Tải bản đầy đủ - 0 (trang)
III. Attaining Maximum Biological Yield per Unit Area

III. Attaining Maximum Biological Yield per Unit Area

Tải bản đầy đủ - 0trang

240



C. L. HEDLEY A N D M. J . AMBROSE



density, however, they will compete so vigorously that the yield per average

plant will be drastically reduced. The density response for an average strongly

competitive plant will therefore be extreme and the slope of this response will be

steep (Fig. 8a). Weak competitive genotypes with the same duration of growth as

strong competitors, will not take full advantage of the resources available at low

planting densities and average plants will therefore have lower biological yields

than strong competitors. When such genotypes are grown as dense populations,

their reduced aggressiveness in competition for resources will reduce the interaction between individuals, and the yield per average plant will be less affected and

may be higher than that of the strong competitor. The density response for an

average weakly competitive plant will therefore be less extreme and the slope of

this response will be shallow (Fig. 8a).

The effect of density on an average individual is determined from measurements made per unit area of the population. The population-density interactions for strong and weak competitors (Fig. Bb), however, will be the inverse of

those for the average individual within a population (Fig. 8a). Yields per unit

area from a monoculture composed of a genotype that is tolerant of high planting

density will therefore show a steep positive response to planting density (Fig.

8b), whereas a population composed of a genotype that is intolerant of competition will show a shallow or even negative response to increased density (Fig. 8b).

Therefore, in theory a genotype's ability to tolerate competition can be determined from a comparison of yields per unit area of populations grown at a low

(noncompetitive environment) and at a high (competitive environment) planting

density. This comparison can be made to test the relative effect of specific

genotype characteristics on tolerance to competition. Comparisons using strongly



a



b

,I



"/



BIl

Unit A r m



Plant



low



High

Plant Density



,

I



low



,,

,,



,,



High



FIG. 8. Biological yield (BY) for weak (WC)and strong (SC) competitors at high and low

planting densities (a) per plant and (b) per unit area.



24 1



DESIGNING “LEAFLESS” PLANTS FOR DRIED PEA CROP



a



b



Strong C o m p e t i t o r s



W e a k Competitors



,,

I



I



,



0

0

0



,



0



,,

0



0



,,

,,

I



,



, ,,

,, ,,,

,,

,

,

0



L



low



I



High



I



tow



I



High



Plant Density



FIG. 9. Biological yield per unit area of (a) strong and (b) weak competitors that are relatively

high- (H) or low- (L)yielding over all planting densities.



and weakly competitive environments will also distinguish between genotypes

that are overall high or low yielding over a range of planting densities, as well as

being strong (Fig. 9a) or weak (Fig. 9b) competitors.

B. THERELATIONSHIP

BETWEEN INITIALSEEDSIZE

A N D PLANTING

DENSITY



The effect of seed size on the response of a genotype to planting density is well

known, especially through the work of Black (1957). Black compared crops of

subterranean clover for the effect of initial seed size on individual plant weight

after 194 days of growth. When grown as spaced plants (weakly competitive

environment) the ratios of the “average” plant weights between the crops were

similar to the ratios between the initial seed weights. At a high population density

(strongly competitive environment), however, the average plant weight from

each crop was similar and not related to the initial seed size. The similarity

between average plants grown in the strongly competitive environment was due

to the fact that plants derived from the small seeds were affected by competition

later than those derived from large seeds. The increased duration of noncompeti-



242



C.



L.HEDLEY AND M . J . AMBROSE



tive growth for plants derived from small seeds resulted in these plants growing

to the same size as those derived from the larger seeds.

This relation between planting density and seed size has been used for determining the optimum planting density for leafed pea varieties for many years. The

recommended planting density for large-seeded varieties has always been lower

than that for small-seeded types (Gane et al., 1971). We have demonstrated

similar relationships for a number of leafless genotypes that differed for seed size

(Fig. 10a and b). We selected two genotypes that had small seeds (BS 20 and BS

41), two that had large seeds (BS 42 and BS 151), and two that had mediumsized seeds (BS 22 and JI 1198). The six genotypes were grown in microplots on

a square planting pattern at two planting densities: 16 and 100 plants/m' (weakly

and strongly competitive environments, respectively). The total above-ground

biological yield per unit area was determined by oven-drying plants taken from

the center of each microplot. The biological yield per unit area of the two

large-seeded genotypes was similar at the two planting densities (Fig. 10a). This

similarity was reflected in an approximate fivefold decrease in the "average"

a



26r BS 42..



I



4



24



-



22



.



20



-



b



18 *



;16-



'B



3



0



'c



14-



312 -



M



U



E

m a



\



10



-



8 -



1



a



Seed Size



6 -



large

-medium



4.



