Epigenetic programming is most likely the least understood part of the control of gene expression and too broad a subject to consider in a single chapter. The difficulty in studying its role in gene expression is that very few Mendelian mutations cause arrest in epigenetic programming and that chromatin changes occurring many cell divisions before transcription starts are difficult to monitor by biochemical means (Lund et al. 1995a). Given this complexity, we shall focus here on one example of epigenetic programming that has already been genetically exploited: parental genomic imprinting. One of the major advantages of studying plant versus animal development is based on the generation of alleles of genes affected in epigenetic programming so that biochemical methods can be applied by comparing tissues of different genetic origin. Here, we consider allelic variations and mutations that specifically focus on epigenetic programming at the gamete level.