casts deposited by Lumbricus terrestris on the soil surface increased with the degree of soil compaction. The bulk density of casts was always less than that of the original soil.     

Evolution of burrow systems after the accidental introduction of a new earthworm species into a Swiss pre-alpine meadow

 The unintentional introduction of a new earthworm species (Aporrectodea nocturna) into a Swiss pre-alpine meadow resulted in a great increase in earthworm density in the newly colonized area (386 m^–2) compared with the density observed in the natural area (273 m^–2) where an earthworm community was already present. To investigate the impact of this introduction on the burrow systems, eight soil cores (length 25 cm, diameter 16 cm) were taken (four in the colonized area and four in the natural area) and analysed with computer-assisted X-ray tomography. The resulting images were processed to obtain the 3D-skeleton reconstructions of the earthworm burrow systems. Due to high variability in these burrow systems, only slight differences were observed between the two areas. The total burrow length and the mean burrow lengths tended to be greater in the colonized area. Moreover, the distribution of pore numbers with depth showed different patterns with a maximum for depths between 10 cm and 15 cm in the colonized area and a maximum for depths between 20 cm and 25 cm in the natural area. These differences may have been related to: (1) the particular behaviour of A. nocturna, which was observed to cast at the surface in this site, and (2) the predominance of juvenile earthworms around the colonization front. These differences were sufficient to create significant effects on the continuity of the burrow systems (assessed by the number of different pathways between virtual horizontal planes) for the two areas. The colonized area was characterized by a greater pore continuity, which could have resulted in enhanced transfer properties. Received: 2 July 1999     

Characterization of the burrow system of the earthworms Lumbricus terrestris and Aporrectodea giardi using X-ray computed tomography and image analysis

 The burrow systems of two earthworm species (Lumbricus terrestris and Aporrectodea giardi) were studied in artificially packed soil columns placed in controlled conditions in the laboratory. At the end of the incubation, which lasted 246 days, the burrow systems were characterized on the undisturbed columns using X-ray computed tomography. This method provided a set of digitized images corresponding to horizontal 3-mm-thick sections. The following parameters were measured using image analysis on each section: number of biopores, their volume, and their individual orientation calculated using an elliptical model of the earthworm channel. The profiles of these parameters through the columns showed that the burrow systems of L. terrestris and A. giardi, which are both anecic species, were very different in terms of total volume, number of burrows, burrow orientation and extension with soil depth. These results led us to conclude that the burrow system of L. terrestris can be considered as a permanent structure whereas that of A. giardi is closer to the burrow system of endogeic species. Received: 12 June 1998     

Novel method for the study of the population dynamics of a genetically modified microorganism in the gut of the earthworm Lumbricus terrestris

Summary A laboratory microcosm study was used to investigate the survival and population dynamics of genetically modified microorganisms (GMM) in the gut of Lumbricus terrestris. Three methods of axenic earthworm production were investigated. An antibiotic mixture of streptomycin and cycloheximide was introduced either passively, mixed with sterile soil or cellulose, or actively, by teflon catheter. Worms treated by all methods lost weight but this was least for the catheter method which was also the only method to produce axenic earthworms. Axenic earthworms were used to determine the effect of competition with indigenous gut bacteria on ingested GMM. The GMM used was Pseudomonas fluorescens, strain 10586/FAC510, with chromosomally inserted Lux genes for bioluminescence, and chromosomal resistance to rifampicin. The bacteria were grown up to the mid-exponential phase before inoculation into earthworms. Bacteria in faecal material were enumerated by dilution plate counting using selective agar. The GMM were re-isolated from the casts of both antibiotic-treated and untreated earthworms. Lower concentrations of GMM and higher concentrations of indigenous bacteria in the casts of untreated compared to antibiotic-treated earthworms suggested that competition is a fundamental control on population dynamics of the introduced bacterial inocula ingested by earthworms.The catheter method, developed in this study, is proposed as a technique to contribute to the risk assessment of environmental release of GMM.     

