Liane B. Russell Biology Division, Oak Ridge National Laboratory Oak Ridge, TN 37830 Composite individuals have always excited mankind's imagina- tion. Even the earliest recorded mythologies are full of fanciful examples of creatures such as centaurs, mermaids, androgynes, and winged horses. While real, naturally occurring, mosaics might not appear as spectacular to the popular mind, their study -- particu- larly in Drosophila and plants -- has made important contributions to genetics, starting relatively early in this century. In mammals, too, examples of mosaics resulting from early or late somatic muta- tions and from abnormal egg maturation and/or fertilization events have been known since the 1930's and exploited for the information they could provide on such subjects as the or1g1n of the germline and the extent of cell mixing (see Russell 1964). Two major, and unrelated, advances, both published 17 years ago, suddenly provided a wealth of material for the study of mam- malian mosaicism. One was the successful manipulation of early embryos to make viable aggregation chimeras (Tarkowski 1961; Mintz 1962) -- a feat made possible by earlier advances in embryo culture and transfer. The other was the hypothesis that only one X chromo- some of a mammal is active (Lyon 1961; Russell 1961) and that, by virtue of the fact that the choice of the active X is made at ran- dom early in development, the normal mammalian female is a mosaic for any X-linked genetic heterozygosity.