Interactions between polymers define many of the physical properties of plant cell walls. The most important of these are available surface area and the related measures of pore size and distribution. Gas adsorption and mercury porosimetry methods for the measurement of these parameters in wheat straw, whole wheat grain and grain fractions were compared. All of the samples examined had a very low surface area regardless of the means of measurement. However mercury porosimetry gave values 10-fold greater than gas adsorption. It is suggested that the assumptions made about pore structure necessary for the use of mercury porosimetry do not hold for plant material and that closed pore systems were distorted by the very high pressures involved. Gas adsorption gave a more realistic assessment of 2–8 m2 g−1 with grain having a lower available surface than vegetative walls. Pore regimes in both grain and straw were very similar and were determined in grain by the bran fraction. Little structure could be found associated with the starchy endosperm. Pores with radii of 1·5–3 nm predominated in wheat cell walls. These are below the size which would allow free penetration of the wall by degrading enzymes. ©1997 SCI