HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by VandenBygaart, A.J. Articles by Protz, R. Search for Related Content PubMed Articles by VandenBygaart, A.J. Articles by Protz, R. GeoRef GeoRef Citation Agricola Articles by VandenBygaart, A.J. Articles by Protz, R.

DIVISION S-5-PEDOLOGY

Estimating Earthworm-Influenced Soil Structure by Morphometric Image Analysis

^a Dep. of Land Resource Sci., Univ. of Guelph, Guelph, ON, Canada N1G 2W1
^b Southern Crop Protection and Food Res. Centre, Agriculture and Agri-food Canada, London, ON, Canada N5V 4T3

bvandenb{at}uoguelph.ca

	ABSTRACT

TOP ABSTRACT INTRODUCTION Methods Results Conclusions REFERENCES

 
Earthworms have a profound influence on soil processes. However, there is generally a lack of adequate means by which to assess the influence of earthworms on soil structure. Not until quantitative methods on undisturbed soil samples are developed will there be any adequate measure of the influence of earthworms on soil structure. This paper describes an extension of an image-analysis method developed for the quantitative determination of the influence of earthworms on soil structural properties. Mammillated vughs are most likely developed by the burrowing of soil macrofauna, in particular earthworms. A learning set of mammillated vughs was compiled with pores taken from a soil developed solely through the channeling and casting of earthworms. This learning set was used to classify soil blocks taken from a no-till and conventionally tilled treated soil. The results indicated that the no-till soils had more than twice the number of mammillated vughs >1000 µm in diameter. This was attributed to the larger earthworm population in the no-till soils, coupled with the change in morphology or destruction of some of the mammillated vugh features caused by disturbance in the conventionally tilled soil. This method should allow for a more effective means to evaluate the influence of earthworms on soil properties within any given soil profile.

Abbreviations: CS, convex shape • CT, conventionally tilled • NT, no-till

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Methods Results Conclusions REFERENCES

 
SOIL ECOSYSTEM ENGINEERS are those large soil fauna that predominantly influence the development and maintenance of the soil structure (Lavelle et al., 1997). Earthworms are identified as the most important soil ecosystem engineers in soils of the temperate regions of the world (Lavelle et al., 1997). This is supported by the many observations of the activity of earthworms and estimates of the rates of soil ingestion and turnover (Darwin, 1881; Tomlin et al., 1995; VandenBygaart et al., 1998). However, there is still a general lack of knowledge about the importance of earthworm activity to the long-term viability and productivity of agricultural soil (Edwards and Shipitalo, 1998).

Many researchers have identified the strong influence of earthworms on the structural genesis of A horizons. Van de Westeringh (1972) observed that up to 50% of the aggregates in surface layers of a temperate orchard soil were recognizable as earthworm casts. Kubiena (1953) and Dobrovol'skii and Titkova (1960) identified nearly all the aggregates of an Ah horizon as earthworm casts in mull-type forest soils of central Europe. Lavelle (1978) determined that the top 15 to 20 cm of wooded savanna soils of the Ivory Coast were composed almost entirely of earthworms castings. Blanchart et al. (1997) went so far as to suggest that in soils that do not have severe episodes of wet–dry cycles, earthworms are the major formers of soil aggregates. However, there is still a lack of truly quantitative means by which to assess the influence of earthworms on soil structure.

Soil micromorphology has been utilized in the past to interpret soil processes and the nature and arrangement of soil features (Miedema, 1997). Previously there have been many qualitative interpretations of the faunal influence on the structure on soil blocks and thin sections utilizing methods in micromorphology (Brewer, 1964; Bal, 1970, 1973; Drees et al., 1994; VandenBygaart et al., 1999a, 1999b). The variations in form and size of aggregates and pores created by macrofauna have been assessed by McKenzie and Dexter (1985). They measured the shape characteristics of earthworm fecal pellets by measuring the area and perimeter of freshly deposited casts from the species Aporrectodea rosea. Although they were able to conclude that the casts were rounder than aggregates taken from a field soil, the method did not account for the undisturbed nature of the packing and distribution of earthworm structural units within the soil profile. Only until undisturbed methods are championed will there be any quantitative determination of the influence of earthworms on soil structure. Kooistra (1991) has attempted the only work to date on undisturbed samples but she used visual identification of faunally-derived morphology on thin sections. Using point-counting techniques, she found that in a minimum-tillage soil, 75 to 80% of the 11- to 25-cm layer consisted of pores developed by soil organisms, including roots. This contrasted with <10% of the volume of pores developed by soil fauna of an Ap horizon modified by annual tillage. Kooistra (1991) suggested that there is a requirement for a quantitative method for assessing the influence of earthworms on soil structure in undisturbed soil samples with varying textures and climates.

