Competitive ability of two grass species: Anthephora pubescens and Eragrostis curvula. 2. Growth analysis

S.Afr.J.Bot., 1994, 60(5): 269 - 275

269

Competitive ability of two grass species: Anthephora pubescens and Eragrostis curvula. 2. Growth analysis Jennifer E. Mynhardt* Roodeplaat Grassland Institute, Private Bag XOS, Lynn East, 0039, Republic of South Africa

M.W. van Rooyen and G.K. Theron Department of Botany, University of Pretoria, Pretoria, 0002, Republic of South Africa Received 16 February 1994; revised 30 June 1994

A replacement series evaluation of the degree of interspecific competition between Anthephora pubescens Nees and Eragrostis curvula (Schrad.) Nees indicated that the former was a poor and the latter a strong competitor. This paper examines the possible correlation between competitive ability and growth and morphological characteristics of these species. The higher leaf area index, leaf area duration, crop growth rate, larger number of lateral tillers and greater tuft height of E. curvula conferred a competitive advantage to that species. The order exhibited by the leaf area ratio and relative growth rate reversed the hierarchy of the replacement series, indicating a lack of correlation between these traits and competitive ability. 'n Vervangingsreeksevaluering van die mate van interspesifieke kompetisie tussen Anthephora pubescens Nees en Eragrostis curvula (Schrad.) Nees het getoon dat eersgenoemde die swakker en laasgenoemde die sterker kompeteerder is. In hierdie artikel word die moontlike korrelasie tussen kompeterende vermoe en groei- en morfologiese kenmerke van hierdie spesies ondersoek. Die hoar blaaroppervlakte-indeks, blaaroppervlakteduur, oesgroeitempo, groter getal laterale lote en hoar polhoogte het aan E. curvula 'n kompeterende voordeel gegee. Die rangorde van die blaaroppervlakteverhouding en die relatiewe groeitempo het die hiarargie van die vervangingsreeks omgekeer en sodoende die gebrek aan korrelasie tussen hierdie plantkenmerke en kompeterende vermoans aangedui. Keywords:

Crop growth rate, lateral tillers, leaf area index, leaf area ratio, relative growth rate, tuft height.

• To whom correspondence should be addressed at: Department of Water Affairs and Forestry, Private Bag X93, Pretoria, 0001, Republic of South Africa.

Introduction Anthephora pubescens Nees is a valuable cultivated pasture in the semi-arid regions of South Africa that receive an annual rainfall of between 250 and 650 mm (Roberts & Fourie 1975; Du Pisani et al. 1986; Fair 1989; Dannhauser 1991). The most outstanding attributes of the species are that it is droughttolerant, very palatable and nutritious and has an excellent "foggage value. Eragrostis curvula (Schrad.) Nees is one of the most useful pasture species because of its adaptability to a range of climatic and edaphic conditions in South Africa. Although E. curvula can be grown successfully in areas with an annual rainfall as low as 450 mm, it is generally recommended for areas with an annual rainfall exceeding 650 rum. Eragrostis curvula pastures are usually ready for grazing before most of the other grass species in spring. However, it has a relatively short summer grazing period, whereafter palatability and nutritional value decline in auturrm resulting in a very poor winter foggage value (Fair 1989). Although A. pubescens and E. curvula do not occur together under natural conditions, the possibility was investigated to grow these two species in mixed pastures, as E. curvula could provide early spring grazing and bulk in summer, while A . pubescens could provide good quality grazing in auturrm and winter. Growth analysis is the first step in the analysis of primary production and is a useful tool to determine net photosynthetic production (Kvet et al. 1971; Causton & Venus 1981; Hunt 1982, 1990). It has also been successfully applied to identify plant characteristics that influence the competitive ability of a species (Roush & Radosevich 1985). In a previous paper

(Mynhardt et al. 1994), the degree of intra- and interspecific competition occurring between A. pubescens and E. curvula was quantified. The objective of the present study was to evaluate the response of Anthephora pubescens and Eragrostis curvula to intra- and interspecific competition by means of growth analysis and to examine the possible correlation between competitive ability and the growth and morphological characteristics of these two species.

