RelativeGrowth Rate

393
RELATIVE GROWTH-RATE: ITS RANGE AND ADAPTIVE

SIGNIFICANCE IN A LOCAL FLORA
BY J. P. GRIME AND RODERICK HUNT
Unit of ComparativePlant Ecology (N.E.R.C.), Departmentof Botany,

The University,Sheffield S10 2TN
INTRODUCTION
The needfor comparativeexperiments
Laboratory experimentation in plant ecology has evolved very largely as an attempt to
pursue investigations which began with fieldwork. With the widespread development of
plant growth-room facilities an alternative approach is possible. This is to measure the
characteristics of plants under a variety of controlled conditions, and to use the results
to predict their field ecology (Grime & Hodgson 1969). One advantage of this approach
is that predictions can be tested against descriptions of the field ecology obtained by
independent field investigation.
In the long term, however, the most important advantage of the predictive approach
is that many growth-room investigations can be reproduced or extended wherever there
are adequate facilities: hence data collected on different species or genotypes and in
various laboratories can be compared directly. When comparable data are available for
a large number of species drawn from a wide range of habitats it may be possible to esti-
mate the limits of variation of a particularplant attribute, to place an individual measure-
ment in context and to attempt to judge its ecological significance.
The investigation describedin this paper is an attempt to examine the range and pattern
of variation in a local flora of one particularplant attribute-the maximum potential rate
of dry matter production. Although most data have been obtained from only one field
population per species the number of species is large and includes representatives from
all the major dry terrestrial habitats of the area. Uncertainty concerning the extent to
which each sample is representative of the species does not, therefore, invalidate the
exercise either as an estimate of the range of variation or as an attempt to recognize
differences between groups of species of contrasted ecology.
Screeningfor relative growth-rate

In a comparison of the physiology of growth among flowering plants of the Britishflora
it may be expected that genetically-determineddifferences will be exhibited, both with
respect to the circumstances optimal for dry matter production and in the maximum
rates of dry matter production actually achieved at the optimum. A number of
comparative studies has revealed such differences in what may be called 'maximum
potential relative growth-rate'* (Blackman & Wilson 1951a, b; Bradshaw et al.
1958; Blackman & Black 1959; Bradshaw, Chadwick et al. 1960a; Bradshaw, Lodge et
* Henceforwardnotated, after the style of Evans (1972, p. 633), as Rmax.
This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

All use subject to JSTOR Terms and Conditions
394 RGR in a localflora
al. 1960b; Rorison 1960; Bradshaw et al. 1964; Jarvis & Jarvis 1964; Jefferies & Willis
1964; Clarkson 1965, 1967; Grime 1965a; Grime & Jeffrey 1965; Myerscough & White-
head 1966, 1967; Hutchinson 1967, 1970a, b; Pollard & Wareing 1968; Rorison 1968;
Higgs & James 1969; Foulds 1970). Despite the facts that (i) the number of species in-
cluded in each investigation was relatively small, and (ii) many were undoubtedly grown
under sub-optimal conditions, there were nevertheless strong indications that species
drawn from contrasted habitats often differedconsistently with respect to Rmx. Hence this
attribute of the flowering plant appears to be a promising candidate for extensive study.
In order to examine further any possible ecological significance in Rmax it would be
necessary, in the first place, to determine it over a wide range of species differing in
ecology. However, it is now well known that relative growth-rate in plants is simultan-
eously subject to genetic, ontogenetic and environmental control. Any attempt to make
meaningful comparisons between species with respect to what is, in effect, a function
close to their 'economy in working' (Blackman 1919) would face the considerable
difficulty of deciding upon what basis and over what time period to study their growth.
Such a study would certainly have to be conducted under controlled conditions in which,
as far as possible, no environmental factors were limiting, in order that the resulting
rates of growth might be internally rather than externally determined (Briggs 1928).
Once these comparisons had been made on a large scale another, more specific,
objective could be achieved; this would be an examination of the relationship between
Rmx and site productivity. The work of Njoku (1959), Jarvis & Jarvis (1964) and Grime
(1965b) supports the statement that high Rmax is associated with success in potentially
productive situations. Conversely, it is possible to deduce from work such as that of
Hackett (1967) and Rorison (1968) that low Rax is associated with a poor performance
in such circumstances. In unproductive situations (and in situations subject to environ-
mental stress) a low Rmx appears to be an advantage to certain species (Bradshaw et al.
1964; van den Driessche & Wareing 1966; Clarkson 1967; Hutchinson 1967; Parsons
1968a; Rorison 1968; Higgs & James 1969; Foulds 1970; Hunt 1970) whereas under
similar conditions a high Rmx may place certain species at a selective disadvantage
(Ashton 1958; Grime & Jeffrey 1965; Loach 1967; Parsons 1968b; Rorison 1968; Hunt
1970). Such hypotheses would be confirmed if it could be demonstrated that Rmaxwas,
per se, of positive or negative adaptive significance. In addition (or alternatively) the
importance of Rma might be seen to derive from its linkage to other variable characteris-
tics which themselves influence the fitness of the plant in the field.
This paper describes an initial attempt to make such comparsions and tests of hypo-
thesis. A variety of native species has been grown in monoculture in a standardized,
potentially-productive environment and measurement has been made of the rates of
growth attained in the phase of more or less exponential growth which follows shortly
after germination. Since growth in this phase is normally the most rapid in the life-
history of the plant (and since the external environmental factors were designed, as far as
possible, to be present in non-limiting quantities) these measurements may be regarded
as provisional estimates which, for the majority of the species, approximate to Rmx.
These comparisons minimize variation resulting from morphological and develop-
mental differences between species. These estimates of Rm, are examined in the light
of the species' known field distribution and an attempt is made to assess their eco-
logical significance. Only data relating to dry matter production are presented and dis-
cussed. Subsequent publications will deal with the influence of seed weight, leaf area
phenomena and the distribution of dry matter between root and shoot.

Table 1. Thefrequency of occurrenceof each
The habitats tabulated are:
1, Lakes, ponds, canals and ditches; 2, Rivers and streams; 3, Unshaded mire; 4, Shaded mire; 5, Rock outcrops; 6, Limestone scree; 7, Cliffs; 8, Walls; 9
stone quarry heaps; 21, Quarry heaps on acidic strata; 22, Scrub; 23, Hedgerows; 24, Limestone woodlands; 25, Woodlands on acidic strata; 26, Broad-lea
In each case column a refers to the percentage occurrence of the species in forty or more 1-m2 samples and column b gives the ranking of the species accor
Habitat 1 2 3 4 5 6 7 8 9 10 11
Species ab ab ab ab ab ab ab ab a b a b a b
Acer pseudoplatanus (seedlings) - - 3 2 1 - - - - 2 -
Achillea millefolium - - - 6 65 3 1 2 2 5 21 21
- 5 60 - 1 - - 4 555 30 23 16 26
Agropyron repens - - - -
Agrostis canina 3 - 19 7 928 1
A. stolonifera 14 5 42 1 40 1 16 13 16 21 - 4 8 14 65 3 37 20 26 18
A. tenuis 1 - 10 25 544 30 8 - 10 14 6 20 646 42 18 70 2
Aira praecox - - - - 580 - - - - -
Alopecurus geniculatus 3 2 9 30 -2 - 9 35
A. pratensis - - - - 2725 1924
Anisantha sterilis - - - - 2 - 4 1 2 2 -
Anthoxanthumodoratum - - 9 30 4 8 41 22 15 4 3 2 45 16 35 12
Anthriscus sylvestris - - - - - 1 5 27 2 22 29 6
Arabis hirsuta - - - 10 34 14 21 4
Arenaria serpyllifolia - - - 26 10 3 2 1 1
Arrhenatherumelatius - - 1 5 44 21 15 97 1 12 10 19 5 6 46 12 35 5 54
Betula pubescens (seedlings) - - 1 4 6 65 - 4 2 1 2
Bidens tripartita - - 1 - - - - 1 - -
- - - 7 52 - 1 - -
Brachypodium pinnatum - - - -
B. sylvaticum - - 1 832 3
Briza media - - 6 65 8 32 - - 5 554
Calluna vulgaris -4 - 21 15 - 11 13 1 - - 2
- - - 18 18 41 6 10 14 1 - 554
Campanula rotundifolia - - -
Cardamine flexuosa - 6 14 6 47 27 3 1 -
C. pratensis 2 - 20 5 928 - - - 1 - 2 16 26
- - 3 - 2 387 - -
Carexflacca
C. panicea - - - 2 - - - - -
Catapodium rigidum - - - - 3 - I - - -
Centaurea nigra - 3 10 34 - 527 1 12 35 5 54
Cerastiumholosteoides - - 7 43 - 25 11 3 1 1 18 21 80 3 49 8
Chamaenerion angustifolium - - 1 4 8 41 - 14 8 16 8 -
- - - - - - 1 21 16 - -
Chenopodium album -
Cirsium vulgare - - 1 6 65 2 1 3 2 12 32
- - - - 1 - 1 - - - -
Cllnopodium vulgare - - - - - - -
Convolvulus arvensis 3 12 32
- - 4 - 3 - - 1 1 47 14 35 12
Cynosurus cristatus
Dactylis glomerata- - 4 - 36 3 14 21 23 2 19 5 11 35 80 3 51 6
Deschampsiacespitosa 1 2 8 3 10 25 1 3 - 4 2 2 14 29
D. flexuosa - - 11 7 544 32 6 261 2 - - 554
- - - 2 1 2 -
Digitalis purpurea - - - - -
Draba muralis - - -
7 12 4 15 10 3 1 - 1 8 14 -
Epilobium hirsutum - - - -
3 - 13 13 2
Eriophorum angustifolium -
Festuca giganteaI - 1 811 - 4 -
F. ovina - - 4 - 38 2 54 4 19 4 4 935
F. rubra - 2 10 5 - 47 1 65 2 21 3 28 2 551 55 9 53 5
2 - 3 1811 - - - - -
Filipendula ulmaria - - - - - -
Galeobdolon luteum- - 732 1
Galium aparine - - 1 1314 - - 3 2 16 24 7
G. palustre 1 - 16 9 1217 - - -
G. saxatile - - 8 33 - 5 80 - 1 1 -9 35
G. verum - - - 6 65 - - - - -
Geraniumrobertianum - - 1 544 7 52 65 2 10 14 8 14 1
Geum urbanum - - - - - - 2 1 - -
Glyceria fluitans 10 8 12 9 10 25 7 32 - - - - - - 5 54
G. maxima 6 14 - 8 33 -
Helianthemum chamaecistus - - - 6 65 22 15 -
Helictotrichon pratense - - 7 52 24 14 3
Heracleumsphondylium- - 3 637 10 34 3 4 1 551 35 21 12 32
Holcus lanatus 2 6 14 23 5 4 14 24 5 41 527 14 9 19 20 70 6 70 2
H. mollis 3 4 12 15 24 4 665 - 4 4 2 541 7 43
Hordeum murinum - - - -- 1 - - -
Juncuseffusus 5 22 2 31 2 10 25 1 -1232
J. squarrosus - - 2 - - -
Koeleria cristata - - - 16 21 19 18 4 -
- - - - - - - - 2
Lathyrus montanus
L. pratensis - 1 - 3 - - 1 27 25 7 43
Lemna minor 53 1 4 13 15 189 1 3 - -
Leontodon hispidus - - - 16 21 2712 1210 - - - 2
Loliumperenne - 2 1 - 10 34 - 2 5 27 16 24 82 2 51 6
Lotus corniculatus - - - 13 26 8 32 3 - 1 541 7 43
Luzula campestris - 3 - 665 8 32 2 1 - 17 33 35 12
- - - - - - - - - - -
Lycopersicon esculentum cv. Ailsa Craig - - -
Matricaria matricarioides - 1 1 - 49 8
- - - 18 18 - 4 1 6 46 2
Medicago lupulina - - - -
Melica nutans 3
Mentha aquatica 4 - 11 17 20 - - - 1
Milium effusum - - - -
Minuartia verna - - 1 -
- - - -
Myosotis sylvatica -
Nardus stricta - 647 1 1
- - -7 52 3 2 1 -
Origanum vulgare - - -
Oxalis acetosella 1 2 1 22 15 2
Phalaris arundinacea 3 8 11 7 43 928 - - - 2
Plantago lanceolata- - 2 - 36 3 5 41 11 13 4 551 50 12 26 18
P. major - - 3 1 665 - - 5 27 2017 10 37 19 24
Poaannua - 8 33 4 12 28 3 11 13 71 2 15 34 21 21
P. pratensis 2 7 43 - 28 9 11 24 719 22 3 6 46 2031 479
P. trivialis 1 15 7 29 3 451 3 5 41 622 40 1 28 12 871 604
Polygonum aviculare- - 1 1 3 - - -1 72 143
P. convolvulus- - - - - - - - 40 8 - -
P.persicaria-
- 2 - - - - 1 506 2
Potentilia erecta- - 833 4 - - - - - - 7 43
Poterium sanguisorba - - 752 11 24 1 - - - -
Prunlln. ,lunriel 7 43 1 9 39 3 1 5 51 27 25 1429

iuency of occurrenceof each of the species includedin the investigation in vegetationsamplesfrom thirty-onehabitats in th
6, Limestone scree; 7, Cliffs; 8, Walls; 9, Arable; 10, Meadows; 11, Enclosed pastures; 12, Limestone pastures; 13, Pastures on acidic strata; 14, Soil heaps; 15, Coal
Woodlands on acidic strata; 26, Broad-leaved plantations; 27, Coniferous plantations; 28, Verges; 29, Paths; 30, Wasteland and heath on limestone; 31, Wasteland and
i b gives the ranking of the species according to percentage occurrence. No ranking is given for species with a percentage occurrence of less than five.
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
a b a b a b a b a b a b a b ab ab a b a b a b a b a b a b ab c
2 - - - 1 2 4 5 56 - - 2 - 3 15 18 S
2 2 5 21 21 16 43 1 7 52 - 15 27 12 32 5 49 5 30 6 61 - -
4 555 30 23 16 26 - - 49 3 14 15 - 7 40 12 32 38 5 3 - - 50 3
1 - - - 22 32 5 22 5 56 - - 2 2 -
8 14 65 3 37 20 26 18 4 - 55 2 30 6 2 45 2 57 2 30 8 31 18 - 2 10 21 4
6 20 6 46 42 18 70 2 62 3 41 5 13 35 364 43 7 21 16 15 25 3 - 14 3 18 11 -
.... - 1 - - 2 2 . .
- 2 - 9 35 - - 4 - 3
27 25 19 24 - - 1 - 5 49 - - - - 5 33 -
1 2 2 - - - 1 - 3 7 42 - - - - 15 18 -
3 2 45 16 35 12 58 4 14 9 3 - 19 21 3 - - 2 - 5 42 -
5 27 2 22 29 6 2 - 11 39 4 - 3 - - 2 - 3 45 7 9
- - - - 4 - - - 2 - 549 - 3 - I - -
1 1 - - 2 - 3 - 11 33 7 40 - - 14 32 - -
19 5 6 46 12 35 5 54 27 25 - 29 10 16 13 34 10 30 11 15 25 10 22 54 4 - 9 22 50 3 13
2 1 2 - 2 1 1 8 21 4 6 48 5 49 - 9 43 - 19 10 - 5
- - - 4 - - - -1235 - 3 - I
- - 7 64 - - - 642 - - - 5 65 - 12 16 10 21 36
5 5 54 27 25 - - - 2 - - - 9 43 - - -
1 - - 2 20 35 24 7 - - - - - - - 21 2 6 36 -
1 - - 554 53 8 1 - - 455 3 - - 14 32 - - - 1
1 - - - - - 3 - 2 . .4
1 - 2 16 26 4 - -
- - - 33 18 - -4 - 11 38 2 - 3
- . . . . 958 . ...- - - - - - -
I - - - - - - 2 - 2 - - 9 43
?- 1 12 35 554 7 64 - I - 642 2 - - 17 28 -
1 18 21 80 3 49 8 31 21 2 15 29 - 455 33 9 15 25 8 28 28 20 - 1 - 2
16 8 - - - 2 2 25 16 44 2 2 - 52 3 18 16 38 11 - 26 5 5 33 4
1 21 16 - - - - 12 37 - - 2 10 38 20 13 - -
1 3 2 12 32 12 53 - 9 46 8 21 - 12 32 12 32 10 22 6 61 - 1
......... -- -3 2
12 32 - - - 4 - - 2 - - 2 - - 10 21
1 1 47 14 35 12. 7 64 1 - - - - 7 42 3 - -
19 5 11 35 80 3 51 6 35 16 - 39 8 22 9 17 25 39 5 32 13 28 11 55 2 - 7 33 25 12 5
4 2 2 14 29 24 30 4 - 12 17 6 42 3 10 38 - 6 61 12 16 - 35
2 - - 5 54 35 16 97 1 3 10 20 15 27 9 37 2 - 37 1 34 1 -
1 2 - - - 3 - 2 - - - 1 -
-4 - -3
8 14 -- - 7 52 4 5 56 5 49 5 36 5 65 -
- - - - 4 2 - -
4 - - - - - - - 2 - - 10 19 - 7
4 - - 9 35 75 1 582 6 26 552 3 2 - 18 26 5 8 5 42 -
28 2 5 51 55 9 53 5 64 2 3 21 19 12 17 51 3 41 27 17 21 5 36 66 1 - 2 5 33 3
- - - - - - 1 - - - - - - - 2 - I
1 - - - - - - - - - - - - - 219 533 36
2 16 24 7 - 4 - 11 39 - - 2 5 36 - - 8 25 60 2 10
1 -i - - -- - - -
1 - - 9 35 25 27 46 3 - - 2 - - - - 9 4 11 18
- - - - 16 43 - - - - 2 -
8 14 1 - - - - - - 19 21 5 56 2 - 12 35 - 2 15 18 20
1 - - - - - - - 5 42 - 12
- - 5 54 -- 2- -
...- 25 27 - - - 37 -
- - - - 45 9 - 642 - - -
1 5 51 35 21 12 32 4 - 15 29 - 2 17 20 - - 32 14 - 4 50 3 12
14 9 19 20 70 6 70 2 36 14 4 48 4 62 1 21 19 45 2 40 8 20 13 48 5 - 9 22 5 33 3
4 2 5 41 7 43 4 9 14 8 50 20 11 4 2 - 3 2 7 6 17 13 30 9 1
1 - - - - - 1 - - 3 10 38 -
1 - - 12 32 - 11 11 5 56 - - - - - - 1
..- - - 522 -. . .
- - - 58 4 - 19 21 - - - 3
- - - 2 7 64 - - - 2- -
- 1 27 25 7 43 5 72 - 4 - 642 2 5 49 - 5
- - - 2 21 32 - - - 26 14 6 48 5 49 - 46 9 - -
5 27 16 24 82 2 51 6 7 64 2 21 19 4 - 8 40 35 12 35 6 8 51 - - 5 33
- 1 5 41 7 43 44 10 2 3 - 26 14 14 29 2 3 32 14 - - -
1 - 17 33 35 12 55 6 12 10 1 2 17 25 2 - - 8 51 6 7 3
....... - - - - - - 30 8 . .. .
49 8 - - 20 23 - 15 23 45 5 18 16 -
1 6 46 2 - 4 - 3 - - 15 23 5 49 - 17 28 -
8 25 4
- - 2 - - - 513 --- -
I - S
- - - 764 463 - - - - - 8 -
1 - - - 4 - I - - - - - 25 22 -
2 - - - 2 - - - - 10 19 7
4 5 51 50 12 26 18 40 12 1 20 23 12 17 3212 17 20 12 32 5 36 38 11 - -
5 27 20 17 10 37 19 24 - - 29 10 2 - 20 17 37 9 10 22 5 65 -
11 13 71 2 15 34 21 21 2 3 4-7 14 15 4 3 6 71 1 48 2 9 43 - 1 -
22 3 6 46 20 31 47 9 33 18 7 19 35 9 6 26 37 35 7 30 14 15 19 48 5 - 4 25 12 1
40 1 28 12 87 1 60 4 9 58 1 63 1 2 2 26 13 50 4 53 1 9 43 - 12 16 50 32
- 72 1 - 7 43 - - 25 16 14 15 15 23 30 14 20 13 2 - - -
- 40 8 - - - - 3 - 2 2 3 .
1 50 6 - 2 - - 11 39 4 - - 2 13 20 -.
- - - 7 43 29 24 1111 - - 2 . .
- 3121 - - - 11 33 3
5 51 27 25 14 29 1643 2 8- - 51 -

