TRANSACTIONS OF SECTION E. 557 
The following Recommendation was passed :— 
‘That the terms First Order, Second Order, Third Order, and Fourth Order 
of Triangulation, as connoting definite degrees of precision, be used to describe 
triangulation, even though the terms now in use (e.g. Major, Minor, &c.), which 
have only a local significance, are also employed.’ 
The following Paper was then read :-— 
New Rainfall Maps of China. By W. G. KenpREw. 
Discussion on Natural Regions of the World. 
Opened by Professor A. J. HERBERTSON. 
The idea of Natural Regions i§ not a new one. It has been used, consciously 
or unconsciously, by every traveller and geographical student with insight. 
What we are attempting now, is (i.) to divide the World into its Natural 
Regions, taking into account all the elements composing them, (il.) to recognise 
and group Natural Regions into different classes and orders, and (iil.) to trace 
the consequences of the recognition of Natural Regions as entities. 
(i) The Natural Region. 
It is unnecessary to review the attempts to divide the Earth’s surface into 
zones or regions each possessing some special property—land, water, igneous 
rocks, little or no rainfall, forests, a round-headed population, &c., &c. In 
this raw material the geographer finds certain relations between different 
elements, certain laws of combination. The elements and the combinations are 
far more complex than those of organic chemistry, but this complexity is no 
reason for not applying scientific methods to their investigation, nor for doubting 
that substantial results can be gained by their use. It is a reason for not 
consigning the study of Geography to the least experienced, but for giving it 
to the man who knows most. The only apology the geographer has to make 
is not for his subject, but for his ignorances. 
The Natural Region is a vital unit as well as a physical one, a symbiosis 
on a vast scale. It is more than an association of plants, or of animals, or of 
men. It is a symbiotic association of all these, indissolubly bound up with 
certain structures and forms of the land, possessing a definite water circulation, 
and subjected to a seasonal climatic rhythm. As each element in a region has 
its own history, and as each varies in its rate of change, so the evolution of thé 
region is highly complex. 
(ii) Types and Orders of Natural Regions. 
The advantage of classifying Natural Regions into types is obvious. Some 
of the larger ones with common morphological or climatic characters have long 
been recognised, e.g. Mountainous Lands, Plains, Monsoon Lands, Deserts, 
the Mediterranean Type of Region, &c. The systematic analysis and classifica- 
tion of all types are more recent efforts of geographers, who encounter two 
difficulties at the outset. One of these is to fix the limit of a natural region : 
Where does the Sahara begin and the Sudan end? The other is to distinguish 
properly between different orders or classes of natural regions: Are there to 
be two, or three, or four, or more? For instance, is it sufficient to divide the 
Monsoon Lands of Asia into great river basins, and each of them again into 
their minor basins, and these again into valleys and plains? Are the classes 
the same in rainy and rainless lands, in mountains and plains? Can we arrive 
at any grouping of order of natural regions as little objectionable as the 
biologist’s organism, organ, tissue, cell? 
Let us consider some of the simpler natural regions. In the Upper Thames 
Plain there are belts of (a) flowing water bordered by (b) meadow flood plains, 
(c) land rising above the flood level. Such belts are also found in the valleys 
