There are many ways of describing; vegetation and discussions on the various methods fill many pages of textbooks and scientific journals. It is sometimes forgotten that the value of a particular method can only be assessed in relation to the purpose for which it is used. The special interests of the ecologist or the group to which he is attached will determine the field of study; once the field has been chosen the scientific method must be followed, but the choice in the first place is a subjective one. In the study of vegetation the rigid following of a particular school is to be avoided; flexibility of approach is essential because one's objects and interests are very probably unique. An example of inflexibility in the study of vegetation is provided by the Braun-Blanquet school of phytosociology, although recently it has been shown that the system can be modified to suit other purposes (Poore's 1955 a, b, c, 1956). (For a description of the Braun-Blanquet system see the first of Poore's articles.)

Anyone's individual approach to the study of vegetation will depend on the people he has talked with, the books and papers he has read, and the types of vegetation before him. For myself, I wanted a method of describing vegetation that would enable a person, working singly, to record as much as possible— including the identity of all the species of higher plants present— during a short stop of say half an hour on reconnaissance survey work. The method had to be applicable to all types of vegetation, and no aspect of the vegetation was to be studied in such detail that other aspects were neglected. The method I shall be describing below has been used for a number of years to record many different types of vegetation and during that time several modifications have been made to it. I would like to make it clear that no part of the method is original; the various pieces have simply been put together to serve my particular purpose. No existing method seemed suitable as it stood— in fact some writers intentionally omit mention of certain attributes of vegetation. Thus Dansereau (1953, p. 110) can say of his method, with its emphasis on structure, that ‘ one of its special advantages is the fact that it does not necessitate a taxonomic inventory‘. Note, too, that Greig-Smith in his recent book (1957, p. 132) dismisses vertical structure (stratification) in half a dozen lines under ‘ physiognomy ‘, yet treats horizontal structure (pattern) in great detail.

Approach to the Study of Vegetation

Before describing the method in detail I would like to draw attention to the work of a few people who have, I think, done much to clarify our approach to the study of vegetation during recent years.

If this is clearly understood then we will have no more circular arguments with the botanist saying that the vegetation of two areas is different because the soils are different, and the pedologist saying that the soils are different because the vegetation is different. The main purpose of vegetation study, then, is to relate the vegetation to the available biota (vegetation history), to the environment (synecology) and to time (vegetation dynamics). But the description of vegetation (plant sociology) must precede the study of its relations. (The description of soils is the pedologist's field and is not discussed here.) Records of stand surveys are the basic data of plant sociology; in the words of Major (1958), ‘unless these stand surveys become part of the literature, the data of vegetation study will disappear along with their interpretation as the interpretation becomes obsolete‘. A stand is any area of vegetation that has been ‘ treated as a unit for purposes of description’ (Greig-Smith. 1957, p. 120), a ‘ concrete natural community ‘ at a particular place and time. It is not an ‘ association ‘, as it is so often called in New Zealand (e.g. Cunningham, 1953); an ‘association is an abstraction, a class produced by choice of class-concept or definition’ (Whittaker, 1957). A stand is composed of a group of species-populations and it is most important to keep in mind that ‘ any ecologic study which treats …combinations of populations as a community without reference to preceding levels and with concepts directly developed for the emergent area is based on an emergentist assumption, whether consciously expressed or not’ (Bray, 1958).

To be able to relate vegetation to the environment and to time, especially in an environmentally diverse and changing country like New Zealand, page 3 it is essential that environmental variation within the sample area be as small as possible and that the vegetation— not the species— be of roughly uniform age. This means that sample areas within stands have to be chosen subjectively. In the region being investigated sample areas (preferably a number for each stand) may be placed in any of four ways: (1) along an environmental gradient, such as altitude, moisture or slope (Whittaker, 1956; McIntosh, 1958); (2) in stands of varying age; (3) in subjectively recognised noda (a nodum is an abstract vegetation unit of any rank and may be compared with a taxon— Poore. 1956); or (4) they may be placed in large numbers without reference to environmental gradients or apparent aggregations of species (Curtis and McIntosh, 1951; McIntosh, 1958). The vegetation of the region may then be analysed and described in terms of sequences— biosequences, climosequences, toposequences, lithosequences, chronosequences (successions), (Jenny. 1946, 1958)— or in terms of continua or gradients (Curtis and McIntosh, 1951; Whittaker, 1956, 1957; Greig-Smith, 1957; McIntosh, 1958). It is not, however, my intention to discuss the classification of communities in this article.

