The mutation in the myotonic dystrophy (DM) gene is an expansion in a triplet (CTG) repeat in the 3' untranslated region of a novel gene that partially encodes a serine-threonine protein kinase (DMPK), with closest sequence homology to a small subgroup of protein kinases involved in the control of proliferation and cell shape. Expansion of the repeat correlates reasonably well with disease severity and offers a plausible molecular explanation for the previously contentious issue of anticipation. There is considerable heterogeneity in CTG expansion size in different tissues of affected individuals. The consensus of data from many laboratories indicates that DMPK mRNA is most probably downregulated as a consequence of the repeat expansion. Two polypeptides (68/78 kDa) have been shown to be absent in mouse knockout mutants and therefore can be considered as bona fide gene products. Previous data suggesting that 52-55 kDa polypeptides were likely candidates, have been firmly ruled out at the same time. Further results from studies of knockout and overexpressing transgenic mice indicate that neither simple loss nor gain of DMPK expression is sufficient to account for the DM clinical phenotype. One of the most pressing questions now being addressed is how expansion of the CTG repeat within the DMPK gene affects gene expression, not only of DMPK, but of all genes at the 19q13.3 locus: is DMPK actually responsible for the clinical phenotype seen in DM? The identification of both immediate upstream and downstream human genes (59 and DMRHP, respectively) has been an important first step to answering these questions. Only when these matters have been dealt with can one reasonably expect to start to delineate the different metabolic and signalling pathways responsible for the diverse phenotypes that make up the complex clinical picture of DM.