There can be little doubt that the opinions expressed by Castleman and Ziem [ 19881 address an important and challenging problem. The evidence submitted must arouse serious doubts as to the scientific reliability of the TLV list published annually and used worldwide as a guideline, if not as law-as it is in Israel. It seems to me, however, that the approach to the problem, as propounded by the authors, is erroneous.
First of all, the definition of TLV seems disputable; quoting from page 3 of the 1982 booklet of ACGIH, it is: "the time-weighted average concentration for a normal 8-hour workday and a 40-hour workweek, to which nearly all (my emphasis) workers may be repeatedly exposed, day after day, without adverse effect." This means that the TLV for a given substance does not claim absolute protection for all exposed workers. With this definition in mind (and to the best of my memory the "late" maximum allowable concentrations (MACs) even defined the (small) percentage of exposed workers who might become ill), one has to come to the same conclusion as W.W. Lowrance in his fine 1976 book, Of Acceptable Risk, that TLVs can only be scientifically based values, agreed upon between workers' and employers' representatives, supported by experienced scientists and government representatives.
It must also be clearly understood, to my mind, that increasingly sophisticated and sensitive electronic and analytical instruments enable us, in 1988, to detect such fine deviations from average physiological values in the levels of poisonous material in tissues or body fluids, or in nerve conduction or behavior, that it is difficult to differentiate between them and "normal" ranges. In this context, Kehoe's almost classical studies of lead content in body fluids of inhabitants of remote areas in Mexico seem significant (1961).
The statement that zero-levels are the only "safe levels" for "allowable exposure levels" for carcinogens seems, of course, scientifically correct, but seems to me equally scientifically true for all toxic exposures. I do not think that with arguments of this kind we can find practical solutions unless we accept what is said to be true for USSR TLVs, separation between the scientific TLV and the real situation in the field.
Problems of animal experiments, which must be long-time, have to be considered. In an unforgettable lecture which I attended (1951/52), Hueper expressed himself approximately as follows: when evaluating diverging results in different