UT EN A Ww
TROPICAL VEGETATION AND CROPS, MILLER 129
ledge and experience, some of which has been directly acquired but much of which may
be derived from the results of earlier workers in the same or allied fields. The greater
the fund of information on which he is able to draw, the sooner he will be able to gain
an understanding of each task he undertakes. Although rapid advances are now being
made, the student of vegetation and crops in the tropics is still less well endowed in that
respect than his counterpart in most other parts of the globe. On the other hand he is
presented with more opportunities for discovery, all the more so because in many parts
of the tropics commanding views are denied to the ground investigator either by the
mature and gentle topography or by the dense vegetation which clothes even the sum-
mits of hills. There are thus probably morc instances in the tropics than elsewhere of
aerial photographs disclosing features in the distribution of vegetation which were pre-
viously unknown and which might otherwise have remained in that state for a long time.
These discoveries may not be of such immediate importance as that of the first rocket-
launching sites at Penemiinde, but they can bring no less a thrill to the interpreter and
may be even more difficult to explain. The effects of human activities in the form of
cultivation patterns may be equally inexplicable at first sight, but are usually more easily
resolved than purely natural features.
Between the extremes of nearly bare desert and luxuriant forest the tropics contain
a very wide range of vegetation forms. These variations in structure provide a first
broad basis for the classification of vegetation by photo-interpretation. At this level
there are few special problems beyond that of assessing the exact limits of the more in-
determinate mixtures of grass and trees. In any part of the world the shorter types of
vegetation such as grassland and scrub can only be dealt with on their appearance in
mass, and attention has never been focussed on single plants. I have no experience of
grassland studies but have found it interesting to note how in both Africa and the Pacific
the shorter and finer montane grassland is clearly distinguishable from the taller and
coarser grasses of the lowlands. Even more remarkable is the similarity in photo appear-
ance of ericaceous communities in Europe and on the mountains of Africa.
With the taller forms of vegetation dominated by woody plants a greater change of
attitude is required by the photo-interpreter embarking on work in the tropics. The main
developments in the use of aerial photographs in forestry took place in parts of the
northern hemisphere where the recent glacial history had produced for each locality a
somewhat limited tree flora. In these conditions the forest communities were dominated
by only one or two species by which they could readily be recognised and labelled. In the
application of aerial photographs it was natural and helpful to attempt to identify these
key species, often as individual trees. In the tropics, however, gregariousness or single
species dominance are the exceptions rather than the rule, and most of the forests, wood-
lands and even the savannas contain a great variety of tree species.
Richards [14] has pointed out that the great bulk of the tropical rain forest can
only be described as à mixed association in which there are local but apparently random
fluctuations in the distribution of the numerous tree species. Presumably this is the result
of uniformly favourable topography and climate and long periods free from marked
geological or climatic changes. These conditions, which are particularly difficult for the
field worker as well as for the photo-interpreter, appear to reach a climax in the low-
land Dipterocarp forests of south-east Asia with their great abundance of closely-related
species. Where utilisation of such forests is fairly intensive, as in parts of Malaya, there
may not in fact be any wide variation in the total stocking of economically valuable
trees, but where only a limited range of timbers can be marketed it would certainly be
valuable if more help could be extracted from aerial photographs. In North Borneo a
classification based on the number and size of emergent crowns is reckoned to give an
indication of economic stocking [8], and it seems probable that other types of correlation
will be found as photographic work is intensified in other places.