Without claiming completeness (biologists may ask other types of questions), we can say that two types of explanatory questions are very common in biology. The format of the first type is: (Q1) How is capacity C brought about in s? s is here a biological system. The questions of the second type are related to the prevalence of a trait, and can be divided into three subtypes: (Q2) Why has trait E remained prevalent in population U during period [t;t']? (Q2') Why has trait E become prevalent in population U during period [t;t']? (Q2'') Why has trait E lost its prevalence in population U during period [t,t']? The two types reflect Ernst Mayr's distinction between functional and evolutionary biology: The functional biologist is vitally concerned with the operation and interaction of structural elements, from molecules up to organs and whole individuals. His ever-repeated question is "How?" How does something operate, how does it function? (...) The evolutionary biologist differs in his method and in the problems in which he is interested. His basic question is "Why?". When we say "why" we must always be aware of the ambiguity of this term. It may mean "How come?" but it mat also mean the finalistic "What for?". When the evolutionist asks "Why?", he or she always has in mind the historical "How come?" (1988, p. 25) Besides the two types of explanatory questions, we can distinguish predictive questions with respect to future prevalence: (Q3) Will trait E remain prevalent in population U during period [t,t']? (Q3') Will trait E become prevalent in population U during period [t,t']? (Q3'') Will trait E loose its prevalence in population U during period [t,t']? The aim of this article is to analyse the role of function ascriptions in answering questions of the three types. As several people have noted (e.g. Mitchell 1995, Walsh 1996) there are many different tasks for which function ascriptions are used. In our view, philosophers of biology must recognise this pluralism and analyse the biological practices in two ways. On the one hand, they should ask what is gained by using functions in biological explanations and predictions. In other words, they should ask whether the different practices are appropriate with respect to the tasks that biologists set themselves. On the other hand, they should try to clarify the relations between the different uses of function ascriptions. Such a classification provides the conceptual clarity which is necessary for avoiding misinterpretations. Our paper contains four sections. In section 2 we will investigate whether function ascriptions are useful for answering questions of type Q1. In sections 3 and 4 we will do the same for explanatory and predictive questions concerning prevalence. While sections 2-4 obviously relate to the justificatory task, in section 5 a classification is offered in which the interrelations between several types of function ascriptions are clarified.