<%BANNER%>

Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2011-05-31.

Permanent Link: http://ufdc.ufl.edu/UFE0024325/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2011-05-31.
Physical Description: Book
Language: english
Creator: Acuna, Carlos
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: Agronomy -- Dissertations, Academic -- UF
Genre: Agronomy thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Statement of Responsibility: by Carlos Acuna.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Blount, Ann R.
Local: Co-adviser: Quesenberry, Kenneth H.
Electronic Access: INACCESSIBLE UNTIL 2011-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024325:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024325/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2011-05-31.
Physical Description: Book
Language: english
Creator: Acuna, Carlos
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: Agronomy -- Dissertations, Academic -- UF
Genre: Agronomy thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Statement of Responsibility: by Carlos Acuna.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Blount, Ann R.
Local: Co-adviser: Quesenberry, Kenneth H.
Electronic Access: INACCESSIBLE UNTIL 2011-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024325:00001


This item has the following downloads:


Full Text

PAGE 1

1 PHYSIOLOGICAL AND GENETIC IMPLICAT IONS TO CONSIDER IN TETRAPLOID BAHIAGRASS BREEDING By CARLOS ALBERTO ACUNA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Carlos Alberto Acuna

PAGE 3

3 To my wife, Lorena

PAGE 4

4 ACKNOWLEDGMENTS I am grateful to God for my existence and for the inspiration and support He has provided throughout my life. I thank my wife and my family for their unconditional company, encouragement, and support. I would also like to thank my grandfather for sharing with me his love for agriculture. I want to thank my supervisory committee chair, Dr. Ann Blount, for providing the opportunity to pursue two graduate degrees at the University of Florida, and for her personal friendship. I especially thank Dr. Kenneth Ques enberry and Dr. Thomas Sinclair for their friendship, professional guidance, and knowledge which enhanced my graduate school experience. I would like to also thank Dr. Cher yl Mackowiak for educating me about soils and for her involvement in designing and executing the experiments under difficult field conditions. I am also grateful to Dr. Wayne Hanna for his professional example and invaluable advice. My thanks also go to Dr. Kevin Kenworthy, for t eaching me about turf breeding and quantitative genetics. I also would like to thank Dr. Brian Schwartz and Dr. Yoana Newman for assisting me with statistical models. I also appreciate the technical support of Judy Dampier, Loan Ngo Hung, Andrew Schreffler, Hua Kang, and Richard Fethiere Finally, I would like to thank the num erous undergraduate students that have helped in each of multiple experiments.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES.......................................................................................................................10 ABSTRACT...................................................................................................................................13 CHAP TER 1 INTRODUCTION..................................................................................................................15 Genetics and Breeding.......................................................................................................... ..15 Ecological Determinants of Growth....................................................................................... 18 Objectives...............................................................................................................................22 2 GENETIC VARIABILITY RESULTI NG FR OM A SECOND CYCLE OF HYBRIDIZATION AMONG BAHIAGRASS TETRAPLOID CLONES............................ 23 Introduction................................................................................................................... ..........23 Materials and Methods...........................................................................................................24 Plant Material and Crosses..............................................................................................24 Progeny Evaluation.........................................................................................................26 Mode of reproduction...............................................................................................26 Field observations....................................................................................................26 Statistical Analysis.......................................................................................................... 27 Results.....................................................................................................................................28 Hybridization Efficiency.................................................................................................28 Segregation for Mode of Reproduction........................................................................... 28 Segregation for Growth Habit, and P roduction of Inflorescences...................................29 Segregation for Cool-season Re growth and Freeze Resistan ce...................................... 29 Discussion...............................................................................................................................30 3 SEASONAL GROWTH, NITROGEN AND PHOSPHORUS UPTAKE OF NOVEL APOMICTI C BAHIAGRASS HYBRIDS............................................................................. 38 Introduction................................................................................................................... ..........38 Materials and Methods...........................................................................................................39 Plant Material..................................................................................................................39 Experimental Design and Plot Management................................................................... 40 Gainesville................................................................................................................ 40 Live Oak and Quincy...............................................................................................41 Statistical Analyses.......................................................................................................... 42 Results.....................................................................................................................................43

