Correlation of the amino acid sequence with the conformation in tobacco mosaic virus protein is considered in this article. After division of the sequence into groups with helical or nonhelical potential, the segments likely to be helical were related to the X-ray diffraction patterns obtained by Franklin, Caspar, Holmes, and Klug. The approximate locations of these segments within the known boundaries of the subunit were predicted from the radial distribution and helical projection of electron density. As a result of these assignments, the number of possible conformations was also reduced for the nonhelical segments. The structure of the subunit was simulated by flexible models of rubber and electrical tubing, as well as by space-filling Corey-Pauling-Koltun models. These models were used to locate the protein segments impinging upon the ribonucleic acid of the virus. The two pairs of carboxyl groups believed to be responsible for the binding of lead were also tentatively identified on these models as aspartic acid residues 64 and 66 (first pair) and glutamic acid residues 131 and 145 (second pair).