Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of density and harvest time on growth and yield of forage corn (Zea mays L.) Tarimo, Akwilin J. P. 1983

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1983_A6_7 T37.pdf [ 5.18MB ]
Metadata
JSON: 831-1.0095703.json
JSON-LD: 831-1.0095703-ld.json
RDF/XML (Pretty): 831-1.0095703-rdf.xml
RDF/JSON: 831-1.0095703-rdf.json
Turtle: 831-1.0095703-turtle.txt
N-Triples: 831-1.0095703-rdf-ntriples.txt
Original Record: 831-1.0095703-source.json
Full Text
831-1.0095703-fulltext.txt
Citation
831-1.0095703.ris

Full Text

THE EFFECTS OF DENSITY AND HARVEST TIME ON GROWTH AND YIELD OF FORAGE CORN (Zea mays L.) by Akwilin J.P. Tarimo B.Sc. (Agr.) University of Dar-es-Salaam, 1980 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in the Department of PLANT SCIENCE We accept this thesis as conforming to the required standard University of British Columbia September, 1983 © Akwilin J.P. Tarimo, 1983 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the library shall make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission Department of Plant Science The University of British Columbia Vancouver, B.C. Canada September, 1983 ABSTRACT A fi e l d experiment was undertaken to investigate the effects of planting density on growth and yield of forage corn (Zea mays L., cv. DK 24) using modern methods of plant growth analysis and yield component analysis. A complete randomized block design was used with four planting densities and five harvests. The four planting densities were 49383, 67204, 87796 and 111111 plants per hectare. Replicate plants were harvested at 21 days after emergence (DAE), 42 DAE, 63 DAE, 85 DAE and 115 DAE. At each harvest, data were recorded of several primary growth characteristics, including plant height, stem diameter leaf areas and dry weights of stems, leaves, leaf sheaths, t i l l e r s and reproductive structures. The recorded data, and indices and ratios derived from recorded data, were analyzed by the analysis of variance, cubic spline regressions and the two-dimensional partitioning technique of yield component analysis. Yield per hectare varied significantly among densities from the second harvest (42 DAE) until maturity. At the crop maturity stage (30.8% crop dry matter content), the yield per hectare increased with increasing number of plants per hectare. The mean yields were: 15.1, 15.9 and 17.1 MT per hectare from the lowest planting density to the highest planting density, respectively. Conversely, yield per plant decreased linearly with increasing number of plants per hectare from 306.4 to 154.1 g/plant. A l l the primary plant growth characteristics were highly affected - i i i -by the planting density treatment, and these effects increased with plant age. Thus, the variability in yield per plant among planting densities was accounted for by the variability of those growth characteristics. The plant growth indices showed that crop growth rate, leaf area index and biomass density were major contributors to yield variability per hectare among planting densities. Yield component analysis showed sporadic contributions by yield components depending on age and the direction of the two-dimensional partitioning technique of yield component analysis. The relative growth rates of yield and yield components did not clearly show the effects of planting density but strongly showed the time course trends of the relative growth rates within stand densities. A l l of the techniques used in this study complemented each other in the analysis of corn growth and productivity. - iv -TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS iv LIST OF TABLES v i i LIST OF FIGURES v i i i LIST OF SYMBOLS x i ACKNOWLEDGEMENTS x i i i 1. INTRODUCTION 1 2. LITERATURE REVIEW 3 2.1 The Corn Plant and Its Cultivation 3 2.2 Effects of Planting Density on the Growth and Yield of Forage Corn 6 2.3 Growth and Yield Analyses 8 3. MATERIALS AND METHODS 10 3.1 Generation of the Primary Data 10 3.2 Growth and Yield Analyses 12 3.2.1 Growth Analysis 12 3.2.2 Yield Component Analysis 14 4. RESULTS 17 4.1 Primary Characteristics of Growth 17 4.2 Plant Growth Indices 31 4.3 Yield Component Analysis 47 4.4 Relative Growth Rates of Yield Components 63 5. DISCUSSION 71 6. CONCLUSIONS 83 7. LITERATURE CITED 85 - V -Page 8. APPENDICES 88 I. Design of experiment - complete randomized block design (CRBD) 88 II. Natural logarithms of yield component ratios - means from each density and harvest.... 89 III. Mean orthogonal yield components at each density and harvest time. 90 IV. Forward-YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages in the early vegetative growth 91 V. Backward-YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages in the early vegetative growth... 92 VI. Forward-YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages in the late vegetative growth.... 93 VII. Backward-YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages in the late vegetative growth.... 94 VIII. Forward-YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages in the early reproductive growth 95 IX. Backward-YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages in the early reproductive growth 96 X. Forward-YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages in the late reproductive growth 97 XI. Backward-YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages in the late reproductive growth 98 - v i -Page XII. Forward-YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages in the maturity stage... 9 9 XIII. Backward-YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages In the maturity stage. 100 - v 1 i -LIST OF TABLES Table Page 1(a) Yield (per plant) of primary plant growth characteristics of forage corn grown at four planting densities and five harvests 18 1(b) Yield (g/plant) of primary plant growth characteristics of forage corn grown at four planting densities and five harvests 19 2 Effects of planting density and harvest time on the primary plant growth characteristics 20 3 Shoot dry matter yield (g/plant) of forage corn at four planting densities and five harvest times... 32 4 Shoot dry matter yield (MT/ha) of forage corn at four planting densities and five harvest times... 33 5 Mean observations from individual morphological characteristics at each density and harvest of forage corn 48 6 Mean yield component values for forage corn at each density and harvest time 49 7 Forward-YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages - data from a l l observations 51 8 Backward-YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages - data from a l l observations 52 9 Yield components that were significant at each harvest time including density effects 61 10 Regression coefficients of the relative growth rates of yield (R w) on the relative growth rates of yield components of forage corn observed in four planting densities 70 — yi i i -LIST OF FIGURES Figure Page 1 Cubic spline regressions describing the time course of plant height (T) in corn at four planting densities 23 2 Cubic spline regressions describing the time course of plant diameter (D) in corn at four planting densities 24 3 Cubic spline regressions describing the time course of number of leaves/plant (Lpj) i n corn at four planting densities 25 4 Cubic spline regressions describing the time course of leaf area/plant (L A) in corn at four planting densities 26 5 Cubic spline regressions describing the time course of dry weight of leaves/plant (W^ ) in corn at four planting densities 27 6 Cubic spline regressions describing the time course of dry weight of leaf sheaths/plant (W^g) in corn at four planting densities 28 7 Cubic spline regressions describing the time course of dry weight of stem/plant (Wg-jO i n corn at four planting densities 29 8 Cubic spline regressions describing the time course of dry weight/plant (W) in corn grown at four planting densities 30 9 Cubic spline regressions describing the time course of crop growth rate (CGR) in corn at four planting densities 35 10 Cubic spline regressions describing the time course of leaf area index (LAI) in corn at four planting densities 37 11 Cubic spline regressions describing the time course of unit leaf rate (ULR in corn grown at four planting densities 38 - ix Figure Page 12 Cubic spline regressions describing the time course of relative growth rate of plant dry weight (R^) in corn at your planting densities AO 13 Cubic spline regressions describing the time course of leaf area ratio (LAR) in corn at four planting densities 41 14 Cubic spline regressions describing the time course of leaf weight ratio (LWR) in corn at four planting densities 42 15 Cubic spline regressions describing the time course of specific leaf area (SLA) in corn at four planting densities 43 16 Cubic spline regressions describing the time course of relative growth rate of leaf area per plant (R^A) i n corn at four planting densities 44 17 Cubic spline regressions describing the time course of relative growth rate of leaf dry weight per plant (Ry^) i n corn at four planting densities.. 45 18 Cubic spline regressions describing the time course of relative growth rate of stem dry weight/plant (RysT^ i n c o r n a t f o u r planting densities 46 19 Cubic spline regressions describing the time course of D/T-yield component of forage corn at four planting densities 55 20 Cubic spline regressions describing the time course of Ljj/D-yield component of forage corn at four planting densities 56 21 Cubic spline regressions describing the time course of L^/Ljj-yie,ld component of forage corn at four planting densities.. 57 22 Cubic spline regressions describing the time course of W L/LA - y i e- L d component of forage corn at four planting densities 58 Cubic spline regressions describing the time course of W/W^-yield component of forage corn at four planting densities Cubic spline regressions describing the time course of relative growth ra^e of yield (R^) in density one (4.9 plants/m ) and the relative growth rates of i t s yield components (RcDl) * n forage corn Cubic spline regressions describing the time course of relative growth ra|e of yield (%) in density two (6.7 plants/m ) and the relative growth rates of i t s yield components (RCD2^  ^ n forage corn Cubic spline regressions describing the time course of relative growth rate of yield (R^j) in density three (8.8 plants/m2) and the relative growth rates of i t s yield components (RcD3^ ^ n forage corn Cubic spline regressions describing the time course of relative growth rate of yield (R^) in density four (11.1 plants/m2) and the relative growth rates of Its yield components (RCD4) ^ n forage corn - xi -LIST OF SYMBOLS A - Land area. ANOVA - Analysis of variance. C - Yield component variable. CGR - Crop growth rate. D - Average plant diameter. DAE - Days after emergence. DM - Dry matter. Ej, - Ear fresh weight. ENT - Total number of ears/plant. FNT - Total number of ears and tassels/plant. H - Harvest index. ha - Hectare. L^ - Leaf area/plant. LAI - Leaf area index. LAR - Leaf area ratio. L F - Leaf blade fresh weight/plant. L^ - Number of leaves/plant. Log e - Natural logarithm. LWR - Leaf weight ratio. MT - Metric tonne. N - Number of plants. R - Relative growth rate of yield component variable. Rp - Remaining fresh weight. Relative growth rate of growth characteristics. Relative growth rate. Relative growth rate of leaf area/plant. Relative growth rate of shoot biomass/plant. Relative growth rate of leaf dry weight/plant. Relative growth rate of stem dry weight. Relative growth rate of shoot yield/plant. Specific leaf area/plant. Plant height. Two-dimensional partitioning. Unit leaf rate. Total shoot dry weight/plant. Dry weight of cobs/plant. Dry weight of grains/plant. Dry weight of husks/plant. Dry weight of leaves/plant. Dry weight of leaf sheaths/plant. Dry weight of stem/plant. Dry weight of tassels/plant. Dry weight of tillers/plant. Total reproductive dry weight/plant. Total vegetative dry weight/plant. Yield of shoot dry matter per plant. Yield Component analysis. -XI 1 1. -ACKNOWLEDGEMENTS I am most sincerely grateful to my supervisor Dr. Peter A. J o l l i f f e for his willingness to supervise this research. The success this thesis owes a great debt to his enthusiasm, understanding and patience. Sincere acknowledgements are also extended to Dr. F.B. Holl fo his good tutorial guidance in the principles of forage corn agronomy which led to the success of this research. Drs. F.B. Holl and V.C. Runeckles are also cordially acknowledged for participating on my supervisory committee. Dr. George W. Eaton and Patricia Bowen are kindly acknowledged for assistance with some of the s t a t i s t i c a l analyses. Special thanks also, are extended to D. Pearce (On retirement leave) for his kind assistance in fi e l d preparation and for some valuable suggestions on the general management of the experimental plots. I gratefully acknowledge the material and technical help of Ashley Herath, Peter Garnett, B. McMillan, Helen Evans, Jean Watters, Andrew Chow and Madhukar Potdar for assisting in the management, harvesting and the hard work of data collection in the f i e l d . - x i v -I also gratefully acknowledge the scholarship granted to me by the International Development Research Centre (IDRC), Ottawa, through the Groundnut and Pulses Project in Tanzania. The generous study leave, granted to me by the Department of Crop Science, University of Dar-es-Salaam, Tanzania, which enabled me to do these studies, I also acknowledge gratefully. I am grateful to Mrs. Jeeva Jonahs for her s k i l l s in typing this manuscript. I also gratefully acknowledge a l l my relatives and friends who either directly or indirectly contributed to the successful completion of this study. - 1 -1. INTRODUCTION Variations in planting density can affect vegetative and reproductive productivity of crop plants. Both yield per hectare and yield per plant can be influenced by the number of plants per unit land area. In most instances, yield per land area increases with increased number of plants per unit land area, while the yield per plant is reduced, until a limit is reached. The analysis of plant response to different planting densities can provide some insight into the biology of plant growth and can contribute to Improved agricultural practices. Corn (Zea mays L.) was used in the present research because i t is a valuable crop, not only for animal feed, but also more directly for human food. The present study, however, is focused on silage corn production for the beef industry. Secondarily, corn grows well in the experimental area available for my research (at the University of British Columbia, Vancouver, B.C.) and corn has growth characteristics which can be studied effectively throughout the growing season. Several approaches to plant growth and yield analysis have been developed and refined in recent years. Such approaches include yield component analysis (Fraser and Eaton, 1983), which has been applied widely in agricultural research, and which subdivides productivity into a set of morphological components whose product is yield. Demographic analysis, a central method in population biology, has been extended to the sub-organismal level in plants to follow the appearance, presence and loss of morphological characteristics (e.g., Bazzaz and Harper, 1977; Lovett Doust and Eaton, 1982). Plant growth analysis (Causton - 2 -and Venus, 1981; Evans, 1972; Hunt, 1982a) includes indices of both the presence and assimilatory performance of morphological characteristics. The chief value of yield component analysis, demographic analysis or plant growth analysis, however, may be that each provides a framework for defining long term relationships among components and overall growth ( J o l l i f f e et_ a l . 