Methods The experiment was carried out in a greenhouse at the Roodeplaat Grassland Institute. Seeds of A. pubescens and E. curvula were sown in both pure and mixed stands in November 1990. In pure stands the planting densities were I, 4, 8, 12 and 16 plants per pot while in the mixed stands (replacement series), the total planting density was kept constant at 16 plants per pot, but the ratios of A. pubescf!ns to E. curvula were varied at 4:12, 8:8 and 12:4 plants per pot. At the end of each consecutive month, commencing January 1991 and terminating May 1991, six replicates per treatment were harvested to determine yield as well as biomass allocation of both species under different degrees of intra- and interspecific competition. A detailed description of the procedure at each harvest was given by Mynhardt (1992). The following growth characteristics were calculated: CGR, crop growth rate; LAR, leaf area ratio; LAI, leaf area index; LAD, leaf area duration; R, relative growth rate of the entire plant; Rio relative growth rate of tillers; Rio relative growth rate of the leaves; and R" relative growth rate of roots. The formulae used to calculate the various rates and ratios are based on those of Kvet et aI. (1971), Hunt (1978, 1982, 1990), Causton and Venus (1981), and Coombs et al. (1986). The morphological development was monitored at fortnightly

S.-Afr.Tydskr.Plantk., 1994, 60(5)

270 intervals, commencing January 1991 and terminating May 1991. At each monitoring date the following values were determined: total number of leaves on the primary tiller, average number of lateral tillers per plant, and average tuft height (mm). In the treatments where there were more than one plant per pot, pigeon rings were placed on the primary tiller of four plants of each species in the centre of each pot. A colour code was used to ensure that the same plant was measured at each monitoring date. The seedling shoot was counted as the primary tiller and secondary and subsequent tillers were counted as soon as they reached the two-leaf stage. Tuft height was measured to the highest living part. Results were analysed statistically using the Statgraphics 5.0 computer program (Statgraphics 5.0, 1989, STSC, Inc. USA) . A one-way analysis of variance (Scheffe) was performed to determine significant differences at IX = 0.05 .

Results and Discussion Due to the high mortality rate of A . pubescens under competitive stress, plants grown at densities of more than 4 plants per pot could only be harvested three times.

Growth analysis Intraspecific competition The relationship between the average number of leaves per plant and per pot, leaf area (LA) per plant, per pot and per leaf of A . pubescens and E. curvula at different densities at the third harvest, is given in Table 1. The average LA per plant decreased significantly with an increase in density. This decrease in LA per plant was not only due to a significant decrease in the number of leaves per plant but also to a significantly decreased leaf area per leaf. Intraspecific competition apparently had a greater negative effect on the LA per plant of E. curvula than A. pubescens, as increasing the density from 1 to 16 plants per pot decreased the leaf area per plant approximately 20-fold in E . curvula as against 13-fold in A. pubescens. Although the total number of leaves per pot tended to increase with increasing density in both species, the LA per pot only increased to a density of 8 and 4 plants per pot for A. pubescens and E. curvula, respectively. Monthly changes in LA for both species are illustrated in Figure 1. In the case of E. curvula, LA per pot at most

densities reached a maximum at the third harvest -- a density of 4 plants per pot producing maximum LA. Although LA per pot was significantly affected by density at the third harvest (Table I), density had no effect on LA per pot at later harvests. Results available for A. pubescens indicate the same trends. Absolute values of LA per pot of E. curvula exceeded those of A. pubescens. The relative growth rate (R) is often used to compare the performances of species or the effect of treatments on a species (Coombs et al. 1986). In both species and at most densities, R declined sharply between the second and third harvests, whereafter it remained fairly constant before showing a slight increase between the fourth and fifth harvests (Figure 2). The initial high R values were due primarily to tiller and leaf growth, while the increase in R towards the end of the season, especially in E. curvula, can be ascribed to an increased R r • The decrease in relative growth rate with time is ascribed to ontogenetic drift (Kvet et al. 1971; Evans 1972; Hunt 1978, 1982; Causton & Venus 1981) which causes a continuously decreasing efficiency of photosynthesis per unit biomass, as well as an increase in the cost to maintain a unit biomass (Cohen 1976). In both species the initial R-values were highest at the lowest densities, but after the second harvest, R was no longer affected by density. Colville and Marshall (1981), however, recorded a 50% reduction in R of Lolium perenne with increasing density. The maximum R-value (Rmax) of A. pubescens (0.679 g g-I week-I) was slightly higher than that of E. curvula (0.546 g g-I

