An optimality model relating the rate of biosynthesis to body temperat
ure and gene duplication is presented to account for several observed
patterns of genome size variation. The model predicts (1) that poikilo
therms living in a warm climate should have a smaller genome than poik
ilotherms living in a cold climate, (2) that homeotherms should have a
small genome as well as a small variation in genome size relative to
their poikilothermic ancestors, (3) that cold geological periods shoul
d favor the evolution of poikilotherms with a large genome and that wa
rm geological periods should do the opposite, and (4) that poikilother
ms with a small genome should be more sensitive to changes in temperat
ure than poikilotherms with a large genome. The model also offers two
explanations for the empirically documented trend that organisms with
a large cell volume have larger genomes than those with a small cell v
olume. Relevant empirical evidence is summarized to support these pred
ictions.