---small



2-



*-**a



t



I



16



High

100



low



0-



I



low

16



Plant Density ( plants/m')



FIG. 10. Biological yield (a) per unit area and (b) per plant of six leafless pea genotypes that

differ for seed size, grown at high andslow planting densities.



1



High

100



DESIGNING “LEAFLESS” PLANTS FOR DRIED PEA CROP



243



plant weight between the low and high plant densities (Fig. lob). These two

genotypes therefore behaved as strong competitors. The biological yield per unit

area of the two small-seeded genotypes responded steeply to increased planting

density (Fig. 10a). At the low planting density the biological yield was approximately half that of the large-seeded types, whereas at high planting density the

biological yield of the small- and large-seeded types was similar. The steep

population-density interaction for the small-seeded genotypes resulted in only a

threefold difference in average plant weight between the two densities (Fig. lob).

The small-seeded genotypes therefore behaved as weak competitors. The

genotypes with medium-sized seeds showed responses that were intermediate

between the large- and small-seeded types.

The association of seed size with tolerance to high planting density is symptomatic of a closer relationship between plant growth rate and plant competition.

The relative growth rate (RGR) is a measure of the efficiency with which cellular

systems, plants, or parts of plants will grow. In a bacterial suspension or in a

plant meristem it is a measure of the rate of cell division, while in the development of a seedling it is a measure of the combined rate of cell expansion and cell

division. Assuming that a developing seedling has ample light, minerals, and

water, then initially its RGR will be linear on a logarithmic scale and the slope

will be dependent on the interaction between the inherent rate of cell division and

expansion and the environment in which the seedling is developing. In order to

main:ain this environmentally determined RGR, the growth rate (GR), which is a

function of the mass of cells that are dividing or expanding at a particular time,

must increase exponentially. The environmental resources required to maintain

this exponential GR will be utilized at a rate equivalent to the GR; therefore if the

GR is low, demand for resources will be less than when the GR is high. It is the

demand that a seedling makes on environmental resources that determines the

competitiveness of the seedling. The GR, after a given duration of time, can

therefore be used as a measure to compare seedlings for their competitiveness;

seedlings with a low GR will be weaker competitors than those with a high GR.

The apparent relationship between seed size and competitiveness exists because

there is usually a good correlation between the size of the seed and the GR of the

seedling.

We have studied the relationship between seedling growth rate and initial seed

size in a wide range of both leafed and leafless peas. Seedlings were grown in

spaced pots in a greenhouse. At regular intervals seedlings were selected at

random and the dry weight was determined. Measurements were only made over

the first 30 days of growth from sowing and were stopped before the formation

of flowers. The plant GR of all the genotypes was exponential over this time

period and the RGR was therefore linear when the growth curve was expressed as

a logarithm. A representative sample of these data is presented in Fig. 11. The

relative growth rates were not significantly different for any of the genotypes,



244



C . L. HEDLEY AND M. J . AMBROSE



2.0

J1463(leafed)



1.0

1.6



1*4

102

1.0



s



t



0.8



D

'-



5



0.6



?



0-4



n

t



c



0.2



0



0



0

P



D

0

4



BS 52 ( l e a f l e s s )



-0.2



- 0.4

- 0.6



J I 463



- 0.8



JI 2

J I 197



- 1.0



BS52



-



-



054



-



28



30



375

127

64



283

227



-192



-1.4

1



12



I



14



16



18 20 2 2 24 26

Days from Sowing



32



FIG.11. Comparison of growth curves (log,) for a range of leafed and leafless pea genotypes

that differ for weight per seed.



including those with the leafless phenotype. There were, however, large GR

differences between the genotypes, as determined from the difference in the

height of the regression lines above the x axis (Fig. 1 1 ) . The difference in GR

correlated, as expected, with seed size, but this correlation only held when



DESIGNING “LEAFLESS” PLANTS FOR DRIED PEA CROP



245



genotypes within a phenotype were compared and not when leafed and leafless

genotypes were compared. Genotypes with the leafless phenotype had, in general, lower growth rates than those of genotypes with the leafed phenotype

irrespective of seed size.