450~------------------------------------~---'

a

400

350 300

'f

250

:5

200

S.

150 100 50 O~--~------.--------r-------'-------'r-~

2

Table 1 The average number of leaves per plant and per pot as well as the average leaf area per plant, per pot and per leaf of Anthephora pubescens and Eragrostis curvula grown at different densities in monoculture at the third harvest a Density (pl./pot)

Number of leaves Per plant

Per pot

Per pot

73.5 b 27.4 a 19.5 a 14.5 a 9.0 a

5

600 500

Per leaf

'f

400

S.

Anthephora pubescens

1 4 8 12 16

4

7001.----------------------------------------b

Leaf area (cm 2) Per plant

3 Harvest

73.5 a 109.4 ab 155.8 b 174.0 b 143.7 ab

227.8 b 77.2 a 53.0 a 33.9 a 20.5 a

277.8 a 308.8 a 424.2 a 406.7 a 328.5 a

3.8 2.8 2.7 2.3 2.3

b a a a a

:5

300 200 100 OL---.------.-------.------~------,_~

2

Eragrostis cUTvuia

1 4 8 12 16

223.5 c 91.8 b 63.7 ab 41.1 a 36.4 a

223.5 a 367.0 b 509.2 c 493.0 c 582.6 c

515.3 c 164.9 b 64.5 a 27.1 a 25.4 a

515.3 bc 659.4 c 516.2 bc 325.0 a 407.0 ab

2.3 c 1.8 b 1.0 a 0.7 a 0.7 a

• Values followed by the same letter do not differ significantly at IX= 0.05.

--- 1 plant per pot

3 Harvest

--+-- 4 plants per pot

4

-

5

8 plants per pot

---""- 12 plants per pot - - 16 plants per pot

Figure 1 Monthly changes in leaf area per pot of (a) Antheplwra pubescens and (b) Eragroslis curvula grown at different densities in monoculture.

271

S.Afr.J.Bot., 1994,60(5) week-I). In their study on 132 species of flowering plants, Grime and Hunt (1975) concluded that fine-leaved grass species exhibited a bias towards a low Rma. (less than 1 g g-l week-I), while grass species with Rma. > I g g-I week-I were invariably broad-leaved species. On the basis of their low Rm.. values, both species in the present study would be classified as 'stress-tolerant' rather than 'competitive' species (Grime & Hunt 1975; Grime 1979). The relative growth rate consists of two components (Coombs et al. 1986), one which measures the efficiency of the plant or crop as a producer of dry mass (nett assimilation rate), and the second the efficiency of the plant as a producer of leaf area (leaf area ratio). In a broad sense, leaf area ratio (LAR) represents the ratio of photosynthetic to respiratory material within the plant (Kvet et al. 1971). Density did not significantly affect the LAR of either species in a particular direction (Figure 3). In general, the monthly changes in LAR of both species indicated a decrease with time, although there was an initial increase at low densities (1 plant per pot) (Figure 3). As in the case of the relative growth rate, LAR is also subject to ontogenetic drift and is influenced by environmental factors (Warren Wilson 1967; Eagles & Ostgard 1971; Woodward 1979; Causton & Venus 1981). On the whole, the LAR values of A. pubescens were higher than the corresponding values of E. curvula. The high leaf content of the foggage (Fair 1989) and palatability of A. pubescens possibly reflect the high LAR values obtained in this species. To express the productivity of crops or natural ecosystems,