This paper describes a method for estimating earthworm-influenced soil structure using techniques in morphometric image analysis of soil micromorphology.

	Methods

TOP ABSTRACT INTRODUCTION Methods Results Conclusions REFERENCES

 
Pore Morphology and Classification
Brewer (1964) set out the most widely accepted classification of pores based on their morphology, and also related their development to specific pedogenic processes. Brewer (1964) identified six main types of pores that can be identified in 2-D sections of soils:

1. Packing pores develop through the loose packing of sand grains or aggregates, and as a result are interconnected in 2-D and 3-D soil space.
   
2. Vughs are relatively large, irregularly shaped pores that are not interconnected on 2-D thin sections.
   
3. Channels are rounded or cylindrical pores that are larger than could be developed by the packing of grains.
   
4. Planes are cracks that develop mainly from shrink–swell processes.
   
5. Chambers form at the ends of earthworm burrows and termite galleries.
   
6. Vesicles are smooth, spherical cavities formed under the surface of crusts created by raindrop impact.
   

Pores Developed by Soil Fauna
Brewer (1964) went further to describe the genesis of pores and to interpret the presence of certain types of pores to pedogenic processes occurring in soil. Channels are created by the burrowing of earthworms and/or large roots. In thin section, channels can appear as circular or ellipsoidal in shape; or if sectioned through their long axis, they appear as long, regular or curved rods. Chambers are domed cavities extending from a channel created by earthworms, or galleries created by termites and ants. They often contain appreciable amounts of fecal material derived from the soil fauna.

Brewer (1964) further subdivided vughs based on their origin. Orthovughs have irregular, rough walls, and are distinguishable from packing pores, which are well interconnected and tend to have smooth boundaries. They are thought to develop from the complex packing of aggregates as a result of the adhesion of flocculated clay to aggregates and mineral grains. Mammillated vughs, however, have smooth walls and are easily differentiated from normal packing pores and orthovughs by the presence of a mammillated conformation. The mammillations are internal protrusions of the pore walls that are rounded and smooth, and are likely created by the packing and coalescing of earthworm fecal pellets (or casts) (Brewer, 1964).

Image Analysis: Background
Automated image analysis has been utilized as a means of quantifying pore structure since the development of the Quantimet image-analysis system (Jongerius et al., 1972). With the recent rapid advancement in digital cameras, computer processing, physical memory, and software, complete image-analysis systems can be readily built for the quantitative image analysis of soil micromorphology (Protz and VandenBygaart, 1998). Protz and VandenBygaart (1998) summarize a methodology for the acquisition of spectral data from soil thin sections and blocks.

Object morphometry by image analysis deals with the measurement of parameters of objects in an image. Although soil pore space in three dimensions is theoretically completely interconnected when sectioned as in thin sections and soil blocks, the soil pores appear as unique, separate objects such that measurements can be performed on them. General parameters used to measure pore morphology in image analysis include area (total number of pixels in the object), perimeter (number of pixels along the periphery of the object), feret diameter (diameter of the smallest circle that can fit completely within the object), and long-axis length (the length in pixels of a line connecting the two farthest pixels in the object). A shape factor can be calculated based on the area and the perimeter (S = area/perimeter²). This measure is often used to differentiate pores that are circular, irregular, or elongated in basic shape. However, the use of this single shape factor is not sensitive enough to differentiate pores of different types as defined by Brewer (1964).

Ringrose-Voase and Bullock (1984) established an image-analysis methodology for the quantitative interpretation of the main soil pore types by image analysis. They introduced a measure of irregularity called the convex shape, to aid in the sensitivity of pore morphometry. The convex shape (CS) measures the perimeter of a convex shell that surrounds the object (Fig. 1) , P_c, and the area of the object, A, such that

(1)

View larger version (15K): [in this window] [in a new window]   Fig. 1 Shape measurements made on objects in an image using morphometric image analysis: (a) area (shaded), (b) perimeter (dashed line), and (c) convex perimeter (dashed line)

(2)

where D = 0.589 and P is the total perimeter. To account for the problem of resolving greater irregularity as the size of the pore increases, 0.589 was used as the exponential coefficient rather than 0.5 (Ringrose-Voase and Bullock, 1984). In other words, the shape factor is more sensitive as the number of pixels along its border increases.

