Crop Science 41:72-77 (2001)
© 2001 Crop Science Society of America
CROP BREEDING, GENETICS & CYTOLOGY
Reproductive Allocation on Branches of Virginia-Type Peanut Cultivars Bred for Yield in North Carolina
Anis-ur-Rehmana,
Randy Wellsa,b and
Thomas G. Isleiba,b
a A.U. Rehman, 602, 70-Parkwood Village Dr., North York M3A 2X7 ON, Canada
b Dep. of Crop Science, North Carolina State Univ., Raleigh, NC 27695-7620
Corresponding author (Randy_wells{at}ncsu.edu)
 |
ABSTRACT
|
|---|
Branching patterns have been altered due to breeding efforts in peanut (Arachis hypogaea L.); however, the effects of these alterations are poorly understood. The objectives of this research were to determine the relationships that exist among branching pattern, dry matter allocation, and yield of 10 genotypes representing the span of breeding efforts in North Carolina from NC 4 to the present. Experiments were conducted in a 2-yr field study at two locations. Numbers of reproductive and vegetative branches of individual plants were recorded for primary (n + 1) and secondary (n + 2) branches. In addition, dry weights (primary branches, cotyledonary branches, and main axis), seed yield, and yield components (seed weight, pod weight, shelling percentage) were determined. Reproductive-to-vegetative ratios (RVR) of n + 1 branch number, cotyledonary branch DW, primary branch DW, and whole plant DW were significantly increased with increased breeding cycles (BC) of hybridization and selection back to an indigenous line. The aforementioned measures of RVR were significantly associated with BC (r2 
0.48, P 
0.027), and year-of-cultivar release (RELYR) (r2 0.57, P 0.013) when analyzed as genotypic means. Mean BC values of n + 1 branch number, n + 1 branch DW, and total stem DW were all significantly correlated with seed yield (r2 0.86, P 0.023). Relatively modest increases in RVR of n + 1 branch number and the greater time and effort required for its measurement renders it less attractive as a selection criterion than merely measuring dry matter ratios.
Abbreviations: BC, breeding cycles • DW, dry weight • RELYR, release year • RVR, reproductive-to-vegetative ratio
|
INTRODUCTION
|
|---|
PEANUT YIELDS have increased over the years because of numerous factors, such as improved cultivars and improved agronomic practices. Mozingo et al. (1987) estimated an 18.5% yield increase in Virginia-type peanut since 1944 because of improved cultivars. This is in agreement with a 19% estimate reported by Wynne and Gregory (1981) from more limited data.
Duncan et al. (1978) proposed three physiological processes to explain most yield variation in peanut. They are (i) partitioning of the assimilate between reproductive and vegetative structures, (ii) length of the pod-filling period, and (iii) the rate of pod establishment. Reproductive efficiency, or the rate of assimilate partitioning into pods, was considered to be an important factor influenced by these physiological processes. Dry matter partitioning into reproductive structures ranged from 41 to 98% for cultivars ranked from oldest to newest (`Early Bunch'), respectively (Duncan et al., 1978).
Ball (1981) reported that 100 genetically diverse genotypes displayed significant positive correlation of fruit weight with total biomass (r = 0.53, P 0.01) and harvest index (r = 0.58, P 0.01). The general conclusion was that either variable could be utilized as a selection criterion in a breeding program to increase seed production. Wells et al. (1991), on the other hand, reported positive correlation between breeding cycles (minimum number of generations back to an introduction from an indigenous germplasm) of lines representing 50 yr of breeding in North Carolina and measures of reproductive growth. Peg number, pod number, pod dry weight, and reproductive-vegetative ratio at 71 d after planting (DAP) were positively associated with both breeding cycle and year-of-genotype release. Conversely, measures of vegetative growth (e.g., main stem length, vegetative dry weight, stem dry weight, and leaf area index) at 133 DAP were negatively associated with breeding cycles and year-of-cultivar release. These data indicate that shifts in dry matter allocation have occurred in response to selection for yield. A resulting question is how have these gross alterations been specifically imposed to dry matter on specific branches throughout the plant.
Harkness and Wright (1980) reported no significant differences in branching pattern in successive generations between variable and normal lines for five cultivars. Wynne (1974) discussed inheritance of branching pattern in crosses between the peanut subspecies where a wide range of patterns appeared. He was not able to postulate any simple genetic models and thought that branching pattern was a quantitative trait. He suggested that studies are needed to optimize branching patterns to attain maximum fruit production.
