218 Diana Anderson Bochkov (17) recommend that the assessment should be based on individual and population prognoses. The individual prognosis should be determined by the quantity of chemical and its mutagenic activity. The population prognosis should be determined by the number of persons of reproductive age who are in contact with the chemical mutagen and the average quantity of this substance for each of them. To make such an assessment. various factors need to be considered: (a) data on the test specimen for which the highest and clearest quantitative dependencies are obtained, (b) the quantity of substance with which an individual comes into contact over a period of a year, (c) the fraction of the population up to the age of 30 who are subject to the action of a mutagen, (d) the mean population dose of substance which can be calculated from the above data, (e) the limit for the admissible level of genetic changes. They suggest prohibiting a substance with mutagenic activity, replacing it with a non- mutagenic compound, or restricting its use to persons of non-reproductive age in cases where its mean population dose causes an increase of 0.1 •o above the spontaneous level and a doubling of the spontaneous level on an individual basis. Again, this approach depends on the reliability of the experimental data obtained and being able to determine information concerning the above-mentioned factors. Auerbach (106) and Sobels (107) also criticise the concept of the doubling dose. Considering the difficulties of interpreting results obtained in a test system, any single test can hardly provide absolute answers. Negative results are often underweighted. Results from several practical test methods should be processed and decisions based on the biological and statistical significance of all the results observed, having regard for the normal range of control values in the test systems used. If a socially and economically useful compound is found to be 'hazardous' for man, then a detailed examination can be made of levels to which workers are exposed and attempts made to improve plant hygiene where the product is manufactured. Further investigations can be made to determine if and how much other groups or people or the general population are exposed, e.g. in the case of vinyl chloride, manufacturing plant exposure levels have been reduced and attention has been focused on whether any free monomer occurs in plastic food wrappings, etc. Auerbach (106) feels that in the benefit/risk calculations of a compound, on the benefit side the calculation should carry a correction factor for the special econo- mic, social or medical situation of the country concerned, e.g. in a country where millions suffer from malaria, the benefits of an efficient preventive or curative drug should be weighted accordingly, or in countries where there is famine problem, pesticides should be considered similarly. CONCLUSIONS Both academic and industrial scientists are well aware of the need for safety evaluation in general toxicological testing and this is certainly true in the field of genetic toxicology. We still do not understand if positive or negative results in a laboratory model test system are really relevant to man because of man's unique metabolism and because of the absence of any convincing 'no-effect' level data for animals or man. Epidemiological evidence for

The current state of mutagenicity testing 219 germ cell mutations after chemical exposure (or in fact any agent) is sadly lacking. In industrial areas it is often hard to pinpoint the exact chemical or agent which may be causing the problem. Unbiased abortion rates are difficult to obtain in interview by comparison with control or unexposed populations. Even if they are taken from hospital and medical practitioners' records not all abortions are recorded. Auerbach (106) does not believe that we shall even be able to identify potential human mutagens with complete confidence and even less shall we be able to feel confident about such quantitative features as thresholds, dose-effect curves and comparisons between mutagens in the human environment. She thinks this is unavoidable when we as a species are both the subject and the object of such investigation. However, we do have thousands of untested chemicals in our environment and some attempt must be made to identify those potentially hazardous to man. The limitations of the simple short-term tests which are more concerned with the concept of somatic mutation are becoming better understood and the problem of false positives and negatives has been discussed earlier (8). At present, we can only do our best with the test systems available and hope that as research progresses our understanding and techniques will improve so that results generated in our model systems will become unequivocal in terms of hazard to man. To achieve this goal, attention will have to be given to studies aimed at assessing the significance to man of positive mutagenic responses produced by a test system for a given chemical, in addition to the search for better assay procedures. ACKNOWLEDGMENTS The author wishes to thank Professor P. Oftedal, Department of Genetics, University of Oslo, and Drs J. A. Styles and J. Ashby, of this laboratory, for reading the manuscript and offering helpful comments. REFERENCES 1 Ames, B. N., Lee, F. D. and Durston, W. E. An improved bacterial test system for the detection and classification of mutagens and carcinogens. Proc. Nat. Acad. $ci. (USA). 70 782-786 (1973). 2 Ames, B. N., Durston, W.E., Yamasaki, E. and Lee, F. D. Carcinogens are mutagens:a simple test system combining liver homogenates for activation and bacteria for detection. Proc. Nat. Acad. Sci. (USA) 70 2281-2285 (1973). 3 Ames, B. N., McCann, J. and Yamasaki, E. Methods for detecting carcinogens and mutagens with the Salmonella/mammalian microsome mutagenicity test. Mutation Res. 31 347-364 (1975). 4 McCann, J., Spingarn, N. E., Kobori, J. and Ames, B. N. The detection of carcinogens as mutagens: bacterial tester strains with R factor plasmids. Proc. Nat. Acad. $ci. (USA) 72 979-983 (1975). 5 McCann, J., Choi, E., Yamasaki, E. and Ames, B. N. Detection of carcinogens as mutagens in the $almonella/microsome test. Part I, Assay of 300 chemicals. Proc. Nat. Acad. Sci. (USA) 72 5135-5139 (1975). 6 McCann, J. and Ames, B. N. Detection of carcinogens as mutagens in the Salmonella/microsome test. Assay of 300 chemicals. Part II. Proc. Nat. Acad. $ci. (USA) 73 950-954 (1976). 7 Coombs, M. M., Dixon, C. and Kissonerghis. Evaluation of the mutagenicity of compounds of known carcinogenicity belonging to the benz[a]anthracene, chrysene and cyclopenta[a]phen- anthrene series, using Ames's test. Cancer Res. 36 4525-4529 (1976). 8 Purchase, I. F. H., Longstaff, E., Ashby, J., Styles, J. A., Anderson, D., Lefevre, P. A. and West- wood, F. R. Evaluation of six short-term tests for detecting organic chemical carcinogens and recommendations for their use. Nature 264 624-627 (1976). 9 de Serres, F. J. The utility of short-term tests for mutagenicity in the toxicological evaluation of environmental agents. Mutation Res. 33 11-15 (1975).