Interspecific competition The average number of leaves produced by A. pubescens and the average LA in a mixture were not significantly lower than those in a pure stand (Figure 5). The effect of interspecific competition on the average number of leaves and LA per plant of A. pubescens therefore did not differ significantly from the effect of intraspecific competition on these values. Eragrostis curvula, however, produced significantly more leaves per plant and had a concomitant greater LA in a mixture at a ratio of

0.1.,..----::---------------------,

45,--;c-------------------,

a

LAR is considered inappropriate and the leaf area index (LAI), that is, the leaf area per unit land area, is used (Coombs et aI. 1986). According to Kvet et al. (1971), leaf area index is the primary factor determining the rate of dry matter production in a closed stand. On the basis of the LAI values the performance per unit land area of E. curvula exceeded that of A. pubescens. The crop growth rate (CGR) serves as a simple index of agricultural productivity. In both species, CGR decreased initially, but increased again between the fourth and fifth harvests (Figure 4). As in the case of LAI, the CGR indicated thatE. curvula was more productive thanA. pubescens. Leaf area duration is a quantitative measure of the time that a plant maintains its assimilatory area and is the growth parameter best correlated with yield (Kvet et aI. 1971; Hunt 1978). In the case of E. curvula, the LAD value reached a maximum at a density of 4 plants per pot (51 826 cm2 day), whereafter it decreased at higher densities. The corresponding LAD value of A. pubescens was only 31 597 cm 2 day.

a

0.09 40 35

EJ 0.06

~ .s::

~

0, G>

.2

iii

0.05 0.04 0.03 20

0.02

£ O.O:I.t-______~~===~C------~ ~.01~--,-----.------r-----,--~

1 -2

2-3

3-4

15 101~---,-----.----.----.----~--

4-5

2

Harvest

4

3 Harvest

5

' 0.081- r - - - : ; ; ; - - - - - - - - - - - - - - - - - - - cb

~,-------------------------------------~b

0.07

40

"C

0.06

35

Cl

0.05

"j~

E! G>

~

! 0, ~ ~

Qj

a:

0.04 0.03 0.02 0.01

10

O+------------~~-------~

-0.01.J.....-_,-----.,------,-------.-2-3 3-4 4-5 1-2 Harvest --- 1 plant per pot ~

-+- 4 plants per pot

-

8 plants per pot

12 plants per pot - - 16 plants per pot

Figure 2 Monthly changes in relative growth rate of (a) Anthephora pubescens and (b) Eragrostis curvula grown at different densities in monoculture.

5

O.J.....-.-----.----.----.----.---2

- - 1 plant per pot

3 Harvest

-+- 4 plants per pot

4

-

5

8 plants per pot

-=- 12 plants per pot - - 16 plants per pot Figure 3 Monthly changes in leaf area ratio of (a) Anthephora pubescens and (b) Eragrostis curvula grown at different densities in monoculture.

S.-Afr.Tydskr.Plantk., 1994,60(5)

272

12 A. pubescens : 4 E. curvula than in a pure stand at the same density or in mixtures at other ratios. In the mixtures an increase in the ratio of E. curvula to A. pubescens resulted in a decrease in the average number of leaves per plant (Figure 5a) and the concomitant LA per plant (Figure 5b) of E. curvula. The presence of individuals of A. pubescens therefore had less effect on the growth of E. curvula than the presence of individuals of E. curvula. An increase in the ratio of A. pubescens in a mixture resulted in an increase in the LAR of A . pubescens in the mixture. The LAR of both species was lower in a mixture than in a pure stand (Figure 5). Morphological development Intraspecific competition The total number of leaves produced on the primary tiller as well as the pattern of leaf death remained unaffected by increasing density in both species. The first leaf death was recorded when the primary tiller had produced five or six leaves. As the appearance of a new leaf was accompanied by the death of the oldest leaf, the number of live leaves on the primary tiller remained more or less constant over the growing season. These results correspond to those of Colville and Marshall (1981) on Loliumperenne. The average number of lateral tillers per plant of A. pubescens sown at a density of 1 plant per pot, increased over the growing season until approximately 29 lateral tillers per plant were reached after 14 - 16 weeks of monitoring, whereafter the

