Heat stability of antibiotics in foods to cooking has been determined by a variety of methods. These include heating in such liquid media as milk, water, buffers and meat extracts, and in solids such as buffered meat homogenates and various sausages. Inactivation of incurred residues in tissues and eggs was also studied. Time and temperature of heating were more easily controlled in liquid media, but results in actual meat products are more indicative of actual cooking processes. Ordinary cooking procedures for meat, even to “well-done” cannot be relied on to inactivate even the more heat sensitive compounds such as penicillins and tetracyclines. More severe heating as for canning or prolonged cooking with moist heat might inactivate the more sensitive compounds.

Experimental observations on thermal injury and death of bacteria in the stationary phase can be explained by assuming that death results from inactivation of (XL) of N critical sites. It is assumed: (i) that inactivation of individual sites occurs at random and follows first-order kinetics, (ii) that the critical sites are identical and of equal heat resistance, and (iii) the bacterial population is homogeneous in heat resistance. A method is described for calculating k (the rate constant for inactivation of individual sites), N (sites per cell), and XL (the number which must be inactivated to cause death under the experimental conditions used) from experimental data. Theoretical curves calculated by using this model are identical with experimental curves, providing support for the assumptions used. Calculated values of N and XL were 130 and 21.7 for Pseudomonas viscosa and 175 and 2.7 for Salmonella anatum. There is considerable uncertainty in the absolute values of N, but they are probably > 100. It is predicted that XL will vary depending on the recovery medium used after heating. This theory is consistent with all experimental observations on thermal injury and death of bacteria.Heat treatment is one of the most widely used methods for destruction of spoilage and pathogenic bacteria. Mild-heat treatment, known as pasteurization, dates from the discoveries of Pasteur in 1860-64 (11), and is widely used for food products which are adversely affected by excessive heating. Despite this long history, no adequate explanation has been developed for the mechanism of thermal destruction of bacteria by heat. An understanding of the mechanism and kinetics of thermal death of bacteria would be helpful in the practical use of heat in processing heat-sensitive foods.Most textbooks of microbiology state that, at constant temperature, thermal death of bacteria is exponential with time, and thermal process calculations used in food processing are based on this assumption (2, 29). However, examples of nonexponential survivor curves, too numerous to list, are found in the literature. They fall into four general types (Fig. 1) which were recognized as early as 1912 by Eikjmann (10). Curve A is the type commonly found by the most careful investigators and shows an initial lag in death rate followed by a logarithmic portion. Curve C is similar to curve A but tails. Wood (31) attributes the tailing to a small population of heat-resistant cells. Concave curves similar to curve D are generally interpreted as indicating that the population is heterogeneous with regard to heat resistance.Rahn (25, 26) gives a good discussion of the early literature on survivor curves. He credits Madsen and Nyman (20) and Chick (6), working independently, with being the first to observe the apparently exponential nature of survivor curves, although Madsen and Nyman present few data to support their conclusions with regard to thermal death. Chick (7) also observed both concave and convex curves. She explained concave survivor curves as indicating a popu...

Complex survivor curves of heated bacteria are interpreted to be composites of several convex survivor cutves that represent populations of different heat resistances in a single culture of bacteria.The variation in heat resistance appears to be physiological rather than genetic since subcultures of heat-resistant cells were no more heat resistant than the parent culture. Composite cures can appear to be nearly exponential.Results support a multiple-site hypothesis of thermal death. Tailing of survivor curves, with small numbers of cells surviving extended heating, wus frequently noted when curves were carried through 6-9 log cycles. Such tailing might be of practical importance because it would predict that small numbers of cells might survive much longer heating than would be predicted from D-values calculated from curves carried through 4-5 log cycles.