ABSTRACT
A biologist might be interested in a particular molecule, a beaker of solution, a cell, or a deciduous forest. A model is a representation, or "mental picture", of the actual physical system and is obtained by stripping the actual system of all behavior except that which bears directly on the problem. The model can then be tested to see if it predicts any of the properties of the "real" system that it represents. Each model is progressively revised, sometimes dramatically, to remain consistent with the outcomes of new experiments suggested by the model itself. Thus, in the long haul, virtually all models prove to be incorrect, but even an incorrect model can suggest the way to a better one. The model of evolution called "Inheritance of Acquired Characteristics" was replaced by the Darwinian model, a basic conclusion of which is that properties favored by selection can be transmitted to offspring.
