This paper describes a set of statistical results concerning vowel and consonant distribution in European Portuguese formal reading, at very slow and normal rates, and discusses some of its implications to the development of several language processing tools.