PAGE 6

6 Seasonal Biomass Yields................................................................................................. 43 Gainesville................................................................................................................ 43 Multi-location biomass production.......................................................................... 44 Rhizome+Root Mass and Nutrient Concentration and Accumulation............................45 Forage Nutrient Composition and Accumulation............................................................ 45 Discussion...............................................................................................................................47 4 ROOT DEPTH DEVELOPMENT IN APOMICTIC BAHIAGRASS..................................62 Introduction................................................................................................................... ..........62 Materials and Methods...........................................................................................................63 Plant Material..................................................................................................................63 Development of a Screening Technique.......................................................................... 64 Effect of Defoliation on the Rate of Root Depth Developm ent...................................... 65 Genotypic Variation for Rate of Root Depth Developm ent............................................ 65 Statistical Analyses.......................................................................................................... 65 Results.....................................................................................................................................66 Development of a Screening Technique.......................................................................... 66 Effect of Defoliation on the Rate of Root Depth Developm ent...................................... 66 Genotypic Variation for Rate of Root Depth Developm ent............................................ 67 Discussion...............................................................................................................................67 5 IMPORTANCE OF RAPID VERTICAL ROOT DEVELOPMENT FOR RECOVERI NG OF SOIL NITROGEN IN TETRAPLOID BAHIAGRASS........................ 76 Introduction................................................................................................................... ..........76 Materials and Methods...........................................................................................................77 Plant Material..................................................................................................................77 Root Development and Harvest Measurements.............................................................. 77 15N Uptake and Partitioning between Roots and Shoots................................................. 78 Statistical Analyses.......................................................................................................... 79 Results.....................................................................................................................................79 Root Depth Development................................................................................................ 79 Plant Mass and Tiller Number......................................................................................... 79 15N Uptake and Partitioning between Roots and Shoots................................................. 80 Discussion...............................................................................................................................81 6 IMPORTANCE OF ROOT MASS AND ROOT LENGTH DENSITY ON FORAGE PRODUCTI ON OF NOVEL APOM ICTIC BAHIAGRASS HYBRIDS.............................. 88 Introduction................................................................................................................... ..........88 Materials and Methods...........................................................................................................90 Plant Material..................................................................................................................90 Experimental Design and Plot Management................................................................... 90 Collection and Analyses of Root Samples......................................................................91 Statistical Analysis.......................................................................................................... 92 Results.....................................................................................................................................92

PAGE 7

7 Forage Production...........................................................................................................92 Root Mass and Root Length Density...............................................................................93 Discussion...............................................................................................................................95 7 CONCLUSIONS.................................................................................................................. 107 Genetics and Breeding.......................................................................................................... 107 Ecological Determinants of Growth..................................................................................... 108 Perspective on Future Research............................................................................................ 111 APPENDIX: SEASONAL FORAGE AND RHIZOM E+ROOT MASS, AND NITR OGEN AND PHOSPHORUS CONCENTRATIONS AND ACCUMULATIONS IN HERBAGE........................................................................................................................ ...113 LIST OF REFERENCES.............................................................................................................120 BIOGRAPHICAL SKETCH.......................................................................................................124

PAGE 8

8 LIST OF TABLES Table page 2-1 Seed set, germination and reproductive efficiency resulting f rom crosses between sexual and apomictic tetraplo ids clones of bahiagrass...................................................... 34 2-2 Bahiagrass tetraploid progeny classifi cation for m ethod of reproduction based on embryo sacs observations.................................................................................................. 34 2-3 Plant diameter and plant height for progeny of 12 com binations of sexual and apomictic clones............................................................................................................... ..35 2-4 Broad sense heritability estimates fo r severa l characteristics measured for 12 combinations of sexual and apomictic bahiagrass tetraploid clones................................. 35 2-5 Number of inflorescences, spring and fa ll regrowth, and freeze resistance of progeny resulting from 12 com binations of sexual a nd apomictic bahiagrass tetraploid clones.....36 2-6 Genetic variance and br o ad sense heritability (H2) estimates for several agronomic characteristics for a F1 population......................................................................................36 3-1 Seasonal and annual biomass accumulation of 12 apom ictic bahiagrass clones grown at Gainesville, FL...............................................................................................................50 3-2 Seasonal biomass accumulation of 13 apom ictic bahiagrass clones grown at Live Oak, FL, 2008. ...................................................................................................................50 3-3 Seasonal biomass accumulation of 13 apom ictic bahiagrass clones grown at Quincy, FL. ......................................................................................................................................51 3-4 Rhizome+root dry mass, N and P concentrations and accumulation (Gainesville)...........51 3-5 Nitrogen concentration in rhizome+root s of 12 bahiagrass clones (Gainesville). ............. 52 3-6 Phosphorus concentration in rhizome+root s of 12 bahiagrass clones (Gainesville). ........ 52 3-7 Seasonal and annual forage N uptake of 12 apomictic bahiagrass clones (Gainesville). ......................................................................................................................53 3-8 Seasonal and annual forage P uptake of 12 apom ictic bahiagrass clones grown at Gainesville, FL................................................................................................................ ...53 6-1 Average biomass produced by of 13 apomictic bahiagrass clones in 2007....................... 98 6-2 Average biomass produced by 13 bahiagrass clones and a high fertilizer rate in 2008. ... 98 6-3 Average biomass produced by 13 bahiagrass clones and low fertilizer rate in 2008. ....... 99

PAGE 9

9 A-1 Biomass of 12 bahiagrass clones gr own at Gainesville in 2006 and 2007. ..................... 113 A-2 Biomass of 12 bahiagrass clone s grown at Gainesville in 2008. ..................................... 113 A-3 Biomass of 13 bahiagrass cl ones grown at Live Oak in 2008. ........................................ 114 A-4 Biomass of 13 bahiagrass cl ones grown at Quincy in 2008. ...........................................114 A-5 Rhizome+root mass of 12 bahiagrass clones (Gainesville)............................................. 115 A-6 Nitrogen concentration in forage of 12 bahiagrass clones grown at Gainesville in 2006 and 2007. .................................................................................................................115 A-7 Nitrogen concentration in forage of 12 bahiagrass clones grown at Gainesville in 2008..................................................................................................................................116 A-8 Phosphorus concentration in forage of 12 bahiagrass clones grown at Gainesville in 2007..................................................................................................................................116 A-9 Phosphorus concentration in forage of 12 bahiagrass clones grown at Gainesville in 2008..................................................................................................................................117 A-10 Nitrogen accumulation in forage of 12 ba hiagrass clones grown in Gainesville in 2006 and 2007. .................................................................................................................117 A-11 Nitrogen accumulation in forage of 12 ba hiagrass clones grown in Gainesville in 2008..................................................................................................................................118 A-12 Phosphorus accumulation in forage of 12 bahiagrass clones grown in Gainesville in 2006 and 2007. .................................................................................................................118 A-13 Phosphorus accumulation in forage of 12 bahiagrass clones grown in Gainesville in 2008..................................................................................................................................119