1983 (in preparation)). While considerable research has been carried out on the effects of planting density on plant performance, much remains to be learned about this interaction. My research had two main objectives: 1. To investigate the time course of forage corn productivity at four different planting densitites using modern techniques of plant growth analysis. 2. To define when and how some morphological and physiological components of forage corn yield respond to different planting densities, and how variation in those components i s associated with variation in yield. In addition, two secondary objectives of the research were: 3. To evaluate the seeding rates currently recommended for forage corn production in this locality. 4. To evaluate the effectiveness of the different approaches to plant growth analysis. - 3 -2. LITERATURE REVIEW 2.1 The Corn Plant and Its Cultivation The corn plant is a t a l l annual grass, with thick, solid stems usually supported by proproots. It is thought to have originated in Central America, although this has not been completely proven. The origin of the plant is more obscure because apparently i t does not occur in the wild form. However, in Mexico there occurs a closely related grass, teosinte (Euchlena mexicana L.), which hybridises freely with Zea. Thus, i t has been suggested that modern corn is either of hybrid origin, or in some other way i t is a derivative of teosinte (Janick, et^ a l . 1981; Langer and H i l l , 1982). Modern cultivars of corn resemble the primordial form only remotely. Moreover modern corn is wholly a ward of humanity since i t is unable to survive and perpetuate i t s e l f without human care in harvesting and planting ( G i l l and Vear, 1980; Janick et a l . 1981). The corn plant does not normally branch. However, a few t i l l e r s do occur. It bears broad, smooth leaves with a conspicuous midrib. The plant is monoecious. It bears the male flowers on the terminal panicle (tassel), and the female flowers are borne on the axils of the middle leaves. Pollen is shed and carried by wind to the female inflorescences (Langer and H i l l , 1982). The cob bears rows of naked caryopses which are protected by the husks. On the other hand, a mass of elongated styles protrude at the end of the husks ready to receive pollen shed from the terminal male inflorescences to effect f e r t i l i z a t i o n of the ovules. One characteristic of effective pollination of the female - 4 -flower is the immediate death of the extruded ends of the s i l k (styles). Grain development and growth then follows and grains mature inside the husks. Unlike other grasses, there is no seed dispersal, which accounts for the fact that corn is not known as a wild plant (Langer and H i l l , 1982). Although predominantly a grain crop, corn is also used as a forage crop. Forage corn is grown in areas where the climate is marginal for grain ripening but where advantage can be taken of the rapid growth of the plant. Forage corn is cut at an immature reproductive stage and is either fed to animals in the fresh state or after having been ensiled (Langer and H i l l , 1982). In a normal silage crop, the ears contribute some 40 to 45% of the dry weight ( G i l l and Vear, 1980), with a digestible dry matter content of about 80%; the remainder comes from the stem and leaves (Stover), in which digestible dry matter content is about 65%. The productivity of forage corn varies with prevailing climatic conditions. In most temperate countries, regions of corn production are limited by low temperature since the corn crop does not grow well under temperatures below about 10°C. Good yields are also very much dependent on the availability of sunshine (Lockhart and Wiseman, 1978). Exposed windy situations are generally not suitable. The corn crop thrives in rich, deep, well drained loam soils. Light soils are reasonably suitable for corn i f they do not dry out. Thin chalky soils are not suitable, nor are heavy soils which are usually very slow to warm up i n spring (Lockhart and Wiseman, 1978). - 5 -Corn seed should be planted on a well prepared seedbed. Thorough ploughing should be done before planting to ensure a good seedbed, and seed should be sown 4-6 cm deep in a level, moist, friable s o i l . Organic manures and f e r t i l i z e r s are required wherever the s o i l f e r t i l i t y i s otherwise insufficient for vigorous growth and high productivity. In order to ensure vigorous growth and high productivity, weeds and pests must be controlled throughout the growing period of the crop. In early crop growth, weeds can smother corn seedlings and impair normal growth. Arrested growth of corn in the early vegetative growth stage is the main cause of poor yields later in the season. Chemical weed control is commonly used in forage corn fields because high corn plant populations restrict the use of mechanized weed control. When, corn is grown for silage, the whole above ground shoot is harvested (Lockhart and Wiseman, 1978). In practise, forage corn makes a highly palatable silage, but i t is better suited for beef than dairy production systems (Lockhart and Wiseman, 1978). The protein content of silage can be improved by addition of non-protein nitrogen compounds, usually when the crop is being ensiled. The ensiled corn should be harvested with a dry matter content of between 25% and 35%. The diges t i b i l i t y value for such silage would be about 68%. Plant maturity at harvest is assessed by observing the kernel development. Plants are normally cut for silage when the kernels become "cheesy" and "doughy" in the early dent stage. The yields realized at this stage of growth depend on the effects of various climatic and agronomic factors during - 6 -the cropping season. Such factors include: the amount and d i s t r i b u t i o n of p r e c i p i t a t i o n , amount and duration of solar r a d i a t i o n , temperature, s o i l c h a r a c t e r i s t i c s planting density per h i l l and per hectare, weed and pest controls and exposure to wind and a i r p o l l u t i o n . 2.2 Effects of Planting Density on the Growth and Yield of Forage Corn As noted above, forage corn y i e l d s may depend on the planting density used during the establishment of the crop. The general e f f e c t s of planting density on the growth and y i e l d of herbacious species were reviewed by Donald (1963) and Holliday (1960). The commonly observed trends of density e f f e c t s on y i e l d are lower y i e l d s per unit land area i n lower densities and higher y i e l d s per unit land area i n higher d e n s i t i e s u n t i l an optimum or plateau i s reached. When high y i e l d / p l a n t i s desirable, the reverse holds true. Y i e l d per unit land area i s , however, the most commonly used agronomic basis for recommending a c e r t a i n planting density for crop species. Optimum planting density may be influenced by environmental hazards. Plants grown at low density could y i e l d lower than otherwise expected due to greater exposure to wind, pests and diseases (Donald, 1963). Corn c u l t i v a r s have been shown to d i f f e r s i g n i f i c a n t l y i n the ways they react to increasing planting density. For example, Buren et  a l . (1974) observed that high density-tolerant genotypes were free from the apparent density e f f e c t s of barrenness. Such c u l t i v a r s were characterized by rapid completion of s i l k extrusion, coincidence of s i l k extrusion and pollen shed, rapid growth i n the f i r s t ear and f i r s t ear s i l k , p r o l i f i c a c y , reduced t a s s e l size and e f f i c i e n t production of grain - 7 -per unit land area. Higher planting densitites have also been found to increase plant quality through increased soluble sugar content and also may advance maturity of silage (McAllan and Phipps, 1977). Some plant morphological characteristics have been found to be more susceptible to density effects than others. Stems are highly susceptible to the effects of planting density (McAllan and Phipps, 1977), mainly being reduced in weight in response to increasing number of plants per unit land area. Corn plants grown at a low planting density produce larger ears, and show slower leaf senescence and more robust stems than corn grown at a high planting density. Stem elongation responds to increased number of plants per unit land area by increased plant height (Lopes and Maestri, 1981). Planting density does not influence the maximum number of leaves per plant. Both plant height and number of leaves per plant reach their maximum values at the time of ear i n i t i a t i o n , regardless of planting density. Dry matter partitioning within a corn plant varies with stage of plant growth and development (Lopes and Maestri, 1981). Leaves and roots are the common sinks in the early stages of growth. At ear ini t i a t i o n , the ear becomes the preferential sink. Such trends have not been shown to be affected by planting density. Higher dry matter yields/ha in plants grown at a density of 8 2 plants/m were found to be due mainly to increased leaf area index (Reraison and Lucas, 1982). Silage yield in corn depends on dry matter accumulation in the shoots (Iremiren and Milbourn, 1978). Increasing number of plants/m reduces dry matter accumulation in individual plants - 8 -but increases dry matter yields/ha asymptotically until a plateau i s reached. Plants grown at a higher density, however, have increased risks of lodging which may interfere with the total recoverable yield at harvest. Time for harvesting forage corn depends on the kernel maturity (Gonske and Keeney, 1969). Such forage w i l l be of better quality when harvested at late dent of kernels than early dent because late dent corn has higher dry matter and protein yields, and lower nitrate nitrogen and soluble nitrogen contents, than does corn harvested at early dent (Gonske and Keeney, 1969). Kernel dry weight accummulation rate is not affected by planting density, but the effective f i l l i n g period may be shortened at high planting density (Poneleit and E g l i , 1979). This effect can also reduce yield per plant to some extent. 2.3 Growth and Yield Analyses Several different approaches have been developed during this century for the quantitative analysis of plant growth and yield. Recent studies have refined many of the main concepts and have provided improved procedures for understanding and analyzing experiments related to crop productivity. Recent reviews and major contributions in this general subject area include publications on plant growth analysis (Causton and Venus, 1981; Evans, 1972; J o l l i f f e et a l . 1983 (in preparation); Hunt, 1982a, 1982b), yield components analysis (Eaton e_t a l . , 1983; Fraser and Eaton, 1983; J o l l i f f e et a l . 1982, 1983 (in preparation)) and demographic analysis (Bazzaz and Harper, 1977; Lovett Doust and Eaton, 1982; J o l l i f f e et a l . 1983 (in preparation)). In this - 9 -thesis, the two approaches used to analyze the response of forage corn productivity to planting density were plant growth analysis and yield component analysis. Plant growth analysis (i.e. the so-called British School of plant growth analysis) includes indices of both the presence and assmilatory performance of morphological growth characteristics (Evans, 1972). Modern plant growth analysis is facilitated by the use of fit t e d curves (Causton and Venus, 1981; Hunt, 1982a; Parsons and Hunt, 1981). The f i t t e d curves describe the quantitative performance of plants or plant parts, integrated both throughout the system under study and across ecologically or agronomically-meaningful intervals of time (Hunt, 1982b) Ratios, rates, compounded rates and integral durations are i t s stock-in-trade and they can a l l be effectively described by cubic spline regressions (Hunt, 1982b). Yield component analysis has been applied in agricultural research to analyze crop productivity in single harvests or In a sequence of harvests during the course of crop growth. In yield component analysis, plant morphological characteristics are selected in a rationalized sequence (e.g. , following the chronological development of plant growth), and are included as a series of ratios (yield components) in a model predicting yield. Yield component models may be analyzed in several ways including simple or multiple regressions of the components in relation to yield. Other analytical procedures involve the use of ordered stepwise multiple regressions (Eaton and Kyte, 1978), analysis of variance (Bowen, 1983) and two-dimensional positioning-yield component analysis (TDP-YCA; Eaton et a l . 