The average tusk size of African elephants has halved since the mid-19th century. A similar effect has been spotted in the Asian elephant population in India. Researchers say it is an example of Darwinism in action, caused by the mass slaughter of dominant male elephants - but whereas evolution normally takes place over thousands of years, these changes have occurred within 150 years. Zoologists at Oxford University fear that poaching and hunting of the largest male elephants, which also have the largest tusks, has changed the natural breeding behaviour of these animals. Their research has shown that the hunting of these large males for their ivory allows smaller males with shorter tusks to produce more calves. Over time the average tusk size decreases.
Of course it's wrong that "evolution normally takes place over thousands of years." Evolution is taking place now in many species, including the human species. Evolution most reductionistically understood is simply the change in allele frequencies over time. This can happen for a variety of reasons. Neutral theory arose to account for the stochastic changes in allele frequencies over time, especially discernible on the molecular level of DNA sequences. On the other hand, the process described above seems to be a rather standard account of evolution occurring via selection. Charles Darwin's original model of natural selection was strongly shaped by analogy to artificial selection imposed by humans upon domesticates; human predation has likely been a far stronger long term factor on the shape of faunal diversity the world over. In any case, one assumes that around 5 generations have elephants have lived over the past 150 years. One may also assume that tusk size is a quantitative character, influenced by many genes of small affect. Therefore, a model based on the breeder's equation seems plausible: Response ~ Heritability X Selection Differential The culling of elephant stocks strikes me as rather similar to truncation selection. A more interesting question to me is how often similar dynamics have been operative among other extant megafaunal species which have managed to survive and not been driven to extinction (yet). Finally, do note that if the change in tusk size is a function of quantitative selection, relaxation of selection (culling through poaching) should result in a bounce back of the trait value to its natural optimum.