The poor correlation between seed size and growth rate in comparisons between leafed and leafless genotypes was not due to a changed relationship between the weight of the embryonic axis and the weight of the seed, since a good

correlation exists between the weight of the embryonic axis and the weight of the

seed irrespective of the phenotype (Fig. 12). Although comparisons between the

developing leafed and “leafless’ ’ phenotype demonstrated no difference in RGR,

it is apparent that the two phenotypes must have differed for RGR earlier in

development, to give the observed differences in GR. The differences may be

due to the composition of the embryonic axes. A difference in the proportion of



-



50



0

c

X



W



r 40

0



c

.



c



s



2 30



n



‘0

Q,

Q,

IA



A



+



20



0



x *



0



A



10



I



I



I



I



I



I



I



1



2



3



4



5



6



7



E m b r y o n i c A x i s D r y Wt.tmgJ

FIG. 12. Relationship between the dry weight of the seed and the dry weight of the embryonic

axis from a range of leafed and leafless genotypes. Leafed: JI 956 (A), JI I194 ( x), J1 321 ( 0 )JI,

813 (W). Leafless: BS 21 (0).I1 1198 (+), BS 5 (*), BS 4 (A). Each point is the mean of ten seeds.



246



C. L. HEDLEY A N D M . J . AMBROSE



dividing to expanding cells in the embryonic axis will give an initial difference in

the RGR of the germinating seedling-the higher the proportion, the greater will

be the initial RGR. Such differences would become less significant as seedlings

develop because the proportion of growth attributed to dividing compared with

expanding cells becomes progressively less significant. The leafed embryonic

axis has leaf meristems that are absent from the leafless phenotype. It is possible

that these additional meristems contribute the dividing cells by which the RGR

could be initially increased.



c. EFFECTO F P L A N T GROWTHR A T E O N THE ACCUMULATION

OF

BIOLOGICAL

YIELDB Y



THE



CROP



The high planting densities required to attain an acceptable yield per unit area

of leafless peas can be interpreted as a requirement to increase crop growth rate

(CGR), especially early in the development of the crop. By definition a crop

growing at a high CGR will have a high leaf area index, and it is the integral of

the leaf area index over the growth period that is related to the biological yield of

the crop (Donald, 1961). In this respect a low-growth rate crop will resemble a

late-sown crop, where the integral of the leaf area index will be reduced by the

reduction in the growing season. This is one reason why pea yields decline with

sowings made progressively later in the season (Kruger, 1973; Milboum and

Hardwick, 1968; Proctor, 1963).

Crop growth rate is determined by the individual plant growth rate and planting density and can be raised therefore by increasing either or both of these

components. Although genotypes with the leafless phenotype have inherently

lower growth rates than those of leafed peas, there is still variation attributable to

differences in seed size. As discussed previously, large seeds will give rise to

plants that have high growth rates and small seeds will give rise to plants that

have lower growth rates. We have studied the interaction between plant growth

rate, as determined by initial seed weight, and planting density, which is the

other component of CGR.

The crop growth rates of two genotypes, one large-seeded (BS 151) and the

other small-seeded (BS 5 ) , were compared when grown at relatively high (100

plants/m') and low 16 plants/m') planting densities. As with all of our crop

experiments, young seedlings selected for uniformity were planted on the square

at both densities, to eliminate any intra- and interrow effects and to maximize

yield per unit area of each treatment. Large plots of each genotype at each density

were used and the whole experiment was replicated three times. Plants were

removed at frequent intervals from random positions within the plots, dryweighted, and used to determine the change in biological yield per unit area of

each crop with time.



DESIGNING “LEAFLESS” PLANTS FOR DRIED PEA CROP



247



Initially environmental resources were not limiting and the individual plants

comprising each crop developed independently of planting density. The relative

growth rate of each crop was therefore directly related to the relative growth rate

of an individual spaced plant. Since it was demonstrated previously (Section

III,B) that pea genotypes grown free of competition have similar relative growth

rates, it follows that during this early phase of crop development the crop relative

growth rates will be similar. A comparison of the linear regressions obtained by

plotting the logarithm of the biological yield per unit area against time (Fig. 13a

and b) confirmed that there was no significant difference between the relative

growth rate of the large- and small-seeded genotypes at either planting density.

Differences between the growth of the crops were therefore due solely to differences in crop growth rate attributable either to initial seed size or to planting

density. Because the regressions were parallel, the differences in growth rate can

be determined from the height of the regression lines from the x axis.

The difference between the regressions of the large- and small-seeded crops

was a constant and independent of planting density. By extrapolating the regression lines to zero time the initial difference between the genotypes could be

determined (Fig. 13a and b). The regression lines cut the y axis, not at an initial

weight equivalent to that of the seed, but at a weight that equates with the weight

of the embryonic axis. This confirmed the suggestion from the previous section

that the embryonic axis must be the unit determining differences in growth rate.

The difference in weights of embryonic axes derived by extrapolation (approximately 3 mg for BS 5 and 5 mg for BS 151) are very similar to those determined

by dissecting dried seed of the two genotypes (2.4 and 4.5 mg, respectively).

This difference in the initial embryonic axis weight determined that the genotype

regressions were separated by 5 days.

