HE ABILITY TO CONCENTRATE certain substancesandto expel others seems to be present in all living cells. Indeed, with the possible exception of certain intracellular parasites, no organism could exist without being able to take up

‘active transport,’ it may be useful to define the term in an unequivocal way. In accordance with Rosenberg (I z I), the term ‘transport’ or ‘active transport’ will be taken to mean a transfer of a substance against a chemical potential gradient. This definition is rather narrow; for instance, cases where a substance is found to pass faster along the concentration gradient than would be expected from simple physico-chemical considerations will not be included. The definition also excludes cases where a substance is concentrated in a cell by being bound to some celluIar constituent as is the case with many dyes. As regards charged par, ticles (ions) it is logical to restrict the term ‘ac tive transport’to cases where the transfer takes place from a lower to a higher electrochemical potential. The electrochemical potential difference for an ion species between two dilute solutions at the same pressure is equal to RT In 3 f1 c2+ RT In fi+ zr; l&-&) where cl and cz are the concentrations and fl and fz the activity coefficients of the ion in the two media, $1-$2 is the electrical potential difference between them, z is the valency of the ion, R is the gas constant and T the absolute temperature. As is well known, the activity coefficient of single ions cannot be measured; but in dilute solutions it is permissible to assume that the activity of an ion is only determined by its charge and the ionic strength of the solution, and so it is possible to compute the activity coefficient. The measurement of the potential difference across living membranes involves the use of liquid junctions, and if the solutions on the two sides of the membrane are not in6nitely dilute or identical there will be an error due to the liquid junction potentials. If the liquid junctions are made through saturated KC1 bridges, the junction potentials are reduced to a very small value. Sever theless, this error makes liquid junctions undesirable in very precise physico-chemical work; but in most ordinary biological and chemical work the error due to liquid junctions is tolerated; as is weII known, the standard methods of PH and redox potential measurements make use of liquid junctions. The direct measurement of electrochemical potential differences in bioIogica1 systems using electrodes which are reversible for the ion in question, is as a rule not feasible. If, for instance, one wishes to use si lver chloride eIectrodes to measure the electrochemical potentia1 difference for ‘Cl, the solutions have to be saturated, with silver chlorid, e; but silver ions, even in very high diIution, are extremely poisonous to most living cells, so that, while avoiding liquid junctions one destroys the system under observation. a comparatively small error due to

This means that we shall only speak of active transport when work has to be done to transfer the ion across the membrane, whether this work is used to overcome a potential difYerence, a concentration difference or a combination of both. This definition has the advantqe that, at least theoretically, a distinction is made