number of lateral tillers decreased (Figure 6a). From the eighth week of monitoring the number of lateral tillers per plant at the higher densities was significantly lower than that of 1 plant per pot. In the case of E . curvula grown at a density of 1 plant per pot, the number of lateral tillers per plant increased to approximately 84 after 14 weeks of monitoring before declining (Figure 6). At higher densities the number of lateral tillers of E. curvula increased at a much lower rate over the growing season, reaching significantly lower maxima. In both species intraspecific competition limited the maximum number of

a C
C.

tv

a. If)

Q)

~ '0

tv

.0

E ::J

Z

8:8 12:4 Ratio A. pubescens : E. curvula

16:0

90~------------------------------~------------.

b 80

o.3,.-----------------------,a

~ 70 C.

0.25

{5 ~

tv

60

l'

50

a.

7>-

0.2

~ 40

0.15

~

.c ~

0.1 o C, ~ 0.05

~

30

~

20 10

U

O+-------~-----------~

0+=== 0:16

~ . 05~---,----------.-----------r---------.---~

1-2

3-4

2-3

4:12 8:8 12:4 Ratio A. pubescens : E. curvula

16:0

4-5

Harvest

80~---------------------.

0.9 , . - - - - - - - - - - - - - - - - - - - - - = : : - - - . , - - , b

O.B

b

C

70 60

0.7 ':">-

~ ~

0.6

'"

0.5

.c ~

0.4

~

e

0.3

20

Ol

g.

0.2

o

0.1

10

O+------------~+_------~

-0.1

~----r_--------___.,._----------__._----------r_-----'

1-2

3-4

2-3

4-5

~

~

4 plants per pot

4:12 8:8 12:4 Ratio A. pubescens : E. curvula

1_

Harvest --- 1 plant per pot

0+=== 0:16

-

A. pubescens _

16:0

E. curvula

8 plants per pot

12 plants per pot --><- 16 plants per pot

Figure 4 Monthly changes in crop growth rate of (a) Anthephora pubescens and (b) Eragrostis curvula grown at different densities in monoculture.

Figure 5 The (a) number of leaves per plant, (b) leaf area per plant and (c) leaf area ratio (UR) of Anthephora pubescens and Eragrostis curvula in a replacement series at the third harvest. Within a species, bars with the same letter do not differ significantly at ex = 0.05.

273

S.AfrJ.Bot.,1994,60(5) lateral tillers produced per plant, while the number of tillers per pot increased with an increase in density. Lateral tiller formation in E. curvula exceeded that in A. pubescens. According to Mitchell (1953) the number of tillers on plants of equal leaf appearance gives a direct comparison of the extent to which the potentialities for tiller formation have been utilized, i.e. whether lateral buds have developed or remained dormant. It is, therefore, an index of fresh active meristematic centres. Since each tiller is capable of rooting and assuming virtually independent existence, it is, within definite limits, an index of the rate of vegetative reproduction. Mitchell (1953) found that the quantity of light energy available appeared to be the chief determinant of the number of tillers produced in ryegrass. Reducing radiation levels tended to inhibit development from the basal nodes (Mitchell 1953). Increased shading due to high densities not only reduces the total irradiance, but the spectral energy distribution is altered by light interception by plant material. The average tuft height of A. pubescens and E. curvula grown at a density of 1 plant per pot increased to 328 and 668 mm, respectively, in the twelfth week of monitoring, whereafter it decreased (Figure 7). The decrease in tuft height was the result of tissue death which commenced at the tips of the plant parts. In the case of A. pubescens, increasing density caused no significant differences in tuft height at a particular monitoring date. In contrast to A. pubescens, a significant difference in the average tuft height of E. curvula, with increasing density, was exhibited between 1 plant per pot and

