[Graham Boak]

Getting out of the boundary layer would have increased the drag, so that argument is self-defeating. In a boundary later the flow is slower, hence any protrusion creates less drag inside than outside. However, this near the leading edge the the boundary layer is insignificant. A gondola is just a large blister, of considerably greater cross-sectional area and poorer aerodynamic profile: the increased surface area is only one contribution to drag.

The prime contribution to drag (in either kind of installation) was likely to be the barrel, with cross-flow separating from the downstream side and creating base drag. Circular sections are poor. This may point at one possible benefit from the Messerschmitt installation. The airflow near the leading edge of the wing is flowing from the stagnation point, which is behind the leading edge, and so flow at the base of the barrel (Spitfire mounting) would have a considerable cross-flow element - around the barrel rather than along it. With the barrel mounted below the wing, it may have avoided this particular condition. I'm not completely convinced by this argument - we are looking for drag forces along the aircraft axis not an extra force upwards. However, the lowest drag for an engine installation, or store carriage, is found below and in front of the wing, and that is where the barrel is on Messerschmitt's installation.

This may have been an inadvertent result of the decision to use easily-removable gondolas, for Messerschmitt's twin and multi-engine types don't demonstrate knowledge of this effect. It was discovered at the RAE between the design of the Halifax and the Lancaster.