Potential interactions between multiple etiologies and pathogenesis of experimental diseases.

At least in so far as the understanding of human disease is concerned, the use of the experimental animal approach is primarily to find a model with sufficient similarity to the abnormalities observable in the human condition to give some assurance that insights obtained from these studies will give valuable clues to the nature of the human process. Certainly from animal models useful information can be obtained which can guide clinical thought at a variety of levels. In order of importance for control and management The worth of an experimental model depends to a great extent on its similarity to the human counterpart and to a lesser degree on the consistency with which it can be produced. When a model is found that rates high in these characteristics, it is often possible, by interrupting the experiments , to describe a sequential development of the processes and thus assign a higher priority to certain steps in the pathogenesis and factors in etiology over that assigned to others. In this way, it.may also be possible to derive rather conclusive evidence as to the truly pathogenic findings and the crucial etiologic factors. Once these qualitative studies have been performed, the next step is usually to find some type of quantitative relationship between the stressors and the endpoint, or endpoints, which appear to be most important. Two features of the quantitative relationship of greatest importance are whether there is a threshold before any effect is seen and whether there is an all-or-none response or a graded one once a sufficient dose has been reached. Certainly for the greatest understanding of the pathogenesis, the earlier in the process and the more discrete the endpoint the better. When multiple agents are used in the experimental situation, the problem becomes more complex. A broad definition of synergy may be used to describe any situation in which two or more agents appear to give some type of result over and above that for either agent alone. A much more exacting definition requires that the result from multiple agents should