the higher densities from the second week of monitoring. It has been shown in many studies with grasses and cereals that density has a major effect on the growth and morphological development of the individual plant (Puckridge & Donald 1967; Kirby & Faris 1972; Kays & Harper 1974). In this study, plants grown at high densities in monocultures were generally characterized by the production of fewer tillers and in the case of E. curvula, decreased tuft height. These features closely correspond to those described by Kirby and Faris (1970, 1972) in barley grown at high densities. I nterspeciJic competition The effect of interspecific competition on the average number of lateral tillers per plant and the average tuft height is illustrated in Figure 8 for both species. Anthephora pubescens produced significantly more lateral tillers per plant and attained a greater tuft height in a pure stand than in a mixture. In the case of E. curvula, tiller formation and tuft height were not affected by the identity of the other plants in the pot. Eragrostis curvula, however, produced significantly more lateral tillers per plant than A. pubescens in every species combination and at each monitoring date.

Conclusions Competition evidently placed constraints on the growth of both species. An increase in the growth characteristics (number of leaves, LA and LAR) of A . pubescens in a monoculture compared to mixtures suggests that interspecific competition by E.

30,-------------·------------·------------------'

a

a

25 en

~

"" ~

300

20

E

15

E Ol 'w

.s ..c

.0

;::: 200

E :J

Z

250

:J

10

f-

150

5

100L-,----.----.----,----.----.---.----,----,~

2

4

6

8 10 12 Weeks of monitoring

14

16

2

18

~

80

650

70

600

E

60

.sE

"0 50

8 10 12 Weeks of monitoring

14

16

18

550 500

Ol

'w ..c 450

Q;

40 E :J Z 30

.0

;:::

:J

f-

400

20

350

10

300

o~,_--_,----,_--_.--_,----,_--_.--_,----,_-

2

4

___ 1 plant per pot ~

6

7001.-----------------------------------------~b'

901.---------------------------------------~b-

~

4

6

10 12 8 Weeks of monitoring -+-- 4 plants per pot

_0-

14

16

250~,_--_.----._--,_--_.--_,----,_--_.--_.--

2

18

8 plants per pot

12 plants per pot - - 16 plants per pot

Figure 6 Fortnightly changes in number of lateral tillers per plant of (a) Anthephora pubescens and (b) Eragrostis curvula grown at different densities in monoculture.

4

--- 1 plant per pot ~

6

8 10 12 Weeks of monitoring --+- 4 plants per pot

~

14

16

18

8 plants per pot

12 plants per pot - - 16 plants per pot

Figure 7 Fortnightly changes in tuft height of (a) Anthephora pubescens and (b) Eragrostis curvula grown at different densities in monoculture.

S.-Afr.Tydskr.P1antk., 1994, 60(5)

274

a a

C
a. Q; 0..

'"Q;

"" a Q; .Q

4

E ~

z

2 0 0:16

4:12

8:8

12:4

16:0

Ratio A. pubescens : E. curvula

b

E

.s 1:

Ol

'iii ..c

b

~ ~

I-

0:16

4:12

8:8

12:4

16:0

Ratio A. pubescens : E. curvula

1_

A. pubescens _

reversed the hierarchy that was established by the replacement series. In contrast to Grime, Tilman's theory (Tilman 1990) predicts that superior nutrient competitors should have relatively low maximal relative growth rates. PlaN populations react to intraspecific competition through either plasticity or mortality or both types of response (Harper 1977; Rai & Tripathi 1985). Only in the case of A. pubescens was mortality affected by higher densities, while both species showed a plastic response to density, which affected their morphological development. Tillering was restricted and as a result the primary tiller played a greater role in the overall carbon economy of the plant compared with its relative importance at low densities. Eragrostis curvula generally grew taller and produced more lateral tillers than A. pubescens. Tillering is a valuable mechanism to enable a crop to fully exploit its environment. The larger number of lateral tillers in E. curvula would therefore enable the species to utilize resources more fully. The greater height of E. curvula led to increased shading of A. pubescens later in the season causing E. curvula to gain resources at the expense of A . pubescens. Keddy (1990) found that simple traits such as biomass, plant height and canopy diameter could account for most of the measured competitive ability. A significant positive correlation between plant height and the position in the competitive hierarchy was also found in studies by Mitchley and Grubb (1986) and Keddy and Shipley (1989). Both the morphological characteristics examined in this study (tillering and tuft height) therefore conferred a competitive superiority to E. curvula.