from thirty-onehabitats in the Sheffieldregion
acidic strata; 14, Soil heaps; 15, Coal mine heaps; 16, Lead mine heaps; 17, Cinders; 18, Bricks and mortar; 19, Manure and sewage waste: 20, Lime-
heath on limestone; 31, Wasteland and heath on acidic strata.
ice of less than five.
20 21 22 23 24 25 26 27 28 29 30 31 Source
b a b a b a b a b a b a b a b a b ab a b a b
3 15 18 9 28 17 6 18 5 14 8 4 - 1 4 Broomhill, Sheffield
61 - - - - - - 14 25 2 16 28 2 Monksdale, Derbys.
50 3 - - 1 - 42 8 10 18 7 64 10 19 Worksop, Notts.
- - - - - - - 6 27 18 24 7 27 Coombsdale, Derbys.
18 - 2 10 21 4 1 1 4 46 6 37 5 6 71 11 17 Wortley, W. Riding
14 3 18 11 - - 5 30 14 9 11 9 21 17 31 6 41 7 38 2 Broomhill, Sheffield
- - - - - 1 2 - 3 Clumber, Notts.
- - -- -- - - - - 1 Maltby, W. Riding
5 33 - 1 1 1 Black Carr Plantn., W. Riding
- 15 18 - - - 1 - - 1 Conisborough, W. Riding
5 42 - - 1 4 - 6 51 2 28 13 630 Bramley, W. Riding
3 45 7 9 28 3 - - 29 12 627 2 4 Dore, W. Riding
I - - - - - 4 - Lathkilldale, Derbys.
32 - - - - - - - - 4 2 2 Coombsdale, Derbys.
4 - 9 22 50 3 13 16 5 30 3 4 39 9 2 37 10 13 14 Via Gellia, Derbys.
43 - 19 10 - 5 38 7 24 4 2 1 - 1 2 Wharncliffe Chase, W. Riding
- - - - - - - - - - Orgreave, W. Riding
35 - 3 - I I - I - 455 - Cadeby, W. Riding
65 - 12 16 10 21 36 3 2 5 28 4 - - 4 - Coombsdale, Derbys.
43 - - - - - - - 8 38 - 13 42 - Via Gellia, Derbys.
21 2 6 36 - 1 - 1 3 6 27 2 18 9 Balderstone, W. Riding
32 - - - 1 - - - 3 2 32 12 2 Markland, W. Riding
4 2 - 1 - - - 5 32 Calver Slough, Derbys.
.- .I- - 1 - - - - 1 1 Cressbrookdale, Derbys.
38 - 2 - 3 - - - 6 51 - 1151 1 Monksdale, Derbys.
....-
- - - - - - 579 - Maltby, W. Riding
43 - - - - - - - Lindrick, W. Riding
28 - - - - - - - 6 51 4 408 1 Maltby, W. Riding
20 - I 2 - 2 - 12 30 18 13 12 47 5 32 Coombsdale, Derbys.
11 - 26 5 5 33 4 13 11 17 6 26 4 12 30 6 27 15 31 629 Walkley, Sheffield
-.- - - - - - - 3 - - Orgreave, W. Riding
61 - 1 - - 1 - - 12 30 6 27 2 1 Maltby, W. Riding
- 2 - - - - 2 - Maltby, W. Riding
10 21 - - - - 1 - 764 1 Orgreave, W. Riding
...-
- - - - ' 18 18 16 13 - - Cressbrookdale, Derbys.
2 7 33 25 12 5 38 1 8 20 4 57 4 25 9 56 2 20 5 Orgreave, W. Riding
61 - 12 16 - 35 5 10 15 2 1 6 51 4 9 55 2 Ashington, Northumberland
37 1 34 1 - - 30 3 31 2 27 3 4 5 27 6 71 54 1 Temple Normanton, Derbys.
1 - - 5 30 6 22 1 - - - I Walkley, Sheffield
- - - - - - - - - Winnats Pass, Derbys.
65 - - - - - 2 3 Maltby Common, W. Riding
- - - - - - - - - - - Pebley Quarry, W. Riding
10 19 - 7 31 8 19 4 1 - 4 - 532 Wharncliffe, W. Riding
26 5 8 5 42 - - 1 - - 7 41 6 27 26 14 195 Two sources*
1 - 2 5 33 3 - 2 2 68 1 22 12 62 1 21 4 Lathkilldale, Derbys.
- 2 - 1 2 2 - - - 2 1 Via Gellia, Derbys.
- 21 9 5 33 36 3 12 12 1 - 3 - - 1 Anston Stones Wood, W. Riding
- 8 25 60 2 10 24 8 19 6 22 2 25 15 2 14 32 630 Via Gellia, Derbys.
- - - - 1 - - - - 2 Orgreave, W. Riding
9 4 11 18 - - 7 24 2 5 14 1 8 59 16 10 Deepcar, W. Riding
- - - - I-- 1
II 1 1151 - Coombsdale. Derbys.
35 - 2 15 18 20 10 4 - 1 6 51 4 4 3 Taddington Wood, Derbys.
5 42 - 12 20 5 30 2 - 1 - - 1 Litton, Derbys.
- - - - - - - I Ashover, Derbys.
.- .- .- .- - - - - - I Cromford Canal, Derbys.
.-
-
... - - - - - -6 71 - Litton, Derbys.
- - - - - - - - - 14 36 - Lathkilldale, Derbys.
14 - 4 50 3 12 20 6 28 3 1 37 10 627 39 9 726 Via Gellia, Derbys.
5 - 9 22 5 33 3 4 8 20 7 11 29 12 12 16 34 11 19 7 Lower Bradway, W. Riding
7 6 17 13 30 9 1 35 1 25 3 24 6 7 14 1 5 79 26 3 Lathkilldale, Derbys.
- - - - - - - - 2 - - Roche Abbey, W. Riding
1 - - 2 - - 3 - 4 Orgreave, W. Riding
.- ...- .- - - - 4 - 2 Broomhead Moor, W. Riding
- - - 12 47 - Millers Dale, Derbys.
.........
- - - - 8 59 - Whitley Wood, Sheffield
12 30 - 14 36 2 Millers Dale, Derbys.
.- .- .- - - - ..- - - Wentworth Woodhouse, W. Riding
.-
9 - - - - - - - 4 - 14 36 - Via Gellia, Derbys.
51 - 5 33 - - - - 36 11 41 3 1 4 Markland Grips, W. Riding
14 - - - 12 30 - 42 6 3 Millers Dale, Derbys.
51 6 7 3 - - - - - 14 25 2 20 19 11 17 Waleswood, W. Riding
- - - - - - Commercial source
33 6 I Roche Abbey, W. Riding
28 - - - - - - - - 6 27 955 1 LathkiUdale, Derbys.
- .1 - 3 - 2 - - - - - Cressbrookdale, Derbys.
.- .- .- . ..-- . -. - - Mansfield, Notts.
8 25 - 4 8 19 - - - 2 -Smithy Wood, W. Riding
--- . . - - - - -
.- Bradwell, Derbys.
I - 5 39 2 2 1 - - 1 1 Calver Slough, Derbys.
5 8 - - - - - 1 4 4 923 Wharncliffe Chase, W. Riding
22 - - - - - - - - - 11 51 - Lindrick, W. Riding
10 19 7 32 16 7 2 612 Anston Stones Wood, W. Riding
- - - - - - - - - 1 1 Mansfield, Notts.
11 - 1 - - - - - 22 16 24 10 17 27 920 Waleswood, W. Riding
65 - - - - - - 15 22 67 2 - 2 Via Gellia, Derbys.
43 - 1 1 2 1 1 17 19 78 I 4 Orgrev, W. Riding
5 - 4 25 12 1 1 4 64 2 31 7 50 4 1610 Grangemill, Derbys.
43 - 12 16 50 3 32 8 16 7 13 10 5 14 1425 39 4 9 55 1512 Cadeby Common, W. Riding
... - - - - -~~7 43 2410 - 2 Orgreave, W. Riding
.- .- - - - - - 1 -1 Orgreave, W. Riding
.- .- - - - - - - - 1
.- Orgreave, W. Riding
..... - - - - 8 38 2 1824 3 Maltby, W. Riding
.....- 59 -
51 - - - -
- -
-
-
-
-
-
-~~8
1036
6" 27 1
Via Gellia, Derbys.
Via ellia.erbrhv.

Carex flacca - - -
C. panicea 2 - - - - - -
Catapodium rigidum - - 1
Centaurea nigra - - 3 10 34 527 1 12 35 5 54
Cerastiumholosteoides - - 7 43 - 25 11 3 1 1 18 21 80 3 49 8
Chamaenerion angustifolium - - 1 4 8 41 - 14 8 16 8 - -
- - - - - - 1 21 16 - -
Chenopodium album -
Cirsium vulgare - - 1 6 65 2 1 3 2 12 32
- - - - 1 - 1 - - - -
Cllnopodium vulgare - - - - - - -
Convolvulus arvensis 3 1 12 32
- - 4 - 3 - - 1 1 47 14 35 12
Cynosurus cristatus
Dactylis glomerata- - 4 - 36 3 14 21 23 2 19 5 11 35 80 3 51 6
Deschampsiacespitosa 1 2 8 3 10 25 1 3 - 4 2 2 14 29
D. flexuosa - - 11 7 544 32 6 26 1 2 - - 5 54
- - - 2 21 -
Digitalis purpurea 2
Draba muralis - - - - 1 - -
7 12 4 15 10 3 1 - 1 8 14 -
Epilobium hirsutum - - - -
3 - 13 13 2 - - -
Eriophorum angustifolium - - -
Festuca gigantea - 1 1811 4 -
F. ovina - - 4 - 38 2 54 4 194 4 - - 9 35
F. rubra - 2 10 5 - 47 1 65 2 21 3 28 2 5 51 55 9 53 5
2 - 3 1811 - - - - -
Filipendula ulmaria - - - -
Galeobdolon luteum- - - 7.32 - 1
Galium aparine - - 1 1314 - - 3 2 16 24 7
G. palustre 1 - 16 9 12 17 - -
G. saxatile - - 8 33 - 5 80 - 1 1 - - 9 35
G. verum - - - 6 65 - - - - -
Geraniumrobertianum - - 1 5 44 7 52 65 2 10 14 8 14 1
Geum urbanum - - -2 1 - - -
10 8 12 9 10 25 7 32 - - - - - - 554
Glyceria fluitans
G. maxima 6 14 - 8 33 - - -
Helianthemum chamaecistus - - - 6 65 22 15 - - -
Helictotrichon pratense - - 7 52 24 14 3 -
Heracleumsphondylium- - 3 6 37 10 34 3 4 1 5 51 35 21 12 32
Holcus lanatus 2 6 14 23 5 4 14 24 5 41 5 27 14 9 19 20 70 6 70 2
H. mollis 3 4 12 15 24 4 665 - 4 4 2 5 41 7 43
Hordeum murinum - - - 1 - - -
Juncuseffusus 5 22 2 31 2 10 25 - - 12 32
J. squarrosus - - 2 - - -
Koeleria cristata - - - 16 21 19 18 4 - -
- - - - - - - - 2
Lathyrus montanus
L. pratensis - 1 - 3 - - 1 27 25 7 43
Lemna minor 53 1 4 13 15 18 -1 3 -
Leontodon hispidus - - - 16 21 27 12 12 10 - - - 2
Loliumperenne - 2 1 - 10 34 - 2 5 27 16 24 82 2 51 6
Lotus corniculatus - - - 13 26 8 32 3 - 1 5 41 7 43
Luzula campestris - 3 - 665 8 32 2 1 - 17 33 35 12
- - - - - - - - - - -
Lycopersicon esculentum cv. Ailsa Craig - - -
Matricaria matricarioides - 1 1 - 49 8
- - - 18 18 - 4 1 6 46 2
Medicago lupulina - - - -
Melica nutans 3
Mentha aquatica 4 - 11 17 20 - - - 1-
Milium effusum - - - -
Minuartia verna - - 1 - -
- - - ---
Myosotis sylvatica -
Nardus stricta - 6 47 1 1
- - - 7 52 3 2 1 -
Origanum vulgare - -
Oxalis acetosella- 1 2 1 22 15 2
Phalaris arundinacea 3 8 11 7 43 9 28 - - 2
Plantago lanceolata- - 2 - 36 3 5 41 11 13 4 5 51 50 12 26 18
P. mnjo- - 3 1 6 65 - - 5 27 20 17 10 37 19 24
Poaannua - 8 33 4 12 28 - 3 11 13 71 2 15 34 21 21
P. pratensis- 2 7 43 - 28 9 11 24 7 19 22 3 6 46 20 31 47 9
P. trivialis 1 15 7 29 3 45 1 3 5 41 6 22 40 1 28 12 87 1 60 4
- - 1 3 - - - 72 1 7 43
Polygonum aviculare - - - - - -
P. convolvulus 408 -
P. persicaria - - 2 - - - - 1 50 6 - 2
Potentilla erecta - - 8 33 4 - - - - - - 7 43
Poterium sanguisorba - - 7 52 11 24 1 - - -
Prunella vulgaris - 7 43 1 9 39 3 1 - 5 51 27 25 14 29
Ranunculusrepens 6 14 6 14 244 244 1 - - 3 20 17 52 10 44 11
8 9 31 2 6 4 12 17 - - - - - - -
Rorippa nasturtium-aquaticum - -
Rumex acetosa I 13 13 7 52 5 41 1 1 4 65 7 47 9
R. acetosella - - - 12 28 - 1 2 2 2 14 29
R. obtusifolius 1 2 4 544 - - - 3 12 22 7 39 2
Salix cinerea ssp. atrocinerea (seedlings) - - 6 47 4 - - - 8 14 - -
Sanicula europaea - - - - - - - -
- - - - - - -
Sarothamnus scoparius
Scabiosa columbaria - - - 9 39 11 24 5 27 -
Sedum acre - - - 12 28 5 41 2 3
Senecio jacobaea- - I - 21 15 35 9 10 14 7 18 3 - 16 26
S. squalidus - - - 7 52 - 7 19 5 27 1 -
S. vulgaris - - - 3 - - 1 19 17 -
- - - 1 3 - - - - 554
Sieglingia decumbens - - -
Silene dioica 1 18 11 2 4 1 - -
Stellaria media - 2 2 - 3 1 65 3 10 37 -
Succisa pratensis 1 1 - - - - 2
Taraxacumofficinale 4 2 31 7 11 24 18 6 22 3 11 35 50 12 28 16
Teucrium scorodonia - - 5 80 32 10 527 1 - -
- - - 13 26 27 13 2 - -
Thymus drucei - - - 2 - - - - - 2
Trifolium medium
T. repens - 2 8 33 - 6 65 5 41 - 1 25 15 72 15 77 1
- 6 47 1 3 8 32 1 4 5 51 -
Tussilagofarfara - -
Ulex europaeus - - - - -
Urtica dioica - 12 15 35 2 8 32 3 13 10 15 26 2 2
Vaccinium myrtillus- - 1 - 8 41 - 2 1 1 - 2
Vaccinium vitis-idaea - 1 - -
Veronica arvensis - 7 52 - 1 6 46 2 2
Viola riviniana- - 1 1 2 43 5 - 1 1 2
Zerna erecta- - - 12 28 - - - - - 2
The data presented in this paper were obtained using seed collected from the sources listed on the right-hand side of the table. In addition to the local sour
* Festuca ovina was collec
(Facing p. 394)