Any description of vegetation will take into account one or more of the following:

• physiognomy: the general external appearance of the vegetation
• composition: the species present; their density, cover, size (height, basal area), and yield
• structure: the horizontal distribution (pattern) and vertical distribution (stratification) of the species
• function: how the species respond to the continuously changing environment (seasonal changes: growth, flowering, fruiting, dispersal, leaf fall and renewal: growth form, life form, leaf form— size, shape, texture)
• dynamics: how any of the above attributes of vegetation change with time.

I have not included frequency in the above list of criteria since it is a complex, non-absolute measure, dependent partly on density and partly on pattern, and has meaning only in relation to the particular size and shape of sampling area used (Greig-Smith, 1957, pp. 2-11).

A statement that attempts to define a vegetation system at all fully will in most cases become very complex. A subjective selection must, therefore, be made of criteria to be used in the characterisation and comparison of stands. In the description of a stand at a particular place and time during a reconnaissance survey few observations need, or can, be made about function or dynamics. As a rule, however, indirect evidence (e.g. the patterns of regeneration of the various species) is used to judge whether the vegetation is changing or not. That leaves us with physiognomy, composition and structure. The most important aspects of these I considered to be, for my purpose, four in number:physiognomy, the species present, their cover or abundance (qualitative rating of density), and stratification. For other purposes other aspects would be emphasised, for instance yield in forest survey work (see Cunningham, 1953, in this journal).

Let us now imagine that we are in a stand of vegetation— a forest community for example. Where shall we sample the vegetation ? What we require, as has been pointed out above, is a relatively uniform area of land; in addition to this, directional change in the vegetation should not be apparent. There is no such thing as uniform vegetation of course, for the individuals of each species form varying patterns within the community, but within the sample area there should be no obvious change of overall pattern, stratification, physiognomy or composition in any particular direction, such as might occur towards a stream or ridge top.

The size of the sample area is determined empirically and will be different for different stands. The larger the area, the greater will be the information obtained and the more will variation in the stand be averaged out. In practice, however, the law of diminishing returns applies so that a compromise is necessary. If, after about five minutes searching over an increasingly large area, no further species can be added for any layer I consider that the list of species will be complete enough for the purpose in hand. We shall not worry about ‘ minimal area’ for the community, since the concept is theoretically indefensible (Poore, 1956; see also Greig-Smith, 1957, p. 137).

The number of samples taken in a stand will depend, in practice, on the time available— usually I have taken only one or a few though obviously it would be better to take a number so that some information on variation within the stand could be obtained. In view of this, and also because the areas are chosen subjectively, it might be better to consider the ‘ samples’ as examples of the vegetation in a stand.

Method of Description

The method involves the use of a guide or master card (Fig. 1) and the filling in of sample cards (e.g. Figs. 2 and 3). The master card is a piece of stiff drawing paper held against a base of aluminium, hardboard or other material by two rubber bands. The card may be sprayed with plastic to protect the surface. The sample cards are blank sheets of cartridge paper, 6 in. by 7 in., that can be filed upright without buckling; before use, each card is folded so as to make two columns 3½ in. wide. The master card contains the instructions for filling in the sample card which is placed— in a folded position— under the rubber bands. The left-hand column of the sample card (Fig. 2) is filled in using the list of headings on the left of the master card, the right-hand column and the two columns on the reverse side of the sample card (Figs. 2 and 3) are filled in using the instructions on the right of the master card (under the sample card).

By listing the species in layers, and rating their cover or abundance, a relatively clear and simple statement can be made about the vegetation of the sample area. Details of the environment are recorded under headings in the upper part of the left-hand column. Other aspects of the vegetation are recorded under headings in the lower part of this column, and by the page break

Fig. 1: Master card showing rubber bands for holding sample card. (Fig. approx. two-thirds actual size.)

use of superscripts placed against the various species in the right-hand column, as follows:

• physiognomy is described under the following headings: name, canopy, trunks, crowns
• composition: size is indicated by the height ranges of the layers and by superscripts (sd, ss, sp, p, 4, 5, 6, 7, juv) placed after the species
• structure: markedly clumped or patchy (contagious) patterns of distribution are indicated by the use of the superscripts C or G placed before the speciespage 6
• function: predominant growth forms are indicated in the name of the stand, and by the names given to each layer; low vitality (poor growth) is indicated by the use of the superscript O placed before the species
• dynamics: evidence of change is recorded under the following headings: dead, logs, browsing, trampling, burning, cutting, age, regeneration, dynamics.

