Surfaces were produced with defined topographical features and surface chemistry. Silicon wafers, and wafers with attached nucleopore filters and quantifoils were coated with titanium using ion beam sputtering technology. Irregularly spaced, but regularly featured surface pits, sizes 0.2 and 0.5 microm, and regularly spaced pits with regular features (1 and 2 microm) diameter were produced. The smallest surface feature that could be successfully produced using this system was of diameter 0.2 microm. Ra, the average absolute deviation of the roughness irregularities from the mean line over one sampling length, Rz, the difference in height between the average of the five highest peaks, and the five lowest valleys along the assessment length of the profile and surface area values increased with surface feature size, with Ra values of 0.04-0.217 microm. There was no significant difference between the contact angles observed for smooth titanium surfaces with 0.2 and 0.5 microm features. However, a significant difference in contact angle was observed between the 1 and 2 microm featured surfaces (p<0.005). Substrata were used in microbial retention assays, using a range of unrelated, differently sized microorganisms. Staphylococcus aureus (cells 0.5-1 microm diameter) were retained in the highest numbers. S. aureus was well retained in the 0.5 microm sized pits and began to accumulate within larger surface features. Rod shaped Pseudomonas aeruginosa (1 microm x 3 microm) were preferentially retained, often end on, within the 1 microm surface features. Some daughter cells of Candida albicans blastospores were retained in 2 microm pits. For S. aureus and P. aeruginosa, the greatest numbers of cells were retained in the largest (2 microm) surface features. The number of C. albicans was similar across all the surfaces. The use of defined surfaces in microbial retention assays may lead to a better understanding of the interaction occurring between cells and surface features.