Many of the hypotheses are clearly correct that are based on the initial observation that TFIIIA has an approximately periodic structure with invariant pairs of cysteines and histidines apparently capable of coordinating metal ions. A startling number of other cDNA clones encode proteins that contain one or more sequences that match the zinc finger consensus, revealing that zinc finger proteins represent perhaps the largest class of DNA binding proteins in eukaryotes and that zinc finger protein-controlled gene expression may be a fundamental aspect of development as well as other processes. A great deal of progress has been made in elucidating the structure of single zinc finger domains. From knowledge of these structures, plausible and testable models can be developed for the complexes between zinc finger proteins and their DNA binding sites. Clearly, one of the most important challenges remaining in this area involves testing and extending these models. Structural data on such protein-nucleic acid complexes derived from NMR or crystallographic studies is tremendously valuable in this regard. Finally, an additional fundamental question is raised by the observation that this family and other important nucleic acid binding proteins contain zinc ions bound in small structural domains. Is zinc binding merely a useful structural strategy for generating domains involved in macromolecular interactions, or are zinc concentration fluctuations used in some manner to regulate gene expression? The biophysical data available to date certainly do not rule out this intriguing possibility.