Since no measure of yield is required and none of the above attributes of vegetation depend directly on size of sample area, it is not necessary to mark out a quadrat; this saves much time, especially in forest.

Filling in the Sample Card

Each item on the master card is discussed below in the order in which it is listed, though in practice only part of the left-hand column can be completed at the start. After the vegetation has been described by layers the remainder of the left-hand column is completed. Provided most of the species can be identified on the spot the majority of stands can be described in under half an hour. Lowland forest communities with many layers may take longer.

No. In. DT.: Number card for filing, initial card and give date.

Name: A name based on the composition of the canopy layer(s) and the predominant growth form(s) is given to the stand. Canopy layers are separated by a bar; species in the same layer are linked by a dash. (For suggesting the use of these symbols in this way I am indebted to Mr. I. A. E. Atkinson.) In open stands (e.g. those of water, sand, gravel and rock), and in some others as well (e.g. bog), it is customary to indicate the site in the stand name. Whatever name is used it should quickly bring to mind the community described on the card. Here are some examples of stand names:

• miro/kamahi forest
• red beech-silver beech forest
• rimu-rata/tawa/mahoe forest
• manuka-Coprosma-Dracophyllum scrub
• Danthonia cunninghamii-Phormium-Dracophyllum shrub-tussock land
• Danthonia flavescens/Celmisia spectabilis tussock grassland
• Danthonia-sweet vernal pasture
• Phyllachne-Raoulia rubra fellfield
• Oreobolus-Carpha-Celmisia bog.

Loc./Grid: Indicate the locality as precisely as possible, give the grid reference (instructions are printed on L. and S. maps), and the number of the map.

Alt.: If you are working over a large altitudinal range and away from trig points it is essential to have a pocket aneroid barometer (altimeter). Enter the actual reading on the card, e.g. 1,050 ft. (an.). The aneroid must be read at all points of known height, also on leaving and arriving at a base, starting point or camp. Later a correction can be worked out, page break

Fig. 2: A completed sample card, front side (redrawn and rearranged from original card). (Fig. approx. two-thirds actual size.)

not uncommonly of several hundred feet, and the corrected height inserted on the card, e.g. 1,290 ft. (corr.).

Slp./Asp.: Estimate the range of slope in degrees and use a small pocket compass to give you the aspect. For a ridge site give the alignment of the ridge.

S.P.M.: Soil parent material is the material from which the soil is formed, i.e. it is the initial state of the soil. It is usually weathered rock, less commonly the rock itself or organic material (peat). The S.P.M. may include any residual part or any transported part of a previous soil. For example, the S.P.M. of much of the soil that supported hard beech forest on the hills to the east of the Hutt Valley consists of the remaining parts of a very old soil, usually many feet deep, formed under a climate warmer than the present. S.P.M. is a very important factor determining vegetation patterns. Record, if possible, therefore:

A very large number of different soil parent materials occur (see N.Z. Geological Survey 1958: Geological Map of N.Z.; Soil Bureau Bull. 5, 1954: General Survey of the Soils of North Island, N.Z.). Not infrequently, especially in areas with well-developed soils, it may be impossible to determine the precise nature of the soil parent material. A few examples of soil parent materials are given below to demonstrate the sort of brief description to be aimed at:

• weathered greywacke
• stony colluvium (greywacke)
• sandy and silty alluvium over boulders (greywacke)
• andesitic ash (sand) over greywacke
• rhyolitic ash and lapilli (gravelly sand)

Drain.: Drainage is judged according to the following scale (American Soil Survey Manual). This is not always easy to do, but in view of the great importance of whether there is too much or too little water in the soil, an attempt should be made. Often the plants themselves will indicate how well drained a site is. Poorly drained conditions result from a high water table, a slowly permeable layer within the profile, seepage, or from some combination of these conditions.

Fig. 3: The reverse side of the sample card shown in Fig. 2.

Canopy: In forest and scrub note the smoothness and continuity of the canopy by such terms as very uneven, uneven, even, smooth (windshorn), discontinuous (few or many canopy gaps), almost completely broken. If certain trees or shrubs project above the general level of the canopy note them as ‘emergent‘.

Trunks: In forest note the form of the trunks of the more important canopy species by such terms as straight, twisted, leaning, much-branched, unbranched.