PAGE 10

10 LIST OF FIGURES Figure page 2-1 Growth habit variation among bahi agrass hybrids generated by crossing F1 sexual and apomictic clones.......................................................................................................... 37 3-1 Biomass of 4 bahiagrass clone s grow n at Gainesville in 2007.......................................... 54 3-2 Biomass of 4 bahiagrass clone s grow n at Gainesville in 2008.......................................... 54 3-3 Biomass of 5 bahiagrass cl ones grow n at Live Oak in 2008............................................. 55 3-4 Biomass of 5 bahiagrass clones grown at Quincy in 2008................................................ 55 3-5 Spring rhizome-root mass of 12 bahiagrass clones grown at Gainesville, FL.. ................. 56 3-6 Spring 2007 rhizome+root nitrogen accum ulation of 12 bahiagrass clones grown at Gainesville, FL. .............................................................................................................. ...56 3-7 Spring 2007 rhizome+root phosphorus accumulation of 4 representative bahiagrass clones grow n at Gainesville, FL. ...................................................................................... 57 3-8 Nitrogen concentration of 4 representative bahiagrass clones during the 2007 growing season...................................................................................................................57 3-9 Nitrogen concentration of 4 bahiagra ss clones during the 2008 growing season. ............. 58 3-10 Phosphorus concentration of 4 bahiag rass clones during the 2007 growing season. ........ 58 3-11 Phosphorus concentration of 4 bahiag rass clones during the 2008 growing season. ........ 59 3-12 Nitrogen accumulation of four bahiag rass clones during the 2007 growing season. ........ 59 3-13 Nitrogen accumulation of four bahiag rass clones during the 2008 growing season. ........ 60 3-14 Phosphorus accumulation of four bahiagra ss clo nes during the 2007 growing season..... 60 3-15 Phosphorus accumulation of four bahiagra ss clo nes during the 2008 growing season..... 61 4-1 Bahiagrass plants g rowing in clear acrylic tubes. A) 3.5-cm diameter tubes filled with potting mix. B) 10-cm diamet er tubes filled with sandy soil.....................................70 4-2 Rate of root depth development for Arge ntine and Tifton 7 grow ing in c lear acrylic tubes (10-cm diameter) filled with potting mix................................................................. 70 4-3 Rate of root depth development for Arge ntine and Tifton 7 grow ing in clear acry lic tubes (3.5-cm diameter) filled with soil............................................................................. 71

PAGE 11

11 4-4 Rate of root depth development, a nd above and below ground m ass for Argentine and Tifton 7 grown in clear ac rylic tubes during spring 2006........................................... 72 4-5 Rate of root depth development a nd above and below ground m ass for Argentine and Tifton 7 grown in clear acrylic tubes during summer 2006........................................ 73 4-6 Effect of defoliation on root depth developm ent for Argentine and Tifton 7.................... 74 4-7 Rate of root depth development ( RRDD), and above and below ground mass for 13 novel apom ictic bahiagrass hybrids growing in clear plastic tubes .................................. 75 5-1 Root depth development for FL-3 and FL-122 bahiagrass growing in clear acrylic tubes filled with sandy soil. ............................................................................................... 83 5-2 Rate of root depth development for Argentine, F L-3, C-65 and FL-122 bahiagrass grown in clear acrylic tubes filled with soil during the spring and summer 2008. ...........84 5-3 Root and shoot mass for Argentine, FL-3, C-65 and FL-122 bahiagrass grown in clear acrylic tubes filled with so il during the spring and summer 2008. .......................... 85 5-4 Atom % 15N abundance in roots and shoots of Argentine, FL-3, C-65 and FL-122 bahiagrass grown in acrylic tubes during the spring and summer 2008............................ 86 5-5 Atom % 15N abundance in roots and shoots of Argentine, FL-3, C-65 and FL-122 bahiagrass grown in acrylic tubes during the spring and summer 2008. .......................... 87 6-1 Root sample collection on bahiagrass pl ots. A hydraulic soil sam pling equipment can be seen at front. A person removing an intern al plastic liner can be seen in the back....100 6-2 Seasonal biomass yields of the cultivar Argentine, the ecotype Common and hybrids FL-13 and FL-14 during 2008. Plots we re fertilized with 120 kg N ha-1 year-1..............100 6-3 Seasonal biomass yields of the cultivar Argentine, the ecotype Common and hybrids C-49 and C-93 during 2008. Plots we re fertilized with 360 kg N ha-1 year-1..................101 6-4 Root length density of 4 bahiagra ss hybrids fertilized with 360 kg N ha-1 yr-1 (high N) or 360 kg N ha-1 year-1(low N). ..................................................................................102 6-5 Relationship between root mass and RLD at the 0to 40-cm soil layer of bahiagrass hybrids in 2007................................................................................................................103 6-6 Relationship between root mass at the 0to 40-cm soil layer and forage annual accum ulation of bahiagrass hybrids in 2007.................................................................... 103 6-7 Relationship between root mass at the 40to 80-cm soil layer and forage annual accum ulation of bahiagrass hybrids in 2007.................................................................... 104 6-8 Relationship between root mass the 80to 120-cm soil layer and forage annual accum ulation of bahiagrass hybrids in 2007.................................................................... 104

