Some scientists believe that although evolutionary theory is explanatory, it does not have, in contrast to the theories of physics, any predictive power. This raises the question of its testability. The analysis given shows that there are good reasons to claim the unpredictability of evolutionary events; nevertheless, the evolutionary theory has potential predictive power. It is argued that the difference between biology and physics lies not in the predictive power of the theories involved, but in the different weight which is (...) lent to the forecasting of particular events in these sciences. A second source of confusion derives from the ambiguity of the term 'prediction'. In order to define 'prediction' for cases in which the term is used to refer to a part of testing procedure, the reference to the time-point "now" is quite irrelevant. Prediction of unknown observational data is sufficient for testing a hypothesis, but such prediction may or may not be identical with forecasting of future events. Different factors that may cause particular difficulties met by biologists in forecasting future events are analyzed subsequently in the second part of the paper. The conclusion is drawn that although particular cognitive situations limiting the ability of forecasting are very frequent in biological sciences, the claim about the peculiar logical status of biological theories is not thereby justified. (shrink)