The first session demonstrated how little useful information has been obtained from cardiac biopsy, open or otherwise, when the conventional techniques are used. We discussed whether cardiac biopsy is useful in diagnosis. At the moment it is useful only when there is an exceptional cause such as amyloid. In the majority of cases, a biopsy which is subjected to light microscopy, ultrastructural studies, histochemistry and so on, will show only non-specific changes. In HOCM the changes are more specific and it seems to be a homogeneous entity. So let us concentrate on congestive cardiomyopathy, which is probably an aetiologically heterogeneous condition in which alcohol, viruses or immunological events following some still unknown antigenic stimulus may all play a part in different patients. When the International Society of Cardiology Council on Cardiomyopathies met a year ago in Singapore, we proposed that an onslaught on cardiac biopsy should be launched, not for diagnosis, but to find out more about causal mechanisms. We suggested that the council might co-ordinate research so that centres could offer their local expertise and pool their resources for most rapid advancement of knowledge. We are fortunate at the Royal Postgraduate Medical School in having Tim Peters to split up cells and look at the biochemistry of the subcellular fractions, the enzyme markers of the different organelles. Maybe that is our main line of progress here, while in New Orleans George Burch has virological know-how. New ways of looking at the biopsy material will emerge, but no one will have the technology to do everything in one place. We would like to suggest a centrally co-ordinated scheme to speed up progress.