Brain-life span conjecture: a reevaluation of the evidence

Empirical evidence for the conjecture that brain weight of mammals is a better predictor of life span than is body weight, is reexamined and evaluated in this paper. The original evidence was that for 63 mammalian species, log brain weight explained 79% of the log life span variance, whereas log body weight explained only 60%; thus, the correlation coefficient rbr for the linear regression of the log life span on log brain weight was 0.88, whereas the correlation coefficient rb for the regression of log life span on log body weight was 0.77. From data on 40 mammalian species (including three primates), we found rbr = 0.81 and rb = 0.75; from data on 35 primate species, we found rbr = 0.68 and rb = 0.65. Correlation coefficients rliv, radr for the regression of log life span on log liver weight or log adrenal weight, respectively, were rliv = 0.78 and radr = 0.81 for the same 40 mammalian species. We conclude that brain weight appears to be a slightly better predictor of life span than body weight but not better than adrenal weight. One primary reason why body weight is a poorer predictor of life span may be a result of its wider range of values compared with brain and adrenal weights.