1983). - 10 -3. MATERIALS AND METHODS 3.1 Generation of the Primary data The experiment was conducted at the Totem Field Laboratory of the University of British Columbia during the summer of 1982. The land was ploughed early in April and was frequently harrowed before the planting date, which was the 26th of May, 1982. Before planting, the f i e l d plots were laid out in a randomized complete block design (RCBD) with 9 blocks and 4 different planting density plots in each block. Five sample sub-plots were randomly marked out within each density plot for the five harvests done in the growing season. Sampled plots were separated by at least one row of unsampled plants to ensure that early harvested plots would not Influence the subsequent growth in remaining sample plots. Four plants were sampled from each planting density plot at each harvest. A complete layout of the experimental plots is shown in Appendix (I). Within the plots, forage corn (Zea mays L.) seed cr. DK 24 was sown in a square planting pattern with the dimensions of the square varying according to the planting density. The square dimensions were 45 cm x 45 cm, 39 cm x 39 cm, 34 cm x 34 cm and 30 cm x 30 cm, representing the following number of plants per hectare: 49383, 67204, 87796 and 111111, for di to di+ (Appendix I), respectively. Bird and weed damage to seedlings and the established crop were kept to a minimum throughout the growing season. Birds were controlled by stretching large pieces of fishnet about 60 cm above the plots immediately after sowing. These nets were removed after the seedlings were fully established. Weeds were constantly eradicated from the - 11 -plots by hand-pulling or by use of a hand hoe. Harvesting of the plots was done when the plants reached the following ages: 21 days after emergence (DAE), 42 DAE, 63 DAE, 85 DAE and 115 DAE, which correspond to growth stages designated as early vegetative growth, late vegetative growth, early reproductive growth, late reproductive growth and forage maturity, respectively. The four plants harvested from each plot were sampled for the following data before drying of samples: Number of leaves per plant (L^) Plant height (T; cm) Plant diameter (D; cm) Leaf area per plant (L^> cm) Leaf blade fresh weight (L F; g/plant) Remaining fresh weight per plant (Rp; g/plant) Total number of tassels and ears/plant (FNT/plant) Total number of ears per plant (ENT/plant) Ear fresh weight per plant (E F; g/plant) Total fresh weight (WF; g/plant) After drying the samples to a constant dry weight at 80°C in a forced air bulk drier, the following data were also recorded: Leaf blade dry weight per plant (WL; g/plant) Leaf sheath dry weight per plant (WLS> g/plant) Stem dry weight per plant (Wg^ ; g/plant) T i l l e r dry weight per plant (WTL; g/plant) Total vegetative dry weight per plant (Vty; g/plant) - 12 -Tassel dry weight per plant (W^AJ g/plant) Husk dry weight per plant (WTJ; g/plant) Cob dry weight per plant (WrjJ g/plant) Grain dry weight per plant (WQ; g/plant) Total reproductive dry weight per plant (Wrep; g/plant) Total shoot biomass yield per plant (W; g/plant) 3.2 Growth and Yield Analyses 3.2.1 Growth Analysis Growth trends and variability in growth characteristics among different planting densitites were developed from the primary plant growth characteristics listed above. Analysis of variance was used to study the variability of means of primary growth characteristics among densities and ages. Whole shoot biomass was also analyzed separately by the ANOVA method. In both cases the ANOVA model was set to partition v a r i a b i l i t y according to blocks, density, harvest time, density x harvest time interaction and error. The respective degrees of freedom were 8, 3, 4, 12 and 696 for those sources of variability. A total of 720 data cases were therefore analyzed in the ANOVA model for each primary growth characteristic. At each harvest, yield, which was defined to be total shoot biomass (W), was derived from the following equation: w = wL + w L S + w S T + w T L + w r e p (1) Fitted curves, describing the time course of most of the primary values described above were generated using a cubic spline regression - 13 -procedure. This procedure ( J o l l i f f e et a l . 1983 (in preparation)) involved the Fortran subroutines DSPLFT and DSPLN available through the University of British Columbia Computing Centre. The procedure also generated fitted curves for the f i r s t derivative (e.g. dW/dt) and the f i r s t derivative of the logarithm (i.e. the relative growth rate) of each characteristic. Smoothing of the spline regressions was determined entirely within the computer program according to the size of the standard deviations of the variates at each harvest. Several derived plant growth indices were developed from the cubic spline regressions of the primary growth characteristics as follows: CGR LAI x ULR (3) Where CGR ULR LAI crop growth rate (g/m /day) 2 2 leaf area index (m /m ) Unit leaf rate (g/m2/day) = RT7/LAR (4) where R^^ = relative growth rate of total shoot biomass (g/g/day). - 14 -LA LAR = (5) Where LAR = leaf area ratio (m2/g) SLA = LA/WL (6) Where SLA = specific leaf area (m2/g). LWR = WT /W (7) Where LWR = leaf weight ratio (g/g). 3.2.2 Y i e l d Component Analys is Plant morphological characteristics were included in a yield component analysis model based on the following developmental order: v _ _ D h LA W L W T" x ~~D x L~T X L~~ x w~ <8> N A L where Y = W = yield. Each term on the right hand side of Eqn. (8) w i l l be called a yield component (C). The yield components were also analyzed by the ANOVA model described above and their means were used to develop cubic spline regressions, using procedures similar to those described above, to trace their trends of growth over time. - 15 -In addition, the yield components were transformed into their natural logarithms. Thus Eqn. (8) was tranformed as follows: The logarithmically transformed data were orthogonalized and transformed according to the method of Eaton et a l . (1983). The analysis then proceeded by the two-dimensional partitioning-yield component analysis (TDP-YCA) method described by these authors. Both forward and backward TDP-YCA analyses were performed on the pooled data from a l l harvests and were also performed on data from individual harvests in order to identify significant yield components and examine the effects of density on yield and yield components. The cubic spline regressions derived for yield components were also used to express the relative growth rate of yield (Ry) according to the equation ( J o l l i f f e et a l . , 1983 (in preparation)): 'N c "A (9) + L °Se ( w-> n RY = E RCi i=l (10) - 16 -where = relative growth rate of each yield component variable (C). n - 6 (the number of yield components). These growth indices were also plotted in time course curves to study their trends at each planting density. 4. RESULTS The results w i l l be presented in four main parts; each part w i l l consider the effects of planting density and harvest time on different aspects of growth and yield of forage corn. Primary characteristics of growth w i l l be summarized in Section (4.1), some derived indices of growth and the analysis of those indices by plant growth analysis w i l l be presented in Section (4.2). Section (4.3) w i l l present and analyze the yield components derived from primary growth characteristics, and Section (4.4) w i l l outline the analysis of relative growth rate of yield using the relative growth rates of yield componenets for each planting density. 4.1 Primary Characteristics of Growth Primary growth characteristics are summarized in Tables (la) and (lb). In general, the primary characteristics increased with plant age un t i l f u l l development was achieved, and they usually decreased with increased number of plants per square metre. An exception to this pattern was stem height (T) which increased with increased planting density. Table (2) indicates that the effects of planting density and harvest time on a l l primary plant characteristics were highly significant (p<.001). The responses of individual characteristics to treatments were further analysed by describing the time and density trends using cubic spline regressions (Figs. 1-7). Such analysis was applied to most of the individual characteristics listed in Tables (la) and (lb). Several - 18 -Table l a Y ie ld (per plant) of Primary Plant Growth cha rac t e r i s t i c s of forage corn at four p lant ing dens i t i es and f i v e harvest t imes. Plant ing Density Plant Growth Charac te r i s t i c s Age T L ) N A FNT ENT (Plants/m 2) (days) (cm) (cm) (# (dm2) (#) (#) 4.9 21 1.81 0 60 7 80 5.012 -2 _ (0.61)1 (0 18) (0.86) (2.16) - -42 25.40 2 11 11.44 36.04 0.06 0.06 (8.64) (0 32) (0.97) (5.96) (0.33) (0.33) 63 156.40 2 50 10 17 55.02 2.56 1.58 (27.70) (0 23) (1.46) (6.73) (0.81) (0.77) 85 194.70 2 64 7 86 46.48 3.25 2.25 (22.08) (0 25) (0.87) (8.31) (1.18) (1.18) 115 195.60 2 79 6 86 38.54 3.36 2.36 (19.79) (0 18) (1 44) (8.99) (1.05) (1.05) 6.7 21 1.62 0 55 7 28 4.16 - -(0.42) (0 I D (0.66) (1.20) - -42 33.00 2 16 11 11 38.37 - -(9.48) (0 30) (1.45) (5.73) - -63 168.30 2 36 9 75 49.64 2.08 1.11 (17.21) (0 25) (1 27) (8.00) (0.77) (0.71) 85 201.80 2 50 7 69 42.94 3.00 2.00 (27.14) (0 25) (1 39) (8.39) (0.89) (0.89) 115 196.40 2 63 6 44 36.20 2.58 1.58 (21.97) (0 20) (1 23) (7.96) (0.77) (0.77) 8.8 21 1.80 0 58 7 .39 4.67 - -(0.48) (0 13) (0 87) (1.88) - -42 27.69 2 03 10 97 33.81 - -(10.80) (0 27) (0 91) (5.77) - -63 153.70 2 15 9 .11 47.34 1.44 0.50 (21.74) (0 20) (1 24) (7.00) (0.65) (0.61) 85 194.90 2 32 6 89 36.30 2.56 1.56 (24.28) (0 24) (1 09) (8.17) (0.88) (0.88) 115 194.20 2 38 5 89 31.13 2.17 1.17 (17.92) (0 24) (1 30) (6.47) (0.45) (0.45) 11.1 21 1.98 0 58 7 22 4.84 - -(0.56) (0 14) (0 80) (2.12) - -42 30.00 1 90 11 19 33.50 - -(9.98) (0 26) (1 19) (6.85) - -63 163.10 2 05 8 75 44.30 1.50 0.50 (25.29) (0 21) (0 97) 5.77) (0.61) (0.61) 85 200.20 2 15 7 03 34.05 2.06 1.06 (20.69) (0 27) (0 97) (6.78) (0.48) (0.48) 115 200.10 2 29 5 50 27.12 2.14 1.14 (23.92) (0 16) (1 06) (6.51) (0.49) (0.49) Mean-* 117.13 1 96 8 32 32.50 1.44 0.843 S.D. 86.57 0 76 (2.13) (16.64) (1.39) (1.011) S i gn i f . *** *** *** *** **# *** values in brackets represent the standard deviat ions represent unrecorded data d i s t r i bu ted throughout the f i v e harvests s i gn i f i c an t at P=.001 - 19 -Table lb • Y ie ld (g/plant) of Primary Plant Growth cha rac t e r i s t i c s of forage corn at four p lant ing dens i t i es and f i v e harvests t imes. P lant ing Density Age (Plants/m 2 ) (days) Plant Growth Charac te r i s t i c s LS 'ST "TL "TA 4.9 6.7 8.8 11.1 21 42 63 85 115 21 42 63 85 115 21 42 63 85 115 21 42 63 85 115 1.55 (0 .82 ) 1 16.38 (4.32) 30.99 (5.39) 31.56 (5.68) 31.92 (5.73) 1.26 (0.44) 17.56 (4.07) 27.91 (4.16) 29.56 (4.65) 29.17 (4.71) 1.44 (0.75) 14.89 (3.46) 25.24 (3.61) 26.32 (5.02) 25.21 (3.35) 1.54 (0.77) 14.15 (3.40) 23.83 (2.98) 23.14 (4.70) 24.01 4.53 0.37 (0.23) 5.47 (1.72) 18.53 (4.69) 20.34 (3.67) 18.17 (3.43) 0.35 (0.38) 6.01 (1.52) 16.25 (3.15) 18.56 (3.70) 15.89 (2.76) 0.32 (0.17) 4.97 (1.42) 14.06 (2.57) 16.23 (3.30) 14.14 (2.31) 0.332 (0.19) 4.94 (1.27) 12.66 (2.23) 13.36 (3.06) 12.93 (2.17) 0.12 (0.11) 5.52 (3.24) 43.63 (14.12) 82.91 (19.62) 73.46 (19.02) 0.09 (0.06) 7.00 (3.16) 38.59 (9.18) 69.99 (18.49) 55.51 (17.39) 0.17 (0.22) 5.19 (2.75) 30.62 (8.23) 58.85 (14.49) 43.17 (10.12) 0.13 (0.09) 5.56 (2.77) 27.45 7.42 49.06 (12.45) 42.50 (10.50) 1.56 (1.92) 3.70 (6.61) 2.91 (6.38) 1.58 (2.09) 0.90 (1.78) 0.98 (2.13) 0.28 (0.68) 0.24 (0.61) 0.05 (0.20) 0.03 (0.09) 0.03 (0.20) 0.00 0.02 8.48 4.41 - -(2.23) (4.47) - -5.73 16.00 30.29 -(1.10) (9.57) (14.50) 5.52 26.91 42.81 107.80 (1.29) (7.46) (18.93) (25.79) 7.68 2.92 - -(2.34) (2.97) - -4.98 11.62 21.49 -(1.37) (6.22) (9.74) -4.69 20.43 24.19 86.22 (0.99) (4.11) (6.88) (16.85) 7.04 0.69 - -(1.77) (1.11) - -4.35 7.94 12.75 -(1.03) (6.51) (8.19) -3.93 15.40 16.75 62.17 (1.07) (4.76) (5.71) (27.03) 6.25 0.37 - -(1.87) (0.57) - -3.56 4.20 6.99 -(1.17 (3.63) (5.28) -3.58 11.03 14.89 45.17 (0.60) (4.66) (6.20) (25.17) 3.28 6.10 8.51 15.07 (3.15) *** (8.90) * * * (14.00) * * * (33.72) * * * Mean3 S.O. S i g n i f . 19.88 (11.26) 10.69 31.98 (7.32) (29.10) 0.61 (2.45) 1. 2. 3. * * * values in brackets represent the standard deviat ions represent unrecorded data d i s t r i bu ted throughout the f i v e harvests s ign i f i c an t at P=.001 - 20 -Table 2 E f fec t s of P lant ing Density and harvest time on the Primary Growth Charac te r i s t i c s of Forage Corn. Treatment Means Plant Growth Cha rac te r i s t i c ~T~. P lant ing Density (Plants/m^) (4.9) (6.7) • (8.8) (11.1) Mean S i gn i f . T (cm) D (cm) L N L f t (dm2) 114, 2, 8. .80 .13 .83 120.20 2.04 8.46 36 .22 34.36 FNT ENT wL (g) 1. 1 22 .84 .25 .48 1.53 0.94 21.09 w L S (g) 12 .57 11.41 w S T (g) 41 .13 34.24 w T L (g) 1 .63 0.69 WTA (9) 3 .91 3.47 WH (9) 9 .46 6.99 wc (g) 14 .62 9.14 WG (g) 21. .56 17.24 114, .50 119, .10 117, .13 *** 1. .89 1. .79 1. ,96 *** 8, .05 7. .94 8. .32 *** 30, .65 28, .76 32, .50 *** 1 .23 1, .14 1. .44 *** 0, .64 0, .54 0, .84 *** 18 .62 17, .34 11, .26 *** 9, .94 8, .85 7, .32 *** 27, .60 24, .94 31, .98 *** 0, .11 0. .01 0, .61 *** 3, .06 2, .68 3, .28 *** 4, .81 3, .12 6, .10 *** 5, .90 4, .38 8, .51 *** 12, .43 9, .03 15, .07 *** - 21 -Table 2 (continued): B. Harvest Time (Age in days). (21) (42) (63) (85) (115) Mean T (cm) 0 (cm) L N L A (dm2) FNT ENT wL (g) 1. 0, 7. .80 .58 .42 29. 2, 11. .02 .05 .18 160. 2, 9. ,40 .26 .44 197.90 . 2.40 7.37 196.60 2.53 6.17 117. 1. 8. .13 .96 .32 *** *** *** 5. .02 36. .04 55. .02 46.48 38.54 32. .50 *** 0. 0. 1. .00 .00 .45 0, 0 15 .01 .01 .75 1. 0. 27. .90 .92 .00 2.72 1.72 27.65 2.56 1.56 27.58 1. 0. 