PAGE 12

12 6-9 Relationship between RLD at the 0to 40-cm soil layer and forage annual accumulation of bahiagrass hybrids in 2007.................................................................... 105 6-10 Relationship between RLD at the 40to 80-cm soil layer and forage annual accumulation of bahiagrass hybrids in 2007.................................................................... 105 6-11 Relationship between RLD at the 80to 120-cm soil layer and forage annual accumulation of bahiagrass hybrids in 2007.................................................................... 106

PAGE 13

13 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PHYSIOLOGICAL AND GENETIC IMPLICAT IONS TO CONSIDER IN TETRAPLOID BAHIAGRASS BREEDING By Carlos Alberto Acuna May 2009 Chair: Ann R. Blount Co-chair: Kenneth H. Quesenberry Major: Agronomy Bahiagrass, Paspalum notatum Flgg, is a warm-season, pe rennial grass extensively cultivated as forage and utility turf in southeastern USA. Th e tetraploid germplasm of this species constitutes an underexploited source of genetic variation. This variation has great potential for bahiagrass improvement because of the possibility of fixing superior hybrids by manipulating apomixis. Physiological and genetic variables that might have a major impact on the process of improving this species were stud ied. The objectives of this research were to evaluate the transmission of apomixis through generations, and estimate the genetic variability for growth habit, cool-season growth, and fr eeze resistance resulting from hybridization of sexual and apomictic F1 clones. Additionally, estimates we re made on the seasonal biomass yields and nitrogen and phosphorus accumulation and the genetic variability for the rate of root depth development, root mass and ro ot length density among these novel F1 hybrids. The relationship between root development and nitr ogen uptake from deep soil layers, and the relationship between biomass yields and root characteristics were also determined. Approximately 20% of the F1 and F2 were classified as apomictic. Eleven percent of the F1 but only 3% of the F2 were classified as highly apomictic. This variable expressivity might be

PAGE 14

14 caused by the genetic background or epigenetic variation. The genetic variation observed for growth habit, cool-season growth, and freeze resi stance remained relatively constant between the F1 and F2. Apomicitc F1 hybrids accumulated more nitr ogen and phosphorus in spring and produced higher cool-season and annual bioma ss yields compared to common bahiagrass cultivars. Little genetic variation was observe d for nitrogen and phosphorus concentrations in foliage among these hybrids. Genetic variation fo r rate of root depth development was observed among hybrids. Higher rates of root depth develo pment resulted in faster access and uptake of nitrogen, and higher root and shoot mass. Vari ation for root length density was found among apomictic hybrids during the second growing se ason. No relationship was found between root length density or root mass, and above-ground bi omass production. The fertil ization rate did not affect root mass or root length density. Root activity might play an important role in nutrient uptake and biomass production.

PAGE 15

15 CHAPTER 1 INTRODUCTION Bahiagrass, Paspalum notatum Flgg, is a warm -season, pe rennial grass native to the Americas. It grows naturally in a vast region th at extends from central Argentina to northern Mexico (Blount and Acua, 2009). This species is especially abundant in native grasslands of southern Brazil, Uruguay, and nor theastern Argentina. Bahiagrass has become widely distributed in almost all tropical and subt ropical regions of the world, particularly in the western hemisphere. Bahiagrass germplasm was introduced into southeastern USA multiple times during the 20th century. Well adapted germplasm and locally released cultivars are now extensively cultivated as forage and utility tu rf in Florida and the southern Coastal Plain region of the USA. Genetics and Breeding As m ost Paspalum species, bahiagrass has a base chro mosome number of x = 10 (Gates et al., 2004). Several ploidy levels are known occu rring in this species, ranging from diploid (2n=2x=20) to pentaploid (2n=5x=50) type s. Pozzobon and Valls (1997) examined 118 bahiagrass accessions collected in Brazil, of these, 108 were tetraploids and th e other 10 were diploids. Dahmer et al. (2008) examined another 65 accessions collected in southern Brazil, of which 64 were tetraploid and 1 was diploid. In bo th studies the authors co ncluded that the rare diploid cytotypes escaped from cultivated Pens acola bahiagrass. Indigenous diploid accessions have only been collected in northeastern Argentina (Burton 1967; Daurel io et al., 2004). Thus, the tetraploid germplasm occupies most of the area where the species is distributed, while the diploid germplasm is restricted to a relatively small region of nor theastern Argentina. Triploids and pentaploids have been occasionally collect ed (Gould, 1966; Quarin et al., 1989; Thischler and Burson 1995). Chromosome pairing behavior, a nd tetrasomic inheritance for most analyzed

