This paper presents a regression model to estimate the total mortality rate (Z) of fish, cetaceans, and mollusks using their maximum age (t_max). The model is based on the assumption that mortality rate is inversely related to maximum age, and that the relationship can be approximated by a linear regression of log(Z) against log(t_max). The model was tested on data from 134 stocks across three taxonomic groups, and the results showed a strong linear relationship between log(Z) and log(t_max), with high coefficients of determination (r²). The combined regression equation, which uses data over the widest possible range of ages, was found to have the highest r² value and is recommended for predictive purposes. The model was applied to data from unexploited or lightly exploited stocks, and it was found to work well for predicting mortality rates in such stocks. However, the technique may underestimate mortality rates in heavily exploited stocks due to age truncation. The regression technique is useful in various applications, including preliminary estimates of mortality rate, when age determinations are limited, when samples are not representative of the population, and when recruitment is highly variable. The technique has limitations, including the lack of consideration of sample size and the potential for overestimation if the mortality rate decreases. The statistical foundation of the technique is weak, which limits the ability to make critical comparisons. More sophisticated methods that consider sample size and assumptions of the exponential model are discussed in other papers. The paper also discusses the relationship between maximum age and sample size, and the effect of sample size on the maximum age observed. The expected maximum age increases with sample size, but the rate of increase slows down as the sample size increases. The paper also suggests that a geometric mean regression may be more appropriate than an arithmetic mean regression for predicting log(Z), as both variables are naturally variable. The geometric mean regression was tested on data from mollusks, fish, and cetaceans, and the results showed that the geometric mean regression line passes through the means of the log-transformed values of Z and t_max.This paper presents a regression model to estimate the total mortality rate (Z) of fish, cetaceans, and mollusks using their maximum age (t_max). The model is based on the assumption that mortality rate is inversely related to maximum age, and that the relationship can be approximated by a linear regression of log(Z) against log(t_max). The model was tested on data from 134 stocks across three taxonomic groups, and the results showed a strong linear relationship between log(Z) and log(t_max), with high coefficients of determination (r²). The combined regression equation, which uses data over the widest possible range of ages, was found to have the highest r² value and is recommended for predictive purposes. The model was applied to data from unexploited or lightly exploited stocks, and it was found to work well for predicting mortality rates in such stocks. However, the technique may underestimate mortality rates in heavily exploited stocks due to age truncation. The regression technique is useful in various applications, including preliminary estimates of mortality rate, when age determinations are limited, when samples are not representative of the population, and when recruitment is highly variable. The technique has limitations, including the lack of consideration of sample size and the potential for overestimation if the mortality rate decreases. The statistical foundation of the technique is weak, which limits the ability to make critical comparisons. More sophisticated methods that consider sample size and assumptions of the exponential model are discussed in other papers. The paper also discusses the relationship between maximum age and sample size, and the effect of sample size on the maximum age observed. The expected maximum age increases with sample size, but the rate of increase slows down as the sample size increases. The paper also suggests that a geometric mean regression may be more appropriate than an arithmetic mean regression for predicting log(Z), as both variables are naturally variable. The geometric mean regression was tested on data from mollusks, fish, and cetaceans, and the results showed that the geometric mean regression line passes through the means of the log-transformed values of Z and t_max.