11. .44 .84 .26 *** *** *** WLS ( g ) wS T (g) wTL (g) wTA (g) 0, .34 5 .35 15, .37 17.12 15.28 7. .32 *** 0. .13 5 .82 35 .07 65.20 53.66 31, .98 *** 0 .00 0 .86 1. .22 0.99 0.00 0. .61 *** .1 7. .36 4.65 4.38 3. .28 *** wH (g) Wc (g) 2, .10 9.94 18.44 6. .10 *** 0 .00 17.88 24.66 8 .51 *** wG (g) - 75.34 15. .07 *** 1. - Values not recorded * * * - S ign i f i c an t at P-.001 according to F-test - 22 -of the dry weight components (V-r^. ^TA' H^* a n d ^G^' however, are not shown individually but are pooled in the total shoot dry weight regressions (Fig. 8). Regressions were also not developed for total number of flowers/plant (FNT) and total number of ears/plant (ENT), since those characteristics were not determined at a l l harvests. The effects of harvest time on primary plant characteristics were always large and for most of the characteristics, density effects were large at later harvests. However, the effects of density on the growth in height (T) of stems were not very large at any time during growth. Thus, while a l l primary characteristics were significantly affected by planting density (Table 2), the magnitude of the effects were variable among the characteristics (Figs. 1-7). Planting density did not greatly affect the early exponential phase of growth of the plant characteristics (Figs. 1-7). The reproductive growth characteristics were not recorded un t i l at the age of 42 days after emergence (DAE) (Table la). Early dry weights of the undeveloped ears were negligible and were not separately Included in the table of dry weights of the ear characteristics (Table lb) at the age of 42 DAE. Undeveloped tassels were also observed at the second harvest (42 DAE), but their separation for individual records was not feasible at that time. Their dry weight values were included with the dry weights for stems at 42 DAE. Also, undeveloped cobs were included in the husk dry weight (WH) at the age of 63 DAE (Table lb). Similarly, the undeveloped grains on the cob were not separated from the dry weights of the cobs (Wc) at the age of 85 DAE (Table lb). - 23 -P l a n t age ( d a y s ) . Figure 1 - Cubic spline regressions describing the time course of plant height (T) in corn at four planting densities. - 24 -Figure 2 - Cubic spline regressions describing the time course of plant diameter (D) in corn at four planting densities. - 25 -Figure 3 - Cubic spline regressions describing the time course of number of leaves/plant (Lj^) i n corn at four planting densities. , 1 I I I 20 40 60 80 100 120 Plant age (days) Figure 4 - Cubic spline regressions describing the time course of leaf area/plant (L A) in corn at four planting densities. - 27 33.0-, 30. (H Plant age (days) Figure 5 - Cubic spline regressions describing the time course of dry weight of leaves/plant (WL) in corn at four planting densities. - 28 -24.0_ Figure 6 - Cubic spline regressions describing the time course of dry weight of leaf sheaths/plant (Wj^ g) in corn at four planting densities. - 29 -4 .9 plants/r 6.7 plants/m 2 8.8 plants/m 2 11.1 plants/n Plant age (days) Figure 7 - Cubic spline regressions describing the time course of dry weight of stem/plant (W S T) in corn at four planting densities. - 30 -Figure 8 - Cubic spline regressions describing the time course of dry weight/plant (W) In corn grown at four planting densities. - 31 -Therefore, grain yields (WQ) w e r e recorded only once, although the recorded grand mean was distributed throughout the harvests (Table lb). It is evident that the reproductive characteristics were more strongly affected by planting density than were many of the vegetative characteristics (Table 2). The curves in Figure (8) show nearly linear trends with age up to the final harvest, indicating that shoot growth continued throughout the study at a l l densities. Ear growth was a major contributor to overall shoot growth at the end of the season. High variability was observed among means recorded from each density, especially toward the reproductive growh period (Table 3). Throughout plant growth total shoot EM production per hectare increased (Table 4). The maximum recorded shoot yield per hectare was 17.1 MT/ha, which was observed from the highest planting density (11.1 plants/m ) at 115 DAE. Thus, despite the decreasing effects of density on many growth characteristics, yield per land area increased slightly with increasing planting density. The average yield/ha for the f i r s t 3 densities were very slightly different from each other (Table 4). These values remained virtually constant despite the observed decreasing 2 trends of shoot dry matter per plant with increasing number of plants/m 2 (Table 4). Thus, increasing the number of plants/m from 4.9 to 8.8 did not affect strongly the yield/ha but strongly reduced the performance of individual plants (Tables 3 and 4; Fig. 8). 4.2 Plant Growth Indices Plant growth indices commonly used in plant growth analysis - 32 -Table 3 Shoot dry matter y i e l d (g/plant) of forage corn at four p lant ing dens i t i es and f i v e harvest t imes. P lant ing Density Harvest Time (days) (Plants/m 2) 1 2 3 4 5 Mean S ign i f . (21) (42) (63) (85) (115) 1. (4.9) 2.04 28.93 109.70 189.70 306.40 127.40 *** ( l . l l ) 1 (9.92) (24.45) (48.72) (61.83) (117.20) 2. (6.7) 1.69 32.15 94.26 157.20 236.10 104.30 *** (0.75) (9.41) (18.25) (36.91) (37.60) (88.69) 3. (8.8) 1.92 25.33 77.89 126.50 180.80 82.48 *** (0.94) (7.65) (15.02) (32.79) (43.43) (70.24) 4. (11.1) 2.00 24.68 70.59 100.30 154.10 70.34 *** (1.02) (7.05) (12.67) (25.41) (36.12) (58.08) Mean 1.95 27.77 88.12 143.40 219.30 96.12 S.D. (0.96) (9.02) (23.52) (49.63) (74.07) (89.01) S i gn i f . N.S. *** *** *** *** 1. - values in brackets are the standard deviat ions * * * - s i gn i f i c an t at p robab i l i t y (P-.001) - 33 -Table 4 Shoot dry matter y i e l d (MT/ha) of forage corn at four p lant ing dens i t i es and f i v e harvest t imes. P lant ing Density (Plants/m 2) 1. (4.9) 2. (6.7) 3. (8.8) 4. (11.1) Mean S.D. S i gn i f . 1 (21) 0.10 (0 .05) 1 0.11 (0.05) 0.17 (0.08) 0.22 (0.11) 0.15 (0.07) N.S. Harvest Time (days) 2 (42) 1.43 (0.49) 2.16 (0.63) 2.22 (0.67) 2.74 (0.78) 4.28 (0.65) 3 (63) 5.42 (1.21) 6.34 (1.23) 6.84 (1.31) 7.84 (1.41) 6.61 (1.70) 4 (85) 9.37 (2.41) 10.54 (2.49) 11.11 (2.88) 11.15 (2.83) 10.54 (3.58) 5 (115) 15.13 (3.05) 15.87 (1.86) 15.87 (3.81) 17.12 (4.01) 16.00 (5.33) 1 - values in brackets represent the standard deviat ions * * * - s i gn i f i c an t at p robab i l i t y (P<.001) N.S. - not s i gn i f i c an t Mean 6.29 (5.79) 7.00 (5.92) 7.24 (6.18) 7.82 (6.45) 9.39 (6.41) S ign i f . - 34 -(Evans, 1972) which w i l l be considered in this section include: crop growth rate (CGR), leaf area index (LAI), unit leaf rate (ULR), relative growth rate of shoot biomass (Ry), leaf area ratio (LAR), leaf weight ratio (LWR) and specific leaf area (SLA). Also, the relative growth rates of individual primary plant growth characteristics were calculated. However, only the relative growth rates of stem dry weights (Ryg T), leaf dry weights (R^x,) and those of leaf area (RLA) in each plant density w i l l be considered in these results, since those characteristics exhibited large density responses (Section 4.1). The effects of plant age and planting density on shoot biomass yield per unit land area (Table 4) must reflect changes in crop growth rate during this experiment. The variability of CGR throughout the growing season could be divided into 3 phases. Phase one, which covers the period between 21 DAE to about 55 DAE, showed increased CGR with 2 increased number of plants/m . However, there was relatively l i t t l e difference in the CGR between densities 2 (6.7 plants/m2) and 3 (8.8 2 plants/m ) during this growth period. The difference of CGR between the f i r s t density (4.9 plants/m ) and the fourth density (11.1 plants/m ) was very large (Fig. 9). During the second phase, which covers the period from about 56 DAE to 98 DAE, the trends in crop growth rate fluctuated and overlapped (Fig. 9). The third phase which covers the period from about 99 DAE to harvest time again showed well defined responses to planting density. In this phase, densities 1 and 4 had steadily increasing values of CGR. The CGR values in densities 2 and 3 were decreasing in most of this period, but exhibited slight increases - 35 -Figure 9 - Cubic spline regressions describing the time course of crop growth rate (CGR) in corn at four planting densities. - 36 -toward maturity (115 DAE; Fig. 9). CGR i s the product of leaf area index (LAI) and unit leaf rate (ULR), so the variability of CGR depends on the variability of LAI and ULR (Figs. 10 and 11). The relative variability of LAI over time and among densities was about six times the variability of ULR. Moreover, LAI increased steadily both between densities and age during phase 1 of CGR (Fig. 10) while ULR increased only slightly with plant age and decreased in both densities later in this phase (Fig. 11). As LAI was decreasing in the period above 55 DAE to 98 DAE, ULR values flactuated with age in the same period (Figs. 10 and 11). In this period the differences among densities were large for both LAI and ULR, but the increase in LAI with decreasing densities (Fig. 10) was counteracted by the decrease in ULR. The values of ULR were increasing in densities 1 and 4 at the end of phase 2, which might have caused the observed increases of CGR in those densities in phase 3. The pattern of vari a b i l i t y of ULR among densities during this period resembled that of CGR. LAI steadily declined after the age of 55 DAE (Fig. 10) and thus could not have contributed to the observed increasing values of CGR in phase 3. The relative growth rate of shoot biomass per plant (Ry) did not show large responses to the planting density (Fig. 12). While the 6.7 plants/m 2 density had the highest Ry in the early vegetative growth period, this difference did not persist in later growth periods. At a l l densities R^  declined rapidly during the f i r s t 70 days of growth. - 37 -Plant age (days) Figure 10 - Cubic spline regressions describing the time course of leaf area index (LAI) in com at four planting densities. - 38 -0.130 -i 0. 105 ro XI 0.080 ro = 0.055H 0.030 Phase 3 4.9 'plants/r lants/m 2 olants/m2 plants/m2 Plant age (days) Figure 11 - Cubic spline regressions describing the time course of unit leaf rate (ULR) in corn grown at four planting densities. - 39 -Figure 12 - Cubic spline regressions describing the time course of relative growth rate of plant dry weight (Ry) in corn at your planting densities. ^ 40 •-. The relative growth rate (R w) is the product of LAR and ULR, so v a r i a b i l i t y of Ry during growth at various densities depended upon the v a r i a b i l i t y of LAR (Fig. 13) and ULR (Fig. 11). The overall relative variability in LAR in this study was about eight times the v a r i a b i l i t y of ULR (Figs. 11 and 13). The relatively high i n i t i a l value of Rw at a density of 6.7 plants/m (Fig. 12) appears to have been caused by the relatively high i n i t i a l ULR at that planting density (Fig. 11). Similarly, the high Ry value at 4.9 plants/m 2 from 40 to 70 days, and for 4.9 and 11.1 plants/m2 at the end of growth (Fig. 12), also seem to have depended on high ULR values (Fig. 11). LAR varied greatly over the course of plant growth, but was not strongly affected by planting density (Fig. 13). However, i t is clear that the declining values in Ry during the course of growth were strongly driven by corresponding declines in LAR. The two components of LAR, leaf weight ratio (LWR; Fig. 14) and specific leaf area (SLA; Fig. 15) were also evaluated. LWR was more responsive to planting density than SLA, which did not seem to vary systematically with planting density throughout most of the plant growth 2 (Fig. 15). LWR responded to increased number of plants/m by exhibiting higher values in higher densities, especially during the late vegetative to maturity growth period (63 to 115 DAE; Fig. 14). Density effects, however, did not alter greatly the time course exhibited by the spline regressions for either LWR or SLA. Both SLA and LWR contribute to the decline in LAR during growth. As with Ry» the relative growth rates of the other primary - 41 -Figure 13 - Cubic spline regressions describing the time course of leaf area ratio (LAR) in corn at four planting densities. - 42 -0.80 0.70 0.60 0.50 00 oo CO 1-1 ob 0.40 •H CU 3 ro QJ 0.30 0.20 11.1 p l a n t s / m 2 8.8 p l a n t s / m 2 ,6 .7 p l a n t s / m 2 .4.9 p l a n t s / m 2 0.10 Figure 14 -60 80 P l a n t age (days) Cubic spline regressions describing the time course of leaf weight ratio (LWR) in corn at four planting densities. 120 - 43 -Figure 15 - Cubic spline regressions describing the time course of specific leaf area (SLA) in com at four planting densities. - 44 -plant growth characteristics showed small responses to planting density (Figs. 1 6 - 1 8 ) . The relative growth rates of leaf area (RL/V F I G * » leaf dry weight ( R ^ J Fig. I 7) and stem dry weight (R W S T5 Fig. 18) showed similar trends in time although they had different quantitative values. The relative growth rate curves generally lacked strong responses to density treatment. From these results, i t is therefore evident that the variability which was visible among different primary growth characteristics as a result of planting density could not clearly be seen in the relative growth rates of those characteristics. The relative growth rates of individual primary characteristics therefore seem to be of l i t t l e value in the interpretation of planting density responses. The time course of spline regression curves of ULR's responded more to plant aging effects than planting density. However, some of the other plant growth indices, eg. CGR, LAI, ULR, LAR and LWR, do reveal the responses to planting density in considerable details (Figs. 9-11, 13 and 1 4 ) . 