«: AGROCHEMICALS: FATE IN FOOD AND THE ENVIRONMENT PROCEEDINGS OF A SYMPOSIUM, ROME, 7 - 1 1 JUNE 1982 JOINTLY ORGANIZED BY IAEA AND FAO l^J I N T E R ...»
This indicates that all agrochemical residues tend to undergo chemical modification, biodegradation, dispersion, etc., but some halogenated ring structures appear to be relatively intractable. When considering the significance of known residues it is important to distinguish between acute toxicological effects indicated in the laboratory and the longer term effects on field populations of both 'target' and 'non-target' organisms. The phenomena of pest resistance to pesticides have important implications in this context.
Estimated world use if at the N. American rate (ha) would be 17 US $ X 10 9.
Estimated world use if at the W. European rate (ha) would be 23 US $ X 10 9.
Estimated world use if at the N. American rate (per capita) would be 38 US $ X 10 9.
Estimated world use if at the W. European rate (per capita) would be 17 US $ X 10 9.
Data and estimates based on GIFAP  and FAO  statistics.
per unit o f agricultural land, or per capita, o f North America or western Europe were matched in other regions this alone would result in a two- to four-fold increase in world usage.
A large fraction, sometimes almost all, of an applied agricultural chemical appears as residues in f o o d, the environment and non-target animal and plant organisms. For example, it was estimated that when lindane was used for c o c o a capsid control in West Africa much less than 0.1% o f the applied insecticide was actually taken up by the insect pest population, the remainder appearing as residues in the atmosphere, crop and soil . Experiments in different countries with 15 N-labelled fertilizer indicate that usually less than half the applied nitrogen is usefully recovered in the crop .
The nature, magnitude and biological significance o f the residue are the net result of the interactions between the residue and its abiotic and biotic environment, especially the latter [7, 8], despite one surprising conclusion to the contrary . However, there are exceptions, e.g. chemical decomposition o f an atrazine residue in soil was evidently more important than that o f microbiological degradation .
Isotope techniques now play a well established role in the study o f these interactions . Moreover, such studies are an essential prerequisite for IAEA-SM-263/32 115 internationally acceptable agrochemical use as implied, for example, by the limits recommended jointly by F A O and WHO for pesticide residues in f o o d.
It is the purpose here to illustrate the wide range o f applicable isotope techniques, to mention some limitations and pitfalls, and briefly to consider the significance o f the interactions studied.
2. N A T U R E A N D MAGNITUDE OF THE RESIDUE
The nature, magnitude, persistence and distribution o f an agrochemical residue depend on many factors . These include the chemical structure o f the original c o m p o u n d, formulation, application regime, and transport processes such as spray drift, crop harvest and export. They also include the 'disappearance' factors that remove or transform the original chemical. Abiotic factors include volatilization, photochemical decomposition, weathering, run-off and soil erosion, leaching down the soil profile and inactivations in situ, e.g. the ion-exchange type adsorption o f bipyridylium herbicide residues. Biotic factors include plant absorption and metabolism, grazing and metabolism by livestock and wildlife, bioaccumulation and degradation by soil microorganisms, fauna and flora.
The time-concentration curve will depend on the pattern and rate o f input and, in its simplest but possibly most useful form, on the effective overall 'disappearance' half-time. It is convenient to recognize the continuous and discontinuous kinds o f input. Continuous input would be represented by a person daily ingesting or imbibing a pesticide residue in f o o d and drink; an annual application o f fertilizer would be discontinuous. In the former case the significant residue, which may be the original chemical alone or a combination with one or more toxicologically significant metabolites, will tend to rise exponentially with time until a steady-state level is reached; that is, when the ' input rate is balanced by the disappearance rate due to, for example, the additive effects o f metabolism and excretion. In the case o f a discontinuous but otherwise equivalent input there will be a series o f initially higher peak concentrations, each peak followed by exponential decay according to the disappearance rate constant or related 'half-time'. In both cases steady-state levels or steady-state maxima tend to be achieved over long periods o f time. Simple equations for the two kinds o f time-concentration curves, and a discussion o f their consequences, have been presented elsewhere . Given time, both otherwise equivalent inputs lead to identical time-weighted mean concentrations.
2.1. Isotope techniques As suggested earlier [11 ], it is convenient to consider isotope techniques as conventional tracers or as monitoring tools. For tracing, the suitably labelled and 116 WINTERINGHAM formulated compound is applied, and its physical and chemical fate followed on the basis o f subjectively timed sampling. In short, the action o f the abiotic or biotic environment on the chemical is studied.
For monitoring the effects o f the chemical on the exposed organism isotopes can be used as labelled reagents or substrates, or for labelling a specified biochemical pathway, which in turn can indicate the effects on the organism or ecosystem.
There are many reviews o f the conventional use o f tracers for pesticides, fertilizers and other agrochemicals (e.g. Refs [11,13—15]) and, more recently, for studying their fate in model ecosystems [ 7, 1 6, 1 7 ]. Less attention has been given to the use o f isotopes as monitoring tools despite their undoubted potential [ 18—20]. For examples o f both kinds o f application attention is drawn to the publications o f the Joint F A O / I A E A Division Chemical Residues Programme [21 ].
There is n o w a wide range o f radioisotopically labelled agricultural chemicals and extensive literature exists on sources, labelled syntheses, sample preparation, extraction, fractionation and assay procedures [ 2 2, 2 3 ]. Against this background only selected topics are discussed here. The discussion is prompted by some 20 years experience at the research bench level and some 15 years o f coordinating studies on behalf o f WHO, F A O and IAEA.