Sublethal effects of imidacloprid on the burrowing behaviour of two earthworm species: Modifications of the 3D burrow systems in artificial cores and consequences on gas diffusion in soil

Earthworms have been termed ‘ecosystem engineers’ (sensu [Jones, C.G., Lawton, J.H., Shachak, M., 1994. Organisms as ecosysem engineers. Oikos 69, 373-386.]) because of the important roles they play in the soil. As a consequence, it is assumed that if earthworms change their behaviour following exposure to pesticides or pollutants this could have a drastic impact on soil functioning. To test this assumption under laboratory conditions, we studied the burrow systems made by two earthworm species (the anecic Aporrectodea nocturna and the endogeic Allolobophora icterica) in artificial soil cores containing imidacloprid, a widely used neonicotinoid insecticide. After 1-month incubation period, the macropores created in the soil core were analyzed by tomography. In order to further characterize transfer properties associated with burrow systems gas diffusion measurements were also carried out. The burrow systems made by the two earthworm species were very different: A. nocturna made more continuous, less branched, more vertical and wider burrows than A. icterica. Some changes to A. nocturna burrow systems were observed after exposure to imidacloprid (they made a smaller burrow system and burrows were more narrow), but only at the highest concentration of imidacloprid used (0.5 mg kg⁻¹). A. icterica worms were more sensitive to imidacloprid and many differences in their burrow systems (length, sinuosity, branching rate and number of burrows) were observed at both concentrations tested (0.1 and 0. 5 mg kg⁻¹). As a consequence, the continuity of the burrow systems made by both species was altered following imidacloprid treatment. Gas diffusion through the A. nocturna soil cores was reduced but no difference in gas diffusion was observed in the A. icterica soil cores.     

A comparison of three earthworm bioassay procedures for the assessment of environmental samples containing hazardous wastes

Summary Three different laboratory earthworm protocols for assessing the potential toxicity of environmental samples were evaluated using Eisenia fetida. The 48-h Contact Test (CT) is a short test and may indicate the presence of water-soluble chemicals. The 14-day Soil Test (ST) is best suited to assess chemical impacts of either known or unknown chemicals in soils. The Giant Axon Conduction Velocity protocol (GACVT) used with either the CT or the ST can detect sublethal neurotoxic effects in shorter periods of time and at lower chemical concentrations than mortality tests. The use of these tests is suggested as an estimate of potential toxicity of environmental soil samples.     

Ecology of the earthworm Allolobophora carpathica in field and laboratory studies

Allolobophora carpathica is an earthworm species found in the Eastern Carpathian mountain region, but little information is available on its ecology. Field sampling in beech woodland of the Bieszczady National Park, SE Poland, found this species in soils with a pH of less than 5.0 and C:N ratios of 12 to 13. From sampling over 2 years, at 4 sites with differing sub-vegetation, a mean density of A. carpathica of 6.75 individuals m^-2 with a biomass of 8.65 g.m^-2 was recorded. The largest mature adults were in excess of 14.0 g but there were significant differences (p < 0.01) between sampling sites. Seasonal patterns of abundance were observed.To obtain more data on the growth and breeding biology of this species, specimens were collected from the field, taken to laboratories and maintained in mesocosms under controlled environmental conditions. Cocoon production was 0.88 cocoons per earthworm per month at 15^oC, but under fluctuating temperature conditions (16 - 24^oC) no cocoon production occurred. The mean cocoon biomass was 83 mg  (n = 104) and incubation took 178 days at 15^oC (n = 14), although hatchability was low (22%) under these conditions. Growth from the hatchling stage, (mean mass 86 mg; n = 18), to maturity (c. 8.5 g) took 8 - 12 months at 12^oC.     

Burrowing ability of the earthworm Aporrectodea longa limited by soil compaction and water potential

Summary Adult earthworms (Aporrectodea longa) were maintained for 199 days in soil columns (h=30 cm; ø=10 cm) where the water potential ranged from -7 to-65 kPa and compaction from 50 to 350 kPa. The weight of casts on the soil surface was measured at the end of two periods of activity (spring and autumn). Cast production increased with bulk density, but the activity of earthworms was limited both by the mechanical strength of the soil and by decreasing water potential. The results obtained in the laboratory conditions of this study were consistent with field observations on casting and burrowing activities. The effects of water potential and soil compaction on these activities were estimated.     

Potential for higher rates of denitrification in earthworm casts than in the surrounding soil

Summary Earthworms (Lumbricus terrestris L.) were cultured in the laboratory and fed on lucerne (Medicago sativa L.). Denitrification rates in the surface casts and the surrounding soil were quantified using C₂H₂-inhibition of nitrous oxide reductase. The investigation also included determination of the N₂O-formation by nitrification as well as CO₂-formation as a measure of respiration. The denitrification rates of wet earthworm casts were found to be significantly higher than those occurring in wet samples from the soil. The low N₂O-formation observed seemed to be due to denitrification. Respiration was higher in casts, indicating higher oxygen demand which resulted in more anaerobic conditions. The energy supply was probably better in casts compared with the surrounding soil.     

