Very true. The article adds: > Another hypothesis known as the Brownian motion model of evolution posits that animals get big and small at random. In this sce

Very true. The article adds: > Another hypothesis known as the Brownian motion model of evolution posits that animals get big and small at random. In this scenario, the descendants of an animal would be both larger and smaller than their ancestor, but this didn’t fit with the data either, since whales got only bigger. The hypothesis only applies to large numbers of species. Of course the largest animals only got larger; otherwise they wouldn't be the largest. If every species only got larger, then the model would be false, but some species get larger and some get smaller. In this case, the article does posits an interesting climactic and fossil-record history of a recent growth spurt that's not purely random Brownian motion. But in general, you cannot look at an outlier and infer whether a general rule is true or false.