Since there is a degree of correlation between S and CS, Ringrose-Voase and Bullock (1984) applied a polar coordinate transformation:

(3)

(4)

where r is a measure of elongation and indentation, and is a measure of the deviation of S to CS due to surface irregularity. Applying the arctan function places more emphasis on small changes in shape when the objects are nearly circular, which is more important for the identification of pore types (Ringrose-Voase and Bullock, 1984).

Using r and , Ringrose-Voase and Bullock (1984) were able to differentiate between vughs, channels, and fissures (planes) based on a classification of a learning set of pores. The learning set consisted of 96 pores; Ringrose-Voase and Bullock (1984) qualitatively identified the pores as vughs, planes, or channels based on Brewer's (1964) classification. A binary bitmap was created with the classified voids taken from the representative thin sections (Fig. 2) . Using this learning set they quantified the r and for each of the pores and plotted them in irregularity–elongation space. Then using a maximum likelihood decision rule, they were able to create rules pertaining to the classification of each pore type in the irregularity–elongation space (Fig. 3) .

View larger version (21K): [in this window] [in a new window]   Fig. 2 Pore learning set used to classify pores as channels, vughs, or fissures on soil thin section (modified from Ringrose-Voase and Bullock, 1984)

View larger version (18K): [in this window] [in a new window]   Fig. 3 The pore learning set plotted in elongation–irregularity space. Contours represent class boundaries as defined in Ringrose-Voase and Bullock (1984) (modified from Ringrose-Voase and Bullock, 1984)

Selection of Test Samples for Methodology Development
To develop the methodology using the proposed image-analysis algorithms, two sets of resin-impregnated soil blocks were chosen from samples originally obtained for a study comparing soil structural changes occurring during a 3-yr period in no-till (NT) and conventionally tilled (CT) systems on a clay loam soil (Typic Haplaquoll). The results of this comparison study will be reported elsewhere. The specifics of the experimental design have been described already in the following: Dwyer et al. (1996), with information on root distribution, bulk density, and soil moisture; Neave and Fox (1998), with information on the impact of tillage on soil fauna population and diversity; and Fox et al. (1999), with information on soil arthropod distribution interactions with soil physical and chemical attributes. The tillage treatments were established from 1988 to 1990 according to a randomized complete block, split-plot design. Only treatment plots of continuous corn on no-till and conventional moldboard plow (fall plowed and spring disked) were sampled and prepared for micromorphological analyses according to methods outlined in Fox et al. (1993). The selected set of resin-impregnated blocks (7.5 x 6.5 x 5 cm) were representative of the NT and conventional-tillage systems: They were sampled from a 0- to 5-cm depth and chosen such that the impacts of the tillage practice on the soil would be maximized. This was achieved in April 1990 by sampling the NT before any seeding (no mechanical disturbance of the soil occurred except at planting with a slit to insert seeds), and in June by sampling the CT soil just after planting (fall plowing with spring disking to prepare the seedbed). All wheel traffic had been confined to the same rows throughout the 3-yr study for NT and only during the growing season for conventional tillage as the plots were plowed each fall, thereby differentiating the tillage practices even further.

The initial field survey of earthworm populations (conducted by hand sorting four replicates of a 30- x 30- x 15-cm deep soil volume in fall 1988) found that the mean number of worms in NT systems was 45, and in CT systems was 9, indicating that within the first year of the study, substantial differences in earthworm activity already existed. Earthworm populations are known to increase markedly when CT soils are converted to minimum or NT in temperate agroecosystems (Ehlers, 1975; Barnes and Ellis, 1979; Lee, 1985; Kladivko et al., 1997; VandenBygaart et al., 1999b). We felt that by using the soil blocks prepared from intact samples taken at specific times during the last year of the study, the probability would be maximized that distinct differences existed in the structural appearance of the morphology. In addition, a separate sample was chosen from the NT set that contained considerable visual evidence of earthworm activity with visible burrows, biopores, and castings in the soil material. Consequently we hypothesized that the images obtained from these samples would provide an appropriate means by which to test the capability of the image-analysis algorithms for differentiating size and shape differences of aggregates and pores, and for assessing whether these differences could be attributed directly to earthworm activity.

