228 Oliver and Tansley. 
Figs. 23, 24. 25 and 26. Various more advanced stages in the same process 
(cf. text p. 219). 
Fig 27. Formation and liberation of gonidia. In three of the cells the 
enveloping membrane has become mucilaginous, in one the contents 
have escaped by a lateral opening from the unaltered membrane of the 
mother cell. Note the well marked wall round the gonidium. 
Fig. 28. Two cells of a filament of four, from which the gonidia have escaped. 
Fig. 29 Chain of cells developing gonidia ; the membranes of the mother-cells 
have all become mucilaginous. 
METHODS OF SURVEYING VEGETATION 
ON A LARGE SCALE. 
By F. W. Oliver 
and 
A. G. Tansley. 
[With Plate XI. and Figs. 77-80]. 
I ^HE object of the present paper is to describe two convenient 
related methods of surveying and mapping the vegetation of 
a small area on a large scale. They are only applicable where the 
ground is fairly flat or gently undulating, and they are essentially 
intended to exhibit the features of an area in which the vegetation 
has a distribution which is complex, but definitely related to the 
variation of physical features within short distances, such as 
variations of soil, surface-level, slope, water-level and the like. 1 
The methods, both of which we have recently used with success 
in the survey of a circumscribed salt-marsh occupying the floor of 
an estuary called the Bouche d’Erquy on the north coast of Brittany, 
may be termed the “Method of Squares” and the “Gridiron 
Method ” respectively. The former is suitable for the purpose of 
constructing a general map of an area comprising a considerable 
number of acres, on a scale of to z or thereabouts, the latter 
1 The “Quadrat method” used by the Nebraska School of Eco¬ 
logists for estimating the number of individuals of different 
species occurring on a small area (5 metres square) appears 
to have little in common with our method, though the quadrat 
maps are in some respects comparable with our “ gridirons.” 
