Abstract: Isolated walls of Bacillus subtilis Marburg, prepared in a manner which avoided metal contamination other than by the growth medium, were incubated in dilute metal solutions, separated by membrane filtration (0.22 mum), and monitored by atomic absorption to give uptake data for 18 metals. Substantial amounts of Mg2+, Fe3+, Cu2+, Na+, and K+ (amounts which were often visible as Au3+, and Ni2+ (the higher atomic-numbered elements also visible as electron scattering), and small amounts of Hg2+, Sr2+, Pb2+, and Ag+ were taken into the wall. Some (Li+, Ba2+, Co2+, and Al3+) were not absorbed. Most metals which had atomic numbers greater than 11 and which could be detected by electron microscopy appeared to diffusely stain thin sections of the wall. Magnesium, on the other hand, partitioned into the central region, and these sections of walls resisted ruthenium red staining, which was not true for the other metals. Areas of the walls also acted as nucleation sites for the growth of microscopic elemental gold crystals when incubated in solutions of auric chloride. Retention or displacement of the metals was estimated by a "chromatographic" method using the walls cross-linked by the carbodiimide reaction to adipic hydrazide agarose beads (which did not take up metal but reduced the metal binding capacity of the walls by ca. 1%) packed in a column. When a series of 12 metal solutions was passed through the column, it became evident that Mg2+, Ca2+, Fe3+, and Ni2+ were strongly bound to the walls and could be detected by both atomic absorption and by their electron-scattering power in thin sections, qhereas the other metals were fisplaced or replaced. Partial lysozyme digestion of the walls (causing a 28% loss of a [3H]diaminopimelic acid label) greatly diminished the Mg2+ retention but not that of Ca2+, Fe3+, or Ni2+, indicating that there are select sites for various cations.