Sample Preparation for Image Analysis
Before impregnating the soil blocks (7.5 x 6.5 x 5 cm) with a polyester resin containing a fluorescent dye (Uvitex OB fluorescent [Ciba-Geigy, Ardsley, NY]), the soil moisture was removed by acetone exchange following the methods of Fox et al. (1993). Once the soil block was hardened, a slab (7.5 cm long x 2 cm wide x 5 cm high) was cut using a rock saw and each face (7.5 x 5 cm) was polished smooth using a series of various-sized grits on a grinding wheel. When the polished face was exposed to ultraviolet light, the fluorescent dye facilitated easy recognition and differentiation between soil pore and solid space as maximum fluorescence occurred in the soil pores. Both polished faces from each soil block were digitally photographed using the Kodak Professional DCS 460 digital camera (Eastman Kodak, Rochester, NY) at 3060- by 2036-pixel resolution under UV illumination (pixel size = 24 µm). The digital images were then segmented into binary (black and white) images where the pores appeared as white and the soil material as black using techniques summarized by Protz and VandenBygaart (1998). The binary images were morphometrically analyzed using Kontron KS 400 version 2.0 image-processing software (Kontron Elektronik, Newport Beach, CA). Measurements made on the pores included shape factor (Eq. [2]) and convex shape (irregularity) (Eq. [1]). The polar coordinate transformations were applied (Eq. [3] and [4]) (Ringrose-Voase and Bullock, 1984) after measurement in a spreadsheet program.

	Results

TOP ABSTRACT INTRODUCTION Methods Results Conclusions REFERENCES

 
Qualitative Interpretation of Earthworm-Influenced Soil Structure
Figure 4 shows the digital image of a soil block from the sample set of soil in NT for 3 yr. We interpreted the soil structure to have been developed solely from the casting and channeling actions of earthworms. This interpretation is supported by the following observations: (i) welded earthworm casts at the soil surface (A), and within the soil matrix at depth (B); (ii) abundant mammillated vughs (Brewer, 1964) developed by the packing of earthworm castings during defecation (G1, G2, G3, G4); (iii) rounded and elliptical large macropores representing cross-sections of earthworm burrows (E); (iv) chamber (C) containing coalesced earthworm casts (D); (v) striotubule (Brewer, 1964) (F) created by the packing of earthworm casts through an earthworm burrow (cross-section).

View larger version (80K): [in this window] [in a new window]   Fig. 4 Image (reflected normal light) from soil block with structure developed solely through the action of earthworms. See text for explanation of annotations

The Mammillated Vughs Learning Set
To undertake the quantitative assessment, a learning set of mammillated vughs was created by cropping 30 pores measuring 1000 µm and greater on the binary image of the section identified as earthworm-derived, based on the presence of the mammillated conformation (Fig. 4G). The cropped sections were combined to create a bitmap image consisting of only mammillated vughs (Fig. 5) . Morphometric image analysis was performed on the image (Fig. 5) with respect to elongation (Eq. [2]) and the convex shape factor (irregularity) (Eq. [1]), and the polar coordinate transformation applied (Eq. [3] and [4]). Figure 6 shows the results of the learning set plotted on the similar irregularity–elongation space as Ringrose-Voase and Bullock (1984). The 95% confidence ellipse of the data is shown along with the probability contours separating the classes planes, vughs, and channels as defined by Ringrose-Voase and Bullock (1984). The mammillated vughs are situated in the lower irregularity portion of the vughs class and overlap the vugh-channels boundary (Fig. 6). The similarity of the mammillated vughs to channels is a consequence of the near-rounded nature of the conformations and the overall shape of the mammillated vughs from the learning set (Fig. 5).

View larger version (22K): [in this window] [in a new window]   Fig. 5 Learning set for mammillated vughs created from image of soil block with structure created by earthworm activity. Numbered objects are related to measurement points in Fig. 6

View larger version (22K): [in this window] [in a new window]   Fig. 6 The learning set for mammillated vughs plotted in elongation–irregularity space showing the 95% confidence ellipse. Numbered points refer to the numbered objects of Fig. 5

Application of the Learning Set
The binary images of pores obtained from the NT and CT treatments were measured using the Kontron software. The polar coordinate values of irregularity and elongation were determined (Eq. [3] and [4]). Table 1 shows the results of the morphometric classification of pores >1000-µm equivalent pore diameter. The classification is based on the vugh, channel, and plane classes as defined by Ringrose-Voase and Bullock (1984) (Fig. 3), along with the proportion of pores identified as mammillated vughs, as defined by the 95% confidence ellipse derived from the learning set (Fig. 6).

View this table: [in this window] [in a new window]   Table 1 Image analysis classification (no. of pores per soil block) of pore morphology >1000-µm equivalent diam. based on Ringrose-Voase and Bullock (1984) and the mammillated vugh learning set.