- 33 la I - I 3I -
958
1 - - - J 0- I*- - 2 - 2 - - 943
1 12 35 5 54 7 64 - I - 6 42 2 - - 17 28 -
1 18 21 80 3 49 8 31 21 2 15 29 - 45 5 33 9 15 25 8 28 28 20 - 1 - 2
16 8 - - - 2 2 25 16 44 2 2 - 52 3 18 16 38 11 - 26 5 5 33 4
1 21 16 - - - - 12 37 - - 2 10 38 20 13 - -
1 3 2 12 32 12 53 - 9 46 8 21 - 12 32 12 32 10 22 6 61 - 1 -
..........- -- -3 2
12 32 - - - 4 - - 2 - - 2 - - 10 21
1 1 47 14 35 12. 7 64 1 - - - - 7 42 3 - -
19 5 11 35 80 3 51 6 35 16 - 39 8 22 9 17 25 39 5 32 13 28 11 55 2 - 7 33 25 12 5
4 2 2 14 29 24 30 4 - 12 17 6 42 3 10 38 - 661 12 16 - 35
2 - - 5 54 35 16 97 1 3 10 20 15 27 9 37 2 - 37 1 34 1 -
1 2 - - - 3 - - 2 - - - 1 -
. - - - 4
--43 . - - -3 - -
8 14 - - 7 52 4 - 556 5 49 5 36 5 65 -
-4 - - - 4 2 - -
4 - - - 2 - - 10 19 - 7
4 - - 9 35 75 1 58 2 - 626 55 2 3 2 - 18 26 5 8 5 42 -
28 2 5 51 55 9 53 5 64 2 3 21 19 12 17 51 3 41 27 17 21 5 36 66 1 - 2 5 33 3
~- - - 1 - - - - - - - 2 - I
1 - - - - - - - - - - - - 219 5 33 36
2 16 24 7 - 4 - 11 39 - 2 - 2 5 36 - - 8 25 60 2 10
1 - - - - - - -
1 - - 9 35 25 27 46 3 - - 2 - - - 9 4 11 18 -
- - - - 16 43 - - - - 2
8 14 1 - - - - -19 21 5 56 2 - 12 35 - 2 15 18 20
1 - - - - - - - 5 42 - 12
- 5 554 2 - - -
...- 25 27 -- 937 -
- - - - 459 -- 642 - - -
1 5 51 35 21 12 32 4 - 15 29 - 2 17 20 - - 32 14 - 4 50 3 12
14 9 19 20 70 6 70 2 36 14 4 484 62 1 21 19 45 2 40 8 20 13 48 5 - 9 22 5 33 3
4 2 5 41 7 43 4 9 14 8 50 20 11 4 2 - 3 2 7 6 17 13 30 9 1
1 - - - - - 1 - - 3 10 38 -
1 - - 12 32 - 11 11 5 56 - - - - - - - 1 -
..- - - 522 - - - - - - - - - - -
- - - - 584 - - - 1921 - - - 3 - - - -
- - - 2 764 - - - 2 - - - - - -
1 27 25 7 43 572 - 4 - 6 42 2 5 49 - 5 -
- - - 2 21 32 - - - 26 14 6 48 5 49 - 46 9 - -
5 27 16 24 82 2 51 6 764 2 21 19 4 - 8 40 35 12 35 6 8 51 - - 5 33
- 1 5 41 7 43 44 10 2 3 - 26 14 14 29 2 3 32 14 - - -
1 - 17 33 35 12 55 6 12 10 1 2 17 25 2 - - 8 51 6 7 3
49 8 - - - 20 23 - - 15 23 45 5 18 16 -
1 6 46 2 - 4 - 3 - - 15 23 5 49 - 17 28 -
............... 1
~ ~ ~ ............ 8 25 4
- - - - 51 3 - - - -
... -- 764 463 .- . 58 - -
1 - - - 4 - - - - - - 25 22 -
2 - - - 2 - -- - - - 10 19 7
4 551 50 12 26 18 4012 1 20 23 12 17 32 12 17 20 12 32 5 36 38 11 - -
5 27 20 17 10 37 19 24 - - 29 10 2 - 20 17 37 9 10 22 5 65 - -
11 13 71 2 15 34 21 21 2 3 47 14 15 4 3 6 1 48 2 9 43 - 1 - 1
22 3 6 46 20 31 47 9 33 18 7 19 359 6 26 9 37 35 7 30 14 15 19 48 5 - 4 25 12 1
40 1 28 12 87 1 60 4 958 1 631 2 2 26 13 50 4 53 1 9 43 - 12 16 503 32
- 72 1 - 7 43 - - 25 16 14 15 - 15 23 30 14 20 13 2 - - -
- 40 8 - - 3 - - 2 2 3 - -
1 50 6 - 2 - - 11 39 4 - - 2 13 20 -
- - - 7 43 29 24 11 11 - - 2 - - -
- - - 31 21 - - - 11 33 3
- 51 27 25 14 29 16 43 - - - 2 -8 51 -
3 20 17 52 10 44 11 - - 29 10 - - 9 37 12 32 8 28 11 38 - - - 1
1 4 65 7 47 9 31 21 3 8 50 - 64 1 11 36 5 49 3 943 - 1 - 1
2 2 2 14 29 - 11 11 56 26 26 - 6 48 2 - - 6 36 -
3 12 22 7 39 2 - 407 8 21 - 11 36 32 13 10 32 - -
8 14 - - - - - 5 56 - 2 9 37 7 42 5 36 8 51 - -
- - - -- - - - - - - - - 3 13
~- - - - 10 55 - - - 6 42 - - - 8 51 - - -
3 - - - 2 - - - 5 65 - -
7 18 3 - 16 26 33 18 - 15 29 2 19 21 8 40 7 42 5 36 46 9 - 2 - 2
5 27 1 - - - - 7 52 22 9 - 23 15 45 5 3 8 51 - -
1 19 17 - - 2 - 28 13 - 2 15 23 42 6 40 4 3 - - 5 33
5 54 16 43 - - - - - - - - - -
1 - - - 57_2 - 4 - - - 1 6 - Q L
1 65 3 10 37 - - - 28 13 - - 5 56 17 21 33 7 - - 1 10 21
- 2 20 35 - - 2
22 3 11 35 50 12 28 16 22 32 - 28 13 2 11 33 35 7 37 9 10 22 38 11 - 6 36 10 21 7
1 - - - 18 41 - - - 1S 27 5 56 - - 8 51 - 6 36 - 2
- - 25 27 - - - 28 13 - 3 - - -
2 2 2
1 25 15 72 15 77 1 36 14 3 24 18 - 23 18 18 19 22 17 8 28 12 35 - 1
4 5 51 - - - - 21 19 38 3 6 42 27 12 42 6 - 48 5 - 2
4 - 1 - - 2 2 -
- -
13 10 15 26 2 2 2 - 446 - - 6 48 15 25 43 3 3 15 14 651 17
1 1 - 2 12 53 415 - - - - - - - 8 5 7 33 -
1111 - - - - - - - 58 3
1 6 46 2 2 7 64 - 3 - 5 65 - -
1 1 - 2 40 12 1 - - 11 33 2 - - 10 19 - 19
- - - 2 4 - -- - - 6 61 - - -
of the table. In addition to the local sources, collections of Brachypodium pinnatum from Crich, W. Riding and Wye and Crundale N. N. R. and of Zerna erecta from Cric
* Festuca ovina
was collected from two sources: Fox House, W. Riding (acidic substratum) and Via Gellia, Derbys. (calcareous substratum).

39 - j - a - - - b1f - 11 31 1 Monksdale, Uerbys.
79 - Maltby, W. Riding
43 - - - - - Lindrick, W. Riding
28 - - - - - - - 651 4 408 1 Maltby, W. Riding
20 - I - 2 - 2 - 12 30 18 13 12 47 5 32 Coombsdale, Derbys.
11 - 26 5 5 33 4 13 11 17 6 26 4 12 30 6 27 15 31 6 29 Walkley, Sheffield
-.- - - - - - - 3 - - Orgreave, W. Riding
61 - 1 - - 1 - - 12 30 6 27 2 1 Maltby, W. Riding
- 2 - - - 2 - Maltby, W. Riding
10 21 - - - - 1 - 764 1 Orgreave, W. Riding
.......- - - -18 18 16 13 - - Cressbrookdale, Derbys.
2 7 33 25 12 5 38 1 8 20 4 57 4 25 9 56 2 20 5 Orgreave, W. Riding
61 - 12 16 - 35 5 10 15 2 1 6 51 4 9 55 2 Ashington, Northumberland
37 1 34 1 - - 30 3 31 2 27 3 4 5 27 6 71 54 1 Temple Normanton, Derbys.
1 - - 5 30 6 22 1 - - - I Walkley, Sheffield
- - - - - - - - - Winnats Pass, Derbys.
65 - - - - - 2 3 Maltby Common, W. Riding
- - - - - - - - - - - Pebley Quarry, W. Riding
10 19 - 7 31 8 19 4 1 - 4 - 5 32 Wharncliffe, W. Riding
26 5 8 5 42 - - 1 - - 7 41 6 27 26 14 19 5 Two sources*
1 - 2 5 33 3 - 2 2 68 1 22 12 62 1 21 4 Lathkilldale, Derbys.
- 2 - 1 2 2 - - - 2 1 Via Gellia, Derbys.
- 21 9 5 33 36 3 12 12 1 - 3 - - 1 Anston Stones Wood, W. Riding
- 8 25 60 2 10 24 8 19 6 22 2 25 15 2 14 32 6 30 Via Gellia, Derbys.
- - - - 1 - - - - 2 Orgreave, W. Riding
9 4 11 18 - - 7 24 2 5 14 1 8 59 16 10 Deepcar, W. Riding
- - - - 1 1 1151 - Coombsdale,
...- Derbys.
35 - 2 15 18 20 10 4 - 1 6 51 4 4 3 Taddington Wood, Derbys.
- 5 42 - 12 20 5 30 2 - 1 - - 1 Litton, Derbys.
.- ..- - - - - - - I Ashover, Derbys.
.-
.- .- - - ..- - - - - I Cromford Canal, Derbys.
- ... - - - - - -6 71 - Litton, Derbys.
- - - - - - - - - 14 36 - Lathkilldale, Derbys.
14 - 4 50 3 12 20 6 28 3 1 37 10 6 27 39 9 7 26 Via Gellia, Derbys.
5 - 9 22 5 33 3 4 8 20 7 11 29 12 12 16 34 11 19 7 Lower Bradway, W. Riding
7 6 17 13 30 9 1 35 1 25 3 24 6 714 1 5 79 26 3 Lathkilldale, Derbys.
- - - - - - - - 2 - - Roche Abbey, W. Riding
1 - - 2 - - 3 - 4 Orgreave, W. Riding
.- ...- .- - - - 4 - 2 Broomhead Moor, W. Riding
- - - 12 47 - Millers Dale, Derbys.
.........
- - - - 8 59 - Whitley Wood, Sheffield
12 30 - 14 36 2 Millers Dale, Derbys.
.- .- .- -.- -. .- - - Wentworth Woodhouse, W. Riding
..-
9 - - - - - - - 4 - 14 36 - Via Gellia, Derbys.
51 - 5 33 - - - - 36 11 41 3 1 4 Markland Grips, W. Riding
14 - - - 12 30 - 42 6 3 Millers Dale, Derbys.
51 6 7 3 - - - - - 14 25 2 20 19 11 17 Waleswood, W. Riding
- - - - - - Commercial source
33 6 - I Roche Abbey, W. Riding
28 - - - - - - - - 6 27 955 1 Lathkilldale, Derbys.
1 - 3 - 2 - - - - - Cressbrookdale, Derbys.
.- ...- .- .- .- - - Mansfield, Notts.
8 25 - 4 8 19 - - - 2 -Smithy Wood, W. Riding
.-
-
-- --.. - - - Bradwell, Derbys.
1 - 5 39 2 2 1 - - 1 1 Calver Slough, Derbys.
5 8 - - - - - 1 4 4 9 23 Wharncliffe Chase, W. Riding
22 - - - - - - - - - 11 51 - Lindrick, W. Riding
10 19 - 732 167 2 612 Anston Stones Wood, W. Riding
- - - - - - - I1 1
I-- Mansfield, Notts.
11 - 1 - - - - - 22 16 24 10 17 27 9 20 Waleswood, W. Riding
65 - - - - - - - 15 22 67 2 - 2 Via Gellia, Derbys.
43 - 1 - 1 2 1 1 17 19 78 1 - 4 Orgreav, W. Riding
5 - 4 25 12 1 1 4 - 64 2 31 7 50 4 16 10 Grangemill, Derbys.
43 - 12 16 50 3 32 8 16 7 13 10 5 14 14 25 39 4 9 55 15 12 Cadeby Common, W. Riding
..- . ... 7 43 24 10 - 2 Orgreave, W. Riding
- .- .- .- .- .- - I1 - - 1 Orgreave, W. Riding
.- .- .- - - - - 1
.- -..- Orgreave, W. Riding
.......- - 838 2 18 24 3 Maltby, W. Riding
-
.... - - - - 8 59 - Via Gellia, Derbys.
51 - - - - - - - 10 36 6 27 1 1 Via Gellia. Derbys.
38 - - 1 2 - 1 522 14 14 2 9 20 Norton, Sheffield
- - .- .- - - - - - - - Bradwell, Derbys.
43 1 - 1 - 2 - 18 18 2 26 14 11 16 Loxley, W. Riding
6 36 1 4 2 1 6 27 2 19 7 Yarncliffe, Derbys.
..-.. . . 6 51 6 27 4 Wortley, W. Riding
51 - - - - - - - - - - I Black Carr Plantn., W. Riding
3 13 16 1 2 Anston Stones Wood, W. Riding
- - - - - - - -1 - -
Thorpe Common, W. Riding
51 - - - - - - - - - 7 64 - Lathkilldale, Derbys.
65 - - - - - I Deepcar, W. Riding
9 - 2 - 2 - - - 17 19 10 18 13 42 4 Deepcar, W. Riding
51 - - - - - - - - 4 - 1 Broomhill, Sheffield
5 33 - - 1 - 3 6 27 2 2 Broomhill, Sheffield
3 4 859 1 Peak Forest, Derbys.
-2 - ^5 IQ&2 161l 12 12 9 17 4 - 4 3 Coombsdale, Derbys.
1 10 21 - 3 9 17 1 8 38 10 18 1 3 Orgreave, W. Riding
....- - - - - I - 764 1 Lathkilldale, Derbys.
11 - 6 36 10 21 7 32 2 2 - 47 5 24 10 14 32 3 Broomhill, Sheffield
51 - 6 36 - 2 2 1 5 14 6 - 16 28 - Bawtry, W. Riding
.- ...- - - - - - - 3 - Cressbrookdale, Derbys.
......- - 12 47 - Maltby, W. Riding
35 - 1 - - - - - 46 6 31 7 2 5 32 Norton, Sheffield
- 2 - - 1 - - 4 6 27 1 3 Central Sheffield
- 1 - - - - - - - 3 2 Norton, Sheffield
15 14 65 1 17 13 10 15 10 14 9 10 29 12 8 22 14 32 9 20 Coombsdale, Derbys.
8 5 7 33 - - 10 15 1 4 - - 3 14 13 Wigtwizzle, W. Riding
5 8 3 - - - - 2 - - - 3 Froggatt Edge, Derbys.
65 - - - - - - - - - Winnats Pass, Derbys.
- 10 19 - 9 12 3 2 1 3 4 21 17 1 Ranby, Notts.
61 - - - - - - - - 2 20 19 - Pleasley Vale, Derbys.
N. N. R. and of Zerna erecta from Crich, W. Riding; Fircester, Wilts.; Juniper Top, Surrey; Markland, Notts. and Monksdale, Derbys. were used.
. (calcareous substratum).

