Whereas the secretory function of the cardiac atrium was the first to be discovered, it is now evident that both atria and ventricles synthesize and secrete several natriuretic peptide hormones-largely in response to increased intracardiac pressure. The predominance of atrial natriuretic peptides (ANP) within atrial granules, and the relatively greater content of brain natriuretic peptides (BNP) within the ventricle, have given rise to the concept of a dual endocrine system [l] where ANP is secreted in response to increase in atrial pressure and BNP, a product largely of the ventricle, is secreted in response to sustained increases of intramural (ventricular) pressure. These products of the heart (ANP and BNP and possibly their prohormone ‘metabolites’) subserve a truly endocrine role and function as hormones with well-defined actions on numerous tissues concerned in the regulation of blood pressure and fluid homeostasis. This relatively simplistic view of the hormonal role of natriuretic peptides is now further complicated by the recognition of potentially important paracrine actions of ANP, BNP and CNP-for example within brain, blood vessel walls and the kidney.

As already indicated (Yandle pp. 561-75), direct cardiac secretion is considered to be the main source of circulating forms of ANP and BNP. The presence of authentic CNP in plasma, and its possible origins, are still controversial [2, 31 but if present it is clear that levels of CNP in peripheral blood are much lower than ANP and BNP. and are not increased in hypervolaemic states such as congestive heart failure [4]. Secretion of natriuretic peptide from non-cardiac sources such as vascular or brain tissue into the blood is possible but difficult to prove and it is improbable that it makes an important contribution to circulating Ievels of ANP or BNP. Urodilatin is not found in the circulation and appears to be synthesized in the kidney, possibly in response to volume