4.3 Yield Component Analysis Further analysis of the morphological basis of dry matter production in forage corn was based upon the yield component analysis equation described earlier (Chapter 3 ) , which is repeated below. The morphological components were selected according to the chronological development of the plant. Thus, W was related to a set of morphological growth characteristics which included T, D, L^» 1>A a n < i W L (Table 5 ) . As detailed in Chapter 3, these characteristics were converted into - 45 -Figure 16 Cubic spline regressions describing the time course of relative growth rate of leaf area per plant (\p) in corn at four planting densities. ~ 1 — AO 60 — r 80 — r ~ 100 - i 120 Plant age (days) Cubic spline regressions describing the time course of relative growth rate of leaf dry weight per plant (R^) in corn at four planting densities. Figure 18 - Cubic spline regressions describing the time course of relative growth rate of stem dry weight/plant (RWST^ i n c o r n a t f o u r P l a n t i n S densities. - 50 -a series of ratios (Table 6) which were then transformed into natural logarithms and analyzed by analysis of variance and two-dimensional forward and backward yield component analysis (TDP-YCA; Appendix II). Thus, W = T x D/T x I^/D x L A / L N x \ / ^ A x W/ (8) Xi The forward and backward TDP-YCA may be used to identify the yield components responsible for yield variability. A l l of the variability in yield can be accounted for by this analysis. The overall forward TDP-YCA results showed that yield v a r i a b i l i t y was largely due to the components T(95.32%), D/T (1.49%) and W/WL (1.96%) (P<.001). In the other dimension, harvest time (95.65%), planting density (0.71%) and density x harvest (0.38%) were a l l significant (P<0.01; Table 7). The effects of planting density, harvest time, and density x harvest time on yield were consistent for both the forward and backward TDP-YCA. The effects of those treatments on individual yield components, however, varied (Tables 7 and 8). While T, D/T and W/WL were the significant yield components in the forward TDP-YCA, L A / L N (15.32%), WL/LA (6.26%) and W/WL (77.35%) were the significant yield components in the backward TDP-YCA (Table 8; P<.001). A l l the significant yield components in both analyses were significantly affected by planting density, except for L A / L N which was not significantly affected by this treatment in the backward TDP-YCA (Tables 7 and 8). The non-significant components in the forward analysis (i.e., L^/D, L A / L N a n d w L / L A ) were - 48 -Table 5 Mean observations from ind iv idua l morphological cha rac t e r i s t i c s of each densi ty and harvest of forage corn. P lant inq Harvest 1 Height Diameter Leaves Leaf area Leaf weight Yield Density (T) l (D) (L N ) (L A ) (WL) (W) (plants/m 2 ) cm cm # dm2 g 9 4. 9 1 1. 81 0. 60 7. 81 5. 16 1. 55 2. 04 2 25. 40 2. .11 11. .44 36. ,04 16. .38 28. ,39 3 156. 40 2. 50 10. .17 55. 02 30. 39 109. 70 4 194. .70 2. .64 7. ,86 46. .48 31. .56 189. ,70 5 195. .60 2. 79 6. ,86 38. ,54 31. 92 306. .40 6. .7 1 1. ,62 0, .55 7, .28 4, .16 1. .26 1. .69 2 33. .00 2. ,16 11. ,11 38. .87 17. .56 32. 15 3 168. .30 2, .36 9, .75 49, .64 27. .91 94 .26 4 201. .80 2. ,50 7. .69 42. .94 29, .56 157. .20 5 196. .40 2, .63 6 .44 36 .20 29, .17 236 .10 8 .8 1 1. .80 0, .58 7 .39 4, .67 1. .44 1 .92 2 27 .69 2 .03 10 .97 33 .81 14, .89 25 .33 3 153, .70 2, .15 9 .11 43, .34 25, .24 77 .89 4 194 .90 2 .32 6 .89 36 .30 26 .32 126 .50 5 194. .20 2 .38 5 .89 31, .13 25, .21 180 .50 11 .1 1 1. .98 0 .58 7 .22 4 .84 1 .54 2 .00 2 30 .00 1. .90 11 .19 33 .50 14 .15 24 .68 3 163 .10 2 .05 8 .75 44 .30 23 .83 70 .59 4 200 .20 2 .15 7 .03 34 .05 23 .14 100 .30 5 200 .10 2 .29 5 .50 27 .12 24 .01 154 .10 Grand Mean 117 .13 1 .96 8 .32 32 .50 19 .88 64 .75 1 - A l l var iab les expressed on a perplant bas i s . - 49 -Table 6 Mean Y ie ld component values for forage corn at each densi ty and harvest t ime. P lant ing Density (plants/m 2 ) Harvest 4.9 6.7 11.1 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 cm/plant cm/pi 1.81 25.40 156.40 194.70 195.60 1.62 33.00 168.30 210.80 196.40 1.80 27.69 153.70 194.90 194.20 1.98 30.00 163.10 200.20 200.10 D/T cm/cm 0.35 0.09 0.02 0.01 0.01 0.36 0.07 0.01 0.01 0.01 0.33 0.08 0.01 0.01 0.01 0.30 0.07 0.01 0.01 0.01 LN/D #/cm 13.96 5.51 4.11 2.99 2.46 13.79 5.18 4.18 3.12 2.46 13.22 5.47 4.28 2.98 2.49 13.12 6.04 4.32 3.33 2.40 L A / L N dm/# w L /L A w/W. 0.63 3.15 5.47 5.94 5.63 0.57 3.54 5.11 5.64 5.65 0.63 3.09 5.24 5.26 5.36 0.66 3.21 5.09 4.88 4.96 g/dm2 g/g g/plant 0.30 0.45 0.57 0.68 0.85 0.31 0.45 0.57 0.70 0.83 0.30 0.44 0.54 0.75 0.84 0.31 0.42 0.54 0.69 0.92 1.31 1.73 3.54 5.99 9.61 1.35 1.82 3.38 5.29 8.15 1.40 1.68 3.08 4.78 7.17 1.30 1.72 2.97 4.34 6.44 9/pl 2.04 28.93 109.70 189.70 306.40 1.69 32.15 94.26 157.20 236.10 1.92 25.33 77.89 126.50 180.50 2.00 24.68 70.59 100.30 154.10 Grand mean 117.13 0.09 5.77 3.98 0.57 3.85 64.75 - 51 -Table 7 Forward - YCA. "Two-dimensional pa r t i t i on ing of the to ta l sum of squares for y i e l d expressed as percentages - data from a l l observat ions. T T e T d Component or Product-Blocks % Density X -Harvest Den. % x Har. % Error % Total C1C1 (T) C 2 C 2 (D/T) C3C3 (LN/D) 0.44*** 0. 03** 93. 67** 0. 07*** 1. 11 95.32*** 0.06*** 0.12*** 0. ,89*** 0. ,02*** 0.41 1.49*** 0.01** 0. ,00 0. ,39*** 0. ,01*** 0. ,35 0.76 C4C4 (L A /L N ) 0.00 0. ,00 0. .00 0. ,00 0. ,01 0.01 C5C5 (W L /L A ) 0.02** 0. .00 0. ,00 0. .00 0. .44 0.46 C 6 C 6 (w/wL) 0.04*** 0. .36*** 0. .40*** 0. .19*** 0. .96 1.96*** C l C 2 C1C3 c l c 4 C1C5 C i C 6 C 2 C 3 C 2 C 4 C 2 C 5 C 2 C 6 C3C4 C3C5 C 3 C 6 C4C5 C4C6 C5C6 0.24 -0. .07 -0. .11 0, .02 -0, .08 0.00 -0.00 -0, .00 0, .24 -0.01 -0 .19 0.00 0.01 -0, .00 -0. .03 -0 .00 0, .01 -0.00 0.07 -0. .01 0, .01 0 .00 -0 .08 -0.00 0.11 -0. .12 0, .17 0, .05 -0, .21 0.00 -0.01 -0 .01 -0 .27 -0 .00 0 .29 0.00 0.01 0. .00 -0, .01 0 .00 -0. .00 0.00 0.04 0 .01 0 .01 0 .00 -0 .06 -0.00 0.02 0 .41 -0, .60 0 .08 0, .09 -0.00 -0.00 -0 .00 0.03 0 .01 -0 .03 0.00 -0.02 -0 .00 0, .05 -0 .05 -0 .04 0.00 -0.04 -0.03 0 .73 -0 .00 -0 .61 0.00 0.00 -0 .00 0, .00 0 .00 -0 .01 0.00 -0.00 0 .01 0 .03 -0 .00 -0 .03 0.00 0.02 0 .02 0 .05 -0 .02 -0 .07 0.00 Ln Y 1.01*** 0 7 ]_*** 95 .65*** 0 .38*** 2 .25 100.00*** TOTAL SS (100) = 2247.6 C5 and LnY are y i e l d components 1 to 6 and y i e l d (w). * t * * ( * * * . s i gn i f i c an t at P = 0.05, 0.01 and 0.001. - 52 -Table 8 Backward - YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages - data from al l observations. talent B l^cTsDensi ty Harvest Den. x Har. Error Total or Product X * * CjCi (T) C 2 C 2 (D/T) C3C3 (LN/D) C4C4 (LA/LN) C5C5 ( V - A * c6c6 (w/wL) ClC 2 C1C3 C1C4 C1C5 ClC 6 C2C3 C2C4 C2C5 C2C6 C3C/4 C3C5 C3C6 C4C5 C4C6 C5C6 Ln Y 0.03*** 0.00 0.02*** 0.16*** 0.29*** 0.29*** 0.01 0.01 -0.06 0.15 0.07 0.01 -0.01 0.04 0.04 -0.01 0.02 0.02 -0.33 0.09 0.16 0.01** 0.02*** 0.01** 0.00 0.15*** 0.85*** 0.03 0.02 -0.01 -0.09 0.21 0.02 -0.01 -0.11 0.26 -0.01 -0.06 0.15 0.02 -0.06 -0.71 0.07*** 0.03*** 0.07*** 11.14*** 0.78*** 72.52*** 0.04 -0.05 1.32 0.37 -0.95 -0.08 -0.20 0.05 -0.53 0.51 -0.07 -0.12 5.00 0.67 5.09 0.01* 0.01*** 0.01 0.08 0.26* 0.64*** 0.00 0.01 -0.04 -0.03 0.08 0.01 0.00 -0.05 0.11 0.00 -0.09 0.13 -0.06 -0.03 -0.67 0.43 0.04 0.33 3.94 4.78 3.04 -0.09 0.01 -1.20 -0.41 0.59 0.04 0.22 0.07 0.12 -0.50 0.20 -0.18 -4.64 -0.67 -3.87 1.01*** 0.71*** 95.65*** 0.38** 2.25 TOTAL SS (100) - 2247.6 _—- C 6 and LnY are yield components 1 to 6 and yield (W). * t * * t *** . significant at P-0.05, 0.01 and 0.001 0.55 0.10 0.43 15.32*** 6.26*** 77.35*** 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00* - 53 -introduced into the multiple regression after most of the total variabi-l i t y in yield had already been accounted for by the earlier components, mainly by T ( 9 5 . 3 2 % ) . Thus they had l i t t l e opportunity to play a signi-ficant role as yield components in this analysis (Table 7 ) . In the other direction, the non-significant components in the backward analysis (i.e., T, D/T and L^/D) were introduced into the multiple regression after most of the total variability in yield had already been accounted for by the (biologically) later components, particularly W/WT, ( 7 7 . 3 5 % ) . Thus they had l i t t l e opportunity to play a significant role as yield components in this analysis (Table 8 ) . W^/LA (6. 2 6 % ) and (15.32%) recovered their significance in the backward analy-sis . The W/WL in the backward analysis was not as large as T (95.32%) in the forward analysis, thus permitting relatively higher contributions of the other yield components to the observed yield variability in the backward analysis. Comparing the results of the forward and backward analyses i t can be concluded that T and D/T may have had their significant contributions to yield variability originating from chronologically subsequent components. Also, i / L ^ and LA/^N "^Y have had their significant contributions to yield variability originating from chronologically earlier components. The fact that L^/D did not account for yield variability in either direction suggests that i t s contribution to yield variability was always assumed by the significant yield components, or i t truly did not contribute to variation in yield. Similar methods of interpretation are applicable for a l l the TDP-YCA analyses that w i l l be presented later in the text. - 54 -Tables (7) and (8) were obtained from the pooled data from a l l planting densities and harvests. Further steps were taken to analyse the effects of density on yield components. The further analyses were essential because of the overwhelmingly large effects of plant age on the variability of yield and i t s components (Tables 7 and 8). One approach used for the further analysis was to describe the time course of yield components at different planting densities using fitted cubic spline regressions. Graphical presentation of the yield component regressions (Figs. 1 and 19-23) showed that, apart from W/WL (Fig. 2 3 ) , the other yield components had no obvious systematic responses to the planting density treatment (although an analysis of variance had earlier indicated significant effects of planting density on a l l the yield components (P<.01; ANOVA Table not presented)). The small though significant effects of density on D/T (P<.01) were evident in Figure ( 1 9 ) and the effects appear to be limited to the early stages of growth. The ranking of D/T among the different densities, however, varied during the stages of growth (Fig. 1 9 ) . Conversely the effects of planting density on L^/D were small during the early vegetative stage of growth and were more pronounced until the f i n a l harvest (Fig. 2 0 ) . The ranking of Lfl/D among the different planting densities, however, also varied during the course of growth. Planting density had larger effects on L A / L N especially in the later stages of plant growth (Fig. 2 1 ) . Throughout that period, L A / L N generally tended to decrease with increasing planting density and this effect had not been previously identified by the TDP-YCA. Planting density had no well - 55 -Figure 19 - Cubic spline regressions describing the time course of D/T-yield component of forage corn at four planting densities. - 56 -- P l a n t age (days) Figure 20 - Cubic spline regressions describing the time course of L^/D-yield component of forage corn at four planting densities. - 57 -4.9 plants/m2 P l a n t age (days ) Figure 21 - Cubic spline regressions describing the time course of L A / L N - y i e l d component of forage corn at four planting densities. - 58 -Figure 22 - Cubic spline regressions describing the time course of W L/L A-yield component of forage corn four planting densities. Figure 23 - Cubic spline regressions describing the time course of W/WL~yield component of forage corn at four planting densities. - 60 -defined effects on the variability of W L/L A (Fig. 22). Figure (23) reveals the large and systematic effects of planting density on W/WL> which was also evident in the TDP-YCA (Table 7). The time course of the fitted cubic spline regression curves were similar to those of W (Fig. 8), although they differed quantitatively. The resemblance of the W/WL and W time course curves suggests the importance of this component in accounting for the observed yield variability among the planting densities. Another approach used to define the effects of density on yield components of forage corn involved the application of TDP-YCA analysis at each harvest. The results of both the forward and backward analyses are summarized in Table (9). The effects of planting density on yield (W) were not significant during early vegetative growth according to both forward and backward analyses (Table 9). Presumably this was because the plants were s t i l l small and had not started to interfere strongly with each other. However, after 21 DAE, the density effects on yield were highly significant (Table 9). The percentage variability among densities increased with harvest time and plant size (Appendices IV-XIII). In the forward TDP-YCA, only %/D was not a significant yield component in the f i r s t two harvests. In the third harvest, L^/B and L A/Lfj did not contribute significantly to the variability in yield. LJJ/D, L A/Lfj and WL/L a were insignificant yield contributors in the fourth harvest. And, in the f i f t h harvest, only L ^ ^ N A N C* - 61 -Table 9 21 42 63 85 115 21 42 63 85 115 Y ie ld components that were s i gn i f i c an t at each harvest t ime, inc luding density e f f e c t s . Plant Age Plant ing (days) Density N.S * * * *** *** *** N.S * * * * * * * * * * * * Y ie ld Component D/T LN/D L A / L N ***+ ***+ ***+ *** *** N.S+ *** ***+ N.S N.S Forward TDP-YCA * * * N.S+ ***+ N.S ***+ N.S ***+ N.S ***+ *** Backward TDP-YCA ***+ N.S N.S N.S N.S+ N.S ***+ N.S+ N.S+ N.S N.S+ ***+ ***+ ***+ W L/L A *** ***+ *** N.S N.S * , * * * - S ign i f i can t at P-.05 and P=.001, r espec t i ve l y . + - inc luding s i gn i f i c an t e f f ec t s of p lant ing densi ty (P=.05). N.S. - Not s i gn i f i c an t (P :0.05) . w/w. *** *+ *** + ***+ ***+ *** *** N.S ***+ *** *** + *** *** ***+ ***+ N.S *** + *** N.S *** + - 62 -WT/LIA did not significantly contribute to the variability in yield. Thus, the general effects of planting density on yield components in the forward TDP-YCA were sporadic depending on age and type of yield component (Table 9). In the backward TDP-YCA, a l l components contributed to yield v a r i a b i l i t y in the third harvest. While L A / L N contributed to yield v a r i a b i l i t y at a l l harvests in the backward analysis, a l l the other yield components were significant only sporadically among the harvests. Also, while WL/L^ . w a s n o t affected by the planting density at a l l harvests, the other yield components (significant or non-significant contributors to yield variability) were affected by planting density at certain times (Table 9). It is evident from Table (9) that significant yield components in the forward analysis were not necessarily the significant yield components in the backward analysis. Also, the amount contributed by a yield component to variation in yield depended on i t s position in the YCA model for both the forward and backward analyses. Components included earlier in a model usually contributed relatively more to yield variability than later yield components (Appendices IV-XIII). Some yield components, however, appeared to be relatively stable in both the forward and backward TDP-YCA. For example, W/WL, which was insignificant only at the 21 DAE harvest in the backward TDP-YCA analysis. WL/L^ showed significant contributions to yield at same harvests in both the forward and backward TDP-YCA. Thus, i t can be suggested that the significant yield components in forage corn depend - 63 -not only on the age of the plant, but also on the direction of analysis of the YCA model. Also, the yield components that are significant at the same age in both forward and backward analyses are considered to be relatively stable contributors to yield variability than those components which had their significance showing up only in one direction of analysis. It is d i f f i c u l t in the present analysis of the yield components using TDP-YCA to attribute a relevant physiological role to the significant yield components which are affected by the planting density variation, because such attributes were not directly assessed by the analysis. 