J. P. GRIME
ANDRODERICK
HUNT 395
MATERIALS AND METHODS

Species
The species were chosen primarily in order to allow comparisons in rates of seedling
growth to be made between groups of flowering plants characteristicof local examples of
the major inland habitats of the British Isles. However, because the number of species
involved was large (132), additional comparisons were possible. A natural consequence
of selection on ecological grounds was the inclusion of species differing widely in other
plant characteristics such as life-history, quantity of seed reserve and morphology of the
root or shoot. In order to facilitate comparisons between groups defined by these secon-
dary criteria, additional species were selected in order that certain under-represented
groups, notably legumes and woody species, became large enough to permit separate
consideration. An important ecological group, that of vernal woodland species, could
not be represented properly because seeds or seedlings were not readily available and
because the experimental conditions appeared to be unfavourable climatically for certain
of these species.
The majority of the species selected were herbaceous plants of widespread occurrence
in Britain. The list (Table 1) includes species characteristicof the more ancient landscape
and also plants associated with habitats which are either disturbed or of recent origin.
With the exception of Brachypodiumpinnatum,*Festuca ovina and Zerna erecta, in which
separate analyses were carried out using seed from different sources, each species was
represented by a collection of seed or seedlings from one local field site. In the case of
tomato (an important naturalized species on sewage waste) and in several coniferous tree
species commercial seed sources were used.
Data from an unpublished vegetation survey (J. P. Grime & J. G. Hodgson, in pre-
paration) have made it possible to estimate the frequency of occurrence of each of these
species in the major habitats in an area of approximately 900 square miles (2400 km2)
around Sheffield, Yorkshire (Table 1, columns headed a). The essential features of the
survey area have been described elsewhere (Lloyd, Grime & Rorison 1971; Grime &
Lloyd 1973). The data from this survey also allow the species recorded in each habitat
to be ranked in order of percentage occurrence in the samples. In Table 1 this ranking
is indicated in columns headed b. The extent to which this sample of species represents
the complete list of plants to be found in each habitat is shown in Table 2. Satisfactory
representation was obtained in all but the first two habitats. These two, both aquatic,
are now excluded from further consideration.
Experimentalconditions
General considerations
Ideally, to ensure measurement of the true Rmax of which each species is capable,
estimations should be made under the growth conditions optimal for that species. In
order to find these optimal conditions it would be necessary to subject each species to a
wide range of factorially-combined levels of the necessary growth-factors (sensu Lockhart
1965) and then to adopt as Rmax the highest value of relative growth-rate, R, obtained
from such a series of experiments. Even on a smaller scale than at present such a task
would be totally impracticable. The environment selected for the measurement of Rmx
in this large number of species therefore had to some extent to be a compromise between
what was theoretically desirable and what was experimentally practicable.
* Nomenclaturefollows that of Clapham,Tutin & Warburg(1962).
B

396 RGR in a local flora
The level of supply of water and mineral nutrients was considerably higher than that
commonly experienced by many of the species in the field. However, the work of Hackett
(1967), Parsons (1968b), Rorison (1968) and others suggests that normally unproductive
species are often able to grow faster in monoculture in a productive environment than
in conditions more nearly approaching those of their natural habitat, although Bradshaw
et al. (1964) recorded reduced yields in greenhouse-grown Festuca ovina and Nardus
stricta at nitrogen levels above 81 parts/106, a concentration which is suboptimal for
some arable species in the field.
Table 2. The representationof species examined among those of common occurrence in

habitats 1-31 (Table 1, facing p. 394)
Percentagerepresentation Number representedfrom
of specieswith an the ten most frequently
occurrence of 5% or more occurring species
Habitat in the habitat
1. Lakes, ponds, canals and ditches 32 4
2. Rivers and streams 44 4
3. Unshadedmire 58 8
4. Shadedmire 46 6
5. Rock outcrops 60 8
6. Limestonescree 60 9
7. Cliffs 53 5
8. Walls 67 9
9. Arable 37 8
10. Meadows 60 9
11. Enclosedpastures 64 10
12. Limestonepastures 66 9
13. Pastureson acidic strata 78 8
14. Soil heaps 64 10
15. Coal mine heaps 52 7
16. Lead mine heaps 73 7
17. Cinders 60 8
18. Bricksand mortar 68 10
19. Manureand sewage waste 58 9
20. Limestonequarryheaps 55 8
21. Quarryheaps on acidic strata 65 10
22. Scrub 61 5
23. Hedgerows 56 7
24. Limestonewoodlands 47 8
25. Woodlandson acidic strata 62 5
26. Broad-leavedplantations 54 6
27. Coniferousplantations 58 6
28. Verges 69 9
29. Paths 81 9
30. Wastelandand heath on limestone 57 10
31. Wastelandand heath on acidic strata 81 9
It seems likely that some species achieve their true Rmax only at rates of supply of
growth-factors which would be harmful to other less demanding species. Comparisons
of Rmaxmade among many species in a single set of environmental conditions therefore
may result in an under-estimation of Rmaxin species adapted to survive in environments
far removed from that used in the investigation. However, there is evidence in the litera-
ture and from this investigation to suggest that, under productive conditions, markedly
sub-optimal growth is restricted to species in which the true Rmaxis low. Hence, despite
inaccuracies, such comparisons would differentiate between species of low and high

J. P. GRIME HUNT
ANDRODERICK 397
Rmaxproviding that the environment selected allows Rmaxto be attained, or nearly so,
in species of high potential growth-rate.
Light regime
The scale of the operation required a cumulative series of standardized experiments.
The controlled-environment rooms described by Rorison (1964) were used for this
purpose. Illumination of the growing plants in these rooms was provided by thirty 5-ft
(152-cm) 80 watt 'warm white' aged fluorescent tubes which supplied a maximum of
38 0 W m-2 visible radiation (0'0545 cal cm-2 min -) at plant height. A long day-length,
18 h, was chosen to encourage a high productivity (Hughes & Evans 1963). The daily
total visible radiation supplied was 2-46 MJ m-2 (equivalent to 58-95 cal cm-2 day-').
This figure is comparable to mean daily outdoor values for only two or three winter
months in the year and is only in the region of 24% of the mean daily value for similar
latitudes to Sheffield in June (de Vries 1955). Although arrangements were made to
ensure that each plant received this level of radiation with no shading by adjacent plants,
it is nevertheless the case that the daily light energy input to plants in the growth-room
was less than that which they would receive in the majority of open field situations at
temperate latitudes.
Table 3. The relationshipbetween temperatureand mean relativegrowth-rate

(week-1) over the period 2-5 weeks after germination in five grass species
(data of Mahmoud 1973)
Temperature (? C) L.S.D.
Species 15 20 25 30 (P< 0-05)
Agrostistenuis 1-31 1-58 1-52 1-36 0'29
Arrhenatherum elatius 0.99 1.31 1.42 1-16 0.37
Deschampsiaflexuosa 0'96 1.22 0.99 0'62 0.11
Festucaovina 1.05 1.06 1.12 0-85 0.23
Zernaerecta 0'72 1 03 0'99 0-82 0 11
Temperatureregime
A uniform day temperature of 20? C was chosen. This is above the normal mean
summer air temperature of most of the field situations near Sheffield but is not outside
the experience of any of the species used. Experiments carried out by Went (1957) con-
firm that high yields may be obtained in a variety of different species grown at this tem-
perature. At night the temperaturewas lowered to 15? C to decrease dry weight losses due
to respiration and to give a crude approximation to a 'normal' diurnal temperature
variation which may be of physiological importance to many species.
Since the present experiments were concluded the work of Mahmoud (1973) has
confirmed the suitability of the chosen temperature regime for near-optimal growth in
several native species. Here, five grasses Agrostis tenuis, Arrhenatherumelatius, Des-
champsiaflexuosa, Festuca ovina and Zerna erecta were grown under conditions, and
using methods closely similar to those of the main series of experiments except that for
each species there was a treatment in which temperaturewas held continuously at 15, 20
25 or 30? C. These results are summarized in Table 3. It seems that in each case a day
temperature of 20? C is at, or near to, the optimum temperature for the realization of
Rmax given the background of the other conditions used. R (mean relative growth-rate)
was calculated from Fisher's (1920) formula: (loge 5W-loge 2W)/T where W and 2 W

are whole-plant dry weights at 5 weeks and 2 weeks respectively and T is the time interval,
3 weeks.
Rooting-medium
The choice of rooting-medium presented few problems. Soil was rejected as a possi-
bility because of its variable and uncharacterizednutrient and water supplying capacity.
Solution culture, although not suffering in this respect, was also rejected because of the
difficulty of devising a single plant-support system suitable for a wide range of plant
sizes and growth forms. The choice therefore rested with sand/solution culture.
Nutrient solution
The choice of a suitable mineral nutrient solution was governed by the need to provide
the more productive species with more than adequate supplies of mineral nutrients. The
'Long Ashton' solution (Hewitt 1966, Tables 40, 41) was selected because of its reported
suitability for the growth of a wide variety of crop species. It was suspected that the levels
of mineral nutrients supplied might be super-optimal for the less productive species. In
order to examine this possibility four species drawn from various types of unproductive
vegetation, Deschampsiaflexuosa, Festuca gigantea, F. ovina and Scabiosa columbaria
were grown in the controlled-environment room in sand/solution culture using both the
standard solution and a solution in which the major mineral nutrients nitrogen, phos-
phorus, potassium, calcium and magnesium were supplied at only one-fifth of the stand-
ard rate. No differences with respect to final yield or R were detectable at P<0.05
between treatments after four weeks' growth.
Rooting-volume
A test was made to determine the minimum volumes of sand and solution necessary
to support the highest yields expected in the projected experiments.Seedlings of sunflower,
Helianthus annuus L., were grown in the controlled environment room for four weeks
under a temperatureregime of 20?/15? C with an 18-h day. Four pot sizes over the range
115-2280 cm3 were used and the sand was supplied daily with an excess of full nutrient
solution. There were clear and significant increases in final dry weight yield with each
increase in pot size. From inspection of the curve of yield v. sand volume it was clear
that yields of sunflower of around 4 g dry matter could be expected under these conditions
from a volume of 500 cm3 of sand. It was thought that few, if any, native species would
approach this yield after only four weeks' growth and, on this basis, the volume of sand
to be used in the experiments was fixed at 500 cm3.
Design of container
Conventional flowerpots use growth-room space inefficiently: at 500 cm3 each, only
128 could be accommodated at once. A special container (Fig. 1) was, therefore, con-
structed from 4-in (10-2-cm) flat, transparent polyethylene tubing. This provided con-
tainers with flexible walls which, when closely packed, assumed the more efficient hexa-
gonal section. With this design it was possible to fit up to 360 into a growth room at one
time and still leave space for a thermohygrograph and dishes of germinating seeds.
Other important design considerations for this container were that it should provide a
standardized rooting and aerial environment in which the roots and shoot of each seed-
ling were completely isolated from those of its neighbour. To provide a shoot enclosure
the polyethylene walls of the container were continued upwards for 10 cm above the

J. P. GRIME AND RODERICK HUNT 399
surface of the sand. At the top of this extension the tubing was turned back on itself for
a further 10 cm and a lining of aluminium foil was inserted into the space between the
two layers of plastic (see Fig. 1). This device ensured a complete isolation of the aerial
Lights 93 cm
from sand
surface
I
I /Seedling
I //
/ /Aluminium
/J/q , /- foil lining
lOAcm / Sj Polyethylene
10/cm /tubing
/ 500 cm3 sand
__ . / Glass-fibre cloth over

' ' / /drainage holes
* */ /
': . / / /Heat-sealed base
/ / /Supporting grille
_
CD 1CD
C Cf _ )O c_ =
|[\~~~~ ~ \ Inverted plastic pot
I \//
Waterproof tray
To waste
FIG. 1. Vertical section through a seedling container.
parts of adjacent seedlings with a drop in light intensity of only 18% between the mouth
of the container and the surface of the sand.
General arrangement
The containers were supported in rows on a PVC-coated heavy-gauge steel grille.

This system allowed completely free drainage from each container and ensured that the
excess solution emerging from the bottom of the containers was not allowed to collect
outside or to contact any other container. It was found that the sand in each tube could
be kept uniformly moist by the addition of 20 ml of solution on alternate days. The
adequacy of this rather low rate of watering was owing to the small exposed-surface/
volume ratio of the sand, the non-porous sides of the containers, the 'sheltered' position
of the sand surface and the relatively high humidity in the growth room (maintained
additively at a level continuously above 60% R. H.). When containers were watered at
this rate solution still emerged from the bottoms and tests indicated that any accumulated
nutrients were flushed out by this process.
A summary of the experimental design and environmental conditions is set out below.
Medium: Sand+ Hewitt nutrient solution Visible radiation at plant height:
38-0 W m-2 (0-0545 cal cm-2 min-')
Day temperature: 20+ 0 5? C Night temperature: 15+0 5? C
Day-length: 18 h to
(06.00 00.00) Relative humidity: always above 60%
Dates of experiments: 24 November 1967-24 May 1973
Design of experiments: 12-18 species x 4 harvests x 5 replicates.
Procedure
Sand treatment
The sand used was a pure, washed silica sand obtained from Messrs Arnolds Quarries
Ltd, Leighton Buzzard, Bedfordshire. It was subjected to further purification in poly-
ethylene bins by an acid-washing procedure similar to that described by Hewitt (1966).
After each experiment the sand was re-used following a repeat of the whole acid-washing
process.
Germination
In order that seedlings of comparable age could be obtained simultaneously for
different species a simple germination test was previously performed on seeds of all
species considered for inclusion in the series. Seeds were set out in Petri dishes on moist-
ened Whatman No. 1 filter paper, in the environment described above and daily counts
of percentage germination were made. On the basis of this information seeds could be
imbibed for different lengths of time so that simultaneous peaks of germination were
attained by up to eighteen species prior to the start of each experiment.
Seedlings of some species which it was desired to include in the experiments could
not be obtained in the normal way either because no seed was available, or because the
seeds had some special or lengthy pre-germination requirements which could not easily
be provided. Several woodland species came into this category. In these cases material
for the experiments was provided in the form of newly-germinated seedlings transferred
directly from the field to the experiment. At the time of collection these seedlings all
possessed cotyledons in good condition and all had either no true leaves or the first one
(or pair) visible. An initial harvest, extra to the main series, was carried out on these
seedlings as they came in from the field. Lemna minorwas grown from single fronds using
the same techniques except that solution culture replaced sand.
Randomization
The containers were placed in two randomized blocks, one on each side of the growth

J. P. GRIMEAND RODERICKHUNT 401
cabinet. This precaution was taken only to minimize the possible effects of differences
between species with respect to maintenance procedure. Rorison (1968) has shown, for
more than one species, that in growth rooms of this design, spatial variation in environ-
ment causes only minor differences in the performance of plants growing in different
positions.
Harvesting
The duration of the experiments was fixed at five weeks with harvests at two, three,
four and five weeks. For a few species it was not possible to harvest exactly at these times
and harvests were taken within three days of the designated dates. In these cases the
actual time intervals were recorded exactly and used in computation. This design allowed
plants to 'settle-in' to the environment before any measurements were taken and was a
sufficiently short period to ensure that even the most productive species were still near
their exponential phase of growth at the end of the measurements. Seedlings were planted
one per container and selected for harvesting at random.
Measurements
Measurements of mean 'seed' weight were made prior to each experiment. One
hundred seeds were weighed together in an air-dried condition and a mean weight was
obtained. This measurement should perhaps more correctly be called a 'disseminule
weight' since in some species (e.g. Poterium sanguisorba)the pericarp was included. At
the time of harvesting a number of measurements was carried out. Although these are
all described here only data on dry weights are dealt with in this paper.
Linear measurements.Plants were removed from the sand and the distances from the
sand surface to the furthest points of the root and shoot were measured. The root measure-
ment was made with roots in a straightened but not stretched condition. The shoot
measurement was made between the sand level and the growing point of the shoot in
species of erect habit and between the sand level and tip of the largest leaf in other species,
including grasses.
Dry weights. Washed plants were placed in paper envelopes and dried at 100? C for at
least 48 h. On removal from the oven the roots and shoots were parted and weighed
separately. The remains of the seed, if any, were included initially in the root weights and
then, on some occasions, detached and weighed separately. Some data have been analysed
both with and without the seed dry weight component.
Leaf areas. Leaves were removed from each seedling before drying and silhouettes
were made on 'UNAX' 3M3 semi-dry diazo blackline paper. Images of the leaves were
then cut out and weighed, a previous calibration giving their equivalent areas.
Whole-plantsilhouettes. A visual record of the sizes of the various species at different

stages of growth was obtained by making silhouettes on 'UNAX' paper of whole repre-
sentative seedlings for each species at each harvest.
Spatial arrangementof containers

