Agriculture Reference
In-Depth Information
OBSERVATION OF EARTHWORM INTERACTIONS WITH SOILS:
METHODOLOGICAL CONSIDERATIONS
F
T
T
D
ROM
WO
TO
HREE
IMENSIONS
Most of the distribution data on earthworm burrows are available in only two dimensions, from
the pioneer work of Ehlers (1975) to the latest computerized axial tomographic (CAT) scan
observations. As connectivity and tortuosity have been reported to be the primary characteristics
of the way earthworm burrows will interact with mass transfer properties, a three-dimensional
reconstruction is a necessary step for studying burrow systems more fully. Several attempts to do
this have been made, and new developments use x-ray CAT scans (Daniel et al. 1997). Although
such three-dimensional reconstruction views the entire earthworm burrow system structure for the
first time, it also raises two essential issues:
¤What can be said about the similarity of two burrow system patterns? What are the
pattern characteristics that describe the burrow system at a given site, and which char-
acteristics can be associated with soil physical properties, such as mass transfer, or with
biological activities such as root distributions?
¤The three-dimensional reconstruction of earthworm burrow systems does not give evi-
dence of their distribution in a three-dimensional connectivity of this system with the
soil matrix. The spatial interactions between biological components (root, earthworms,
microorganisms) requires that the distribution of any of these components should be
described in three dimensions; because they develop at different scales, description
methods should be compatible.
T
UNCTIONAL
P
ROBLEMS
IN
C
OMPARING
E
ARTHWORM
B
URROW
P
ATTERNS
F
OPOLOGY
AND
Topological and stereological analyses (Kretzschmar 1988) seem to be insufficient to describe
the specificity of earthworm burrow system patterns. The simplest case has been described for
Lumbricus
species (Lamparsky et al. 1987) for which individual earthworms lived in a single
burrow and developed a few branches around it. The necessity to describe the whole burrow
system is the major difficulty faced in this case. It has been shown that complex and continuous
burrow patterns could be developed by single individual
Megascolides australis
(Kretzschmar
and Aries 1992).
When the earthworm burrow system results from the activity of a monospecific population
(without the possibility of identifying individual earthworm territory) or from multispecific
earthworm populations, where even the territory and pattern specificity of the burrows do not
make it possible to distinguish burrows from each species, comparison of burrow system patterns
requires that the topological and functional characteristics should not only rely on their geomet-
rical distribution, but also take into account the earthworm behavior that governs such patterns.
New approaches using Co-labeled earthworms together with x-ray tomography have opened new
ways to understand, under artificial conditions, the differences between behavioral trajectories
and the resulting burrow system (Jegou et al. 1999; Capowiez et al. 2001). Behavioral studies
of earthworm movements and related functions under natural conditions are extremely rare and
should be one of the research priorities for a better understanding of the development of
earthworm populations in soils.