PAGE 16

16 loci indicated that tetraploid races originat ed by autoploidy (Forbes and Burton, 1961; Stein et al., 2004). The genetic diversity of bahiagrass has been evaluated using simple sequence repeats markers (Cidade et al., 2008). After evalua ting genetic polymorphisms among 91 accessions from Brazil, Argentina and Uruguay, the authors concl uded that wide variability is present in this species. The genetic variability f ound in tetraploid populations growi ng in the vicinity of diploid populations was higher than that observed within isol ated populations (D aurelio et al., 2004). These results indicate that the center of genetic dive rsity of this species is located in northeastern Argentina. Ploidy levels in this species are linked to c ontrasting reproductive ch aracteristics. Diploid cytotypes reproduce sexually and set low amounts of seed when self-pol linated due to selfincompatibility (Burton, 1955; Acua et al., 2007). Tetraploid ecotypes reproduce asexually by apomixis and set the same amount of seed when selfor cross-pollinated (Burton, 1948; Acua et al., 2007). Apomixis in bahiagrass includes the formation of unreduced embryo sacs from nucellar cells (apospory), the pa thenogenetic development of em bryos, and the development of the endosperm following fertilization of the pol ar nuclei (pseudogamy). These differences on mode of reproduction indicate that different br eeding approaches are needed for diploid and tetraploid races. The genetic control of apomixis in bahiagrass is still not well understood. Two out of three autotetraploids obtained by chromosome doubling of sexual diploids we re classified as facultative apomictic (Quarin et al., 2001) indica ting that the genetic de terminants for apospory were present at the diploid level, or that novel variation resulting fr om chromosome doubling was responsible for the expression of the trait at the tetraploid leve l. Segregation ratios resulting

PAGE 17

17 from crossing sexual induced and apomictic tetrap loid clones always showed an excess of sexual progeny (Martnez et al., 2001; Acua et al., 2009). Since self-pollination of sexual induced tetraploids, previously classified as sexual, only produced sexual indi viduals, it is believed that apospory is inherited as a single dominant Mendeli an factor with distor ted segregation. This factor is located in a large genomic region characterized by suppression of recombination and preferential chromosome pairi ng (Martnez et al., 2003; Stein et al., 2004; Stein et al., 2007). Results from a recent comprehensive transcriptome survey of genes differentially expressed in inflorescences of aposporous and sexual tetrap loid genotypes, indicated that apomixis in Paspalum involves the altered expressi on of a signal transduction cascade that seems to be triggered by the silencing of a large genomic regi on, including the apospory locus (Laspida et al., 2008). It was also shown in this study that se veral genes, which are involved in aposporous development, are ploidy-regulated. Although a remarkable amount of information about apomixis in Paspalum has been generated in the last decade, little is known about how this knowledge applies to breeding tetraploid bahiagrass. Theoretically, new apomic tic genotypes could be readily produced through hybridization of sexual induced and apomictic tetraploids (Ha nna and Bashaw, 1987). Continued breeding should be possible by repe ated crossing of superior sexual plants with superior apomictic male pollinators. However, it is unk nown if the segregation ratios observed for apomixis in the progeny of sexual induced and apomictic clones will remain constant across successive cycles of hybridization. This is es pecially unpredictable in a species showing distorted segregation and variab le expressivity. Only 11% of a progeny resulting from crossing sexual induced and apomictic tetrap loid clones was classified as highly apomictic (Acua et al.,

PAGE 18

18 2009). If this percentage is reduced through succ essive cycles of hybridization, this breeding approach would become impractical. Sexual apomictic crosses usually release a la rge amount of genetic variation because of the heterozygosity of the apomictic parents (Hanna and Bashaw, 1987). This phenomenon was observed when several induced sexual and apomic tic bahiagrass clones were crossed (Acua et al., 2009). However, it is questionable if this gene tic variation is releas ed from the apomictic parent or mainly results from crossing two isolated races (diploid and tetraploid). Variability for traits of agronomic importance, such as cool -season growth and freeze resistance are thought to be quantitatively inherited. The ge netic variability for these traits is expected to decline through successive cycles of hybridization as other tra its with quantitative inheritance (Poehlman and Sleper, 1995). Ecological Determinants of Growth Bahiagrass is a C4 species adapted to tropical and subtropical regions. When grown in subtropical regions, it shows a marked seasonality of growth in response to photoperiod and temperature (Sinclair et al., 2001; Gates et al., 1999). In Florida, most of bahiagrass production occurs between April and October (Gates et al., 2004). Fi ve months with no forage available to cover animal needs represents the most serious limitation for cattle production systems in southeastern USA. It is also a serious limitation when bahiagrass is cultivated as utility turf since its quality stays low with no growth. Plants with an extended growing se ason would be able to produce more biomass during the cool-season and probably during each year. An extended growing season would allow plants to capture so lar radiation and utili ze nutrients for a longer period each year resulting in higher annual biomass production. Genetic variability for cool-s eason growth has been observe d in the diploid bahiagrass germplasm (Gates et al., 2001). Several cycles of recurrent restricted phenotypic selection