Towards the end of each experiment a majority of the more productive species pro-

duced sufficient shoot material to emerge from the mouths of their containers. By this
time, however, the removal of early harvests had created sufficient space in the growth-
room to allow spacing out of the remaining containers to ensure that no shading of
adjacent smaller plants occurred.
In order to increase the effective capacity of the cabinet in some of the experiments a
'staggered' planting schedule was adopted. The initial planting was of material for the
three-, four- and five-week harvests. When the three-week harvest was removed from the
growth-room the plants for the two-week harvest were introduced. These new plants
then grew for the remaining two weeks of the experiment and were harvested at the same
time as the five-week harvest.
Special problems
During the course of the experiment, a small proportion of the seedlings suffered
setbacks in the form of desiccation, fungal attack or disturbance during the application
of nutrient solution. Seedlings showing severe stunting and/or death of leaves were
removed from the experiment and discarded. It seems likely that among the plants which
were allowed to remain in the experiment and which therefore contributed to the total
variability of the data there were some individuals which suffered less obvious checks
on their growth. Such a situation is unavoidable in this type of experiment where one
cannot entirely separate the variability inherent in the seed population from that en-
gendered by experimental conditions or techniques.
A number of small-seeded species, e.g. Thymus drucei, experienced a high rate of
seedling mortality during the first week after planting. This was apparently related to
drying of the surface layers of the sand and was rectified by repeating the experiments
using containers of sand which were allowed to stand in 10 cm of deionized water during
the first week.
When some containers had been in the growth room for two to three weeks it was
found that a crust of algae up to 2 mm thick had built up over the exposed upper surface
of the sand. In one way this was an advantage because it prevented the displacement of
the sand which tended to occur when nutrient solution was applied. Where it was sus-
pected that this crust caused asymmetrical penetration of the solution into the surface
layers of the sand it was broken into small sections with a mounted needle and good
drainage was restored.
Repeatability
If data from a series of experiments are to be combined it is advisable, even where
there has been strict control of the environment, to establish that the experiments have
been carried out in a comparable manner. The environment itself may be monitored by
instruments which ensure an exact repetition of environmental conditions from experi-
ment to experimentbut the technical and procedural repeatabilityis not so easily checked.
It was, therefore, decided that at intervals throughout the period of the investigation
experiments would be repeated on one species as a check on the constancy of the experi-
mental procedure and environmental conditions. Silene dioica was chosen for this
purpose. No significant differencesin Rma could be detected at P < 005 between experi-
ments. The final value of Rmx presented for this species is a combined value obtained
from a pooling of all such data (105 plants in all).

QUANTITATIVE ANALYSIS
Generalrequirements
The analysis of the large body of data obtained from these experiments presented
problems of its own. Because of the scale of the operation a single, computerized pro-
cedure was desirable, but the data themselves required a variety of analytical approaches.
Although in the great majority of the present species, the period two to five weeks after
germination encompassed some part of an exponential phase of growth, the precise
patterns of growth obtained varied considerably from species to species. Many showed
true exponential growth over the whole of the period studied, with no statistically-
significant departuresfrom log.-linear dry weight increases with time. Other species had
a high initial R which, as a result of various internal and/or external influences, declined
towards the end of the period of observations. In some cases this decline was smooth, in
others it was of increasing steepness. Still other species began growth with low values of
R which increased, either uniformly or at a decreasing rate, over the period studied.
A few species exhibited combinations of these trends: R began with a low value which
then increased, was held at a steady level for some time and then finally decreased (a
log.-sigmoidal growth pattern). In the remaining species, growth apparently possessed
no clear-cut pattern and often appeared to be following an intermediate path between
various of these trends.
Approachesthroughregression analysis
Rather than adopt the more conventional harvest-interval approach to the growth-
analysis it was decided to calculate growth-functions from regressions fitted to the raw
data. Among others, Hughes & Freeman (1967), Radford (1967) and Evans (1972, p.
338) have described the advantages which this approach enjoys (see Hunt & Parsons
(1974) for a full bibliography). In this particular case the method was desirable because
of its ability to smooth out any small deviations from general trends with time and to
draw on all of the relevant data for each species in calculating individual values of R.
As a first attempt at an analysis of the data the computer program described by Hughes
& Freeman (1967) was used. This program derives R as the differential (slope) of a
third-order (cubic) polynomial fitted to the natural logarithms of successive estimates
of whole-plant dry weight.
In general, this analysis was not successful. Although any log.-sigmoidal trends in the
data were described adequately, the cubic's approximations to the simpler trends resulted
in an artificially erratic behaviour of R (and other functions) with time. Unacceptably
large standard errors also resulted, even from reasonably uniform raw data. Much of this
problem, no doubt, stemmed from the application of relatively high-order regressions
to data covering only a small number of harvest occasions: the program was originally
constructed to serve a system of frequent small harvests (in this context it can produce
generally-acceptableresults as shown by Hunt & Burnett (1973)). With only four or five
harvest occasions, however, this particularform of analysis often imposed its own pattern
on the values of R obtained. Therefore development of a further computer program,
which suffered less in this respect, was undertaken.
Working along broadly the same lines as those of Hughes & Freeman (1967), Hunt &
Parsons (1974) constructed a program principally for the analysis of the present data,
although other applications were borne in mind. In general terms, plant growth-analysis
functions of the form 1/ Y.d Y/dX, Z/ Y and 1/Z.d Y/dX were derived from lines or curves

fitted to the natural logarithms of variates Y and Z v. time, X. A 'stepwise' regression
procedure was incorporated in which the most apposite of a series of polynomials
from the first to third order (inclusive) could be fitted independently to loge Y and to
logeZ according to statistical tests built into the program. A further facility giving
only a mean and standard error was included for cases where no significant trends with
time existed in the (logged) raw data.
Analysis via this program removed much of the artificial pattern and variability
imposed on most of the present results by the Hughes & Freeman program (see Hunt
& Parsons (1974), cf. their Figs 1 and 2). In preparation for a further publication on leaf
area phenomena, progressions both in W and in leaf area, LA, have been analysed by the
Hunt & Parsons program in each of 192 sets of data; a total of 384 fitted regressions
(some subsequently rejected). To quote Hunt & Parsons, 'Of this total, 210 were linear,
104 were quadratic, sixty-eight were cubic and two were "no-fit", all decided at P <0-05.
In these 192 sets of data the orders of polynomial fitted to loge W and to logeLAwere the
same in 114 cases, were one degree apart in forty-nine cases, two degrees apart in twenty-
eight cases and three degrees apart ("no-fit"/cubic) in one case.' Thus, in the present body
of data the simpler trends with time are the most frequent and the logarithms of W and
LA follow trends that are often similar but are by no means exclusively so.
Comparisonswith conventionalanalyses
Table 4 contains relative growth-rates calculated for Holcus lanatus. Here, values have
been calculated by a number of methods as a specific example of the ways in which
different methods of treating the data can yield different results.
The application of Fisher's (1920) formula (p. 397) gave the overall value of R as
1-61week- 1, but similarcalculations made between individual harvest occasions indicated
values that initially were higher and finally were lower than this. This value for 11,35R
(prefixes indicate the period in days, following Evans (1972)) and that given by the log.-
linear regression must, therefore, be viewed with some suspicion. The two higher-order
regressions indicate substantially higher initial values of R but only in the case of the
log.-quadratic regression is this value significantly higher than that of the overall mean
(P <0 05). In this case the quadratic regression gives a significantly better fit to the data
than does the linear.* The cubic, however, gives no significant improvement over the
quadratic, whilst adding substantially to the size of the confidence limits on the fitted
values of R. For plots of logeW and R in this case see Hunt & Parsons (1974), their
Fig. 2(a) and (c) for the log.-quadratic regression and their Fig. l(a) and (c) for the
log.-cubic regression.
Derivation of Rmax
When an analysis had been established which provided reasonably accurate descrip-
tions of the experimental data a problem remained in deciding on which of the fitted
values of R best represented the concept of Rmaxand provided the most legitimate and
informative basis for the comparisons between species discussed in the introduction.
The possibility of making estimates across the board at a common time or at a common
dry weight was rejected because this would often involve comparisons between plants
at very different stages of development and might, in some species, apply to phases of
growth in which R was limited by external factors. A simple approach which has been
* Significancedecidedat P < 005 by a test of the ratio regressionmean square/residualmean square.

explored here is to take the highest value of R obtained for each species during the period
of observations.
This procedure is not without its limitations, however, and these are reviewed in the
discussion. In particular, where log.-curvilinear growth is exhibited, Rmaxis an instan-
taneous rate and is not necessarily sustained for any substantial period of time. Neither
is this estimate of Rmaxinvariably the value of the true maximum slope of the log W
curve. It is merely the fitted slope at a point close to this maximum (although for all
Table 4. Relative growth-rate calculatedfor Holcus lanatus by various

methods
Relative growth-rate(week-1)
Fisher's(1920) formula:
11-35R 161
11-20R 1-75
20-28R 1 59
28-35R 1.46
LogeW v. time, linear regression:

11-35R 1-56
(0-10)
LogeW v. time, quadraticregression:
11R 2-01
(0-28)
2oR 1-70
(0-12)
28R 1-40
(0-12)
35R 111
(0-28)
LogeW v. time, cubic regression:
1R 2-08
(0-69)
20R 1-68
(0-22)
28R 1-38
(0-21)
35R 1-18
(0-69)
The notation follows that of Evans (1972) in which R and R respectively
representinstantaneousrelative growth-rateand mean relative growth-rate
overa statedperiodof time.The prefixesindicatethe timesor periodsinvolved
(measuredin days from planting). The figure in bracketsis the 95% con-
fidencelimit.
of the quadratic curves and for some of the cubic curves the maximum slope was, in
fact, at the first harvest occasion). Furthermore, in species exhibiting exponential growth
Rmaxis equal in value to R.
Notwithstanding the foregoing difficulties these approximations to Rmaxhave been
used in a first attempt to explore the possible ecological significance of this quantity,
since the capacity, or lack of capacity, for rapid growth (over however brief a period)
is itself a phenomenon of physiological and ecological interest. Cases in which the analy-
ses have produced what might be considered as anomalous results will be discussed
individually.

Table 5. Values obtainedfor Rma, and R (week-l); confidence limits (CL) are given at
P<0-05
Rmax R
Value Standard Lower Upper Value Standard Lower Upper
error CL CL error CL CL
Acerpseudoplatanus 0-34 0'09 0.16 0-52 0.34 0.09 0.16 0-52
Achilleamillefolium 1.96 0.11 1-71 2-21 1-70 0.09 1 49 1.90
Agropyronrepens 1-21 0-05 1.11 1-33 1-21 0-05 1.11 1-33
Agrostiscanina 1-41 0-08 1-23 1-59 1-41 0'08 1-23 1-59
A. stolonifera 1-48 0.10 1-26 1-70 1-48 0.10 1-26 1-70
A. tenuis 1-36 0-12 1.11 1-61 1-36 0-12 1.11 1-61
Airapraecox 0-87 0-07 0-73 1 00 0-87 0-07 0-73 1 00
Alopecurusgeniculatus 1-24 0-07 1.10 1-38 1 24 0-07 1.10 1-38
A. pratensis 1-29 0.10 1-07 1-51 1-29 0.10 1-07 1-51
Anisanthasterilis 2-28 0-12 2-03 2-53 1 43 0-07 1 29 1-57
Anthoxanthum odoratum 0-94 0-08 0-77 1.11 0-94 0-08 0-77 1.11
Anthriscussylvestris 0-52 0-07 0-38 0-66 0-52 0-07 0-38 0-66
Arabishirsuta 1-32 0-12 1 05 1-60 1-32 0-12 1 05 1-60
Arenariaserpyllifolia 1-22 0-14 0-92 1-52 1-22 0-14 0-92 1-52
Arrhenatherum elatius 1-30 0-07 1-15 1 46 1-30 007 1-15 1 46
Betulapubescens 0-80 0-06 0-68 0-93 0-80 0-06 0-68 0-93
Bidenstripartita 1-89 0.09 171 2-06 1-30 0-12 1-06 1-55
Brachypodium pinnatum* 1-03 009 0-85 1-20 1-03 0.09 085 1-20
B. sylvaticum 1-35 0-13 1-06 1-63 1-02 0-04 0-94 1-12
Briza media 1.11 0-06 0-97 1 24 1.11 0-06 0-97 1 24
Callunavulgaris 0-35 0-05 0-25 0-45 0-35 0-05 0-25 0-45
Campanularotundifolia 0-81 0-17 0-45 1-17 0-81 0-17 0-45 1-17
Cardamineflexuosa 1.01 0.09 083 1.19 1.01 0.09 083 1.19
C. pratensis 1-15 0-08 0-98 1-33 0-76 0-12 0-50 1-03
Carexflacca 1-38 0-05 1-28 1-47 1-26 0-04 1-17 1-35
C. panicea 0-68 0.11 0-42 0-93 0-68 0.11 0-42 0-93
Catapodiumrigidum 1 60 0-18 1 23 1-98 1 09 0-06 0-95 1-22
Centaureanigra 1-13 0-07 0-98 1-29 1-13 0-07 0-98 1-29
Cerastiumholosteoides 1 46 0.09 126 1-65 1 46 0.09 126 1-65
Chamaenerion angustifolium 1-44 0-08 1-27 1-61 1-44 0.08 1-27 1-61
Chenopodium album 2-12 0-34 1-39 2-85 1-25 0.10 1-03 1-47
Cirsiumvulgare 1-59 0-25 1-03 2-14 0-89 0-14 0-58 1-21
Clinopodiumvulgare 0-71 0-04 0-62 0-81 0-71 0-04 0-62 0-81
Convolvulusarvensis 2-44 0-26 1-84 3 03 1-36 0-13 1 07 1-64
Cynosuruscristatus 1-54 0-06 1-41 1-67 1-54 0-06 1-41 1-67
Dactylisglomerata 1-31 0-07 1-16 1-45 1-31 0-07 1-16 1-45
Deschampsiacespitosa 1-45 0-06 1-33 1-57 1-45 0-06 1-33 1-57
D. flexuosa 0-81 0-08 0-63 0.99 0-81 0-08 0-63 0'99
Digitalispurpurea 1-21 0-15 0-87 1-54 1-21 0-15 0-87 1-54
Draba muralis 1-24 0-07 1 09 1-39 1-18 0.10 0-97 1 40
Dryas octopetala 1-12 0-22 0-65 1-59 0-58 0-07 0-43 0-74
Epilobiumhirsutum 1-83 0-06 1-70 1-96 1-83 0-06 1-70 1-96
Eriophorumangustifolium 0-71 0-14 0-41 1.01 0-67 0-12 0-39 0-94
Festucagigantea 1-44 0-20 1-02 1-85 1-04 0-06 0.91 116
F. ovina 1 00 0-07 0-86 1-14 1 00 0-07 0-86 1-14
F. rubra 1-18 0-07 1 03 1-34 1-18 0-07 1 03 1-34
Filipendulaulmaria 0-95 0-12 0-64 1-21 0-95 0-12 0-64 1-21
Fraxinusexcelsior 0.90 0-21 0-47 1-33 0-48 0-07 -0-34 0-62
Galeobdolonluteum 0-75 0-06 0-63 0-88 0-75 0-06 0-63 0-88
Galiumaparine 1-51 0.10 1-30 1-72 1-17 0-04 1-10 1-25
G.palustre 1-16 0-07 1.01 1-32 1-16 0-07 1.01 1-32
G. saxatile 1-52 0-25 0-97 2-06 0-79 0'10 0-59 1 00
G. verum 1-12 0-14 0-82 1-41 1-12 0-14 0-82 1-41
Geraniumrobertianum 1-14 0-06 1 01 1 27 1-14 0-06 1.01 1 27
Geumurbanum 0-73 0-15 0-35 1'11 0-73 0-15 0-35 1.11
Glyceriafluitans 1-33 0.09 114 1-52 1-33 0.09 114 1-52
G. maxima 1-08 0-12 0-82 1-33 1 08 0212 0-82 1-33

