Because it was impossible to use the process level (Level D) for the PNG
project, methods were devised at the empirical level (Level C) for gathering
information about PNG's crops held by experts and for predicting the crops'
performance. Much to the surprise of myself and my colleagues, the
methods were successful and considerable thought was given to the reasons
why that was so. The scientific reasons have been discussed at some length
already [15, Chap. 10], but it was noted that the methods devised at Level
C appeared to mimic methods which mankind has probably been using for
thousands of years. The purpose of this present paper is to explore the
implications of this finding for improving the gathering, use and mobilisation
of environmental information about lesser-known plants.