Discusses in theoretical terms complexity of interactions of weather-driven air infiltration by 1) wind and 2) convection induced by indoor/outdoor temperature difference. Notes implications for practice of this complexity such as near impossibility of achieving accurate computer models. Treats flow through a single crack. Illustrates diagrammatically and discusses nature of the interaction of the 2 effects for several idealised examples. In an appendix proves mathematically the subadditivity of the effects for a wide class of situations.

Discusses use of tracer gases for the measurement of natural ventilation rates States advantages of using radio isotopes are increased speed and sensitivity. Gives expressions for calculating air change rates using radio isotopes from thedecrease in signal. Suggests use of krypton 85 or Xenon 133 as tracers. Discusses errors in the method. Reports study of air quality in a naturally ventilated building in Yakutsk. Air change rates, temperatures and concentrations of carbon monoxide were measured in kitchens with gas stoves.

Derives mathematical relationships for the connection between pressure loss and volume flowrate using simple crack models and applying known laws of similarity for flow in pipes or gaps. Demonstrates how these relationships permit more exact determination of the permeability of cracks in normal building structural components than has been possible hitherto with the use of a few approximate average values for crack permeability coefficients and pressure exponents.

Describes behavioural studies of the window opening habits of families in 123 houses to show strong seasonal pattern. During winter, window opening is closely related to moisture levels in the external air. In summer it is most closely linked to outside temperature. Larger families have more open windows. Re-examines ventilation criteria to suggest 3 seasons : deep winter with minimum ventilation for body odour removal ; spring/autumn for controlling moisture and summer for cooling.