We can classify traits into three general groups. Adaptive traits are those that, when present increase the organism’s reproductive success. These are the traits we normally think about when we think about evolutionary processes. Non-adaptive traits are those generated by stochastic (random) processes, like drift, linkage, and bottlenecks. These traits become established not because they improve reproductive success but simply because they happened to be fixed randomly within the population. If an allele is extremely deleterious independent of its environment, it will be expected to rapidly disappear from the population. Such strongly deleterious alleles are, most likely, the result of a new mutation that occurred within the affected individual or the germ line of its parents.
When we consider a deleterious allele we mean it in terms of its effects on reproductive success. An allele can harm the individual organism carrying it yet persist in the population because it improves reproductive success in some measurable way. Similarly, there are traits that can be seen as actively maladaptive, but which occur because they are linked mechanistically to some other positively selected, adaptive trait. Many genes are involved in a number of distinct processes and their alleles can have multiple phenotypic effects. Such alleles are said to be pleiotropic, meaning they have many distinct effects on an organism’s phenotype. Not all of the pleiotropic effects of an allele are necessarily of the same type; some can be beneficial, others deleterious. As an example, a trait that dramatically increases the survival of the young, and so increases their potential reproductive success, but leads to senility and death in older adults could well be positively selected for. In this scenario, the senility trait is maladaptive but is not eliminated by selection because it is mechanistically associated with the highly adaptive juvenile survival trait. It is also worth noting that a trait that is advantageous in one environment or situation can be disadvantageous in another (think the effects of diet on the effects of the gulo1 mutation). All of which is to say that when thinking about evolutionary mechanisms, do not assume that a particular trait exists independently of other traits, that it functions in the same way in all environments, or that the presence of a trait is evidence that it is beneficial.
So, naturalists observe, a flea has smaller fleas that on him prey; and these have smaller still to bite ’em; and so proceed