The present study was performed to test the hypothesis that specific alterations have been made to branching pattern in genotypes selected for greater seed yield. The objectives of this study were (i) to determine if relationships between branching pattern and dry matter allocation exist in genotypes representative of intense selection for seed yield and (ii) to ascertain the relationship between branching pattern and yield.
|
MATERIALS AND METHODS
|
|---|
Field experiments were conducted in 1992 and 1993 at the Peanut Belt Research Station, Lewiston, NC, and the Upper Coastal Plains Research Station, Rocky Mount, NC. Soil types were Goldsboro sandy loam (aquic paleudult, fine-loamy, siliceous, thermic) and Norfolk sandy loam (typic Kandiudult, fine-loamy, siliceous, thermic), respectively. Ten genotypes spanning more than 50 yr of breeding effort in North Carolina were planted at Rocky Mount by conventional planter on 18 May 1992 and 14 May 1993 (Table 1). At Lewiston, planting occurred on 13 May 1992 and 10 May 1993.
Genotypes were grown in plots of four rows 7.0 m long and 0.9 m apart. Plant spacing within the row was 25 cm. Other cultural practices followed during the growing season were standard for the region. Plots were irrigated as needed. Five randomly selected plants were harvested at maturity from the outer rows of each plot and the number of reproductive and vegetative branches were recorded for primary (n + 1) and secondary (n + 2) branches. Branch type determinations were terminated at the point where no branch appeared in the axil of the leaf. Reproductive-to-vegetative ratios for branch number were calculated by dividing number of reproductive branches by the total number of branches (vegetative and reproductive). Cotyledonary branches (lateral branches that arise from the respective cotyledonary axils), the remaining n + 1 branches, and the main axis, and their respective pods, were bagged, dried at 70°C for 72 h, and weighed. Leaves were not included in the vegetative sample. Reproductive-to-vegetative branch dry weight ratios for cotyledonary branches, n + 1 branches, main axis, and the total stem (the summed dry weights of the aforementioned segments) were calculated by dividing the pod dry weight by the branch dry weight.
At harvest, the two middle rows were mechanically inverted, allowed to air dry, and the pods were subsequently harvested mechanically. Pods were dried and yield was recorded. A 20-pod sample was collected for each plot and the pod and seed weight, and seed number were determined. Shelling percentage was calculated by dividing weight of seeds by pod weight and multiplying by 100.
Experimental design was a randomized, complete block with two replications in each of the four environments. Analysis of variance was determined by PROC GLM (SAS, 1987). The effect of breeding cycle (maximum number of cycles of hybridization and selection back to an introduction from an indigenous germplasm) and genotype within breeding cycle was tested by type I mean squares and the genotype x environment interaction as error. Regression analysis was performed by PROC GLM (SAS, 1987). Regressions of variables against year-of-cultivar release did not include NC Ac 17921 because it is an advanced line that has not been released.
|
RESULTS AND DISCUSSION
|
|---|
Genotype main effect was significant for yield and shelling percentage (Table 2). The maximum number of BC exhibited significance for yield, 20 pod weight, and 20 pod seed weight. Only shelling percentage showed significant variation among genotypes within BC.
View this table:
[in this window]
[in a new window]
|
Table 2. Analysis of variance mean squares for peanut yield, yield components, and reproductive-to-vegetative ratios (RVR) for branch number and dry weight
|
|
Yield ranged from a low of 4229 kg ha-1 for `NC 5' (Emery and Gregory, 1970) to a high of 5390 kg ha-1 for `NC 6' (Wynne et al., 1977)( Table 3). The relationship between seed yield and individual genotypes within BC (Fig. 1A)
was weak and nonsignificant (r2 = 0.16). Similarly, no components of yield were significantly associated with BC, with r2 values of 0.28, 0.22, and 0.02 for the relationship between BC and seed weight, pod weight, and shelling percentage, respectively.