PAGE 19

19 (Burton, 1974) for higher forage yields has resu lted in open-pollinated populations with greater cool-season growth. A large variation for cool-season growth and freeze resistance was observed among progeny generated by crossing sexual and a pomictic tetraploid clones (Acua et al., 2009). A relatively high proportion of this variation was attributed to genetic variability based on heritability estimates. However, variation for c ool-season growth and freeze resistance was only observed among progeny cultivated as individual plants. An extended growing season could compromise the long-term survival of this crop in a real production system. Research is needed to determine if this new tetraploid germplas m can produce more bioma ss during the cool-season and annually when grown in swards. Although rainfall is relatively high in the region where bahiagrass is cultivated, water is considered one of the main fact ors limiting its growth (Gates et al., 2004). Frequent droughts, high vapor pressure gradients between leaves and the environment during the growing season, and low water holding capacity of light textured soils result in water limited bahiagrass growth in most situations. Availability of essential minerals in the soil also limits bahi agrass production. This crop is mainly cultivated in soils with inherent lo w fertility and low cation and anion exchange capacities. Therefore, fertilizer application is essential for crop production in most situations. The main essential nutrient limiting bahiagrass production is nitrogen (Gates et al., 2004). Bahiagrass production dramatically increased as annual rates of nitrogen fertilization increased (Beaty et al., 1960; Blue, 1973). A single annual application of fertilizer is a common and recommended practice for bahiagrass pastures (Gates et al., 2004). The hi gh precipitation of the region and the low cation and ani on exchange capacities of the soils could result in large amounts of fertilizer bein g lost in percolated water. A 7-year experiment with Pensacola

PAGE 20

20 bahiagrass showed that the above-ground recovery of applied nitrogen increased from 30% during the first year to 70% during the 7th year. Part of the non-reco vered nitrogen was used to build the rhizome-root system, but a consider able amount was lost, especially during the establishment years. Therefore, genotypes that can rapidly access stored soil water and nutrients from fertilization would have an advantage that could result in higher biomass yields, while better protecting the environment. Rapid establishment and utilization of available resources is especially important when bahiagra ss is used in rotation with ot her crops that leave high amounts of residual fertilizer in the soil. Genotypes with early vigor and good seedling establishment tend to enhance transpiration at the expense of direct soil evaporation (L udlow and Muchow, 1990). Another characteristic that might be especially important for a quick es tablishment is the recovery of stored water due to rapid root penetration. Since nut rients readily migrate to deep soil layers in light textured soils, superior vigor and rapid root penetration would be also important for the recovery of nutrients left in the soil from previous crops or from early fertilizer applications. Thorup-Kristensen (2001) reported variation among plant families for ra te of root penetration including crucifer and grass crops. The crucifer N-catch crops were fa ster in developing deep rooting and depleting nitrogen from the subsoil than grass crops. Only small differences for rates of root penetration were found within botanical groups. In contrast, variability for rate of root penetration was observed among perennial grass spec ies (Burton et al., 1954). Highe r rates of root penetration were observed for Coastal bermudagrass ( Cynodon dactylon Pers.) than for other subtropical grasses like bahiagrass, Pangola ( Digitaria eriantha Steud.), Dallis ( Paspalum dilatatum Poir.), and carpetgrass (Axonopus fissifolius Raddi). Intraspecific variabi lity for maximum root depth was also observed in annual crops such as rice (Shen et al., 2001), sunflower and soybean

PAGE 21

21 (Dardanelli et al., 1997), and also for perennial crops such as Zoysiagrass (Marcum et al., 1995). It is unknown if there is genetic variability for rate of root pe netration within the bahiagrass germplasm. However, Burton (1943) reported th at two apomictic bahiagrass ecotypes differed for first years root production. These result s indicate that genetic variability for root characteristics may exist within the species. Root mass and root length at di fferent soil depths are other pl ant characteristics that might be related with bahiagrass biomass production. Se lection for greater deep-root mass to shoot mass ratio in tall fescue (Bonos et al., 2004) resu lted in genotypes with higher drought tolerance (Karcher et al., 2008). While root depth might be considered an important characteristic to capture nutrients that move readily with water flow (such as nitrogen), va riation in root length density in the top soil layers is expected to be more important for the uptake of nutrients that have low mobility in the soil, such as phosphor us. It is crucial that the phosphorus absorbing surfaces in the soil are extensiv e and prolific to make contact with the available phosphorus (Sinclair and Valdez, 2002). Ludlow and Muchow (1990) have questioned the relation between root depth or length, and biomass production under water st ress situations because of the cost of root growth and maintenance represent clear divers ions of assimilates, which might be used for shoot growth, and thus may decrease yield potential. Th e fact that root length densities can vary from 0.3 to 6 cm cm-3 among temperate cereals and legumes with no effect on soil water ex traction suggest that root length densities may be in excess of requirement in some crops. The study of root mass and length at different soil depths, and their relati onship with bahiagrass biomass production may add light to this issue. It might al so indicate the potentia l use of these root traits for breeding bahiagrass.

PAGE 22

22 Objectives Determ ine the stability of the inheritance a nd expression of apomix is across cycles of hybridization between sexua l and apomictic tetraplo id clones (Chapter 2). Quantify the genetic variability for growth habit, cool-season growth, and freeze resistance generated by a second cycle of hybridization between sexual and apomitic tetraploid clones (Chapter 2). Determine seasonal biomass production, nitr ogen and phosphorus concentrations and contents of novel apomictic hybrids grown in swards (Chapter 3). Evaluate the genetic variation among novel a pomictic hybrids for rate of root depth development (Chapter 4). Determine the relationship between rate of root depth development and nitrogen uptake (Chapter 5). Evaluate the genetic variability for root ma ss and root length density at different soil depths among novel apomictic hybrids (Chapter 6). Establish the relationship between root ma ss and length, and biomass production (Chapter 6). Determine the effect of nitrogen fertilizati on on root mass and length at different soil depths (Chapter 6).

