It has been shown (e.g. Davies & Francis, 1941a) that, contrary to common belief, the hearts of poikilothermal vertebrates (fish, Amphibia, reptiles) are devoid of those specialized muscular tissues-nodal and Purkinje tissues that are responsible for the initiation and propagation of the stimulus for contraction in the hearts of mammals and birds, and that in each of these coldblooded animals the musculature of the heart has the same histological characters throughout its several chambers. Nevertheless, when each chamber of the heart is separated from its neighbours it will continue to contract, but at a different rate from the others. This phenomenon is not confined to the poikilothermal vertebrates, and it is well known that the chambers of the mammalian heart beat at different intrinsic rhythms when the specialized tissues are inactivated. Since these different rhythmical rates cannot be explained on an anatomical basis, the possibility of chemical differences in the muscle fibres has been explored.The role of adenosinetriphosphate (ATP) in the contraction of voluntary muscle has been established as a result of the work of Meyerhof, Parnas and their schools (for review, see Lipmann (1941)) and more recently of SzentGy6rgyi (1942SzentGy6rgyi ( , 1943. The ATP content of voluntary muscle cannot be directly estimated, but an indication of its distribution can be obtained by estimating separately adenosine and pyrophosphate, since these substances do not exist in the free state but only in combination as adenine nucleotide (e.g. ATP) (Lohmann & Schuster, 1934; Umschweif & Gibaylo, 1937). In cardiac muscle, ATP exists in combination with adenosinediphosphate (ADP), forming a dinucleotide (Embden, 1932;Beattie, Milroy & Strain, 1934;Ostern, 1934), but there is no evidence that this dinucleotide differs significantly from the mononucleotide ofvoluntary muscle in its physiological properties. The separate NUCLEOTIDE ETC., IN HEART estimation of adenosine and pyrophosphate should likewise give an indication of the distribution of the nucleotide complex in the heart. In voluntary muscle, ATP is present as the magnesium salt (Szent-Gy6rgyi, 1943; Biel- Gaddie & Stewart (1938).
METHODSHearts from the frog, pigeon, rabbit and ox were used. Except where specially mentioned, the rabbits and pigeons were killed by a blow on the head and the frogs were pithed. The results have been expressed in mg./g. wet weight. Owing to the small amounts of tissues available, it was not possible to determine the dry weight in each analysis. The water content of the atria and ventricles has, however, been determined in a separate series of animals for each species, and the differences between the two chambers were found to be negligible in every case.Nudleotide estimations. In the frog, the heart was perfused (in situ) with isotonic NaCl solution and the ventricle and atria removed in one piece and dropped on to filter paper; the atria were then cut apart from the ventricle and each of these two segments was weighed on a torsion balance ...