View this table:
[in this window]
[in a new window]
|
Table 3. Seed yield, yield components, and reproductive-to-vegetative ratios (RVR) for branch number, and dry weight of ten peanut genotypes averaged across two years and two locations
|
|
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 1. The relationship of (A) yield, (B) reproductive-to-vegetative ratio (RVR) of n + 2 branch number, and (C) RVR of n + 1 branch number to the maximum number of breeding cycles of hybridization and selection back to an indigenous peanut line. Symbols represent genotype means across year and location (n = 8)
|
|
Genotype effects were significant for RVR of the n + 1 branch number, cotyledonary branch dry weight (DW), primary (n + 1) branch DW, and total stem DW (Table 2). Most of this difference was accounted for by the linear effect of BC, which was highly significant (P 0.01) for all of these traits. Reproductive-to-vegetative ratio for the n + 1 branch number ranged from a low of 0.46 for `NC 4' (Mozingo et al., 1987) to a high of 0.50 for `NC-Fla 14' (Emery et al., 1974), `NC 9'(Wynne et al., 1986), and `NC-V 11' (Wynne et al., 1991) (Table 3). Reproductive-to-vegetative ratios for DW of the cotyledonary branches, n + 1 branches, and total stem showed similar patterns among the genotypes, with simple correlation coefficients of 0.92 and 0.83 (P 0.01) for the relationship between RVR of the total stem and RVR for cotyledonary branch DW and RVR for n + 1 branch DW, respectively. Total stem RVR for DW ranged from 1.11 to 1.75 for NC 5 and NC 9, respectively (Table 3). The largest values of RVR for DW was found in the n + 1 branches, with values ranging from 1.29 for NC 5 to 2.20 for NC 9. The lowest values of RVR for DW were found in the main stem, which ranged from 0.54 for NC 5 to 1.0 for NC 9.
Seed yield of the individual genotypes (Fig. 1A) was not significantly associated with the maximum number of BC back to an indigenous population (r2 = 0.16). The relationship between the mean seed yield within each BC, however, was significantly associated (r2 = 0.96, P = 0.003) with the number of BC. Reproductive-to-vegetative ratio of the n + 2 branch number of the genotypes was not related to BC (Fig. 1B), while RVR of the n + 1 branch number was significantly associated with BC (Fig. 1C). Genotypic mean total stem (Fig. 2A) , cotyledonary branch (Fig. 2B), and n + 1 branch (Fig. 2C) DW RVR were all significantly related to BC (r2 0.56, P 0.013), indicating that increases in BC were responsible for at least 50% of the realized increase in RVR of the various morphological categories. Reproductive-to-vegetative ratios of the n + 1 branch number (Fig. 3A)
, n + 1 branch DW (Fig. 3B), and total stem DW (Fig. 3C) were all highly significant (r2 0.86, P 0.023) in their association with seed yield when the mean BC values were correlated. The relationship between the RELYR was associated with the various parameters of growth in a manner very similar to the relationships with BC (Fig. 4) . The r2 values for the relationships between individual genotype RELYR and RVR of n + 1 branch number (Fig. 4A), RVR of cotyledonary branch DW (Fig. 4B), RVR of n + 1 branch DW (Fig. 4C), and RVR of total stem DW (Fig. 4D) were 0.61, 0.57, 0.67, and 0.66 (P 0.019), respectively. The relationship between BC and RELYR resulted in a significant r2 of 0.86.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 2. The relationship of peanut reproductive-to-vegetative ratio for dry weight of the (A) total stem, (B) cotyledonary branch, and (C) n + 1 branch to the maximum number of breeding cycles of hybridization and selection back to an indigenous peanut line. Leaves were not included in the vegetative DW. Symbols represent genotype means across year and location (n = 8)
|
|
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 3. The relationship between peanut reproductive-to-vegetative ratio for dry weight of the (A) n + 1 branch number, (B) n + 1 branch DW, and (C) total stem DW to yield. Leaves were not included in the vegetative DW. Symbols represent breeding cycle (maximum breeding cycles of hybridization and selection back to and indigenous peanut line) means across year and location (n = 8, 16, 24, 24, and 8 for BC = 0, 2, 3, 4, and 5, respectively)
|
|
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 4. The relationship between peanut reproductive-to-vegetative ratio for dry weight (DW) of the (A) n + 1 branch number, (B) cotyledonary branch DW, (C) n + 1 branch DW, and (D) total stem DW to year-of-cultivar release. Leaves were not included in the vegetative DW. Symbols represent year-of-cultivar release means across year and location
|
|
Wynne (1974) demonstrated that improved RVR lines could be selected from crosses of Virginia- with either Valencia- or Spanish-type peanuts. He examined the ratios of average contiguous reproductive sequences to the sum of the average contiguous vegetative sequences and the average contiguous reproductive sequences. He suggested that studies for optimum branching pattern were required for maximum fruit productivity and to guide the peanut breeder in the selection of genotypes after crosses between subspecies. The present study exhibits that positive improvements have occurred in reproductive component allocation (i.e., n + 1 branch number, cotyledonary branch DW, n + 1 branch DW, and total stem DW) because of selection for yield.