E. curvula

Figure 8 The (a) number of tillers per plant and (b) tuft height of Anlhepfwra pubescens and Eragroslis curvula in a replacement series at the third harvest. Within a species, bars with the same letter do not differ significantly at (Y = 0.05 .

curvula had a greater effect on the growth of A. pubescens than intraspecific competition. In contrast, E . curvula grew better in a mixture than in a monoculture, suggesting that intraspecifi" competition was more detrimental than interspecific competition by A. pubescens. In a previous paper in which the results of the replacement series were analysed (Mynhardt et al. 1994), the values of the relative yield per plant and aggressivity indicated that A. pubescens was a poor competitor, while E. curvula was a strong competitor. Two of the reasons cited for the superior competitive ability of E . curvula were (a) that the high densities were beyond the limits of A. pubescens and (b) the fact that E. curvula seedlings emerged three to five days earlier than those of A. pubescens. The growth analysis could identify several further plant characteristics which could give E. curvula a competitive advantage. The higher leaf area index (LAI) , leaf area duration (LAD) and crop growth rate (CGR) values of E. curvula gave that species a greater carbon assimilation capacity. According to Grime (1979), one of the key characteristics of plants positively correlated with competitive ability is the maximum relative growth rate. The values of the relative growth rate (R) as well as leaf area ratio (LAR) seem to suggest the superiority of A. pubescens. However, other studies (Roush & Radosevich 1985; Van Rooyen 1988) indicated the lack of correlation between competitive ability and R and also demonstrated that the order exhibited by the LAR

Acknowledgements The authors thank Carel Moolman and Frans Mashabala for technical assistance as well as the Roodeplaat Grassland Institute for financial support and provision of facilities . References CAUSTON, D.R. & VENUS, J.e. 1981. The biometry of plant growth. Edward Arnold, London. COOMBS, J., HALL, D.o., LONG, S.P. & SCURLOCK, J.M.O. 1986. Techniques in bioproductivity and photosynthesis. Pergamon Press, Oxford. COHEN, D. 1976. The optimal timing of reproduction. Am. Nat. 11 0: 801 - 807. COLVILLE, K.E. & MARSHALL, e. 1981. The pattern of growth, assimilation of 14COZ and distribution of 14C-assimilate within vegetative plants of Lolium perenne at low and high density. Ann. appl. Bioi. 99: 179 - 190. DANNHAUSER, e.S. 1991. Die bestuur van aangeplante weiding in somerreenvaldele. Review Printer, Pietersburg. DU PISANI, L.G., VAN RENSBURG, W.L.J. & OPPERMAN, D.PJ. 1986. Influence of soil pH and fertilization on the dry matter production, chemical composition and digestibility of perennial grasses: 1. Anthephora pubescens Nees. J. Grassld. Soc. sth. Afr. 3: 52 - 55. EAGLES, e.P. & OSTGARD, O. 1971. Variation in growth and development in natural populations of Dactylis glomerata from Norway and Portugal. 1. Growth analysis. J. appl. Ecol. 8: 367 - 381. EV ANS, G.e. 1972. The quantitative analysis of plant growth. Blackwell Scientific Pub!., Oxford. PAIR, J. 1989. Guide to profitable pastures, 2nd edn. M. & J. Publications, Harrismith. GRIME, J.P. 1979. Plant strategies and vegetation processes. Wiley, London. GRIME, J.P. & HUNT, R. 1975. Relative growth rate: its range and adaptive significance in a local flora. J. Ecol. 63: 393 - 422. HARPER, J.L. 1977. PopUlation biology of plants. Academic Press, London.