Table 5 (contd)
Rmax R
Helianthemumchamaecistus 0-70 0.11 0-46 0-93 0-70 0.11 0-46 0-93
Helictotrichonpratense 0-75 004 0-66 0-84 0-75 004 0-66 0-84
Heracleum sphondylium 0-58 0-06 0-47 0-70 0-58 0-06 0-47 0-70
Holcus lanatus 2-01 0-13 1 73 2-28 1-56 005 1 46 1-66
H. mollis 1-44 007 1-28 1-60 1 44 007 1-28 1-60
Hordeummurinum 1-76 0-16 1-41 2-10 1-18 0-08 1 01 1-35
Juncuseffusus 1-03 0-08 0-85 1-21 1-03 008 0-85 1-21
J. squarrosus 1-62 0-08 1-46 1-77 1-42 007 1-27 1-57
Koeleria cristata 094 005 0-84 1 04 094 005 0-84 1 04
Lathyrusmontanus 0-49 0-06 0-37 0-61 0-46 0-06 0-34 0-58
L.pratensis 073 010 051 095 073 010 051 095
Lemnaminor 2-45 008 229 2-61 206 044 1-62 2-50
Leontodonhispidus 0-89 007 0-75 1-03 0-89 007 0-75 1-03
Loliumperenne 1.30 006 1-17 1-43 1-30 006 1-17 1-43
Lotus corniculatus 1.05 006 0-92 1.19 1 05 006 0-92 1.19
Luzulacampestris 1.01 006 0-88 1-14 1.01 006 0-88 1-14
Lycopersiconesculentum 1-64 0-15 1-33 1-95 1-02 007 0-87 1-16
Matricariamatricarioides 1.17 0.10 0-95 1-39 1-17 0.10 0-95 1-39
Medicagolupulina 1'14 0-06 1.01 1-27 1-14 0-06 1.01 1-27
Melica nutans 099 005 0-88 1.10 099 005 0-88 1.10
Menthaaquatica 1.30 008 1-13 1-46 1.30 008 1-13 1-46
Miliumeffusum 1 11 007 0-96 1-27 1.11 007 0-96 1-27
Myosottssylvatica 1-15 0-06 1-03 1-28 1-15 0-06 1-03 1-28
Nardusstricta 0-71 0-07 0-55 0-86 0-71 0-07 0-55 0-86
Origanumvulgare 1-46 0-10 1-25 1-68 1-46 0 10 1-25 1-68
Oxalis acetosella 0-51 0-04 0-42 0-60 0-51 0-04 0-42 0-60
Phalarisarundinacea 1-24 0-12 098 1-50 1 24 0-12 0-98 1-50
Picea abies 0-42 0.09 0-23 0-60 0-33 0-08 0-16 0-50
P. nigra ssp. laricio 0-33 0-05 0-21 0-44 0-29 0-05 0-18 0-40
P. sitchensis 0-22 0-05 0.09 0-35 0-22 0-05 0.09 0-35
P. sylvestris 0-36 0-04 0-28 0-44 0-36 0-04 0-28 0-44
Plantagolanceolata 1-70 0.10 1-50 1.90 1-40 008 1-24 1-56
P. major 1-61 009 1-42 1-80 1-61 009 1-42 1-80
Poa annua 2-70 0-20 2-28 3-11 1 74 0.09 1-56 1-92
P. pratensis 1-26 007 1-12 1-40 1-26 007 1-12 1-40
P. trivialis 1 40 0-12 1-15 1-66 1-40 0-12 1-15 1-66
Polygonumaviculare 1-43 006 1-32 1-55 1-43 006 1-32 1-55
P. convolvulus 1-92 0-26 1-37 248 1-35 008 1-18 1-52
P. persicaria 1-29 006 1-17 1-42 1-29 006 1-17 1-42
Potentillaerecta 0-83 0.11 0-59 1-06 0-83 0'11 0-59 1-06
Poteriumsanguisorba 1-35 0-18 0-97 1-72 0-92 0-06 0-80 1-04
Prunellavulgaris 0-86 0-07 0-72 1.00 0-86 0-07 0-72 1'00
Ranunculusrepens 1-39 0-16 1-05 1-74 0-93 0-06 0-81 1-04
Rorippanasturtium-
aquaticum 1-05 006 0.91 1'19 1-05 006 091 119
Rumexacetosa 1-71 0-15 1-39 204 1-36 005 1 26 1 47
R. acetosella 1-55 008 1-39 1-72 1-55 008 1-39 1-72
R. obtusifolius 1-49 0-14 1-20 1-78 1-49 0-14 1-20 1-78
Salix cinereassp.
atrocinerea 1-06 009 0-88 1-25 1-06 009 088 1-25
Saniculaeuropaea 0-77 0-13 0-50 1-05 0-47 0-05 0-38 0-57
Sarothamnusscoparius 0-63 005 0-52 073 063 005 0-52 073
Scabiosacolumbaria 1-26 0-07 1.11 1 42 1-13 0-06 1 00 1 26
Sedumacre 0-71 0-08 0-54 0-88 0-71 0-08 0-54 0-88
Seneciojacobaea 1-24 0-06 1-12 1-38 1-24 0-06 1-12 1-38
S. squalidus 228 008 2-12 2-45 1-40 009 1-21 1-59
S. vulgaris 1-63 0-17 1-27 1-98 0-84 006 0-72 0-97
Sesleriaalbicans 0-75 0-05 0-65 0-86 0-75 0-05 0-65 0-86
Sieglingiadecumbens 0-60 006 046 073 060 006 046 073

Table 5 (contd)
Rmax R
Silene dioica 2-27 0-15 1-98 2-56 1-32 0-07 1-18 1 46
Stellariamedia 2-43 0.11 2-20 2-66 2-09 0'10 1-88 2-31
Succisapratensis 0-76 0-07 0-61 0.91 0-76 0-07 0-61 0.91
Taraxacumofficinale 1.19 0-06 1-07 1-32 1.19 0-06 1-07 1-32
Teucriumscorodonia 1 08 0'09 0-88 1-29 1 03 0-08 0-85 1-21
Thymusdrucei 0-72 0'09 0-53 0-92 0-72 0.09 0-53 0-92
Trifoliummedium 0-82 0-04 0-74 0'90 0-82 0-04 0-74 0'90
T. repens 1-26 0-08 1 10 1-42 1 26 0-08 1 10 1-42
Tussilagofarfara 1-32 0-08 1-15 1-49 1-32 0-08 1-15 1-49
Ulex europaeus 0-74 0-06 0-61 0-87 0-74 0-06 0'61 0-87
Urticadioica 2-35 0-06 2-22 2-48 2-20 0-05 2-10 2-31
Vacciniummyrtillus 0-52 0-08 0-35 0-70 0-52 0-08 0-35 0-70
V. vitis-idaea 0-23 0-07 0.10 0-37 0-23 0-07 0'10 0-37
Veronicaarvensis 1-18 0-13 0'90 1-45 1-18 0-13 0'90 1 45
Violariviniana 0-65 0-07 0-50 0-79 0-65 0-07 0-50 0-79
Zernaerecta* 1 04 0-04 0-96 1.11 1-04 0-04 0-96 1.11
* Includingdata suppliedby R. Law.
RESULTS
Values for Rmaxand R (calculated from the linear regression of loge W on time) are given
in Table 5.
Standard errors are included to provide a simple index of the variability of these
estimates for each species and 95%/confidence limits allow individual tests of significance
to be made.
Distributions of Rmaxwithin the present sample of species are given in Fig. 2 as histo-
grams with a class-interval of 0-25 week-'. The range of values, extending from Picea
sitchensis to Poa annua, represents more than a twelve-fold difference in values of Rmax.
The overall distribution (Fig. 2(a)) is single-humped, but the right-hand tail is slightly
more extensive than that on the left. Errors in the estimation of Rmaxwere not taken into
account in preparingthese histograms and some of the higher values had wide confidence
limits. Hence this right-hand tail may not be genuine and it would be unwise to speculate
on the slight skewness of this distribution.
Little difference emerged between Dicotyledones and Monocotyledones with respect
to the distribution of Rax. Both groups showed a distribution which was strongly
single-humped (Fig. 2(b) and (c)) and which extended over a wide range. The very
lowest classes in Fig. 2(b) are occupied by woody species (q.v.), so there is no real differ-
ence between the lower limits of the distributions shown in Fig. 2(b) and (c).
All fourteen of the woody species had low values of Rax (Fig. 2(d)). Deciduous and
coniferous tree species, shrubs and undershrubs, e.g. Dryas octopetala, came into this
category. This result agrees well with that of Jarvis & Jarvis (1964) who collected infor-
mation on R for eleven woody species and reported values equivalent to 0-053-0-822
week- for growth in the seedling phase.
Fig. 2(e) shows the frequency-distribution for species which are normally annual in
habit. The distribution is strongly biased towards the right, completely absent from the
bottom three class-intervals and virtually absent from the fourth (Aira praecox only).
All of the legumes studied had o1w to medium values of Rax (Fig. 2(f)). This group
included woody shrubs of low Rax (Sarothamnusscoparius, Ulex europaeus),grassland

30 (
(a) (b)
20
I0-
30
(d) (e) Lf)
20
10-
0 I 2 3 0 I 2 3 0 I 2 3
Rma (week-')
FIG. 2. The frequencydistributionof Rmaxin (a) the whole sampleand (hatchedarea),(b)

Dicotyledones,(c) Monocotyledones,(d) woody species, (e) annual species, (f) Papiliona-
ceae (legumes).Class-intervalsare 0-25 week-1.

species of low Rmax(Lathyrus spp.) and grassland species of low to moderate Rmax
(Lotus corniculatus,Medicago lupulina, Trifolium spp.). In view of the small size of this
sample it would be unwise to attach a general significance to the distribution of Rmax
within legumes.
No significant differences in Rmax were revealed within the three species for which
more than one seed source was used (Brachypodiumpinnatum, Festuca ovina, Zerna
erecta, see Table 1 footnote). For these species the values listed in Table 5 are derived
from pooled data of all populations used.
DISCUSSION
Accuracy of the data
Before considering the physiological and ecological significance of the data it is neces-
sary to comment on the reliability of the estimates of Rmaxobtained. Two possible sources
of error can be identified. The first is inaccurate curve-fitting owing either to variability
in the data or to methodological limitations and the second is restriction of growth by the
experimental conditions.
Curve-fitting
Some insight into the variability of the data was obtained by examination of the fitted
estimates of final whole-plant dry weight (W): the variability of values obtained showed
no systematic trend with absolute values when plotted on a logarithmic scale. In other
words, the transformation of all values of W to natural logarithms before analysis had
removed the systematic increases in variability which occurred with increases in W,
leaving only the contributions of random and experimental errors. Certain species also
had relatively variable final yields where there had been seedling mortality during the
experiment, viz. Digitalis purpurea, Geum urbanum,Origanumvulgare, Scabiosa colum-
baria and Veronicaarvensis.
The variability of the estimates of Rmaxgiven in Table 5 increases with the absolute
value of this function. This is an inevitable consequence of the quantitative analysis
adopted, since the highest values of R are more likely to be obtained from the early parts
of log.-curvilinear regressions than from the 'overall' log.-linear model. The estimation
of slopes from curvilinear regressions becomes progressively less certain as the order of
polynomial increases (see Hunt & Parsons (1974), cf. their Figs 1(c) and 2(c)). It is in the
elimination of non-significant and unnecessary high-order terms in the regressions of
loge W and logiLA on time that Hunt & Parsons' analysis represents an improvement on
that of Hughes & Freeman (1967). Nevertheless, comparisons made across a large body
of data such as this, depending as they do on a mixture of analytical methods, face new
problems of heterogeneity of variance in the fitted growth-functions. It is considered
that this is a small price to pay for the more realistic description of the trends in each of
the sets of data that this analytical procedure provides.
Many of the values of Rmax in Table 5 are high in comparison with other values
reported in the literature. For example, Evans (1972) reported '[the dry weight of]
seedlings of Epilobium hirsutum increased... 85-fold in 15 days, corresponding to a
mean RGR over the period of 2-07 [week-l]' (work of Shamsi 1970). This value was
considered by Evans to be among the highest values of R ever recorded and yet, in the
present investigation, this value is exceeded by Rmx in nine different species. Variations
in environment apart, many of these differences are due largely to the present methods

of estimating R. The values of Rmaxreported here are instantaneous values and are not
necessarily sustained for any substantial period of time. Hence, although many of these
values are high, they give way to more moderate values when calculated over a more
extended period. Even so, Shamsi's result for E. hirsutumis still exceeded by R in two
species here (Stellaria media and Urtica dioica), although our value of R for Epilobium
hirsutumitself is lower (1-83 week- ). The size of the values obtained thus suggests that
the experimental conditions were suitable for near-maximal growth-rates in the more
productive species even if only for a relatively brief period.
Elkington (1971), in presenting physiological data for a Biological Flora account on
Dryas octopetala, quoted results obtained for the growth of this species in the present
investigation. A value of R covering the whole period of observations (14-35 days after
planting) and equivalent to 0-58 week-l was reported. It was compared with values for
Poa annua and Urtica dioica respectively equivalent to 1-90 and 2-27 week-l, again
calculated from the present data over the whole period of growth. The values given by
Elkington (1971) for these two latter species were based on preliminary analyses only
and should be replaced by the values listed in the present Table 5.
Restriction of growth by externalfactors

In certain species growth curves flattened considerably towards the end of the experi-
ment. Where this coincided with the onset of flowering, e.g. in Bidens tripartita, Senecio
vulgaris,Polygonumconvolvulusand Chenopodiumalbum,the effect appearedto be a normal
feature of development. In other species, however, this phenomenon was most probably
due to external constraints. These were apparent in species such as Poteriumsanguisorba,
Plantago major and Centaureanigra, each of which ultimately formed a rosette of large
leaves which tended to experience mutual shading within the container. A decline in
relative growth-rate occurred in many of the very highest-yielding species, e.g. Hordeum
murinumand Lycopersicon esculentum.In these species the growth form was such that
mutual shading could probably be discounted. A more likely explanation is that the roots
became 'pot-bound'.
All of these cases exhibited significant log.-curvilinear progressions of W on time and
values of Rmx were taken from the early parts of the curves, before flattening had taken
place. However, more recent experiments indicate that some of the most rapidly-growing
species achieve Rmaxprior to the 2-5 week period covered by this investigation. The over-
all spread of final yields obtained in these experiments has, therefore, suffered a check or
truncation at the higher end of the scale. This illustrates one of the natural disadvantages
of the present 'broad-front' style of investigation. To have increased the number of
harvests and the space available to each seedling would have reduced substantially the
number of species which it was possible to include in the investigation. Despite these
sources of error and for the reasons discussed on pp. 395-7 it seems reasonable to con-
clude that broad comparisons of R are possible from the present body of data. It is neces-
sary, however, to bear in mind that the values of Rmaxpresentedhere are first estimates and
particularly in the case of large, fast-growing species, are likely to fall marginally below
the true maxima.
Thephysiological basis for differencesin Rmax

Rma and 'woodiness'
In Fig. 2(d) it was clearly established that seedlings of the tree species included in the
investigation yielded consistently low values of Rma. Comparatively slow growth in tree
c

seedlings has been recorded previously (Jarvis & Jarvis 1964; Loach 1970) and has been
attributed to the expenditure of photosynthate on woody tissue, a process concomitant
with a slow rate of expansion of leaf area. It is tempting to apply the same explanation
also to the slow growth-rates of the six undershrubs examined (Calluna vulgaris, Helian-
themum chamaecistus, Dryas octopetala, Thymus drucei, Vaccinium myrtillus and V.
vitis-idaea). However, it remains to be confirmed that a significantproportion of the dry
matter of these species is, in fact, devoted to woody tissue during early seedling develop-
ment. Moreover, since all these species may be broadly described as microphyllous other
explanations for the low rates of dry matter production may be involved.
Rmaxandplant morphology
Several plant attributes are known to exercise a controlling effect on relative
growth-rate and, in a subsequent paper, techniques of growth-analysis will be used to
assess the relative importance of leafiness and leaf efficiency in determining Rmx in each
species. However, as a preliminary to formal analysis some general points can be made
with regardto the relationships between Rax and plant morphology. With few exceptions
the herbaceousspecies of low Rmax(i.e. < 1 0 week- 1) are small in stature both as seedlings
and as mature plants. Among the slow-growing dicotyledons the majority of the species
have small leaves, often in low rosettes, e.g. Viola rivinianaand Campanularotundifolia,or
on creeping shoots, e.g. Lotus corniculatusand Prunella vulgaris.A further observation is
that many of the slower growing species, e.g. Lotus corniculatus,are those which as estab-
lished plants in the field are known to develop very long, often swollen, tap-root systems
(see, for example, those illustrated by Salisbury (1952)). The grasses of low Rmaxinclude a
high proportion of small tussock species. With the exception of Festuca rubra, all of the
narrow-leaved grasses examined (Nardus stricta, Festuca ovina, Deschampsiaflexuosa,
Aira praecox and Koeleria cristata) had low values of Rmx.
The slow-growing species include two which are exceptional in that they attain a large
stature at maturity. These are the Umbellifers Heracleum sphondyliumand Anthriscus
sylvestris. In these plants, low Rmaxappears to be due to the translocation of photo-
synthate into a swollen rootstock, a process which, perhaps surprisingly, appears to
commence with the appearance of the first leaf. The ecological significance of this pheno-
menon will be considered later.
The group of species which achieves values of Rmaxof 1-5 week- or higher is hetero-
geneous with respect to seedling morphology, leaf shape and the form of the mature
plant. The grasses are invariably broad-leaved but both erect forms (e.g. Holcus lanatus
and Bromus sterilis) and low-growing forms (e.g. Cynosurus cristatus and Agrostis
stolonifera) are represented. The fast-growing Dicotyledones include some of the tallest
species (e.g. Epilobiumhirsutumand Urtica dioica) and the shortest (e.g. Plantago major
and Cerastium holosteoides). Leaf form varies from large, simple leaves (e.g. Rumex
obtusifolius)to tripinnate structures (e.g. Achillea millefolium).
The ecological significance of differencesin Rmax
Extrapolations to field conditions
The estimations of Rmaxin this investigation were carried out on seedlings growing
under conditions far removed from those obtaining in the natural habitats of the species
concerned. It would be naive, therefore, to expect that the values measured are a reliable
guide to the rates of dry matter production commonly achieved in nature. Even where
seedlings are growing under productive conditions and in the absence of competition