Increased RVR of total stem DW (Fig. 2A), cotyledonary branch DW (Fig. 2B), and n + 1 branch DW (Fig. 2C), with increased BC and/or RELYR, is consistent with earlier reported alterations of reproductive allocation associated with selection for increased seed yield (Ball, 1981; Coffelt et al., 1989; Duncan et al., 1978; Wells et al., 1991). These shifts in DW allocation are related to significant, albeit relatively small, alterations in branch number RVR on n + 1 branches (Table 3, Fig. 1C, Fig. 3A). Mean dry weight RVR values for genotype were significantly associated with the n + 1 branch number RVR, exhibiting r-square values of 0.31, 0.70, and 0.49 (P 0.096) for the relationships with RVR for DW of cotyledonary branches (Fig. 5A)
, n + 1 branches (Fig. 5B), and the total stem (Fig. 5C), respectively. This report is the first to link increased plant RVR to alterations of specific branch RVR, be it DW or number.
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 5. The relationship between peanut reproductive-to-vegetative ratio (RVR) for dry weight of the (A) cotyledonary branch DW, (B) n + 1 branch DW, and (C) total stem DW to RVR for n + 1 branch number. Leaves were not included in the vegetative DW. Symbols represent genotype means across year and location
|
|
Coffelt et al. (1989) found increased RVR measured as increased mature pod dry weight divided by the plant dry weight in more recent cultivars. Similarly, Emery et al. (1973) reported that `NC 17' had larger harvest indices measured by a number of methods when compared with `Florigiant' (Carver, 1969), the leading North Carolina cultivar at that time. In addition, Duncan et al. (1978) found greater reproductive partitioning coefficients associated with newer, higher yielding cultivars. Similarly, Wells et al. (1991) reported a positive correlation between BC of lines representing 50 yr of breeding in North Carolina and peg number, pod number, pod dry weight, and RVR. Newer genotypes tended to have smaller vegetative mass and shorter main stem lengths, and increases in reproductive allocation.
Trends in many crops show similar alterations between dry matter allocation and yield. Examinations of wheat (Triticum aestivum L.) and cotton (Gossypium hirsutum L.) cultivars since the turn of century found increased dry matter allocation to reproductive organs relative to vegetative organs. Selection for yield has created shorter plants with larger harvest indices in wheat (Austin et al., 1980). In cotton, selection for yield has created shorter plants with larger RVR (Wells and Meredith, 1984). Transition from vegetative to reproductive growth in cotton was earlier and more complete thus attributing to earlier maturity of more recent cultivars (Meredith and Wells, 1989).
On the basis of these data, plant breeders could make advances in seed yield through selection for greater reproductive dry matter and reproductive branch number allocation. Similar conclusions were presented by Ball (1981) and Bera and Das (1997). Our data indicate that this approach could be accomplished using measurements of the cotyledonary or n + 1 branch dry matter alone (Fig. 3B and C), thus reducing resource expenditures. The relative modest increases in RVR of n + 1 branch number (Table 3, Fig. 1C) and the greater time and effort required for its measurement renders this variable less attractive than merely measuring dry matter. Whether "optimum" RVR ratio has been obtained, as suggested by Wynne (1974), appears doubtful on the basis of the absence of a flattening in the relationships of yield to RVR for n + 1 branch number, n + 1 branch dry weight, and total stem dry weight (Fig. 3).
The data contained herein show several concurrent alterations in dry matter allocation and branching patterns. There was a modest range in RVR for branch number (approximately 0.46–0.50) of n + 1 branches with increasing BC and RELYR. Further, positive effects of BC and RELYR were observed on the RVR for dry weight of cotyledonary branches, primary branches, and total stem. These data give evidence of morphological changes in both the occurrence of reproductive and vegetative branches and of their respective dry weights in response to selection for increased seed yield. The incorporation of selection for variation in dry matter allocation in current breeding programs appears worthy of further investigation.
|
ACKNOWLEDGMENTS
|
|---|
The authors wish to thank John B. Graeber, Gary A. Little, and Phillip M. Rice for their assistance in completing this study.
