What can observations of young binaries tell us about binary formation?

In this paper I discuss three areas in which observational data is likely to have a strong impact on theoretical ideas about binary star formation: the collation of binary statistics for different primary masses, the acquisition of spectral energy distributions for individual components within pre-main sequence binaries and the use of binary statistics to constrain the mix of star forming regions that must combine to form the field. Theoretical and observational work is ongoing in each of these areas. It is suggested that for observers an important future task is to remeasure the wide binary population among local field stars, whilst for theorists a problem yet to be addressed is how the mass of a molecular cloud core affects its binary fragmentation properties.