PAGE 23

23 CHAPTER 2 GENETIC VARIABILITY RESULTING FROM A SECOND C YCLE OF HYBRIDIZATION AMONG BAHIAGRASS TETRAPLOID CLONES Introduction Approxim ately 125 grass species form their se eds by an asexual proc ess called apomixis (Bashaw and Hanna, 1990). This characteristic offers a unique opportunity for developing and using superior genotypes. Seed of any superi or obligate apomict could be increased through open-pollination for an unlimited number of generati ons without loss of vigor (heterozigosity) or change in genotype (Hanna and Bashaw, 1987). Apomixis is the predominant mode of reproduction among the polyploid germplasm of Paspalum (Quarin, 1992). Bahiagrass, Paspalum notatum Flgg, has become one of the most economically important species of this genus mainly because of its use as forage and utility turf (Blount and Acua, 2009). This species include s polyploids that repr oduce by apomixis and diploids that reproduce sexually (Burton, 1948; Burton, 1955). Apomix is in bahiagrass includes the formation of unreduced embryo sacs from nucellar cells (apospory) the pathenogenetic development of embryos, and the development of the endosperm following fertilization of the polar nuclei (pseudogamy). Apospory in this species is inherited as a single dominant Mendelian factor with distorted segrega tion (Martnez et al., 2001). This f actor is located in a genomic region characterized by suppression of recombin ation and preferential chromosome pairing (Martnez et al., 2003; Stein et al., 2004; Stein et al., 2007). In nature, the tetraploid (2n=4x=40) cytotype s of bahiagrass are predominately facultative or obligate apomictic, and able to produce redu ced pollen (n=2x=20) (Gates et al., 2004). Sexual tetraploids have been generated by treating bo th diploid (2n=2x=20) s eed and tissue cultured calluses with chromosome duplication treatments (Burton and Forbes, 1960; Quarin et al., 2001; Quesenberry and Smith, 2003). Crosses between sexua l induced tetraploid clones used as female

PAGE 24

24 parents and apomictic tetraploid clones used as pollen donors resu lted in the release of a large genetic variability for traits of agronomic importance among the F1 hybrids (Acua et al., 2009). When the progeny were classified by mode of reproduction, 80% of the hybrids were sexual, 11% highly apomictic, and 9% facultative apomictic. Continued improvement should theoretically be possible by repeated crossing of s uperior sexual plants wi th superior apomictic male pollinators. If the apospory locus has a plei tropic lethal effect with incomplete penetrance as stated by Martnez et al. ( 2001), the proportion between sexua l and apomictic progeny could vary among hybridization cycles. The genetic variab ility for the selected agronomic traits with quantitative inheritance is expected to be redu ced by each cycle of hybridization and selection. The objectives of this research were to create a bahiagrass segregating F2 population by crossing F1 sexual hybrids and unrelated apomictic F1 or natural hybrids, determine the segregation for mode of reproduc tion, and estimate the resulti ng genetic variation for growth habit, cool-season regrowth, production of inflorescences and freeze resistance. Materials and Methods Plant Material and Crosses Crosses were m ade between sexual and apom ictic tetraploid clones during the summer 2006 to generate a segregating bahiagrass populati on. Twelve sexual tetraploid clones were used as female parents (Table 2-1). These 12 clones were hybrids generated at the University of Florida by crossing induced tetraploids (derived from seeds of the diploid cultivar Tifton 9) and the apomictic tetraploid clones, Argentine a nd Tifton 7 (Acua et al., 2009). The induced tetraploids were selected for this study becau se they were identified as sexual and crosspollinated, based on two years of observations, an d were selected out of several hundred clones based on growth habit, cool-season regrowth and freeze resistance. Seven highly apomictic clones were used as pollen donors (Table 2-1). Five of them were also hybrids generated by

PAGE 25

25 crossing induced sexual and apomictic tetraploid s, and later identified as highly apomictic (Acua et al., 2009). They were also selected ba sed on superior spreading, cool-season regrowth and freeze resistance. The other two were the cu ltivar Argentine, which is the best adapted tetraploid ecotype to southeastern USA, and the experimental hybrid Tifton 7. Crosses were made by enclosing one infloresce nce from the sexual female and one or two inflorescences from the apomictic male in a glas sine bag prior to anthesis. Care was taken to select inflorescences at the same stage of ma turity. All bags were shaken each day during anthesis. Five days after anthesis the inflores cences from the apomictic parent were removed from the bags, leaving only inflorescences fr om the sexual parent until seed maturity. At maturity, each head from the sexual parents was threshed separately, the number of florets was counted, empty florets were removed, and the number of florets co ntaining caryopses was determined. Seed were scarified using concentrated sulfuric acid for 10 min, and were sown in flats containing sterile germination medium in Fe bruary 2007. Individual seedlings were later transplanted to seed ling flats in a greenhouse. Plants were transplanted into a field located at the Agronomy Forage Research Unit near Hague, Florida on 11 May 2007. The soil classification at this location was a loamy, sili ceous, subactive, thermic, Arenic Endoaquult. Parents were asexually propagated in the gree nhouse and transplanted into th e field with the progeny. While the apomictic parents were propagated by seeds, sexual parents were vegetatively propagated using short pieces of rhizomes containing the ap ical meristems. Progeny from each cross were planted in 55-plant rows where individual pl ants were spaced 1 m x 1 m. A row containing multiple replications of the two parents involved in the specific cross was planted next to their progeny. The field was fertilized with 290 kg ha-1 of 21-3.1-11.6 (N-P -K) in June 2007.

