In order to achieve reliable and reproducible analysis of biological materials by SIMS, it is critical both that the chosen specimen preparation method does not modify substantially the in vivo chemistry that is the focus of the study and that any chemical information obtained can be calibrated accurately by selection of appropriate standards. In Oxford, we have been working with our new Cameca NanoSIMS50 on two very distinct classes of biological materials; the first where the sample preparation problems are relatively undemanding - human hair - but calibration for trace metal analysis is a critical issue and, the second, marine coccoliths and hyperaccumulator plants where reliable specimen preparation by rapid freezing and controlled drying to preserve the distribution of diffusible species is the first and most demanding requirement, but worthwhile experiments on tracking key elements can still be undertaken even when it is clear that some redistribution of the most diffusible ions has occurred. (c) 2006 Elsevier B.V. All rights reserved.