The 50+ known spider silk cDNAs and genes encode fibroins that all have a similar organization consisting of a long tandem array of repeats followed by a short, carboxy-terminus region. This shared organization, sequence similarities among the carboxy-termini, and the presence of divergent genes within individual genomes are strong evidence for the existence of a silk gene family. Each silk fibroin is thousands of amino acids in length and over 90% of the protein consists of the tandem arrayed repeats. Depending on the paralog, an individual repeat may be composed of twenty to several hundred amino acids. The level of sequence similarity among repeats in a single fibroin can be extraordinarily high. Only a few other proteins are known to be as internally repetitive and as homogenized as spider fibroins. In contrast to the marked similarity of repeats within a fibroin, repeats from paralogous fibroins can differ substantially. These paralogous repeats can vary not only in sequence, but also in length and amino acid composition. These characteristics are used to quantify the sequence complexity of the different types of repeats. When these complexities are mapped on a phylogeny of the gene family, it appears that the diversification of spider silks has involved sequence simplification.