Received for publication February 28, 2000.
|
REFERENCES
|
|---|
- Austin, R.B., J. Bingham, R.D. Blackwell, L.T. Evans, M.A. Ford, C.L Morgan, and M. Taylor. 1980. Genetic improvement in winter wheat yields since 1900 and associated physiological changes. J. Agric. Sci. 94:675–689.
- Ball, S.T. 1981. Physiological basis for yield differences in peanuts (Arachis hypogaea L.) M.S. thesis. North Carolina State Univ., Raleigh.
- Bera, S.K., and P.K. Das. 1997. Studies on the selection for yield in groundnut (Arachis hypogaea L.). J. Oilseed Res. 14:153–156.
- Carver, W.A. 1969. Registration of Florigiant peanuts. Crop Sci. 9:849–850.[Free Full Text]
- Coffelt, T.A., M.L. Seaton, and S.W. VanScoyoc. 1989. Reproductive efficiency of 14 Virginia-type peanut cultivars. Crop Sci. 29: 1217–1220.[Abstract/Free Full Text]
- Duncan, W.G., D.E. McCloud, R.L. McGraw, and K.J. Boote. 1978. Physiological aspects of peanut yield improvement. Crop Sci. 18:1015–1020.[Abstract/Free Full Text]
- Emery, D.A., and W.C. Gregory. 1970. Registration of NC 5 peanuts. Crop Sci. 10:460.[Free Full Text]
- Emery, D.A., A.J. Norden, J.C. Wynne, and R.W. Mozingo. 1974. NC-Fla 14: An early maturing large seeded Virginia bunch peanut variety. North Carolina Agric. Exp. Stn. Bull. 448.
- Emery, D.A., J.C. Wynne, and P.W. Rice. 1973. Can reproductive efficiency in cultivated peanuts be improved? Oleagineux 28: 399–403.
- Harkness, C., and D.J. Wright. 1980. Branching pattern in successive generations of peanuts, Arachis hypogaea L. Peanut Sci. 7:41–45.
- Meredith, W.R., Jr., and R. Wells. 1989. Potential for increasing cotton yields through enhanced partitioning to reproductive structures. Crop Sci. 29:636–639.[Abstract/Free Full Text]
- Mozingo, R.W., T.A. Coffelt, and J.C. Wynne. 1987. Genetic improvement in large-seeded Virginia-type peanut cultivars since 1944. Crop Sci. 27:228–231.[Abstract/Free Full Text]
- SAS Institute. 1987. SAS/STAT Guide for personal computers. 6th ed. SAS Inst., Cary, NC.
- Wells, R., T. Bi, W.F. Anderson, and J.C. Wynne. 1991. Peanut yield as a result of fifty years of breeding. Agron. J. 83:957–961.[Abstract/Free Full Text]
- Wells, R., and W.R. Meredith, Jr. 1984. Comparative growth of obsolete and modern cultivars. II. Reproductive dry matter partitioning. Crop Sci. 24:863–868.[Abstract/Free Full Text]
- Wynne, J.C. 1974. Breeding potential of the subspecies of Arachis hypogaea L.: Inheritance of branching pattern, estimates of combining ability and photoperiodic response, Ph. D. Diss.(Diss. Abstr. AAI 7507698), North Carolina State Univ., Raleigh, NC.
- Wynne, J.C., W.V. Campbell, D.A. Emery, and R.W. Mozingo. 1977. NC 6: A southern corn rootworm-resistant peanut variety. North Carolina Agric. Exp. Stn. Bull. 458.
- Wynne, J.C., T.A. Coffelt, R.W. Mozingo, and W.F. Anderson. 1991. Registration of `NC-V 11' peanut. Crop Sci. 31:484–485.[Free Full Text]
- Wynne, J.C., and W.C. Gregory. 1981. Peanut breeding. Adv. Agron. 34:39–72.
- Wynne, J.C., R.W. Mozingo, and D.A. Emery. 1986. Registration of `NC 9' peanut. Crop Sci. 26:197.[Free Full Text]
TOP
ABSTRACT
INTRODUCTION
