The generation of a polarized cell type requires the asymmetric distribution of many cellular proteins. While a great deal is known about the vesicular trafficking that targets membrane-associated proteins to particular compartments of a cell (Pelham and Munro, 1993), the mechanisms underlying the localization of cytoplasmic proteins are less well understood, and it has generally been assumed that these proteins simply diffuse to the places in the cell where they are needed. However, one way that proteins can be targeted to discrete subcellular locations is to localize the messenger RNAs that encode them. The first evidence for the intracellular localization of specific mRNAs was discovered almost twelve years ago, with the demonstration that actin mRNA is enriched in the myoplasm of Ascidian eggs (Jeffery et al., 1983). Since that time, many other maternal mRNAs have been shown to localize to specific regions of eggs or oocytes (for example, Figures 1C-1F). As eggs are generally much larger than most somatic cells, it is easier to detect localized transcripts within them, but this phenomenon is not some special property of germ cells, as first shown by Lawrence and Singer (1986), who observed that 13-actin mRNA localizes to the leading edge of migrating fibroblasts (Figure 1B). Indeed, as the sensitivity of in situ hybridization techniques has improved, examples of localized transcripts have been found in an increasing number of somatic cell types, including fibroblasts, myoblasts, neurons, oligodendrocytes, and epithelial cells (reviewed by Wilhelm and Vale, 1993). Thus, it is now reasonable to propose that the intracellular localization of specific mRNAs is a general mechanism for protein targeting that probably occurs in all polarized cell types.