We plan to continue our investigations of X inactivation and genomic imprinting in human females by exploring the molecular mechanisms. As X inactivation is the means of X dosage compensation in all mammals and features that differ among mammals are useful in dissecting components of the mechanisms involved, we will continue studies of X inactivation in marsupial and mouse as well as human cells. 1. Our previous studies of X inactivation in germ cells suggest that gametic differences in DNA. methylation could provide the *molecular basis for at least some phenomena attributed to parental differences in genomic imprinting. We will extend these studies by examining the methylation status of pre-meiotic and meiotically arrested female germ cells, and primordial germ cells of both sexes. We will analyze methylation using antibodies to 5-methylcytosine in histological preparations of gonads, and develop a method for PCR amplification from small numbers of germ cells, so as to examine specific HpaII sites. Using in situ hybridization we will determine if methyltransferase is expressed in unmethylated cells. 2. Our studies of marsupial cells show inactivity of the paternal G6PD allele despite absence of DNA methylation and raise the possibility that an inhibitor complex may be silencing the G6PD locus. We will look for such proteins using in vivo footprinting of the G6PD-CpG island by, ligationmediated PCR. 3. We will explore in more detail the role of DNA replication in reactivation and reversal of X inactivation by examining replication of inactive and reactivated genes on X chromosomes in mousehuman hybrid cells. 4. In search of sequences involved in early steps of X inactivation, we will identify the X chromosome segment required for inactivation using mouse-human hybrid cells containing X chromosomes with deletions in the region considered to be the putative X inactivation center in mouse and man. To identify the "essential" segment we will define the minimal area of overlapping sequences present on deleted human X chromosomes that are inactive, and the limits for the region of homology with the putative inactivation center on the mouse X, and will look for conserved cDNAs, expressed in early mouse embryos and mapping to this segment.