Fate and dispersal of recombinant bacteria in a soil microcosm containing the earthworm Lumbricus terrestris

Summary A model ecosystem was designed containing Lumbricus terrestris in soil. The microcosm is suitable for assessing the dispersal and survival of genetically engineered microorganisms. After the inoculation of genetically engineered microorganisms their number in the worm excretions decreased continuously to reach the level of the detection limit within 50 days.     

Reproduction and growth of three deep-burrowing earthworms (Lumbricidae) in laboratory culture in order to assess production for soil restoration

Laboratory experiments were conducted to examine the growth and reproduction of three deep-burrowing lumbricids, Aporrectodea longa, Lumbricus terrestris, and Octolasion cyaneum. The reproductive output was recorded as 18.8, 38.0, and 32.3 cocoons per worm per year for A. longa, L. terrestris, and O. cyaneum, respectively. For the same species, maturity was reached at a mean mass of 3.9, 5.0 and 2.4 g, within 3 months from the hatchling stage by L. terrestris and within 4 months by the other two species. The hatching success of cocoons at 15 and 20°C was within the range of 70–80% for each species, except A. longa at the higher temperature, where a viability of 47% was recorded. Twenty percent of viable O. cyaneum cocoons produced twin hatchlings, compared with only one percent for A. longa and L. terrestris. A combination of these results suggests that a complete life-cycle for each species could be achieved within 6 months (L. terrestris and A. longa) or 7–8 months (O. cyaneum). Each species has particular life-cycle strategies that would aid survival and colonisation, under field conditions, if inoculated into restored soils.     

Nitrogen mineralization and reorganization in casts of the geophagous tropical earthworm Pontoscolex corethrurus (Glossoscolecidae)

Summary Mineral N concentrations ranged from 133.1 to 167.8 g g^-1 dry soil in fresh casts of the endogeic earthworm Pontoscolex corethrurus fed on an Amazonian Ultisol; this was approximately five times the concentration in non-ingested soil. Most of this N was in the form of NH _inf4 ^sup+ . N also accumulated in microbial biomass, which increased from a control value of 10.5–11.3 to 67.5–74.1 g g^-1 in fresh casts. During a 16-day incubation, part of the NH _inf4 ^sup+ -N was nitrified and/or transferred to the microbial biomass. Total labile N (i.e., mineral+biomas N) decreased sharply at first (ca. 50% in the first 12 h), and then more slowly. The exact fate of this N (microbial metabolites, denitrification, or volatilization) is not known. After 16 days, the overall N content of the casts was still 28% higher than that of the control soil. Incubation of the soil before ingestion by the earthworms significantly increased the production of NH _inf4 ^sup+ in casts. We calculate that in a humid tropical pasture, 50–100 kg mineral N may be produced annually in earthworm casts. Part of this N may be conserved in the compact structure of the cast where the cast is not in close contact with plant roots.     

Earthworms of different functional groups affect the fate of the Bt-toxin Cry1Ab from transgenic maize in soil

The fate of the insecticidal Cry1Ab protein from crop residues (leaves and roots) of the transgenic maize variety MON810 was studied in the presence and absence of two earthworm species (Lumbricus terrestris, Aporrectodea caliginosa; separate incubations) in soil microcosms. The recombinant Cry1Ab protein was quantified using a highly sensitive ELISA. Control microcosms received corresponding non-transgenic plant material. All earthworms survived in the microcosms over a period of 5 weeks, irrespective of whether they received MON810 or non-transgenic plant material. Weight loss was observed for both earthworm species, independent of the plant material or transgenic modification. A strong decline of immunoreactive Cry1Ab in plant residues (mean initial concentration approx. 5000 ng g⁻¹) of MON810 was observed in all treatments, but in microcosms with earthworms this decline was significantly higher with less than 10% of the initial Cry1Ab concentration remaining after 5 weeks. Cry1Ab concentrations in casts were only 0.1% of those found in remaining plant material of the respective microcosms. No immunoreactive Cry1Ab proteins were found in earthworm tissues (threshold of detection: 0.58 ng g⁻¹ fresh weight). No further decline was found for Cry1Ab concentrations in casts of A. caliginosa during a subsequent period of 3 months of incubation in bulk soil (<0.1 ng g⁻¹) after removal of the earthworms from the microcosms, while in casts of L. terrestris the concentration decreased from 0.4 to below 0.1 ng g⁻¹. In conclusion, this study demonstrates that earthworms enhance the decline of immunoreactive Cry1Ab proteins from maize residues.