4.4 Rela t ive Growth Rates of Y i e l d Components Relative growth rate has long been used in plant growth analysis as an index of the efficiency of plant growth. Ordinarily, relative growth rate has been applied to dry matter accumulation, but i t has also been used as an index of performance of other plant characteristics (e.g. Hunt, 1982b). If V is a quantitative variate of a plant, then 1 d d(Log V) The more generalized use of relative growth rate leads to i t s application in assessing the performance of yield components. As J o l l i f f e et a l . (1983 (in preparation)) have shown, a yield component series may be transformed into an additive series of relative growth rates of yield components (e.g. see the previous Eqns. repeated below). - 64 -W = T x D/T x LJJ/D x L A/L N x W^/L^ x W/WL (12) Log e (W) = Log e(T) + Loge(D/T) + Loge(L^/D) + Log^L^/L^) + (13) Log E(W L/L A) + LogE(W/WL) And = R T + R D / T + R L N / D + \ A / m + RtyL/LA + VWL-*-* ( 1 4 ) Equation (14) has been shortened as * Y = j x R C i <10> The relationships among the relative growth rates of yield (Ry = Ry) and yield components were analysed graphically (Figs. 24-27) and by multiple regressions. At a l l densities there was a similar patern in the time course of the relative growth rates of yield and yield components (Figs. 24-27). The i n i t i a l l y high values of Ry were supported by positive relative growth rates for T, LA/IN, WL/^A a n <^ W/ WL> while the rates for D/T and LN/D were negative. The decline in Ry during the f i r s t 80 DAE was associated with large decreases in R-r a n d RLA/LN A N (* A decrease in Rn/x UP t o a D O u t 50 DAE. After about 80 DAE the relative growth rates of the yield components became relatively stable and small. Most of the components had positive relative growth rates during - 65 -Plant age (days) Figure 24 - Cubic spline regressions describing the time course of relative growth rate of yield (R w) in density one (4.9 plants/m ) and the relative growth rates of its yield components (RcDl^ * n forage corn. - 66 -0.20-, Figure 25 - Cubic spline regressions describing the time course of relative growth rate of yield (Ry) in density two (6.7 plants/m ) and the relative growth rates of i t s yield components ( R C D 2 ^ I N forage corn. - 67 -Figure 26 - Cubic spline regressions describing the time course of relative growth rate of yield (Ry) in density three (8.8 plants/m 2) and the relative growth rates of i t s yield components (RQD3^ * n forage corn. - 68 -Figure 27 - Cubic spline regressions describing the time course of relative growth rate of yield (Rjp in density four (11.1 plants/m ) and the relative growth rates of i t s yield components (Rr;n4^ * n forage corn. - 69 -that period, and this sustained a positive relative growth rate of yield. It is clear from Figs. (24) to (27) that the contributions of different yield components to Ry shift during growth. However, these graphs do not clearly reveal the manner in which planting density affects yield in forage corn. A supplementary analysis involved the development of multiple regressions of % as a dependent variable on the combined relative growth rates of a l l the yield components (Rrj's) throughout plant growth for each planting density. As suggested by Eqns. (8) and (13) the Rrj's accounted for 100% of the coefficient of determination of Rw in every density (Table 10). These multiple regressions confirmed that Ry was the sum of the relative growth rates of the yield components. The variability of Rw among densities, therefore, depended on the variability of Rrj's among densities. The regression coefficients alone, however, could not be used to explain differences of Ry and i t s components among the planting densities. Table 10 Regression coefficients of the relative growth rates of yield (Ry) on the relative growth rates of yield components of forage corn observed in four planting densities. Planting Denisty Constant (a) Regression Coefficient (b)) for Each Yield Component Coefficient of Determination R C 1 Rc2 Rc 3 RC4 Rc5 Rc6 4.9 6.7 8.8 11.1 0.0144*** 0.0100*** 0.0123*** 0.0112*** 0 478*** 0.086 -0.506*** 0.434** 0.217 -0.180 0 194 -0.129 -0.599** 0.681** 0.459** -0.223 0 488** 0.119 -0.250 0.502* 0.506* -0.313 0 742*** 0.401*** -0.286 0.412*** 0.096 -0.155 100*** 100*** 100*** 100*** * t » * t «** . significant at P<0.05, 0.01 and 0.001, respectively. 5. DISCUSSION The effects of planting density on the time course of growth and yield of corn were the main focus of this study. The planting densities used in this study ranged from 59% (49383 plants/ha) to 134% (111111 plants/ha) of the average planting density currently recommended for forage corn in the locality (83,000 plants/ha). The experiment has provided information on overall shoot dry matter yield and on many primary characteristics of growth. In addition,the primary results have been used to derive various indices of growth which help in the interpretation of the planting density effects as they arose during growth. A secondary aspect of the study is that i t allows some assessment of the value of different methods for analyzing plant growth (i.e., plant growth analysis and yield component analysis). Harvestable yield per hectare has always been the basis of agronomic planting density recommendations (Donald, 1963; Holliday, 1960; Iremiren and Milbourn, 1978; Remison and Lucas, 1982). Total 2 yield per hectare increased with increased number of plants/m overcoming the reduced yield per plant. Thus, the performance of individual plants, or the performance of individual plant parts, cannot solely be used to account for trends in yield per hectare with changing stand densities. It is well known that higher planting densities can have substantial yield advantages, but such yield advantages must be weighted against the i n i t i a l costs of crop establishment and subsequent f i e l d management costs. It was visually evident, however, that weed problems were reduced in the highest planting density plots in my study, - 72 -and the reduction in weed control requirements was one beneficial aspect of the high density. At the forage corn maturity stage (30.8% crop dry matter content), the variability among the different planting densities in shoot dry matter yield per hectare were not extreme. The yields ranged from 15.1 to 17.1 MT/ha from the lowest to the highest planting densities. The intermediate densities both yielded 15.9 MT/ha. In comparison, the average yield of corn (cv. DK 24) grown in local growing areas was 14.9 MT/ha (mean based on observations from 6 t r i a l locations; B.C. Ministry of Agriculture and Food, 1982). This average yield i s based on recommended planting densities for the early maturity corn cultivars (e.g. DK 24) which range from 75000 to 90000 plants per hectare. The specific densities in the t r i a l locations were not shown in the Field Corn Recommendations Summary (B.C. Ministry of Agriculture and Food, 1982). However, direct comparisons are limited by the planting configuration used in this study. Forage corn is generally grown closely spaced between plants within rows, but widely spaced (70-80 cm) between rows in order to achieve a desired planting density. The planting configuration used in my study was square (i.e. , equidistant spacing in and between rows) for each planting density. This planting configuration was chosen to maintain a consistent geometry among the plants in treatments at different planting densities. Therefore, the difference between the square planting pattern and the pattern commonly used in forage corn production must be kept in mind in - 73 -relating the results in this thesis to normal farming conditions. The increase in forage corn yield per hectare with increased planting densities suggests that i t may be worthwhile for farmers to use greater than currently recommended seeding rates where conditions of growing corn are good. It should also be noted, however, that this suggestion is based on only one season's results from one growing site. Crop growth rate is a widely used index of production efficiency in plant stands (Sestak et a l . , 1971). The relatively stable shoot dry matter yields obtained at the lowest three planting densities can be attributed to the similar value of crop growth rates (CGR) observed at those densities during the course of plant growth (Fig. 9). Compared to the next two densities, the lowest planting density had i n i t i a l l y lower values of CGR, up to about 55 DAE, and then relatively higher values toward crop maturity. The highest planting density on the other hand, maintained higher CGR both during vegetative growth (up to about 55 DAE) and at maturity, which could have accounted directly for the observed higher yields of shoot dry matter per hectare in this planting density (Fig. 9; Table 4). The CGR for the 2nd and 3rd planting densities were quantitatively similar at most times in the early vegetative growth (Fig. 9), a period during which vegetative dry weight is steadily generated by rapid crop growth rates. Warren Wilson (1981) partitioned CGR according to the following sequence: - 74 -CGR = LAI x ULR = ! x W x R w = E x W x ULR x LAR A = x x W x ULR x SLA x LWR (14) A It is clear that crop growth rate (CGR) may be dependent on a variety of indices, one of which is planting density (N/A). The contribution of the indices to the effects of planting density on CGR w i l l be discussed in turn. Leaf area index (LAI) is used to describe the extent of the assimilatory (i.e. photosynthetic) apparatus of a plant stand (Watson, 1952) and i t serves as a key index for interpreting variations in yield of plants grown at different stand densities. Unit leaf rate (ULR) Is an index of the efficiency of dry matter accumulation on a leaf area basis. ULR is strongly dependent on net photosynthesis rate, but is also dependent on the rates of respiration and inorganic accumulation. Both LAI and ULR were highly affected by the planting density but the effects varied during the growth of the crop (Figs. 10 and 11). Leaf area Index (LAI) increased strongly and systematically with increased number of plants per square metre (Fig. 10). This suggests - 75 -that LAI made an Important contribution to increased dry matter yield per hectare by increasing CGR values. Unit leaf rate (ULR) was always lowest in the highest planting density throughout most of the growing period of the crop (i.e., up to about 104 DAE), except the later values of ULR were higher at the lowest and highest planting densities than in the two intermediate planting densities toward crop maturity. Thus, although ULR tended to be reduced by increased planting density, this effect on CGR was overcome by increased assimilatory surface (LAI) of the crop stand. The relatively small differences In shoot yield and CGR observed at the lowest three densities seem to have been caused by responses in ULR and LAI which nearly cancelled each other. Relative growth rate of shoot biomass (Ry) is an index of the efficiency of dry matter accumulation per unit dry matter. As Eqn. (14) indicates Ry contributes to CGR in conjunction with total biomass density (W x N/A), and % in turn depends on the interactive effects of ULR and leaf area ratio (LAR). Ry was a more important determinant of the time course of dry matter accumulation in shoots than i t was at the response of shoot biomass productivity to planting densities (Fig. 12). It was observed that plants invested relatively more of their accumulated dry matter for self-generation during vegetative growth than during later on in development. Relative growth rate was low, but positive after 55 DAE (Fig. 12). CGR was maintained at high levels from 55 DAE onwards (Fig. 9), however, because of the high biomass densities maintained over the later stages of growth in a l l treatments. Unit leaf rate (ULR) and leaf area ratio (LAR) are the components - 76 -of Ry» and i t is clear that the time course of Ry was dominated by the influence of LAR (Figs. 11 to 13). In contrast with ULR, LAR generally tended to increase with increasing planting densities, and the counteracting responses of ULR and LAR tended to dampen planting density response of Ry. While the effect of density on LAR was small throughout growth (Fig. 13), i t appeared to arise at about the late vegetative growth (42 DAE) stage, making i t an early reaction to the density treatment. The large decrease in LAR during the course of crop growth was in turn due to joint contributions of declines in leaf weight ratio (LWR) and specific leaf area (SLA) (Figs. 14 and 15). The tendency of LAR to increase with increasing planting density, which i s suggested in Figure (13), was entirely due to LWR since SLA exhibited an inconsistent response to planting density. In summary, the effect of planting density on overall forage corn dry matter yield per hectare was significant and was the result of changes in many growth indices. Changes in CGR parallelled the changes in overall yield. The largest contributors to this density response, however, were made by alterations in LAI and biomass density (Eqn. 14). Yield per plant decreased with increasing planting density (Fig. 8). Plant height (T; Fig. 1) was the only primary characteristic which increased with increasing planting density. A l l the other primary characteristics of individual plants decreased with increasing planting density (Figs. 2-7) and they combined to produce the decreasing yield per plant with increasing planting density. A l l the primary plant - 77 -growth characteristics were significantly affected by planting density by the second harvest (42 DAE; Figs. 1-7). The differences among densities of yield per plant and the yields of individual plant parts widened with plant age. The analysis of relative growth rates of shoot biomass (Ry) and the relative growth rates of individual morphological characteristics (RQ^; Figs. 12, 16-18) did not reveal much about the effects of planting density. The major value of the relative growth rates of the primary growth characteristics is to express the dynamic partitioning of activ i t i e s within the plant. Although the responses to planting density by the relative growth rates of primary growth characteristics (RQ) were not distinctive, the use of such indices should be continued in future work. Other factors besides planting density may cause characteristic changes in relative growth rates, and these indices are a meaningful expression of important biological characteristics. Yield component analysis provided additional information on the growth and yield responses of forage corn to planting density. The two-dimensional partitioning of yield and yield components (TDP-YCA) analysis revealed that the variability in yield per plant not only depended on the stand density, but also depended on some morphological yield components (Tables 7-9). It is clear that yield variability in the early harvest (21 DAE) was independent of the density treatment, but depended very strongly on the yield components T, D/T, L^/IN» W L / L A A N D W / W L (Table 9; P < . 