Three hundred and sixty-one biopsies obtained by bioptome from 201 patients with suspected cardiomyopathy were analysed morphologically, semiquantitatively, quantitatively, and statistically to ascertain whether or not histological examination provided a measurable pattern characterising the different forms of cardiomyopathy. In addition, we have tried to find out whether or not morphometry increases the diagnostic reliability. Heart from 29 cases (10 female and 19 male) who died as a result of road traffic accidents or other extracardiac causes were sampled, selecting specimens from sites usually biopsied during life. Of these biopsies, 16.1 per cent were of such poor quality that morphological categorisation was not possible. The mean size of biopsies was 0.79 mm2, and on average 1.38 mm2 of tissue was available per patient. The histological descriptions were grouped in eight entities; morphometric analysis of the endocardium, myocardium, and interstitium resulted in a more accurate evaluation of the morphological changes. Muscle fibre diameter showed distinct variance between several morphological groups, particularly between normal and pathological groupings and between right and left sided ventricular samples. The applied scoring system included the histological HOCM index (hypertrophic cardiomyopathy) which separated clearly this type of cardiomyopathy from other disorders. Small foci of chronic inflammatory cells were seen in 25 per cent of the biopsies and did not indicate myocarditis. This study did not confirm the possible existence of small vessel disease. Morphometric data permitted distinction between several histological groups but this did not apply to individual patients in veiw of a considerable degree of overlap of the pathological changes. Combining the various morphological features and applying measurements failed to establish a definite pattern for congestive cardiomyopathy (COCM), and alcoholic heart muscle disease could not be separated from COCM. It is, however, suggested that semiquantitative and morphometric methods are mandatory, not only to define more accurately the morphological changes, but also to provide the basis for additional biochemical, immunological, and virological analysis.