There were also significantly more vughs in the NT than in the CT soils (Table 1). Ringrose-Voase and Bullock (1984) suggest that the majority of vughs are likely formed by natural or artificial compaction of packing structures. The greater number of vughs in NT relative to conventional tillage could be the direct result of increased compaction associated with the NT soil management. This is supported by Dwyer et al. (1996) who found for this soil that bulk density was significantly greater in the NT relative to the conventional tillage. However, we also cannot discount the possibility that there are more vughs at this size created at the expense of voids of smaller sizes (<1000 µm).

	Conclusions

TOP ABSTRACT INTRODUCTION Methods Results Conclusions REFERENCES

 
Although many researchers have determined that earthworms play a major role in impacting soil structural properties such as pore-size distribution, pore continuity, water-holding capacity, and aggregate stability (Edwards and Shipitalo, 1998), considerable research is still necessary before the influence of earthworms on soil structure is known on a volumetric basis. The method presented here is a movement toward this goal. It facilitates an effective means by which to estimate in two dimensions the influence of earthworm activity on soil structural attributes of shape and size. This methodology is aimed toward better characterization of the soil structure that is observed in the various horizons of the soil profile. It is not to be thought of as a direct measure of earthworm populations; this information is only obtained from direct field measurements (i.e., formalin extraction, hand sorting). Instead, it has been developed to provide an estimate of earthworm influence or impact on the overall soil structure that has developed and is visible from field observations of the soil profile. In the future, we anticipate that further refinement of algorithms available in image-analysis technology, especially those that can be related to defining morphometric pore characteristics and delineating aggregate shapes, will eventually lead to complete automatic determination of all soil structural attributes and objective evaluation of their origins. A thorough understanding of the development, genesis, and impacts on soil structural quality will further our efforts toward managing our vital soil resource. Such methods give us the means to assess whether soil structure is optimized for the functioning of the soil with respect to crop growth in providing adequate porosity, and water and air infiltration.

	ACKNOWLEDGMENTS

 
Funding for this research was provided by Agriculture and Agri-Food Canada and the University of Guelph. Thanks to Dr. Anthony Ringrose-Voase of CSIRO Land & Water, Canberra, Australia, for his helpful suggestions and guidance with the methodology.

Received for publication May 12, 1999.

	REFERENCES

TOP ABSTRACT INTRODUCTION Methods Results Conclusions REFERENCES

 