Table 6. Themeanfrequency of occurrenceof species belongingtofour classes

of Rma, in habitats 3-31 (Table 1)
For each habitatand classof Rmaxthe upperfigurerefersto the meanpercentageoccurrenceof
thesespeciesin quadratsof the habitat.The lowerfigureis the sameafterangulartransformation
and with 95% confidencelimits added.
Habitat Rmaxclass (week-1)
< 1-0 1-0-1-24 1-25-1-44 > 144
3. Unshadedmire 4-77 7-87 8-85 700
11-92 14-88 15-33 13-79
+ 263 3-92 + 3-66 + 474
4. Shadedmire 4-33 8-44 12-24 10-63
10-79 15-27 18-56 16-66
+3-64 +6-33 +5-11 +?904
5. Rock outcrops 6-45 12-11 7-92 6-73
13-28 18-53 14-60 13-59
+ 3-19 + 488 + 3-28 3-98
6. Limestonescree 18-86 26-55 16-50 5-20
23-96 28-54 22-06 13-02
+6-47 + 1149 + 10-50 2-82
7. Cliffs 5-13 825 560 407
11-65 14-99 12-14 10-70
+ 359 +523 +3-21 +2-83
8. Walls 1-83 5-18 6-80 4-11
7-42 10-91 12-78 10-31
+1-54 +4-14 +3-69 +2-91
9. Arable 2-43 11-00 18-55 19-69
8-44 15-54 22-83 23-54
+2-99 +8-61 +7-02 ?9-80
10. Meadows 44-40 34-33 43-69 59-00
41-78 34-22 40-61 50-42
+11-67 +20-14 +10-81 +32-73
11. Enclosedpastures 7-27 15-36 28-86 29-43
14-12 21-31 30-31 31-33
+4-23 +6-31 + 758 + 14-10
12. Limestonepastures 21-88 22-94 18-83 16-75
25-79 25-99 23-88 21-34
+ 5-83 + 8-46 + 5-06 11-48
13. Pastureson acidic strata 22-64 17-20 10-31 2-83
25-25 21-48 15-75 9-30
+15-10 20-65 +7-31 314
14. Soil heaps 6-17 9-88 21-13 17-41
13-23 15-27 25-35 22-02
+6-80 6-24 + 554 6-83
15. Coal mine heaps 6-38 6-50 14-30 7-80
12-82 13-78 20-35 13-35
+ 6-77 + 3-66 + 499 + 7-71
16. Lead mine heaps 11-65 23-00 11-89 2029
17-71 27-59 17-74 24-10
+ 545 7-91 + 546 + 14-78
17. Cinders 5-22 9-88 18-16 13-06
12-25 16-29 2350 19-16
+ 424 + 499 + 5-25 + 5-69

Table 6 (contd)
Habitat Rmaxclass (week-1)
<1-0 1 0-1 24 1-25-1-44 > 144
18. Bricks and mortar 4-00 14-33 20-38 23-00
11-30 19-63 24-57 26-81
+6-88 +9-83 +?558 +8-13
19. Manureand sewage waste - 1343 17-78 22-85
20-16 23-65 26-67
+ 854 + 509 ? 820
20. Limestonequarryheaps 10-54 18-88 20-65 12-54
16-76 23-81 24-76 18-77
+6-40 + 683 ? 644 + 655
21. Quarryheaps on acidic strata 15-20 550 1000 -
21-40 13-55 18-27
?13-36 +8-26 8-48
22. Scrub 8-07 2-62 8-21 5-25
14-73 8-78 15-08 11-71
+ 429 + 197 + 3-70 + 5-73
23. Hedgerows 33-33 17-00 27-00 17-14
33-33 22-40 29-66 22-50
+4411 + 16-51 +?981 + 1312
24. Limestonewoodlands 12-22 5-10 11-00 700
18-73 11-97 16-68 13-36
+7-41 +?415 + 656 + 1090
25. Woodlandson acidic strata 8-60 2-73 6-82 6-20
15-21 9-03 13-02 13-66
+6-29 +2-10 +4-41 +6-67
26. Broad-leavedplantations 5-20 2-11 7-50 5-71
10-96 7-98 14-46 12-76
+ 600 + 200 + 420 5-49
27. Coniferousplantations 6-00 2-14 685 440
11-74 7-90 13-12 11-14
+854 +2-86 +5-13 +6-72
28. Verges 7-59 15-93 21-14 14-27
13-83 20-48 25-28 20-47
+4-69 +822 + 596 ? 657
29. Paths 3-77 985 17-14 17-36
10-87 16-46 22-61 20-80
+ 170 + 550 + 5-01 + 10-54
30. Wastelandand heath on limestone 10-60 17-50 12-56 11-36
17-18 21-79 18-20 17-55
+3-77 +7-10 ?4-71 +6-90
31. Wastelandand heath on acidic strata 7-00 455 9.19 460
12-40 10-56 15-86 11-17
+5-31 +3-20 +3-47 +3-20

such as in the early stages of recolonization of disturbed ground it is necessary to recog-

nize that plant characteristics apart from Rmaxwill exert a major influence upon dry
matter production. In the early seedling phase the most important of these is the initial
capital for growth provided by the seed reserves. The relationships between Rmax,seed
weight and yield after five weeks' growth in the present experiments will be examined in
a later publication, but it is already clear that high yields may coincide either with a
combination of high seed weight and high Rmax(e.g. Hordeummurinumand Polygonum
convolvulus)or with that of small seed weight and exceptionally high Rmax (e.g. Urtica
dioica and Stellaria media).
The indicatorvalue of Rax

It has been acknowledged already that the intrinsic and extrinsic constraints on R in
plants growing under natural conditions are such that it is pointless to extrapolate to the
field in absolute terms from laboratory determinations of Rax. However, the possibility
remains that values of Rm,, are a clue to the relative productivity of species growing under
field conditions. Such a relationship is suggested by the fact that species associated with
fertile habitats such as arable fields (e.g. Chenopodiumalbum, Poa annua and Polygonum
spp.) or manure heaps (e.g. Stellaria media and Urtica dioica) have higher values of Rmax
than species characteristic of unproductive habitats such as unfertilized pastures (e.g.
Festuca ovina, Nardus stricta, Sieglingia decumbens,Helictotrichonpratense and Sesleria
albicans). Correlations between potential growth-rate and site fertility have been com-
mented upon or are evident in a number of publications (e.g. Bradshaw et al. 1964;
Grime 1966). The data collected in the present investigation provide an opportunity to
test the strength of this correlation.
In Table 6 species have been classified according to Rax into four equally-populated
class-intervals and the mean frequency of occurrence of the species in each class has been
calculated for twenty-nine of the thirty-one habitats represented in Table 1. Angular
transformations have been added to permit statistical tests. A significant reduction
(P<0-05) in the frequency of slow-growing species (i.e. < 10 week-1) was detected in a
number of productive habitats. These include various types of spoil, enclosed pasture
and arable land. It is apparent also that species of low Rmaxattain only low frequency
on paths. From Table 6, it is clear that there is a statistically-significantreduction in the
frequency of species of high Rma (i.e. > 15 week- ) in the flora of two unproductive
habitats, limestone scree and unenclosed pastures on acidic strata. Fig. 3 allows a closer
inspection of the nineteen habitats in which there is evidence of a skewed distribution in
Rmax.
In exploring the correlation between site fertility and Rmaxa complication must be
recognized. This arises from the fact that in many of the fertile habitats the preponderance
of fast-growing species also coincides with physical disturbance and the presence of a
high proportion of annual plants which, it has already been noted (Fig. 2), include
many species of high Rmx. In the histograms of Fig. 3 the contribution of annuals to each
class of Rax in selected habitats has been distinguished by hatching. These figures show
clearlythat in each of the habitats in which there is a bias among the commoner constituent
species towards high Rmax(Habitats 9, 10, 11, 14, 15, 17, 18, 19, 23, 29 and 31) the in-
crease in the contribution of annuals, where it occurs, is generally of secondary importance
to the increase in abundance of perennial plants of high Rmax.
From Table 6 and Fig. 3, therefore, there is confirmation that productive and un-
productive vegetation types differ with respect to the potential maximum growth-rates

(6) (9) (10)
I0
20 (I )2) (13)
10-
(14) (15) j (16)
t (17) (18) (19)
(20) (21) (23)

10 II^ 7
,,10,,,, I DI i w
A B C D R A B C D R
(31)
FIG. 3. The distribution of Rmax between four equally-populated class-intervals plotted

for nineteenof the thirty-onehabitatsgiven in Table 1. All specieswhich occurredin 10%
or more of the quadrats are included. The class-intervals used are A, < 1-0; B, 1-0-124;
C, 1 25-1 -44;D, > 1-44week-'1.ColumnR shows specieswhichweresimilarlyabundantin
the habitat but which were not screenedfor Rma,,.Throughout,the hatchedareasreferto
annual species.The system of numberingof the habitatsfollows that used in Table 1:
6, Limestonescree;9, Arable; 10, Meadows; 11, Enclosed pastures;12, Limestonepas-
tures; 13, Pastures on acidic strata; 14, Soil heaps; 15, Coal mine heaps; 16, Lead
mine heaps; 17, Cinders; 18, Bricks and mortar; 19, Manure and sewage waste; 20,
Limestonequarry heaps; 21, Quarryheaps on acidic strata; 23, Hedgerows;28, Verges;
29, Paths; 30, Wasteland and heath on limestone; 31, Wasteland and heath on acidic
strata.

ANDRODERICK
J. P. GRIME HUNT 417
of the major constituent species. Moreover, account must be taken of the fact that the
habitat types recognized for the purposes of these comparisons are extremely broad and
are likely to be rather heterogeneous both in vegetation and in productivity. More
precise comparisons on the basis of individual vegetation samples and in which the
frequency of each species within the sample is used to calculate weighted means for
Rmax(Grime 1974) suggest that in natural vegetation there may be a sensitive adjustment
of general levels of Rmaxprevailing in a community in response to variation in site
fertility.
In order to explain the marked differences between common vegetation types with
respect to the potential growth-rates of the component species it is necessary to consider
the significance of Rmaxin relation to the very different processes which may determine
the success or failure of species in different types of habitat.
Earlier in this discussion it was pointed out that high Rmaxcoincides with a wide
range in morphology. It is apparent that at the two extremes of this morphological
range the capacity to achieve a high rate of dry matter production under favourable
circumstances is incorporated into distinct strategies. At one extreme high Rmaxis
associated with tall stature, the capacity for extensive lateral spread above and below
ground and the tendency to deposit a dense layer of litter on the ground surface. This
combination of attributes is conducive to the efficient capture of light, water, mineral
nutrients and space and is particularly well developed in species such as Epilobium
hirsutum, Chamaenerionangustifolium and Urtica dioica which frequently occupy ex-
tensive areas of productive, relatively undisturbed, vegetation apparently excluding the
majority of other plants by means of their superior competitive ability above and below
ground (but see Pigott (1971) for an alternative interpretation of the performance of
U. dioica under productive conditions).
A marked contrast to this 'competitive' strategy is provided, at the other extreme,
by annuals and short-lived perennials. Here, high Rm,, occurs in association with a
relatively short life-history and a growth pattern in which a large proportion of the
photosynthate is directed into seeds (Salisbury 1942; Harper 1961). In the ruderal, the
significance of high Rma appears to be rather different from that which it has in com-
petitive species. In the ruderal, high Rmaxfacilitates rapid completion of the life-cycle
rather than exclusive occupation of the habitat. It is likely, however, that in ruderals of
productive, arable land high Rmx will confer a considerable advantage in situations
where fertility allows rapid growth and the onset of competition occurs at an early stage
of colonization.
With respect to both the competitive and the ruderal strategy the significance of high
Rma may extend beyond the capacity for rapid dry matter production. It may be helpful
to recognize Rma as one of the more easily quantified expressions of a group of physio-
logical attributes which are under common genetic control or show parallel responses
to natural selection. Hence, it may be suggested that species of high Rmax possess other
attributes which place them at an advantage in productive or disturbed environments.
These may include, for example, great plasticity in deployment of photosynthate between
root and shoot and a rapid rate of response to environmental variation especially in the
extension growth of stems, petioles and roots, and in the expansion of leaf area.
The need to consider the concomitants of high Rmaxis even more obvious when an
attempt is made to explain the low incidence of potentially fast-growing species in
unproductive habitats. It is possible that genetic characteristicsconducive to rapid growth
in productive conditions become disadvantageous when the same plants are subjected

to environmentalextremes.The plant characteristics which cause speciesof high Rmax
to be vulnerableto environmental stressesare,in manycases,stillunidentified.Moreover,
there is a strong probabilitythat the pathologydiffersbetweenstressesand possibly
also betweenspeciesexperiencingthe samestress(Loach1970).
Among the documentedinvestigationsof the susceptibilityof fast-growingspeciesto
stressare the resultsobtainedby Grime(1965b, 1966)and Loach (1967)whichsuggest
that inflexiblyhigh respirationratesare implicatedin the rapiddeteriorationand mor-
tality of deeply-shadedseedlings.With respectto failureundercertainother forms of
stress, such as mineralnutrientdeficiencyand heavy metal toxicity, the investigations
whichhavebeencarriedout so far havenot identifiedthe featureswithinthe physiology
of fast-growingspecieswhichaccountfor theirgreatersusceptibility.
As a corollaryto thesecommentson the ecologyof fast-growingspecies,it is necessary
to referto the adaptivesignificanceof low valuesfor Rmax. The resultsof the present
investigationare consistentwith those of previousworkerswho have drawnattentionto
the fact that habitatssubjectto environmentalstress,in particularnutrientdeficiencies
(Kruckeburg1954; Beadle 1954; Bradshawet al. 1964; Hackett 1967; Clarkson1967;
Parsons 1968b;Higgs & James 1969)and severeshading(Grime 1965a;Loach 1970)
tendto be colonizedby speciesof low potentialgrowth-rate.
Thereareat leastthreewaysin whichinherentlylow ratesof growthcouldbe adaptive
to conditionsof stress.In the firstplace,slow-growingplantsmakemodestdemandsand
arethereforelesslikelyto exhausttheresourcesof theirimmediateenvironment.Secondly,
theremay be lower rates of incorporationof photosynthateand mineralnutrientsinto
structure,thus allowingthe build-upof reserveswithinthe plant. Thirdly,and perhaps
most important,underconditionsof protractedand extremestress,speciesof low Rmax
may be betterfittedto surviveperiodsin whichlittle or no growthis possible.Investiga-
tions into the physiology of shade-tolerantplants (Grime 1966; Hutchinson 1967;
Loach 1967; Mahmoud& Grime 1974)suggest that species of low Rmaxpossess two
characteristicslikely to preventrapiddeteriorationof plants in whichgrowthhas been
arrestedby some form of environmentalstress.The first is a low respiratoryrate and
the second is the extendedlife-spanof individualleavesandroots.
Therelativescarcityof speciesof low Rmax in productivevegetationsuggeststhat these
plants have a lower competitiveability. Slow-growingspecies are also infrequentin
habitatssuchas pathsin whichthe vegetationexperiencesa high intensityof damageor
disturbance.The explanationfor this findingappearsto be that species of low Rmax
are ill-adaptedeitherfor the replacementof foliageby regrowthof establishedplantsor
for rapidseedlingestablishmentin areasof baregroundcreatedby disturbance.
Special cases
The mainconcernin this paperhas been to examinethe generalrelationshipbetween
Rmax and plant distributionand it wouldbe unwise,particularlyin the absenceof more
extensiveevidenceon the lack of intraspecificvariationin Rm,, to regardthe single
determinationgiven here for each speciesas definitive.For this reasonno attempthas
generallybeen made to commentupon the significanceof differencesin Rmaxbetween
speciesof similardistribution.However,two particularcasesmeritfurtherinvestigation.
The first concerns the species Galium saxatile, Scabiosa columbaria and Poterium
sanguisorba,each of which has a moderatevalue of Rmax (in the range 1-3-1-6).These
species are all plants of relativelyunproductivevegetation and are to be found in
the field in associationwith speciesof ratherlow Rmax.
In this case it appearsthat the

