The mammalian toll-like receptors (TLRs) are products of an evolutionary process that began prior to the separation of plants and animals. The most conserved protein motif within the TLRs is the TIR, which denotes Toll, the Interleukin-1 receptor, and plant disease Resistance genes. To trace the ancestry of the TLRs, it is desirable to draw upon the sequences of TIR domains from TLRs of diverse vertebrate species, including species with known dates of divergence (i.e., representatives of Mammalia and Aves) in order to establish a relationship between time and genetic divergence. It appears that a gene ancestral to modern TLRs 1 and 6 duplicated approximately 130 million years ago, only shortly before the speciation event that led to humans and mice. Though it is not represented in mice, TLR10 split from the TLR[1/6] precursor about 300 million years ago. The origins of other TLRs are more ancient, dating to the origins of vertebrate life, and some present-day vertebrate species appear to have many more TLRs than others. Moreover, the patterns of TLR expression are quite variable at the level of tissues, even among closely related species. A given TLR in species that are related by descent from a common ancestor may acquire different duties within each descendant line, so that some microbial inducers are avidly recognized in one species but not in others; likewise the intensity and the antomic location of an innate immune response may vary considerably. In this review, we discuss the computational methods used to analyze divergence of the TIR, and the conclusions that may be safely drawn.