• Bal L. Morphological investigation in two Moder-humus profiles and the role of soil fauna in their genesis. Geoderma 1970;4:5-36.
• Bal L. Micromorphological analysis of soils: Lower levels in the organization of organic soil materials. Soil Survey Pap. No. 6. Wageningen, the Netherlands: Netherlands Soil Survey Inst, 1973.
• Barnes B.T., Ellis F.B. Effects of different methods of cultivation and direct drilling, and disposal of straw residues, on populations of earthworms. J. Soil Sci. 1979;30:669-679.
• Blanchart E., Lavelle P., Braudeau E., Le Bissonais Y., Valentin C. Regulation of soil structure by geophagus earthworm activities in humid savannas of Ivory Coast. Soil Biol. Biochem. 1997;29:431-439.
• Brewer R. Fabric and mineral analysis of soils. New York: John Wiley & Sons, 1964.
• Darwin C.R. Formation of vegetable mould, through the action of worms, with observations on their habits. London: J. Murray, 1881.
• Didden W. Involvement of Enchytraeidae (Oligochaeta) in soil structure evolution in agricultural fields. Biol. Fertil. Soils 1990;9:152-158.
• Dobrovol'skii G.V., Titkova N.F. Characteristics of the structure of soils of bottomland oak stands. Pochvovedenie 1960;1:15-25.
• Drees L.R., Karathanasis A.D., Wilding L.P., Blevins R.L. Micromorphological characteristics of long-term no-till and conventionally tilled soils. Soil Sci. Soc. Am. J. 1994;58:508-517.[Abstract/Free Full Text]
• Dwyer L.M., Ma B.L., Stewart H.N., Balchin D., Culley J.L.B., McGovern M. Root mass distribution under conventional and conservation tillage. Can. J. Soil Sci. 1996;76:23-28.
• Edwards W.M., Shipitalo M.J. Consequences of earthworms in agricultural soils: Aggregation and porosity. In: Edwards C.A., ed. Earthworm ecology. Boca Raton, FL: CRC Press, 1998:147-161.
• Ehlers W. Observations on earthworm channels and infiltration on tilled and untilled soil. Soil Sci. 1975;119:242-249.
• Fox C.A., Fonseca E.J.A., Miller J.J., Tomlin A.D. The influence of row position and selected soil attributes on Acarina and Collembola in no-till and conventional continuous corn on a clay loam soil. Appl. Soil Ecol. 1999;13:1-8.
• Fox C.A., Guertin R.K., Dickson E., Sweeney S., Protz R., Mermut A.R. Micromorphological methodology for inorganic soils. In: Carter M.R., ed. Soil sampling and methods of analysis. Boca Raton, FL: Lewis Publ, 1993:683-709.
• Jongerius A., Schoonderbeek D., Jager A., Kowalinski S. Electro-optical soil porosity investigation by means of Quantimet B equipment. Geoderma 1972;7:177-198.
• Kladivko E.J., Akhouri N.M., Weesies G. Earthworm populations and species distribution under no-till and conventional tillage in Indiana and Illinois. Soil Biol. Biochem. 1997;29:613-615.
• Kooistra M.J. A micromorphological approach to the interactions between soil structure and soil biota. Agric. Ecosyst. Environ. 1991;24:315-328.
• Kubiena W.L. The soils of Europe. London: Thomas Murby, 1953.
• Lavelle P. Les vers de terre carnivore des savane de Lamto (Cote d'Ivoire). Pleuplements, populations et fonctions d'ecosysteme. Publ. Lab. Zool. E.N.S. 1978;12:85-101.
• Lavelle P., Bignell D., Lepage M., Wolters V., Roger P., Ineson P., Heal O.W., Dhillion S. Soil function in a changing world: The role of invertebrate ecosystem engineers. Eur. J. Soil Biol. 1997;33:159-193.
• Lee K.E. Earthworms: Their ecology and relationships with soils and land use. New York: Academic Press, 1985.
• McKenzie B.M., Dexter A.R. Physical properties of casts of the earthworm Aporrectodea rosea. Biol. Fertil. Soils. 1985;5:152-157.
• Miedema R. Applications of micromorphology of relevance to agronomy. Adv. Agron. 1997;59:119-169.
• Neave P., Fox C.A. Response of soil invertebrates to reduced tillage systems on a clay loam soil. Appl. Soil Ecol. 1998;9:423-428.
• Protz, R., and A.J. VandenBygaart. 1998. Towards systematic image analysis in the study of soil micromorphology. Sci. Soils 3:4. (available online at http://link.springer.de/link/service/journals/10112/bibs/8003001/80030004.htm) (Verified 1 Dec. 1999).
• Ringrose-Voase A.J., Bullock P. The automatic recognition and measurement of soil pore types by image analysis and computer programs. J. Soil Sci. 1984;35:673-684.
• Tomlin A.D., Shipitalo M.J., Edwards W.M., Protz R. Earthworms and their influence on soil structure and infiltration. In: Hendrix P.F., ed. Earthworm ecology and biogeography in North America. Boca Raton, FL: Lewis Publ, 1995:159-183.
• Van de Westeringh W. Deterioration of soil structure in worm free orchards. Pedobiologia. 1972;12:6-15.
• VandenBygaart A.J., Protz R., Tomlin A.D. Changes in pore structure in a no-till chronosequence of silt loam soils, southern Ontario. Can. J. Soil Sci. 1999;79:149-160 a.
• VandenBygaart A.J., Protz R., Tomlin A.D., Miller J.J. Tillage system effects on near-surface soil morphology: Observations from the landscape to micro-scale in silt loam soils of southwestern Ontario. Soil Tillage Res. 1999;51:139-149 b.
• VandenBygaart A.J., Protz R., Tomlin A.D., Miller J.J. ¹³⁷Cs as an indicator of earthworm activity in soils. Appl. Soil Ecol. 1998;9:171-177.

This article has been cited by other articles:

				S. S. Lee, C. J. Gantzer, A. L. Thompson, S. H. Anderson, and R. A. Ketcham Using High-Resolution Computed Tomography Analysis To Characterize Soil-Surface Seals Soil Sci. Soc. Am. J., August 20, 2008; 72(5): 1478 - 1485. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by VandenBygaart, A.J. Articles by Protz, R. Search for Related Content PubMed Articles by VandenBygaart, A.J. Articles by Protz, R. GeoRef GeoRef Citation Agricola Articles by VandenBygaart, A.J. Articles by Protz, R.