0 0 1 ) . By the use of yield component analysis, i t was also possible - 78 -to trace some of the early sources of density effects on plant growth. In the early vegetative growth (21 DAE), the density treatment had insignificant effects on the overall observed yield variability, but plant height (T), which was a significant yield component, was affected by the planting density treatment (P < .001). Thus, the early susceptibility of T to planting density reveals the possible importance of this component in determining subsequent effects of planting density on yield. This view is supported by the forward TDP-YCA results (Table 9) which shows that the variability in yield was highly correlated with variability in T. The importance of T as a yield component was observed by Buren et a l . (1978) in forage corn; Douglas et a l . (1958) in Crested wheat grass and J o l l i f f e et a l . (1982) in f i e l d beans (Phaseolus  vulgaris L.). D/T was a significant component throughout a l l the harvests, and was also highly susceptible to the effects of planting density after early vegetative growth (21 DAE). D/T is an indicator of the storage capacity of stems. Since a large proportion of W was obtained from the stems, the significant relationship between var i a b i l i t i e s D/T and yield is perhaps not surprising. D/T decreased with increased number of plants/m which showed that plant diameter (D) was more susceptible to planting density than plant height (T; Figs. 1, 2 and 19). Lfl/D was a significant yield component only at the fi n a l harvest in the forward TDP-YCA. This was due to the more rapid senescence of leaves in the older plants which was visible in the higher planting density especially during the reproductive growth of the - 79 -plant. The backward TDP-YCA showed that L N/D was a significant yield component from a l l harvests from 63-115 DAE. Thus, the variability of LJJ/D before the fi n a l harvest was taken over by T and D/T in the forward analysis. The var i a b i l i t i e s of W/Wj,, w L ^ L A a n d LA^ LN could not eliminate the contribution made by variability of L^/D in the backward analysis (Table 9). The significance of different yield components during growth was sporadic; individual components varied widely in their contributions at different plant ages and different directions of the TDP-YCA analysis. Yield components were both important contributors to yield variablity and sensitive detectors of planting density effects. More studies using similar components, however, are necessary to verify the yield component responses found in this study which involved data from a single cultivar, season and location. The analysis of the relative growth rates of yield components (Rrj) revealed large disparities among themselves during vegetative plant growth (Figs. 24-27). Toward crop maturity, however, the differences among them were relatively small. Also, i t was noteworthy how the Rrj's tended to concentrate toward Ry as crop maturity approached. The Rrj's added up to the Ry, but there was very l i t t l e effect of planting density on their trends. In view of some of the large changes in primary characteristics previously discussed, the sta b i l i t y in the contributions of Re's to Ry is remarkable. This is one of the f i r s t studies of the relative growth rates of yield and yield components. While the biological aspects of Rrj's can - 80 -be d i f f i c u l t to assess, the approach offers a direct and clear view of the contributions by yield components to the relative growth rate of yield. Some of the advanced approaches to growth and yield analyses in forage corn were made possible by the refinement of procedures in tradi-tional plant growth analysis (Hunt, 1977, 1982 a & b) and yield component analysis (Eaton et a l . 1983; J o l l i f f e et a l . 1983 (in preparation)). The analysis of variance (ANOVA) of a l l the primary growth characteristics studied including yield per hectare and yield per plant, gave the st a t i s t i c a l significance of responses to age and density. However, given the large number of degrees of freedom for error, the significant differences of the means among treatments were very large, although the actual quantitative differences among means were not necessarily very large. The ANOVA results provided mean values and standard deviations for each of the four planting densities at a l l five harvests. Traditional methods of plant growth analysis, using trends and indices calculated from two consecutive harvest, would have provided a limited view of the time and density responses from this data. Analysis of the data using the method of cubic spline regressions (Parsons and Hunt, 1981) provided closely spaced interpolated data points throughout the time course of crop growth. This technique not only gave smooth growth curves, but also described the growth rates of the individual plants, plant parts and yield components. Thus, the time course curves generated by these cubic spline regressions gave a clearer picture of the effects of density and age on plant growth - 81 -characteristics and growth indices than was available from the primary data or ANOVA results. Another technique used for analysing growth and yield was the two-dimensional partitioning technique (Eaton et a l . 1983), and this proved to be highly valuable. The technique provided weighted variances among treatments and yield components, taking the total variability in yield to be 100%. The relative amount of variability contributed by a source in the analysis was influenced by the variability of other sources. Plant age dominated variability which caused relatively small var i a b i l i t i e s to be contributed by the density treatment (Tables 7 and 8). This problem, however, was solved by two approaches. F i r s t l y , a l l of the individual yield component means from the untransformed data were re-analysed using the cubic spline regression technique and plotted as described above. The variability in yield components among densities was then observed from the time course curves. A second approach was the use of TDP-YCA analyses involving data from each individual harvest in succession. The latter approach not only specified the yield components responsible for yield variability at each harvest, but also traced the early sources of the effects of planting density on the crop. Because the dominating effect of time was removed, the interpretation of the TDP-YCA analysis using data from individual age groups was clearer than when the analysis used pooled data among a l l densities and harvests. A supplementary analysis of the whole data used the relative growth rates of the individual yield components, and this analysis was - 82 -also facilitated by use of cubic spline regressions. This method clearly defined the contributions by yield components to the relative growth rate of yield throughout crop growth. The effects of planting density on the relative growth rates of the yield components, were not clearly defined from the time course curves. In summary, the analysis of the present results on forage corn was facilitated by recent improvements in methods for the analysis of growth and productivity. It is clear that our ab i l i t y to interpret the physiology of yield in crop plants relies heavily on the refinements of mathematical procedures available for plant growth analysis, and i t is now possible to gain better insight into the management or improvement of crop growth. Growth analysis and yield component analysis are complementary approaches since they both aim to interpret productivity. Both approaches can be effectively used in the analysis of effects of treatments, such as planting density treatments, investigated in the present research. - 83 -6. CONCLUSIONS 1. The observation that yield per hectare of forage corn was significantly increased by increasing planting density suggests that planting densities higher than those currently recommended may be worthwhile. This is a tentative conclusion based on a limited study; i t presumes that good circumstances w i l l prevail for crop growth. 2. Yield per plant is reduced by increasing planting density because of decreases in most primary plant characteristics, and many of these effects originate within a few weeks of emergence. The relative growth rates of primary characteristics, however, did not reveal much about the effects of planting density on growth. 3. The effects of planting density on yield per hectare were correlated with changes in crop growth rate (CGR). The responses of crop growth rate in turn are dependent mainly on changes in leaf area index (LAI) or biomass density. Unit leaf rate (ULR) contributed to a lesser extent. Other components of CGR, including leaf area ratio (LAR) leaf weight ratio (LWR) and specific leaf area (SLA) did not strongly influence the yield: density relationship. 4. A l l the yield components studied contributed significantly to either the total variability in yield or the effects of planting density. The effects of yield components on yield, however, were sporadic depending on age and the direction of TDP-YCA analysis. 5. The relative growth rate of yield and the relative growth rates of the yield components did not reveal much about the effects of planting densities. However, they showed distinctive relationships - 84 -among the relative growth rate of yield and the relative growth rates of yield components throughout the course of yield formation. 6. Modern methods of growth and yield analysis used in the study give complementary information about the generation of yield and the responses to planting density. - 85 -L I T E R A T U R E C I T E D Bazzaz, F.A. and Harper, J.L. (1977). Demographic analysis of the growth of Linum usitatissimum. The New Phytol., 78: 193-208. B.C. Ministry of Agriculture and Food (1982). Field corn recommendations summary. Victoria, B.C. Bowen, P.A. (1983). The effect of oxygen fumigation of sawdust medium on the yield and yield components of greenhouse cucumbers. Scientia Horticulture 20: (in press). Buren, L.L., Mock, J.J. and Anderson, I.C. (1974). Morphological and physiological traits in maize (Zea mays L.). Crop Science, 14: 426-429. Causton, D.R. and Venus, P.C. (1981). The biometry of plant growth. ArnoId, London. Donald, CM. (1963). Competition among crop and pasture plants. Adv. Agron., 15: 1-118. Douglas, R.D. and Lu, K.H. (1958). A correlation and path coefficient analysis of crested wheat grass seed production. Agron. J., 51: Eaton, G.W., Bowen, P.A. and J o l l i f f e , P.A. (1983). Two-dimensional partitioning of yield variation. Biometrics (submitted for review). Eaton, G.W. and Kyte, T.R. (1978). Yield component analysis in the cranberry. J. Amer. Soc. Hort. Sci., 103: 578-583. Evans, G.C. (1972). The quantitative analysis of plant growth. Blackwell Scientific Publications, Oxford. Fraser, J. and Eaton, G.W. (1983). Application of yield component analysis to crop research. Field Crop Abstr., 36: (in press) G i l l , N.T. and Vear, K.C. (1980). Agricultural botany. 2. monocotyledonous crops, pp. 118-127. Duckworth, London. Gonske, R.G. and Keeney, D.R. (1969). Effect of f e r t i l i z e r Nitrogen, variety and maturity on the dry matter yield and Nitrogen fractions of corn grown for silage. Agron. J., 61: 72-76. Holliday, R. (1960). Plant population and crop yield. Field Crop Abstr., 13: 159. Hunt, R. (1982a). Plant growth analysis: second derivatives and compounded second derivatives of splined plant growth curves. Ann. Bot., 50: 317-328. - 86 -Hunt, R. (1982b). Plant growth curves: functional approach to plant growth analysis. Edward Arnold, London. Hunt, R. and Parsons, I.T. (1977). Plant growth analysis: further applications of a recent curve f i t t i n g program. Ibid. 14: 965-968. Iremiren, G.O. and Milbourn, G.W. (1978). The growth of maize. IV. Dry matter yields and quality components for silage. J. Agric. Sci. Comb., 90: 569-577. Janick, J., Schery, R.W., Woods, F.W. and Ruttan, V.W. (1981). Plant science. An introduction to world crops. 3rd Ed., pp. 512-516. W. H. Freeman, San Fransisco. J o l l i f f e , P.A., (et a l . 1983 in preparation). Plant growth analysis additive and multiplicative components of growth. Dept. Plant Science, U.B.C, Vancouver, B.C. J o l l i f f e , P.A., Eaton, G.W. and Lovett Doust, J. (1982). Sequential analysis of plant growth. New Phytol, 92: 287-296. Langer, R.H.H. and H i l l , G.D. (1982). Agricultural plants Ed., pp 104-109. Cambridge University Press, Cambridge. Lockhart, J.A.R. and Wiseman, A.J.L. (1978). Introduction to Crop Husbandry. 