Abstract
In certain non-ventilating insects, the metabolic CO2, is impounded for long periods alternating with brief, sudden "bursts" of rapid release. During the burst the spiracular valves are open wide, and during the interburst period they are constricted. During interburst the rate of CO2 output is so low, in comparison with the concurrent rate of O2 uptake, that it cannot be explained on the basis of pure diffusion. Instead it is proposed that interburst valve area is sufficiently restrictive that (a) O2 cannot at first diffuse inward as fast as it is being consumed, hence (b) a slight negative intratracheal pressure develops which in turn (c) induces a bulk in-flow of air that (d) both augments the inward transfer of O2 and impedes the out-diffusion of CO2. In support of this thesis it is shown by computation the rates of air in-flow exist which (a) account not only for the degrees of interburst CO2 retention actually observed in pupae of the saturniid moths Agapema galbina and Hyalophora cecropia, but for a wide range of theoretical steady-state/O2/CO2 transfer ratios, (b) provide an interburst equilibrium between in-flowing N2 from the air and out-diffusing N2 that accumulated in the tracheae during early stages of air in-flow, (c) involve values for tracheal pO2, pN2, spiracular valve area and trans-spiracular pressure head which are physiologically and anatomically reasonable, and (d) are compatible with changes in pupal CO2 release rate caused by changing ambient PO2, temperature and metabolic rate, and with the observed time courses of both the burst and interburst periods of the cycle. Examples are cited of possible instances of flow-diffusion in other organisms, and attention is called to the parallel between the process and active transfer in liquid systems.