During virus assembly, viral precursor proteins form an empty capsid (procapsid) that undergoes large structural changes to become a mature capsid. In the case of Escherichia coli bacteriophage k, an icosahedral virus of triangulation number T=7 with a non-contractile tail, the procapsid, is comprised mainly of the major capsid protein gpE. The expansion of the prohead occurs upon DNA packing. Without the subsequent binding of the gpD protein, the capsid cannot accommodate the full length DNA (Sternberg and Weisberg, 1977). Hence, gpD has been proposed to function in stabilizing the k-head against the pressure imposed by packaged DNA. Thimble-shaped protrusions on the expanded head visible by cryo-electron microscopy were interpreted as trimers of gpD (Dokland and Murialdo, 1993). This was confirmed at higher resolution: the crystal structure of gpD is found to be a homo-trimer, apparently identical to the capsid-bound gpD trimers observed by cryo-electron microscopy of empty capsids at 15 Å resolution (Yang et al., 2000). However, the N-terminal region of gpD that is required for the binding to the capsid could not be observed in the crystals. In contrast to this trimeric capsid-bound form and to the trimeric form observed in the crystal structure, gel-filtration experiments suggested that gpD is a stable monomer in solution (Imber et al., 1980; Forrer et al., 2004). It was therefore of interest to examine this monomeric form in solution in order to determine whether there are any differences between the structures that might prevent its trimerization. As a first step, we have determined the NMR structure of the monomeric form of gpD in solution to help understand the structural basis of the trimerization during capsid assembly as well as the mechanism of capsid stabilization by gpD.