This paper surveys the influence and causes of pores in cemented carbide and describes methods used for its description and measurement. Porosity scales, counting and measuring methods, radiography, and ultrasonic inspection are described and discussed. The quantitative data are used to develop statistical concepts that make possible more meaningful descriptions of porosity and extend the practical use of quality-control data. The statistical data obtained from the observation of porosity on a polished surface are used to predict the probability of pores in terms of number and size on the basis of the volume of a component· and of -its surface. area. In this way a correlation can be obtained between quantitative data from the examination of polished surfaces and dat8: obtained from radiography or ultrasonics. Practical results from production experience are presented and the influence of environmental cleanliness, hot isostatic treatment, and grades and types of powder granule are illustrated. A short comment is made on the metallurgical behaviour of pores and a new hypothesis is presented on the influence of pores on the probability of fracture.