2.2. Bound residue problem
Once the likely nature o f the residue in soil, water or tissue has been established (usually aided by tracer techniques) it is possible to develop analytical chemical procedures for routine monitoring and control purposes, e.g. for pesticide residues in f o o d as studied and reported by the joint FAO/WHO meetings on pesticide residues in f o o d and the environment. However, chemical analysis o f organic residues almost invariably involves solvent extraction and clean-up. Part of the total residue may be 'bound', chemically or physically, and will not be recovered by solvent extraction. Although such bound residues may be neither detected nor determined by chemical analysis, they can be o f toxicological significance . Alternatively, being undetected they may be wrongly assigned to some other disappearance factor when trying to account for the total residue.
Bound residues formed as a result o f exposure to a radiolabeled chemical can be readily detected, either by non-destructive assay o f radioactivity o f the fully extracted sample or after total oxidative combustion o f 35 S- or 14 C-labelled residues. Use o f stable 15 N-labelled fertilizer residues similarly provides for the assay o f 'bound' nitrogen in the form o f the soil immobilized element, this despite the likely great excess o f existing native soil nitrogen . However, in this case samples must be chemically prepared for mass or emission spectrometric assay.
By chemical fractionation procedures the nature and magnitude o f chemically bound residues in wheat after exposure to 14 C-labelled methylbromide as an insect IAEA-SM-263/32 fumigant were successfully determined [25—27]. Bound residues have similarly been detected and studied in wheat fumigated with 32 P-labelled phosphine gas [28—30] and in 14 C-phorate- or parathion-exposed soils [31 ].
2.3. Isotope techniques as monitoring tools As monitoring tools isotope techniques appear to be relatively unexploited, despite their sophisticated development and use as diagnostic tools in medicine .
Use o f labelled reagents and enzyme substitutes in the laboratory and determination of natural or environmental isotopic ratios as ecological indicators have been reviewed elsewhere .
Use o f 14 C- or 3 H-labelled acetylcholine as the substrate for acetylcholinesterase is probably the most sensitive method available for enzyme assay o f carbamate or organophosphorous insecticide residues. It provides for minimal sample dilution, rapid assay and use o f very low substrate concentrations. It is, therefore, a very sensitive method for carbamate residues. Higher substrate concentrations, enzyme dilution, etc. dictated by conventional methods tend to reverse the very inhibition it is sought to detect. The techniques have been successfully applied to residues in human blood after occupational exposure, in agricultural products and in aquatic ecosystems .
In the labelled p o o l technique  entire metabolic pools are labelled in vitro by feeding or injecting a suitably labelled precursor; for example, 14 C-acetate for labelling the ketogenic amino acid and acetylcholine pools o f insects, 32 P-phosphate for the phosphorylated intermediates and nucleotides o f insects, and 3S S-sulphate for the sulphur amino acid and protein pools o f plants. Extraction, fractionation and comparative radioassay o f the labelled metabolites from control and f r o m exposed organisms then provide a quantitative indicator o f metabolic disturbances as a result o f toxic action or some other undesirable side effect, e.g.
formation o f undesirable volatile sulphides as a result o f fumigating wheat with methylbromide .
Related applications include use o f labelled DNA-precursors, e.g. 3 H-thymidine, for studying D N A repair inhibition by agrochemicals [19, 20], animal hormone disturbances by using radioimmunoassay techniques , and the effects o f agrochemical residues, e.g. in soil run-off, on the net primary productivity o f aquatic ecosystems, or in the microbiological activity o f soils [8, 35].
2.4. Some limitations and pitfalls
Isotope techniques, like all others, have their limitations, and certain pitfalls await the investigator . Some points bear reiteration here.
Costs and lack o f availability o f suitably labelled compounds can be a handicap for workers in less industrialized countries . The possibility o f 118 WINTERINGHAM establishing an international bank or centre for the conservation, storage and provision o f labelled agrochemicals and their significant derivatives has been considered, but at the time ( 1 9 7 6 ) it was not felt to be a feasible undertaking for the Joint F A O / I A E A Division .
Pitfalls are usually, but not invariably , due to data misinterpretation, or to an inadequate understanding o f the labelled system. Isotopic exchange can sometimes occur between quite different forms o f the same element; for example, between the bromine of an organic alkylbromide and the inorganic bromide naturally present in all plant tissues. Unless recognized this can lead to an overestimate o f the bromide residue formed in cereals under the conditions o f 82 Br-labelled methylbromide fumigation. Effective isotope exchange can also occur when 15 N is used to follow the leaching o f a labelled nitrogen fertilizer down the soil profile. This is due to the slow but sure mineralization o f the unlabelled soil organic nitrogen and the reciprocal microbiological immobilization o f the labelled mineral nitrogen. Unless taken into account, the tracer alone would underestimate the loss o f mineral nitrogen to the root zone [ 3 2, 3 6 ].
A pitfall in the context o f model ecosystem studies is due to the possible confusion o f primary catabolites with secondary anabolites. Many model or simulated ecosystems have been described for studying the fate o f an added agrochemical residue [ 7, 1 6, 1 7, 3 7, 3 8 ]. For this purpose the labelled c o m p o u n d, usually labelled with 14 C, is added to the model ecosystem. Clearly, if biodegradation in one species leads to a labelled but quite innocuous fragment, such as 14 C-acetate, 32 P-phosphate, 35 S-sulphate or 14 C-carbon dioxide, the label will almost certainly reappear in a range o f anabolites o f other exposed organisms.
For example, some 14 C-ring or labelled aromatic herbicides can undergo opening o f the benzene ring through microbiological degradation, presumably via adipic acid, so that 14C02 could reappear as anabolites in photosynthesizing algae or plankton. Unless recognized this could be wrongly interpreted as undesirable persistence or bioaccumulation in secondary biota.