Crowns: In forest indicate the shape and depth (spreading, rounded, deep, conical, etc.), diameter (small, medium, large), and density (compact, diffuse) of the crowns of the more important canopy species and note page 10 any conspicuous damage. In scrub and tussock grassland the diameters of the important species can be measured with an expanding rule, e.g. Danthonia flavescens (2), 3, (4) ft. diam., indicating that most tussocks have a spread of three feet.

Dead: Note here any standing dead or dying plants.

Logs: In forest note the size and abundance of stumps (in position or overturned) and logs on the floor. Note their presence in scrub, tussock grassland, etc., as evidence of vegetational change.

Brows., Trampl., Burn., Cutt.: Browsing, trampling, burning and cutting are four common disturbance factors. Record any observations you can make, including the presence of animals in the vicinity. In particular watch out for charred wood, even small pieces of which are preserved for a long time.

Age: In secondary vegetation the age of the oldest pioneer trees and shrubs will indicate the age of the stand. In primary vegetation even if the maximum age of the trees and shrubs is known it can only indicate a minimum age for the stand.

Regen.: Note which important species appear not to be regenerating and which are regenerating profusely.

Dynam.: Note whether the stand is primary or secondary vegetation, its approximate position in any succession, and. if possible, to what extent it is modified.

Listing the Species by Layers: The instructions for filling in the rest of the sample card are on the master card (Fig. 1). In practice it is quite impossible to start filling in the card at the top and then proceed to the bottom. Especially in forest you cannot see very much from any particular spot, so the card is filled in a bit here, a bit there, as you move about to get as many different views as possible. First of all each layer is named by the predominant growth form, or forms, and the species listed under the apropriate layer (Figs. 2 and 3: note that all names are abbreviated— often more so than shown here). Usually little trouble is experienced in distinguishing either canopy layers or the lowest layers, but sometimes in forest the subcanopy shrubs and trees vary continuously in height; it is better, then, to consider all the species as belonging to a single diffuse layer rather than list them in arbitrary height ranges. In compiling the lists those species which occur only as juveniles in a layer. or which have already been listed for a higher layer. are placed to the right (Fig. 3); regeneration of a species is more easily followed in this way. Combines, such as are formed by northern rata with many other trees, are listed like this: D.cupr. + M.rob. Bryophytes on the floor are usually treated as a single entity, though sometimes important species (e.g. Sphagnum spp.) are listed separately. Lichens are always lumped together. Juvenile trees are given a superscript according to the size class they fall into, as shown on the master card (Fig. 1) (sd— seedling, ss— small sapling, sp— sapling, p— pole). Juveniles of other plants may be designated sd or juv according to their size.

The next step is to estimate the cover, and measure— estimate if out of reach— the height range of each layer, again by walking about in the sample area. Cover is used here in the usual sense as the percentage area covered by the various species, not the percentage of the ‘ground occupied by perpendicular projection on to it of the aerial parts of individuals of the species’ (Greig-Smith, 1957, p. 5). How nearly the two correspond depends on the proportion of gaps in the leaf mosaic. For example, a beech forest having 90% (canopy) cover may have only 50% cover of leaves, twigs, etc.

During a third inspection of the area underline or place brackets round each species, when necessary, to indicate its cover or abundance rating as shown in the figures (use numbers instead if you like). Add any species seen that have not already been listed. Mark those species with low vitality (O). and those with obviously clumped or patchy (contagious) distributions (C).(the distribution of the individuals of most species is contagious to some extent— Greig-Smith, 1957). Species occurring only in canopy gaps, thus having markedly contagious distributions, can be marked (CG) or just (G). Finally, in forest, note the D.B.H. (diam. breast height) of the trunks of the more important trees. The D.B.H. classes are shown on the master card. The less common classes are placed in brackets, e.g. rimu (5), (6), (7). A graduated stick, slasher, expanding rule or diameter tape may be used to measure the trees.

Listing Epiphytes and Lianes: Low epiphytes are those on logs, stumps, tree-ferns and the bases of trees, high epiphytes those on tree trunks and branches. When the epiphytes and lianes have been listed dashunderline or place in brackets those species which are prominent or inconspicuous respectively. Prominence takes into account both size and cover: thus a few clumps of Collospermum hastatum may make the species prominent, but a very large number of fronds of a filmy fern needs to be present for the fern to be rated prominent. Inconspicuous high epiphytes will almost certainly be incompletely listed, unless a fallen tree happens to be in the sample area.