Agriculture Reference
In-Depth Information
TABLE 17.2
Effects of Benomyl after Surface Application or Total
Mixture with the OECD Artificial Soil Test
Control
0.5 kg ha
−
1
2.5 kg ha
−
1
Number of juveniles
Surface application
11.8
E
3.6
8.7
E
1.7
1.8
E
0.4
Total contamination
11.8
E
3.6
6.9
E
2.5
0
Body weight
Surface application
124.4
E
0.3
130.1
E
5.3
65.0
E
1.3
Total contamination
124.4
E
0.3
136.7
E
3.6
49.9
E
2.0
Source: Adapted from Kula (1994).
The use of biomarkers is a new concept in earthworm toxicity testing. These have been developed
based on enzyme functions (Goven et al. 1993, 1994), for coelomycetes with a cytometric assay
(Brousseau et al. 1997), for hemoglobin content (Rozen and Mazur 1997), and for sperm production
and fertility (Cikutovic et al. 1993; S.A. Reinecke and Reinecke 1997). For an overview of the potential
uses of biomarkers in earthworm toxicity testing, see the work of Scott-Fordsman and Weeks (1998).
A major challenge to the adoption of earthworm toxicity tests is still the question why tests
using earthworms should be better than tests using other organisms. There is also the question
whether earthworm biomarkers should be generally applicable to other soil organisms. For further
improvement of earthworm toxicity testing, attention is needed to the statistical interpretation of
dose-effect curves, intercompound testing, interspecies testing, testing real (low and repeated) field
doses, and artificial vs. natural soil.
In interpreting the results of earthworm ecotoxicity tests, it must be realized that, when using
the LC
(the concentration that results in toxicity to 50% of the test animals), the form of the dose-
effect curves is not critical. However, in using more sensitive end points such as reproduction and
use of end points like the no observed effect concentrations (NOECs), their form certainly matters.
Spurgeon et al. (1994) reported a dose-response step function for Cd and a gradual decrease for
Zn. Posthuma et al. (1993) reported a linear logarithmic model as the best description for the
combined effects of Cd/Cu and Cd/Zn on the reproduction of earthworms and a hormesic model
for the effect of Cu/Zn on earthworms.
50
Especially when deriving NOEC values, the model that is
used and the statistical variation affect the results considerably. Therefore, the design of the
experiment and the variability in the resulting data set can influence the NOEC derivation greatly.
Moreover, it is only an approximation, and as a consequence (e.g., for step functions) it is not
always possible to derive NOECs. Hence, to understand and interpret fully the different response
patterns of earthworms to chemicals, much more work is necessary.
Studies of intercompound effects of organic compounds that influence the overall toxicity (e.g.,
Neuhauser et al. 1986; Van Gestel and Ma 1993; Kula 1994; Larink and Kula 1994; Callahan et
al. 1994) and on the effects of heavy metals (Posthuma et al. 1993; Diaz-Lopez and Mancha 1994)
showed that, for a number of compounds, generalized relationships can be observed. For generally
acting (narcotic) organic compounds (Van Gestel and Ma 1993), it is possible to derive a QSAR,
as shown in
Figure 17.4
.
For assessing the toxicity of heavy metals to earthworms, Posthuma et al. (1993) developed an
approach in which they ÑsummedÒ heavy metals on the basis of their LC
values expressed in
toxicity units. Diaz-Lopez and Mancha (1994) investigated the effects of different anions and the
additions of different fertilizers on the toxicity of copper. Various copper salts (sulfate, nitrate, and
chlorides) have different toxicities to earthworms; moreover, the combination of NH
50
NO
and
4
3
CuSO
produces a much greater toxic effect than these chemicals tested separately. Fischer and
4