The difference between the regressions of the high- and low-density crops

within each genotype was proportional to the initial weight or number of embryonic axes per unit area. The difference between 16 and 100 axes per unit area

resulted in a difference of 18 days between the regression lines for crops growing

at the two planting densities.

The period of exponential growth during which plants grew free of interplant

competition ended when resources for plant growth became limiting. Interplant

competition was then initiated, resulting in a progressive decline in the relative

growth rate of all the crops (Fig. 13a and b). Competition was initiated when a

specific biological yield per unit area had been attained that was independent of

genotype and planting density and varied only with the time from sowing. The

crop growth rates were therefore similar at this specific biological yield per unit

area. Within a given density the plant growth rate, when competition was initiated, was therefore similar for both genotypes (Fig. 13c and d). The plant

growth rates differed between planting densities and were higher in the lower

density by an amount that was proportional to the decrease in plant number per



248



C. L. HEDLEY AND M. J . AMBROSE



-



13



C.



12

11

10

9



8

7

6

5

4

41

3



20



40



60



80



1

100 120



3

20



40



60



80



loo



120



Days From Sowing



d



11



C



10



-C.



-c.



9

8



-C.



7

6



5

4

3

2



FT



I



i

20



40



60



80



.



100 120



1

0



I



'

i

20



Days From Sowing



40



60



80



100 120



DESIGNING “LEAFLESS” PLANTS FOR DRIED PEA CROP



249



unit area. More simply, when competition was initiated in the high-density

crops, it was between plants growing at relatively low growth rates, while in

low-density crops competition was between plants growing at relatively high

growth rates. Competition was first observed in the large-seeded crop (BS 151)

grown at the high planting density, occurring after 58 days from germination.

The same biological yield per unit area was attained by the small-seeded

genotype (BS 5 ) after 63 days. At the low planting density competition was

initiated between plants of the large- and small-seeded genotypes after 76 and 81

days, respectively.

It is apparent from this experiment that crop growth rate can be varied equally

well by initial seed weight, or more correctly the weight of the embryonic axis,

and by planting density. In theory it is therefore possible to produce an identical

crop growth rate by sowing large seeds at low planting density or by sowing

small seeds at a proportionally higher planting density (Fig. 14). Although it has

been suggested that such crops would be similar (Donald, 1961), it is apparent

that the effect on individual plants will be very different (Fig. 14). Competition

would occur between the plants of both crops after the same time from germination. The large-seeded plants, however, would have high growth rates when

competition was initiated, whereas the growth rate of the small-seeded plants

would be low. The consequences of these different crop structures are more

significant to the partitioning of biological yield into economic yield and will be

discussed in Section IV.

In the experiment (Fig. 13a-d) the duration of crop growth was similar for the

two genotypes at both planting densities and so the amount of growth that

occurred after competition had been initiated within the crops was limited not

only by diminishing resources, but also by a finite time in which the plants could

utilize these resources. The average plant weight of the two genotypes at 100

plants/m’ was approximately 6 g/plant, which gave a crop yield of about 600

g/m2 for both crops. The similarity between the final yield of the two genotypes

at this planting density suggests that both crops had become limited by competition at this maximum biological yield and that this yield was therefore an

environmentally determined optimum.

The two genotypes differed for final biological yield at the low planting



FIG. 13. Crop (a, b) and plant (c, d) growth curves of two leafless pea genotypes that differed for

weight per seed (BS 5 , 130 mg; BS 151, 300 rng) grown at two planting densities. (a) BS 5 (0)and

BS 151 ( + ) at 100 plants/rn2; (b) BS 5 (0)

and BS 151 (63)

at 16 plants/m’; (c) BS 151 plants at 100

(+) and 16 plantsh’ (@

(d) BS

I);

5 plants at 100 (0)and 16 plantdrn’ (0).Horizontal arrows

indicate the biological mass at the onset of interplant competition (C). Symbols 0, 0,

+, @ with

vertical arrows indicate the time from sowing to the onset of interplant competition. fl with vertical

arrows indicates flowering time.



C. L. HEDLEY AND M. J. AMBROSE



250



Biomass

Ceiling

Onset of Plant

Competition



t

T



Time



b



............................................



C



/



..............................................



//



............................................



0



0

0



0



'...................../(............"'....*.

/



,

8



,'



Small



High



/



0



t



t



T



T



Time

FIG. 14. (a) Theoretical growth curves of two contrasting crops growing at the same crop growth

rate. One crop composed of a large-seeded genotype sown at low planting density and the other of a

small-seeded genotype sown at high planting density. (b,c) Growth curves of single plants from each

of the two crops. T is the time of onset of interplant competition.



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

III. Attaining Maximum Biological Yield per Unit Area

Tải bản đầy đủ ngay(0 tr)

×