The test series for the investigation of the energy consumption of growing pigs of the breeds large white and improved land race pig as well as cross breeds of the two breeds in a total of 369 metabolism periods (as described in the first two pieces of information of this publication series -- Hoffmann and others, 1977 and Jentsch and Hoffmann, 1977) were statistically analysed for the purpose of the derivation of the energy requirement for maintenance and the partial energy requirement for growth in order to test the possibilities of the factorial analysis for the derivation of energy requirement values of growing pigs. The dependence of the maintenance requirement of growing pigs (investigations in the live weight range of 10 to 40 kg -- see 1st information--were made with boars those in the live weight range of 30 to 120 kg were made with gelded boars, 2nd information) on the live weight can best be characterised by applying a power exponent of 0,61 or 0,62 for the live weight. A definition is offered to be discussed for the energetic maintenance requirement of productive live stock and laboratory animals as a conventional value. The energy requirement values derived from the doubly-factorial statistical analysis show a satisfactory adaptation to the measured values as such concerning energy intake and observed growth performance of the test animals. The conclusion is drawn that the factorial analysis of the energy requirement (maintenance plus partial performances) results in a better estimate of the requirement of growing animals than the assessment according only to live weight and live weight increase without characterising the energy requirement for partial performances. This is important for the further working on and more exact definition of requirement norms.