Professional Documents
Culture Documents
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.
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
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 -
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.
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)
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).
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
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
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.
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
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
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
/ / /Supporting grille
_
CD 1CD
C Cf _ )O c_ =
I \//
Waterproof tray
To waste
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.
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
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.
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
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
11-20R 1-75
20-28R 1 59
28-35R 1.46
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.
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-')
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
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 (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
I0
20 (I )2) (13)
10-
,,10,,,, I DI i w
A B C D R A B C D R
(31)
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-
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.