relatively high values of the three species are due to the occurrence soon after germina-
tion of a very brief period of rapid growth which then gives way to rates more typical
of species from unproductive habitats.
A second anomaly is provided by the two biennial Umbellifers, Heracleumsphondylium
and Anthriscussylvestris. Despite their extremely low relative growth-rates both of these
species achieve high frequencies of occurrence in three of the habitats dominated by
species of high Rma (meadows, hedgerows, road verges). Earlier in this discussion it was
pointed out that the low relative growth-rate of the seedlings of these two species may
be due to the translocation of photosynthate into the rootstock, a process which begins at
a very early stage of seedling development and which provides the capital for production
of the flowering shoot in the second year. The remarkableability of the seedlings of these
two species to establish in vegetation composed of potentially fast-growing perennial
species and simultaneously to build up a root storage organ, doubtless at the expense of
leaf development, may in part be related to the relatively large seeds which confer an
impetus to seedling development which is unusual among native British herbaceous
plants.
ACKNOWLEGMENTS
The authors are grateful to Dr J. G. Hodgson and R. Law for permission to quote their
unpublished data and to R. A. Allen, Janet Buckhorn, A. V. Curtis, J. Hargreaves,
A. M. Neal and Carmen Rathey for technical assistance. Thanks are due to I. T. Parsons
of the University of Bristol Computer Centre for computing assistance and to Professor
E. W. Yemm for providing facilities for one of us (R.H.) in the Department of Botany
at the University of Bristol during part of the preparation of this paper. The work was
supported, in part, by the Natural Environment Research Council.
SUMMARY
Estimations have been made of the maximum potential relative growth-rate (Rma)
attained in the exponential phase by 132 species of flowering plants including representa-
tives from each of the dry terrestrialhabitats of the Sheffieldregion. The period of growth
between two and five weeks after germination was studied in a standardized, productive
environment and fitted growth-curves were used to derive various growth-analysis
parameters.
Woody species exhibited a bias towards low values of Rmaxand a similar trend was
evident among fine-leaved grasses. Annual plants were most frequent in the high Rmax
category. Grasses and forbs included a wide range of growth-rates and in both, high
values of Rmaxwere associated with a variety of growth forms. With the exception of the
woody plants and the biennial herbs, Anthriscussylvestris and Heracleum sphondylium,
all the species of low Rmaxexamined were species which as seedlings and mature plants
tend to be small in stature.
The possibility that Rmaxis of adaptive significancein the field was tested by examining
the frequency of species of low or high Rmaxin vegetation samples from a range of habitat
types. In several disturbed and/or productive habitats fast-growing species were pre-
dominant and species of low Rmaxwere virtually or completely absent. The reverse was
true of several stable, unproductive habitats. Species of moderate Rmaxwere ubiquitous.
The adaptive significance of Rmaxand its contribution to the determination of her-

baceous vegetation are explored by recognizing three primary strategies in herbaceous
plants. In the 'competitive' strategy, high Rmaxcoincides with tall stature, extensive
lateral spread and the tendency to accumulate leaf litter, all characteristicswhich facilitate
the exclusive occupation of productive, undisturbed habitats. The 'ruderal' strategy
also involves high Rma but here it is associated with a short life-history in which much
photosynthate is directed into seeds. The potential for rapid growth allows such plants
an opportunist exploitation of disturbed habitats. The third strategy, that of the 'stress-
tolerant' plant, is characterizedby a low potential relative growth-rate and small stature.
Maximum potential relative growth-rate thus appears to be of general significance
in the determination of vegetation composition, but in individual habitats this composi-
tion is also considered to be dependent upon additional plant characteristics.
REFERENCES
Ashton,D. H. (1958). The ecology of EucalyptusregnansF. Muell. The species and its frost resistance.
Aust. J. Bot. 6, 154-76.
Beadle,N. C. W. (1954). Soil phosphateand the delimitationof plant communitiesin easternAustralia.
Ecology, 35, 370-5.
Blackman,G. E. & Black, J. N. (1959). Physiologicaland ecological studies in the analysis of plant
environment.XI. A furtherassessmentof the influenceof shadingon the growthof differentspecies
in the vegetative phase. Ann. Bot. N.S. 23, 51-63.
Blackman,G. E. & Wilson, G. L. (1951a). Physiologicaland ecological studies in the analysis of plant
environment.VI. The constancy for differentspecies of a logarithmicrelationshipbetween net
assimilationrate and light intensityand its ecological significance. Ann. Bot. N.S. 15, 63-94.
Blackman,G. E. & Wilson, G. L. (1951b).Physiologicaland ecological studies in the analysis of plant
environment.VII. An analysisof the differentialeffectsof light intensityon the net assimilationrate,
leaf area ratio and relativegrowthrate of differentspecies. Ann.Bot. N.S. 15, 373-408.
Blackman,V. H. (1919). The compoundinterestlaw and plant growth. Ann.Bot. 33, 353-60.
Bradshaw,A. D., Chadwick,M. J., Jowett, D., Lodge, R. W. & Snaydon,R. W. (1960a). Experimental
investigationsinto the mineralnutritionof severalgrass species. III. Phosphatelevel. J. Ecol. 48,
631-7.
Bradshaw,A. D., Chadwick,M. J., Jowett,D. & Snaydon,R. W. (1964). Experimentalinvestigationsinto
the mineralnutritionof severalgrass species. IV. Nitrogen level. J. Ecol. 52, 665-76.
Bradshaw,A. D., Lodge, R. W., Jowett, D. & Chadwick,M. J. (1958). Experimentalinvestigationsinto
the mineralnutritionof severalgrass species. I. Calciumlevel. J. Ecol. 46, 749-57.
Bradshaw,A. D., Lodge,R. W., Jowett, D. & Chadwick,M. J. (1960b).Experimentalinvestigationsinto
the mineralnutritionof severalgrass species.II. Calciumand pH. J. Ecol. 48, 143-50.
Briggs,G. E. (1928). A considerationof some attemptsto analysegrowthcurves. Proc. R. Soc., B, 102,
280-5.
Clapham,A. R., Tutin, T. G. & Warburg,E. F. (1962). Flora of the BritishIsles, 2nd edn. Cambridge
UniversityPress, London.
Clarkson,D. T. (1965). Calciumuptake by calcicole and calcifuge species in the genus AgrostisL. J.
Ecol. 53, 427-35.
Clarkson,D. T. (1967). Phosphorussupplyand growthrate in speciesof AgrostisL. J. Ecol. 55, 111-8.
Dreissche, R. van den & Wareing,P. F. (1966). Dry matter production and photosynthesisin pine
seedlings. Ann. Bot. N.S. 30, 673-82.
Elkington,T. T. (1971). Biological Flora of the BritishIsles: Dryas octopetalaL. J. Ecol. 59, 887-905.
Evans, G. C. (1972). The Quantitative Analysis of Plant Growth. Blackwell Scientific Publications, Oxford.
Fisher,R. A. (1920). Some remarkson the methods formulatedin a recent article on 'The quantitative
analysisof plant growth'. Ann. appl.Biol. 7, 376-72.
Foulds, W. (1970). A comparison of the reaction to moisture stress in three species of legumes with particular
reference to cyanogenesis. M.Sc. thesis, University of Sheffield.
Grime,J. P. (1965a). Shade tolerancein floweringplants. Nature,Lond.208, 161-3.
Grime,J. P. (1965b).Comparativeexperimentsas a key to the ecology of floweringplants. Ecology,45,
513-5.
Grime,J. P. (1966). Shade avoidance and tolerancein floweringplants. Light as an EcologicalFactor
(Ed. by R. Bainbridge,G. C. Evans & O. Rackham),pp. 281-301. BlackwellScientificPublications,
Oxford.
GrimeJ. P. (1974).Vegetationclassificationby referenceto strategies. Nature,Lond.250, 26-31.

Grime,J. P. & Hodgson,J. C. (1969). An investigationof the ecologicalsignificanceof lime-chlorosisby

means of large-scalecomparativeexperiments.EcologicalAspectsof the MineralNutritionof Plants
(Ed. by I. H. Rorison), pp. 67-99. BlackwellScientificPublications,Oxford.
Grime,J. P. & Jeffrey,D. W. (1965). Seedlingestablishmentin verticalgradientsof sunlight. J. Ecol.
53, 621-42.
Grime,J. P. & Lloyd,P. S. (1973).An EcologicalAtlas of GrasslandPlants.EdwardArnold, London.
Hackett,C. (1967).Ecologicalaspectsof the nutritionof Deschampsiaflexuosa (L.) Trin.III. Investigation
of phosphorusrequirementand responseto aluminiumin water culture,and a study of growth in
soil. J. Ecol. 55, 831-40.
Harper,J. L. (1961).Approachesto the study of plant competition. Symp.Soc. exp. Biol. 15, 1-39.
Hewitt, E. J. (1966). Sand and WaterCultureMethodsused in the Studyof Plant Nutrition.Commonw.
agric. Bur. tech. Commun.No. 22 (revised2nd edn), FarnhamRoyal, Bucks., England.
Higgs, D. E. B. & James,D. B. (1969). Comparativestudieson the biology of uplandgrasses.I. Rate of
dry matterproductionand its control in four grass species. J. Ecol. 57, 553-63.
Hughes,A. P. & Evans, G. C. (1963). Plant growth and the aerial environment.IV. The effects of day-
length on Impatiensparviflora. New Phytol. 62, 367-88.
Hughes,A. P. & Freeman,P. R. (1967). Growthanalysisusing frequentsmall harvests. J. appl.Ecol. 4,
553-60.
Hunt, R. (1970). Relativegrowth-rate:its range and adaptivesignificancein a local flora. Ph.D. thesis,
Universityof Sheffield.
Hunt,R. & Burnett,J. A. (1973).The effectsof light intensityand externalpotassiumlevel on root/shoot
ratio and ratesof potassiumuptakein perennialryegrass(LoliumperenneL.). Ann.Bot. 37, 519-37.
Hunt, R. & Parsons,I. T. (1974). A computerprogramfor derivinggrowth-functionsin plant growth-
analysis. J. appl. Ecol. 11, 297-307.
Hutchinson,T. C. (1967). Comparativestudiesof the ability of speciesto withstandprolongedperiodsof
darkness. J. Ecol. 55, 291-9.
Hutchinson,T. C. (1970a). Lime chlorosis as a factor in seedling establishmenton calcareoussoils. II.
The developmentof leaf waterdeficitsin plants showinglime chlorosis. NewPhytol. 69, 143-57.
Hutchinson,T. C. (1970b).Lime chlorosis as a factor in seedlingestablishmenton calcareoussoils. III.
The ability of green and chloroticplants fully to reverselarge leaf water-deficits. New Phytol. 69,
261-8.
Jarvis,P. G. & Jarvis,M. S. (1964).Growthratesof woody plants. PhysiologiaPl. 17, 654-66.
Jefferies,R. L. & Willis, A. J. (1964). Studies on the calcicole-calcifugehabit. I. Methods of analysis of
soil and plant tissuesand some resultsof investigationson four species. J. Ecol. 52, 121-38.
Kruckeberg,A. R. (1954). The ecology of serpentinesoils. III. Plant species in relation to serpentine
soils. Ecology,35, 267-74.
Lloyd, P. S., Grime,J. P. & Rorison,I. H. (1971). The grasslandvegetation of the Sheffieldregion. I.
Generalfeatures. J. Ecol. 59, 863-86.
Loach, K. (1967). Shade tolerance in tree seedlings. I. Leaf photosynthesisand respirationin plants
raisedunder artificialshade. New Phytol. 66, 607-21.
Loach, K. (1970). Shade tolerancein tree seedlings.II. Growth analysis of plants raised under artificial
shade. New Phytol. 69, 273-86.
Lockhart,J. A. (1965). The analysis of interactionsof physical and chemical factors on plant growth.
A. Rev. Pl. Physiol. 16, 37-52.
Mahmoud,A. (1973). A laboratoryapproachto ecologicalstudiesof the grasses: Arrhenatherumelatius
(L.) Beauv.ex J. & C. Presl, Agrostistenuis Sibth.andFestuca ovina L. Ph.D. thesis, University of
Sheffield.
Mahmoud,A. & Grime,J. P. (1974). A comparisonof negativerelativegrowthrates in shadedseedlings.
New Phytol. 73, 1215-9.
Myerscough,P. J. &Whitehead,F. H. (1966).Comparativebiology of Tussilagofarfara L., Chamaenerion
angustifolium(L.) Scop., EpilobiummontanumL. and EpilobiumadenocaulonHausskn. I. General
biology and germination. New Phytol. 65, 192-210.
Myerscough,P. J. & Whitehead,F. H. (1967).Comparativebiology of TussilagofarfaraL., Chamaenerion
angustifolium(L.) Scop., EpilobiummontanumL. and EpilobiumadenocaulonHausskn. II. Growth
and ecology. New Phytol. 66, 785-823.
Njoku, E. (1959). An analysis of plant growthin some West Africanspecies. I. Growth in full daylight.
Jl W. Afr. Sci. Ass. 5, 37-56.
Parsons,R. F. (1968a). The significanceof growth-ratecomparisonsfor plant ecology. Am. Nat. 102,
595-7.
Parsons,R. F. (1968b).Ecologicalaspects of the growthand mineralnutritionof threemallee species of
Eucalyptus. OecologiaPlantarum,3, 121-36.
Pigott, C. D. (1971). Analysis of the response of Urticadioica to phosphate. New Phytol. 70, 953-66.
Pollard,D. F. W. & Wareing,P. F. (1968). Rates of dry matterproductionin forest tree seedlings. Ann.
Bot. 32, 573-91.

Radford,P. J. (1967). Growth analysisformulae-their use and abuse. Crop.Sci. 7, 171-5.
Rorison,I. H. (1960). The calcicole-calcifugeproblem. II. The effects of mineralnutrition on seedling
growthin solution culture. J. Ecol. 48, 679-88.
Rorison,I. H. (1964). A double shell plant growthcabinet. New Phytol. 63, 358-62.
Rorison,I. H. (1968). The responseto phosphorusof some ecologicallydistinctplant species. I. Growth
rates and phosphorusabsorption. New Phytol. 67, 913-23.
Salisbury,E. J. (1942). TheReproductiveCapacityof Plants. Bell, London.
Salisbury,E. J. (1952). Downsand Dunes: TheirPlantLife andits Environment. Bell, London.
Shamsi,S. R. A. (1970). ComparativebiologyofEpilobium hirsutumandLythrumsalicaria.Ph.D. thesis,
Universityof London.
Vries, D. A. de (1955). Solar radiationat Wageningen. Meded.LandbHoogesch.Wageningen,55, 277-
304.
Went, F. W. (1957). The experimentalcontrol of plant growth. Chronicabot. 17, 1-343.
(Received 6 June 1974)


RelativeGrowth Rate

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

RelativeGrowth Rate

Uploaded by

Copyright:

Available Formats

393

RELATIVE GROWTH-RATE: ITS RANGE AND ADAPTIVE

BY J. P. GRIME AND RODERICK HUNT

Unit of ComparativePlant Ecology (N.E.R.C.), Departmentof Botany,

Screeningfor relative growth-rate

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

MATERIALS AND METHODS

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

Table 2. The representationof species examined among those of common occurrence in

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

Table 3. The relationshipbetween temperatureand mean relativegrowth-rate

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

/ 500 cm3 sand

__ . / Glass-fibre cloth over

|[\~~~~ ~ \ Inverted plastic pot

FIG. 1. Vertical section through a seedling container.

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

Whole-plantsilhouettes. A visual record of the sizes of the various species at different

Spatial arrangementof containers

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

Table 4. Relative growth-rate calculatedfor Holcus lanatus by various

LogeW v. time, linear regression:

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

FIG. 2. The frequencydistributionof Rmaxin (a) the whole sampleand (hatchedarea),(b)

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

Restriction of growth by externalfactors

Thephysiological basis for differencesin Rmax

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

Table 6. Themeanfrequency of occurrenceof species belongingtofour classes

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

such as in the early stages of recolonization of disturbed ground it is necessary to recog-

The indicatorvalue of Rax

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

(14) (15) j (16)

t (17) (18) (19)

(20) (21) (23)

FIG. 3. The distribution of Rmax between four equally-populated class-intervals plotted

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

Grime,J. P. & Hodgson,J. C. (1969). An investigationof the ecologicalsignificanceof lime-chlorosisby

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM

(Received 6 June 1974)

This content downloaded from 69.167.65.194 on Mon, 2 Dec 2013 13:17:26 PM