S.AfrJ.Bot., 1994, 60(5) HUNT, R. 1978. Plant growth analysis. Arnold, London. HUNT, R 1982. Plant growth CUIVes. The functional approach to plant growth analysis. Edward Arnold, London. HUNT, R 1990. Basic Growth Analysis. Unwin Hyman, London. KA YS, S. & HARPER, J.L. 1974. The regulation of plant and tiller density in a grass sward. 1. Ecol. 58: 681 - 698. KEDDY, P.A. 1990. Competitive hierarchies and centrifugal organization in plant communities. In: Perspectives on Plant Competition, eds J.B. Grace & D. Tilman, pp. 265 - 290. Academic Press, San Diego. KEDDY, P.A. & SHIPLEY, B. 1989. Competitive hierarchies in plant communities. Oikos 49: 234 - 241. KIRBY, E.J.M. & FARIS, D.G. 1970. Plant population induced growth correlations in the barley plant main shoot and possible honnonal mechanisms. 1. expo Bot. 21: 787 - 798. KIRBY, E.J.M. & FARIS, D.G. 1972. The effect of plant density on tiller growth and morphology in barley. 1. agric. Sci. Camb. 78: 281 - 288. KVET, J., ONDOK, J.P., NECAS, J. & JARVIS, P.G. 1971. Methods of growth analysis. In: Plant photosynthetic production: Manual of methods, eds. J. Sestak, J. Cat sky & P.G. JaIVis, pp. 343 - 385. Junk, The Hague. MITCHELL, K.J. 1953. Influence of light and temperature on the growth of ryegrass (Lolium spp.). 1. Pattern of vegetative development. Physiol. PI. 6: 21 - 46. MITCHLEY, J. & GRUBB, P.J. 1986. Control of relative abundance of perennials in chalk grassland in southern England 1. Constancy of rank order and results of pot- and field-experiments on the role of interference. 1. Ecol. 74: 1139 - 1166. MYNHARDT, J.E. 1992. An ecophysiological study of two key grass species, Anthephora pubescens Nees and Eragrostis curvula

275 (Schrad.) Nees. MSc thesis, University of Pretoria, Pretoria. MYNHARDT, J.E., V AN ROOYEN, M.W. & THERON, O.K. 1994. Competitive ability of two grass species: Anthephora pubescens and Eragrostis curvula. 1. Yield and biomass allocation. S. Afr. 1. Bot. 60: 261 - 268. PUCKRIDGE, D.W. & DONALD, C.M. 1967. Competition among wheat plants sown at a wide range of densities. Aust. 1. Agric. Res. 18:193-211. RAI, J.P.N. & TRIPATHI, RS. 1985. Effects of density and soil nitrogen levels on growth of Galinsoga quadriradiata and G. parviflora in pure and mixed stands. Can. 1. Bot. 64: 2708 - 2715. ROBERTS, B.R. & FOURIE, I .H. 1975. Algemene grasse van Noord-Kaapland. N.G. Sendingpers, Bloemfontein. ROUSH, M.L. & RADOSEVICH, S.R 1985. Relationships between growth and competitiveness of four annual weeds. 1. appl. Ecol. 22: 895 - 906. . TILMAN, D. 1990. Mechanisms of plant competition for nutrients: The elements of a predictive theory of competition. In: Perspectives on Plant Competition, eds. lB. Grace & D. Tilman, pp. 117 - 141. Academic Press, San Diego. VAN ROOYEN, M.W. 1988. Ekofisiologiese studies van die efemere van Namakwaland. Ph.D. thesis. University of Pretoria, Pretoria. WARREN WILSON, 1. 1967. Effects of seasonal variation in radiation and temperature on net assimilation and growth rates in an arid climate. Ann. Bot. 31: 41 - 57. WOODWARD, F.I. 1979. The differential temperature response of the growth of certain plant species from different altitudes. 1. Growth analysis of Phleum alpinum L., P. bertolonu D.C., Sesleria albicans Kit. and Dactylis glomerata L. New Phytol. 82: 385 - 395.