The policies proposed above for work on the environmental relationships of
lesser-known plants may seem extremely radical to research workers who
have spent decades using conventional research approaches; but the proposals
amount to little more than suggesting that we begin for the computer-literate
what was begun centuries ago for the language-literate. All that is different
about the current proposals is that the methods will permit a great increase
in the mobilisation of personal knowledge about plant/environment
relationships and greater interaction with people outside formal research
organisations.
There could be a fear too that concentration on coarse levels of description
and prediction for lesser-known species may tend to down-value critical
enquiry into the properties of these plants, but that is not expected to be the
case. Knowledge of the practical qualities of these plants will still be needed,
and quite detailed investigations will be required before investment is made
in major commercial projects. The proposals in this paper are therefore not
aimed at stifling enquiry but at showing that when one cannot use the most
professionally-admired methods, there are honourable alternatives to giving
up altogether.
It also needs to be appreciated that the approaches proposed here have the
potential for prompting some satisfying and useful research. For example,
the methods offer workers the chance, if they wish, to get a holistic view of
a species' environmental relationships instead of having to be satisfied with
detailed knowledge for some factors and no knowledge for others. Having
this holistic knowledge for a large set of species could also lead to ideas
about new mixes of species.
Another potentially fruitful line of enquiry relates to the use of soil
descriptors for predictive purposes. Because in many field contexts only
elementary soil survey data can be obtained, and because it is impossible in
any case to use more sophisticated data due to the lack of knowledge about
lesser-known plants, there is a need to know how to get the greatest value
from routine soil analytical data (see Table 1, Level C for examples of the
measurements normally available).