Mammalian sperm chromatin is the most highly condensed eukaryotic DNA known.   The major function of this tight packaging is most likely to protect the chromatin while the sperm is motile in transit during fertilization.  Indeed, sperm chromatin is resistant to external stresses such as exogenous nuclease attack and even brief sonication.  However, our laboratory has long held the hypothesis that the evolutionary pressures that selected highly compact chromatin largely devoid of histones retained aspects of the parent chromatin structure that were required for proper embryogenesis.  Recent data from several laboratories has demonstrated that this is the case for histones that are selectively retained on a few genes, indicating that epigenetic marks are inherited by the embryo.  We have focused on another epigenetic component of sperm chromatin, the organization of the DNA into loops that are attached at roughly 25 kb intervals to a proteinaceous sperm nuclear matrix.  We have suggested that this organization is also inherited by the embryo, and that it is required for proper function of the paternal genome after fertilization.  We demonstrated that the sperm nuclear matrix participates in two different functions that may be closely related.  Sperm cells isolated from the epididymus and the vas deferens can be induced to fragment their DNA into loop-sized pieces with divalent cations.  These double stranded breaks can be reversed with EDTA treatment, suggesting that topoisomerase II or a similar enzyme is involved in cleaving the DNA.  Several lines of evidence suggest that these breaks occur at the sites of DNA attachment to the sperm nuclear matrix.  After fertilization, the sperm nuclear matrix is required for paternal DNA replication in the first cell cycle of the embryo.  We have recently shown that DNA replication licensing, the process by which the cell marks replication origins during G1 in preparation for DNA synthesis, is different for the maternal and paternal genomes in the mouse zygote.  The maternal DNA appears to be licensed by metaphase II, while the sperm DNA is licensed de novo after fertilization.  We hypothesize that the sperm nuclear matrix marks the origins of replication, directing the origin recognition complex proteins to the proper sites.  These data suggest that the sperm nuclear matrix is the site of two major functions in sperm chromatin, and that both functions share molecular components.   This work is supported by U.S.A NIH grant number 1R01HD060722.