[Bill Walker]

One third is the modern factor for converting range into radius, commonly used on turbine powered aircraft. Perhaps the older aircraft had a less direct relationship between higher power and higher speed. With two wings and lots of struts and wires you may reach a point where more power just burns fuel and makes more noise, without moving you much faster.

Working with USN specs since the 1980s I know that they now have very detailed conditions defined for calculating both range and radius. The radius calculation does, as Graham and Joe indicated, include more time for form up, more time at high power, plus time over the target (i.e., burning fuel but not going anywhere). They probably had similar detailed definitions back in the 1930s, which could explain the numbers we are seeing.

In reality, all these published numbers are just guidelines. The actual usable numbers would be worked out flight by flight, based on the circumstances. I find it interesting that back in the 1930s and 1940s it was common to perform fuel consumption tests on individual aircraft at operational units. This suggests to me that there could be significant aircraft to aircraft variation in range and radius numbers, even within a given type.