Unfortunately I can't an online reference -- somewhere I have hard copies of 
articles by Kekwick and Pawan and about Pennington's experiments with 
diabetics. They're too old to even have an abstract at pubmed -- at least 
that's what I remember. Here's a summary I found at atkinscenter.com of one 
of Pekwick and Pawan's studies -- they did lots -- right through the 1960s. 
Until the government anti-fat propaganda started in the late seventies there 
was actually a lot of research on high protein, high fat diet; ketones, sugar 
, etc. 
Calorie Intake in Relation to Body-Weight Changes in the ObeseReference:
Kekwick, A., Pawan, G.L.S., "Calorie Intake in Relation to Body-Weight 
Changes in the Obese," The Lancet, July 28, 1956, pages 155-161.
Summary:
The purpose of this study was to evaluate whether caloric restriction or 
alterations in the proportions of dietary fat, protein and carbohydrate had a 
greater effect on changes in body weight. This study was comprised of three 
different series. Using six obese individuals, the first series had all 
subjects consume a different diet every 7 to 9 days. The ratio of 
carbohydrate (47%), fat (33%), and protein (20%) was kept constant while 
varying total calories (2,000, 1,500, 1,000, 500). Researchers concluded that 
the fewer calories subjects consumed, the greater their weight loss. Half of 
the weight loss was attributed to water loss while the remainder was 
concluded to be fat tissue. Series 2 examined 14 subjects on three 
1,000-calorie diets. On each of the three diets, 90% of the calories came 
from one of the following: protein, fat or carbohydrate. After seven days, 
subjects consuming the very-high-protein and the very-high-fat diets lost 4.2 
pounds and 6.3 pounds respectively. The very-high-carbohydrate group had a 
1.2-pound weight gain. In Series 3, five subjects ate 2,000-calorie “mixed 
diets” for seven days. At the end of the week, total calories were increased 
to 2,600, with more calories coming from dietary fat and protein and less 
coming from carbohydrate. During the first week on the 2,000-calorie “mixed 
diet,” subjects gained an average of 1.1 pounds. Although eating more total 
calories and more fat calories the second week on the 2,600-calorie “mixed 
diet,” 4 of the 5 subjects lost weight (an average of 2.8 pounds).
Commentary:
The findings from Series 2 and 3 of this study can be applied to the Atkins 
Principle of Weight Loss. Individuals taking in more calories from fat and 
protein lost a significantly greater amount of weight than individuals eating 
a greater proportion of carbohydrate, even when consuming only 1,000 
calories. Eating a diet comprised of 90% fat, subjects still lost weight. It 
becomes apparent that dietary fat is not the culprit in weight gain, but 
excess carbohydrate is. Even when calories are increased to 2,600, 
individuals on a restrictive carbohydrate intake can lose weight. With excess 
carbohydrate in the body, fat burning is prevented, carbohydrate conversion 
to fat occurs, fat is allowed to accumulate and weight gain ensues. 
http://atkinscenter.com/Archive/2002/1/11-175782.html

For more summaries in this vein go to : 
http://atkinscenter.com/science/researchsummaries/body_composition.html

Namaste, Liz
<A HREF="http://www.csun.edu/~ecm59556/Healthycarb/index.html">
http://www.csun.edu/~ecm59556/Healthycarb/index.html</A>