4th Ed., pp. 108-110. Pergamon Press, New York. Lopes, N.F. and Maestri, M. (1981). Growth, morphology, assimilate distribution and dry matter production in Maize (Zea mays L.) grown at 3 population densities. Field Crop Abstr., 35_(9): 7208. Lovett Doust, J. and Eaton, G.W. (1982). Demographic aspects of flower production in bean plants (Phaseolus vulgaris L.) Amer. J. Bot. 69: (in press). McAllan, A.B. and Phipps, R.H. (1977). The effect of sample date and plant density on the carbohydrate content of forage maize and changes that occur on ensiling. J. Agric. Sci. Camb., 89: 589-597. Parsons, I.T. and Hunt, R. (1981). Plant growth analysis: curve f i t t i n g using the method of B-Splines. Ann. Bot., 48: 341-352. Poneleit, C.G. and E g l i , D.B. (1979). Kernel growth rate and duration in maize as affected by plant density and genotype. Crop science, 19_: 389-388. Remison, S.U. and Lucas, E.0. (1982). Effects of planting density on leaf area and productivity of the maize cultivars in Nigeria. Exp. A g r i c , 18(1): 93-100. Sestak, Z., Catsky, J. and Jarvis, P.G. (1971). Plant photosynthetic production, a manual of methods (ed), Dr. W. Junk, N.V., The Hague. - 87 -Warren Wilson, J. (1981). Analysis of growth, photosynthesis and Light interception for single plants and stands Ann. Bot., 48: 507-512. Watson, D.J. (1952). The physiological basis of variation in yield Adv. Agron., 4: 101-145. - 88 -APPENDIX I E x p e r i m e n t a l , d e s i g n Randomized complete b l o c k d e s i g n (RCBD) 4 t 1 5 . ' k m m B3d 2 2 4 . 6 m 8 ^ 3 B3d 4 1 . ' 5 m B3CI2 B2d 3 7 6 . 8m B2d 4 B2di Bid! 10 Bid 2 11 " ¥ 4 Bid 3 12 _ _ y _ > 5 . ' 4 m ' l l " 5 m B = b l o c k 1'9 . 2 m d = d e n s i t y 13 B5d2 25 B8d3 14 B5°3 26 Bgdi 15 B5dl 27 Bad2 16 B 5d 4 28 B8d4 17 B 4d 3 29 Bgd2 18 B 4d 1 30 Bgd4 19 B 4d 4 31 Bgd3 20 B4d2 32 Bgdi 21 B6d4 33 B7d3 22 Wl 34 B 7d! 23 Beds 35 B7d2 24 Bgd2 36 B7CJ4 Appendix II Natural logarithms of yield component ratios - means for each density and harvest time. Planting Density (Plant/m2) Harvest Ln (T) Ln(D/T) Ln(LN/D) Ln(LA/LN) Ln(w L/lA) Ln(W/LL) Ln Y 4.9 1 0.54 -1.09 2.60 -0.52 -1.22 0.27 0.58 2 3.17 -2.44 1.70 1.14 -0.81 0.54 3.31 3 5.03 -4.12 1.40 1.69 -0.58 1.25 4.67 4 5.27 -4.30 1.09 1.77 -0.39 1.78 5.21 5 5.27 -4.25 0.89 1.72 -0.18 2.26 5.70 6.7 1 0.44 -1.07 2.60 -0.60 -1.23 0.29 0.44 2 3.45 -2.69 1.64 1.25 -0.81 0.59 3.43 3 5.12 -4.27 1.42 1.62 0.57 1.21 4.53 4 5.30 -4.39 1.11 1.72 -0.37 1.65 5.03 5 5.27 -4.31 0.86 1.72 -0.20 2.09 5.45 8.8 1 0.55 -1.13 2.57 -0.53 -1.23 0.31 0.55 2 3.25 -2.55 1.69 1.12 -0.83 0.51 3.19 3 5.03 -4.27 1.44 1.65 -0.63 1.12 4.34 4 5.26 -4.43 1.08 1.65 -0.31 1.55 4.80 5 5.27 -4.40 0.89 1.67 -0.20 1.95 5.17 11.1 1 0.64 -1.22 2.56 -0.50 -1.18 0.26 0.55 2 3.33 -2.70 1.73 1.08 -0.87 0.54 3.16 3 5.08 -4.37 1.45 1.62 -0.62 1.08 4.24 4 5.29 -4.55 1.18 1.57 -0.39 1.46 4.58 5 5.29 -4.46 0.86 1.59 -0.11 1.85 5.01 Grand Mean 3.89 -3.35 1.54 1.12 -0.64 1.13 3.70 Appendix III Mean orthogonal yield components at each density and harvest time. Planting C 4 C5 C6 Ln Y Density Harvest Cl C2 C3 (Plants/m2) 1 1 0.54 -0.10 -0.02 -0.02 0.01 0.09 0.58 1 2 3.17 0.40 0.08 0.02 0.03 -0.08 3.31 1 3 5.03 0.03 0.22 0.03 0.00 0.04 4.67 1 4 5.27 0.02 -0.02 0.01 -0.01 0.22 5.21 1 5 5.27 0.07 -0.21 -0.10 0.02 0.48 5.70 2 1 0.44 -0.15 -0.06 -0.03 -0.01 0.07 0.44 2 2 3.45 0.35 0.08 0.05 -0.00 -0.13 3.43 2 3 5.12 -0.05 0.24 -0.00 -0.02 -0.08 4.53 2 4 5.30 -0.05 -0.01 -0.01 -0.00 0.06 5.03 2 5 5.27 0.01 -0.23 -0.05 0.01 0.28 5.45 3 1 0.55 -0.13 -0.06 0.02 -0.02 0.07 0.55 3 2 3.25 0.35 0.08 0.01 -0.01 -0.18 3.19 3 3. 5.03 -0.12 0.21 0.12 -0.01 -0.13 4.34 3 4 5.26 -0.11 -0.08 -0.01 0.01 -0.13 4.80 3 4' 5 5.27 -0.09 -0.26 -0.02 0.01 0.08 5.17 1 0.64 -0.17 -0.05 0.00 0.02 -0.00 0.55 4 2 3.33 0.25 0.15 0.01 -0.02 -0.13 3.16 4 3 5.08 -0.19 2.21 0.12 -0.01 -0.21 4.24 4 4 5.29 -0.20 0.01 -0.03 -0.03 -0.21 4.58 4 5 5.29 -0.13 -0.30 -0.08 0.04 -0.11 5.01 Grand Mean 3.89 -0.00 0.00 -0.00 0.00 0.00 3.70 Ci - Cc represents the orthogonal variables of components 1-6. Ln Y is the natural logarithm of total plant DM yield (W). - 91 -Appendix IV Forward - YCA. "Two-dimensional Partitioning of the total sum of squares for yield expressed as percentages - 1n the early vegetative growth. Yield Component Block Density Error Total or Product Ci (T) 11.27** 1.85* 22.14 35.25*** C 2 (D/T) 12.54* 0.75 20.76 39.47*** C 3 (LN/D) 0.52*** 0.18** 1.42 2.12 1.48** 0.01 6.50 8.00*** C 4 (LA/LN) C 5 (WL/LA) C6 (W/HL) C,C? 11.03 -0.74 10.29 0.00 1 - 0.25 0.16 0.08 0.00 1 57 0.13 1.44 0.00 ClC 3 C1C4 C1C5 0.86 0.16 9.82 10.84*** 0.24 0.07 4.02 4.33*** 0.96 0.37 -1.33 0.00 dcfi 0.70 -0.12 -0.58 0.00 C2C3 0.20 -0.32 0.52 0.00 CoCi -0.81 -0.05 0.86 0.00 del 1.47 -0.34 -1.12 0.00 C ,Ct 0.48 0.03 -0.50 0.00 Clca 0.30 0.02 -0.32 0.00 C3C5 -0.39 0.13 0.26 0.00 del -0.20 0.04 0.16 0.00 C4C5 -0.47 0.02 0.44 0.00 C4C6 0.09 0.00 0.09 0.00 C5C6 6!ll -0.03 -0.09 0.00 Ln Y 70.40*** 1.12 28.49 100.00 Total SS (100X) - 36.824 C\ - - - C5 and Ln Y are Yield components 1 to 6 and natural logarithm of yield (W), * . * * . *** - Significant at P • 0.05, 0.01 and 0.001. - 92 -Appendix V Backward - YCA. Two-dimensional Partitioning of the total sum of squares for yield expressed as percentages in the early vegetative growth. Yield Component Block or Product Denisity Error Total Ci (T) C 2 (OA) C 3 (LN/D) C 4 (LA/LN) C 5 (WL/LA) C 6 (H/WL) CiC 2 C1C3 C1C4 C1C5 CiC 6 C2C3 C2C4 C2C5 C2C6 C3C4 C3C5 C3C6 C4C5 C4C6 C5C6 0.45** 0.03 0.12*** 30.26*** 8.16 0.06 0.04 0.01 1.46 1.43 0.05 0.01 0.23 0.23 -0.01 -0.25 0.02 0.02 11.63 0.65 0.18 0.17* 0.03** 0.00 0.81 0.13 0.02 0.04 0.03 0.11 -0.09 -0.01 0.01 -0.09 -0.05 0.01 0.01 0.02 -0.00 0.09 -0.08 0.01 2.23 0.33 0.28 26.75 29.11 1.02 -0.08 -0.04 -1.57 -1.34 -0.04 -0.02 -0.14 -0.17 0.00 0.24 -0.01 -0.02 -11.72 -0.56 -0.19 2.86 0.40 0.40 57.82*** 37.41*** 1.10 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Ln Y 70.40*** 1.12 28.49 100.00 Total SS (100X) - 36.824 Ci — Cs and Ln Y are yield components 1 to 6 and natural logarithm of yield (W), * t * * t *** . Significant at P » 0.05, 0.01 and 0.001. - 93 -Appendix VI Forward - YCA. Two-dimensional Partitioning of the total sum of squares for yield expressed as percentages in the late vegetative growth. Yield Component or Product Block Density Error Total Cl (T) C 2 (DA) C3 U H D) C< (LA/LN) C5 (WL/LA) C 6 ( M / H L ) CiC 2 C1C3 C1C4 C1C5 ClC 6 C2C3 C2C4 C 2 C 5 C2C6 C3C4 C3C5 C3C6 C4C5 C4C6 C5C6 41.84*** 0.70* 0.02*** 1.01*** 1.41* 0.41* 0.89 0.06 -2.93 7.58 0.50 -0.19 -0.92 0.53 -0.36 0.26 -0.17 0.07 -1.75 0.30 0.10 5.37** 1.56*** 0.00 0.21* 1.22*** 0.31** -2.25 0.22 0.69 -2.54 0.01 0.02 0.60 68 64 01 -0.19 -0.03 0.57 0.45 0.73 28.05 4.72 0.09 2.71 6.96 3.31 1.36 -0.28 2.24 -5.05 -0.50 0.17 -0.32 -3.21 -0.27 -0.26 0.19 -0.04 1.18 -0.75 -0.83 75.26*** 6.97*** 1.18 3.92*** 9.59*** 4.04* 0.00 -0.00 -0.00 0.00 -0.00 0.00 -0.00 0.00 -0.00 0.00 0.00 0.00 -0.00 0.00 -0.00 Ln Y 49.40*** 10.48*** 40.12 100.00 Total SS (100X) - 16.101 Ci --- C6 and Ln Y are yield components 1 to 6 and natural logarithm of yield (W), * t ** > ••* . significant at P « 0.05, 0.01 and 0.001. - 94 -Appendix VII Backward - YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages in the late vegetative growth. Yield Component Block or Product Density Error Total Ci (T) C 2 (D/T) C 3 (LN/D) C 4 (LA/LN) C5 (H L/LA) c 6 (w/wL) CiC 2 C1C3 C1C4 C1C5 CiC 6 C2C3 C2C4 C2C5 c2c6 C3C4 C3C5 C3C6 C4C5 C 4 C 6 C5C6 1.08** 0.00 0.04*** 2.79*** 8.33*** 21.03*** 0.01 -0.30 -1.17 3.48 -0.54 0.00 0.06 0.05 0.15 0.37 -0.45 0.68 -3.24 7.16 9.88 0.21 0.00 0.05** 1.32*** 0.41 4.99*** -0.04 -0.05 0.48 0.39 0.49 0.01 -0.05 -0.04 -0.03 -0.30 -0.24 0.32 0.74 2.79 -0.94 5.77 0.00 0.83 12.06 7.72 33.37 0.04 0.35 0.70 -3.86 0.06 -0.01 -0.00 -0.00 0.12 0.07 0.69 -1.00 2.50 -9.95 -8.93 7.06*** 0.00 0.92 16.17*** 16.46**" 59.39*** 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Ln Y 49.40*** 10.48*** 40.12 100.00 Total SS (100X) - 16.101 - - - Cs and In Y are yield components 1 to 6 and natural logarithm of yield (W). * > * * t «.* . Significant at P - 0.05, 0.01 and 0.001. - 95 -. Appendix VIII Forward - YCA. Two-dimensional Partitioning of the total sum of squares for yield expressed as percentages in the early reproductive growth. Yield Component Block or Product Density Error Total Ci (T) C 2 (D/T) C 3 (LN/D) C 4 UA/LH) C5 (M L/L A) C 6 (W/W ) CiC 2 C1C3 C1C4 C1C5 ClC 6 C 2 C 3 C2C4 C2C5 C2C6 C3C4 C3C5 C3C6 C4C5 C4C6 C5C6 Ln Y 9.41*** 8.92*** 0.04** 0.27* 0.15 2.45** 8.25 -0.37 -0.86 -0.10 -2.87 -0.35 -1.79 -1.44 -8.42 -0.01 0.06 0.20 0.10 0.64 0.B3 15.11*** 1.65** 21.81*** 0.01 0.05 0.05 2.56*** -1.99 0.11 -0.50 -0.33 -1.70 0.85 08 26 14.43 -0.00 0.01 0.23 0.09 0.50 0.51 40.67*** 18.38 17.47 0.20 1.50 5.99 9.09 -6.27 0.26 1.36 0.43 4.57 -0.50 0.71 0.19 -6.02 -0.00 -0.07 -0.43 -0.18 -0.50 -0.51 44.22 29.44*** 48.21*** 0.24 1.82 6.20*** 14.19*** -0.00 0.00 0.00 -0.00 -0.00 0.00 -0.00 0.00 0.00 0.00 -0.00 0.00 -0.00 -0.00 -0.00 100.00 Total SS (100%) • 9.8749 C 1 —- C6 and In Y are yield components 1 to 6 and natural logarithm of yield (H). * > «* > *** . significant at P - 0.05, 0.01 and 0.001. - 96 -Appendix IX Backward - YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages in the early reproductive growth. Yield Component or product Block Density Error Total Ci (T) C 2 (D/T) C3 (LN/D) Ca (LA/LN) 1. 78*** 0.40** 4.33 6.50*** 0. 69** 2.48*** 3.07 6.24*** 2. 11*** 0.82** 6.86 9.79*** 1. 24* 0.26 8.55 10.05*** C5 (WL/LA) 1. 26* 0.26 8.55 10.05*** C 6 (W/HL) 8. 76*** 11.38*** 37.34 57.47*** C1C2 0. ,20 0.30 -0.50 0.00 CjC 3 C1C4 C1C5 C1C6 -2. 25 0.38 18.78 -0.00 0. 29 -0.35 0.05 0.00 2. 25 0.32 -2.57 0.00 3. ,50 2.26 -5.76 -0.00 C2C3 C2C4 -0. 54 2.76 -2.24 0.00 -0. ,59 1.10 -0.51 0.00 C2C5 C2C6 C3C4 C3C5 C 3 C 6 C4C5 C4C6 C5C6 -0. 19 0.09 0.10 0.00 -3. ,38 9.76 -6.37 -0.00 -1. 17 0.55 0.62 -0.00 -2. ,10 0.15 1.95 -0.00 -2. 53 5.91 -3.38 0.00 -0. ,14 0.12 0.26 0.00 1. 47 1.28 -2.76 -0.00 4. 46 0.85 -5.31 -0.00 Ln Y 15. ,11*** 40.67*** 44.22 100.00 Total SS (10W) - 9.8749 Ci —- Cg and In Y are yield components 1 to 6 and natural logarithm of yield (W). * > * * t *** . Significant at P • 0.05, 0.01 and 0.001. - 97 -Appendix X Forward - YCA. Two-dimensional Partitioning of the total sum of squares for yield expressed as percentages in the late reproductive growth. Yield component or product Block Density Error Total 4.07*** 0.27 10.98 15.33*** 1.31 25.71*** 38.21 65.22*** 0.35*** 0.16** 1.50 2.01 0.02* 0.00 0.10 0.12 0.25** 0.04 1.42 1.71 0.85 2.21*** 12.54 15.61*** 0.70 -2.19 1.49 -0.00 0.27 0.02 -0.29 0.00 -0.14 -0.00 0.14 -0.00 -0.12 -0.07 0.20 0.00 1.17 -0.79 -0.38 0.00 -0.15 3.04 -2.89 -0.00 -0.10 0.38 -0.28 0.00 0.33 0.91 -1.24 0.00 -0.03 14.76 -14.73 -0.00 0.10 0.03 -0.13 -0.00 -0.27 -0.03 0.30 -0.00 -0.72 0.92 -0.20 0.00 -0.05 0.00 0.05 0.00 -0.06 0.11 -0.05 0.00 -0.00 0.19 -0.19 -0.00 Cl (T) C 2 (D/T) C 3 UN/D) C4 (LA/LN) C 5 (WL/LA) C 6 W\) CiC 2 C l C 3 C1C4 C1C5 C l C 6 C2C3 C2C4 C 2 C 5 c2c6 C3C4 C3C5 C3C6 C4C5 C 4 C 6 C5C6 Ln Y 7.77** 45.68*** 46.55 100.00 Total SS (100X) - 17.862 Ci — C6 and In Y are yield components 1 to 6 and natural logarithm of yield (W). * t * * t *** . significant at P « 0.05, 0.01 and 0.001. - 98 -Appendix XI Backward - YCA. Two-dimensional partitioning of the total sum of squares for yield expressed as percentages in the late reproductive growth. Yield component Block Density Error Total or product Ci (T) C 2 (D/T) C 3 (LN/D) C 4 (LA/LN) C5 (WL/LA) c6 (w/wL) CiC 2 CjC 3 C1C4 C1C5 ClC 6 C2C3 C2C4 C2C5 C2C6 C3C4 C3C5 C3C6 C4C5 C4C6 C5C6 0.41* 0.07 2.80 3.28 0.13*** 0.10*** 0.59 0.83 1.15* 0.53* 8.67 10.35*** 1.01 1.35** 14.12 16.48*** 5.28*** 0.06 1.09 1.68 5.87* 23.67*** 37.85 67.39*** -0.20 -1.18 0.32 -0.00 0.68 -0.09 -0.58 0.00 -0.65 -0.37 1.02 -0.00 -0.10 -0.04 0.15 0.00 1.68 -1.51 -0.16 0.00 -0.15 0.39 -0.25 0.00 0.05 0.70 -0.76 -0.00 -0.11 -0.01 0.12 0.00 -1.42 3.10 -1.69 -0.00 0.58 1.43 -2.01 0.00 -0.40 -0.13 0.53 -0.00 -0.14 6.44 -6.30 0.00 -0.19 0.09 0.10 -0.00 -0.74 10.49 -9.00 -0.00 0.53 -0.48 -0.05 0.00 Ln Y 7.77*** 45.68*** 46.55 100.00 Total SS (100%) - 17.862 Ci — Cg and Ln Y are yield components 1 to 6 and natural logarithm of yield (W), * j * * t *** . significant at P « 0.05, 0.01 and 0.001. - 99 -Appendix XII Forward - YCA. Two-dimensional Partitioning of the total sum of squares for yield expressed as percentages at forage maturity stage. Yield Component Block Density Error Total or product Cl (T) C 2 (D/T) C 3 (LN/D) Ca (LA/LN) C5 (WL/LA) C6 (W/WL) ClC 2 C1C3 C1C4 C1C5 ClC 6 C2C3 c2c4 C2C5 C2C6 C3C4 C3C5 C3C6 C4C5 C4C6 C5C6 Ln Y 13.06*** 59.65*** 27.29 100.00 Total SS (100%) • 17.137 Ci — C5 and Ln Y are yield components 1 to 6 and natural logarithm of yield (W). * t * * t *•* . significant at P « 0.05, 0.01 and 0.001. 2.13*** 0.04 3.62 5.79*** 4.52** 40.14*** 27.05 71.70*** 0.76*** 0.07 2.53 34.35*** 0.04* 0.02 0.35 0.41 0.26 0.01 2.52 . 2.79 1.48 1.57** 12.89 15.95*** 4.77 -1.20 -3.57 0.00 -0.23 -0.08 0.31 -0.00 0.06 -0.05 -0.01 0.00 -0.15 -0.00 0.15 0.00 -0.09 -0.27 0.36 -0.00 -1.28 3.08 -1.79 -0.00 0.16 0.63 -0.79 0.00 -0.37 1.21 -0.85 -0.00 -1.02 13.55 12.52 0.00 -0.10 0.05 0.06 -0.00 0.47 0.04 -0.51 -0.00 1.36 0.55 -1.91 0.00 -0.19 -0.01 0.20 -0.00 -0.02 0.07 -0.06 -0.00 0.50 0.23 -0.73 -0.00 - 1 0 0 -Appendix XIII Backward - YCA. Two-dimensional Partitioning of the total sum of squares for yield expressed as percentages at forage maturity stage. Yield component or product Block Density Error Total Cl (T) C 2 (D/T) C 3 (LN/D) C 4 (LA/LN) C5 (ML/LA) C6 (W/WL) ClC 2 C1C3 CiC 4 C1C5 ClC 6 C2C3 c2c4 C2C5 C2C6 C3C4 C3C5 C3C6 C4C5 C4C6 C5C6 0.68*** 0.09*** 4.24*** 0.60** 0.17* 5.71** 0.36 -0.26 0.71 0.04 1.55 -0.37 -0.16 0.04 -1.12 -2.30 0.05 3.30 -0.14 1.29 -0.73 0.08 0.05** 1.42** 0.15 0.05 33.58*** 0.06 0.30 0.07 0.06 1.38 0.51 16 02 40 0 0 2 0.76 0.21 13.76 -0.03 3.47 1.17 2.28 0.39 14.93 3.12 1.26 31.18 0.30 -0.05 -0.79 -0.11 -2.93 -0.14 -0.00 -0.06 -1.28 1.54 -0.26 -17.06 0.17 -4.76 -0.45 3.04 0.54 20.58*** 3.88*** 1.48 70.47*** -0.00 -0.00 -0.00 -0. 0. 00 00 -0.00 0.00 00 00 00 00 00 -0.00 0.00 -0.00 Ln Y 13.06*** 59.65*** 27.29 100.00 Total SS (100X) - 17.137 Ci --- Cs and Ln Y are yield components 1 to 6 and natural logarithm of yield (W). * . ** , *** - Significant at P - 0.05, 0.01 and 0.001. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0095703/manifest

Comment

Related Items