The measurement of the physical properties of a gravel-bed river is important in the calculation of sediment transport and physical habitat values for aquatic animals. These properties are not always easy to measure. One recent report on flushing of fines from the Klamath River did not contain information on one location because the grain size distribution of the armour could not be measured on a dry river bar. The grain size distribution could have been measured using a barrel sampler and converting the measurements to the same as would have been measured if a dry bar existed at the site. In another recent paper the porosity was calculated from an average value relation from the literature. The results of that paper may be sensitive to the actual value of porosity. Using the bulk density sampling technique based on a water displacement process presented in this paper the porosity could have been calculated from the measured bulk density. The principle topics of this paper are the measurement of the size distribution of the armour, and measurement of the porosity of the substrate. The 'standard' method of sampling of the armour is to do a Wolman-type count of the armour on a dry section of the river bed. When a dry bar does not exist the armour in an area of the wet streambed is to sample and the measurements transformed analytically to the same type of results that would have been obtained from the standard Wolman procedure. A comparison of the results for the San Miguel River in Colorado shows significant differences in the median size of the armour. The method use to determine the porosity is not 'high-tech' and there is a need improve knowledge of the porosity because of the importance of porosity in the aquatic ecosystem. The technique is to measure the in-situ volume of a substrate sample by measuring the volume of a frame over the substrate and then repeated the volume measurement after the sample is obtained from within the frame. The difference in the volumes is the volume of the sample.