PAGE 26

26 Progeny Evaluation Mode of reproduction Two hundred and eleven plants were selected from 2,700 total generated plants. Selection was based on good general vigor. This subsam pl e was used to estimate the proportion of apomictic and sexual progeny. Alth ough this selection based on vigor might introduce bias in the results because vigor could be linked to sexua lity or apomixis, it was used to reduce the probability of including progeny resulting from accidental self pollination among the evaluated plants. Inflorescences from this group were fixed at anthesis (when the embryo sacs are usually fully developed) in FAA (18 Ethanol 70 %: 1 Form aldehyde 37 %: 1 glacial acetic acid). Pistils were dissected out of the florets and cleared using the method of Young et al. (1979). Ovules were observed using a differential in terference contrast microscope. A minimum of 20 ovules were observed from at least two different inflorescences. Plants bearing ovules with single embryo sacs containing the egg apparatus, th e bi-nucleated central cell, and a mass of antipodals at the chalazal end we re classified as sexual. In contrast, plants bearing ovules with multiple or single embryo sacs with the egg apparatus, the central cell, no antipodals, and variable size and position, were cl assified as apomictic. Plants producing ovules with either sexual or apomictic embryo sacs were classified as facultative. Field observations Plant diam eter was estimated using the aver age between the longest and the shortest diameter of a given plant on 2 October 2007, and 26 September 2008. Plant height was measured from the base of the plant to the top of the canopy on 21 September 2007 and 12 September 2008. The number of inflorescences per plant was counted on 22 September 2007.

PAGE 27

27 On 23 September 2007 all plants were defoliate d to approximately 6cm above the soil level and regrowth was visual ly estimated on 28 October 2007, and 6 May 2008 using a 1 to 5 scale, where 1 = plants showing the lowest am ount of herbage and 5 = plants showing the highest amount of herbage. Plants were de foliated on 19 September 2008 and regrowth was estimated on 7 November. Also, freeze resistan ce was visually estimated on 21 November 2008 after one freeze event on 19 November, with temper ature of -6 C, using a 1 to 5 scale, where 1 = the least freeze resistant, and 5 = the most freeze resistant plant. Statistical Analysis Chi-square tests were u sed to compare the observed apomixis/sexuality segregation ratio with expected ratios and with previously report ed ratios. Field observati ons were analyzed using PROC GLM of PC SAS (SAS Institute, 2004) as a completely randomized design. When significant differences among families were detect ed for one variable, the Waller-Duncan Test was used for mean separations. Unless otherwise stated in the text, all differences refer to significance at P < 0.05. Broad sense heritability (H2) estimates were calculated using the following formula: Where 2 p equals the phenotypic variance among the progeny, 2 sp equals the phenotypic variance among clonal replications of the sexual parent, and 2 ap equals the phenotypic variance among replications of the apomictic parent. 2 p includes the additive, dominance, and epistatic genetic variance, variation due to interactions between genotypes and environment, and variation due to environmental effects. The environmenta l variation is estimated based on the variation among clonally propagated parents. Varian ces and means were obtained using PROC

PAGE 28

28 UNIVARIATE of SAS. The vari ance used to calculate H2 for fall regrowth was calculated across years, while the variance for plant diameter and height is the average variance between years. Results Hybridization Efficiency In 2006, 2,700 progeny were generated using 12 com binations of selected sexual and apomictic bahiagrass clones. Parental clones we re selected based on their mode of reproduction (highly sexual or highly apomictic) and superior agronomic charac teristics including cool-season regrowth, spreading and leaf tissue freeze resistance (Acua et al., 2009). Sexual clones were used as female parents and highly apomictic clones as pollen donors. Th e average seed set was 30% varying from 13 to 51% (Table 2-1). Gr eat variation was obser ved among crosses for germination. The average germination was 82% varying from 0 to 99%. With the exception of seeds from FL-33 FL-53 which did not germinat e, germination can be considered high varying from 59 to 99%. The average reproductive effici ency resulting from these crosses was 25% varying from 9 to 46% indicating that the outcome from this type of hybridization is highly dependant on the genotypes selected as parents. Segregation for Mode of Reproduction Two hundred and eleven progeny were selected between 2007 and 2008 based on their superior vigor to study the segregation for m ode of reproduction. The use of this approach was expected to reduce the probability of including progeny resulting from self-pollination of sexual parents in the study. Although different numbers of progeny were selected from each family, all families were represented in this group (Table 2-2). One hundred and seventy three plants were classified as sexual because only single reduced embryo sacs were observed in their ovules. Seven were classified as highly apomictic becau se aposporous embryo sacs were observed in no

PAGE 29

29 less than 65% of their ovules. Th e remaining 31 plants were classified as facultative apomictic because aposporous embryo sacs were observed in no more than 30% of their ovules. Segregation for Growth Habit, and Production of Inflorescences Plant diam eter and plant height were m easured in 2007 and 2008 to characterize the growth habit of the progeny. Significant diffe rences were observed among families for plant diameter in both years. A range of family mean s from 28 to 42 cm in diameter was observed in 2007 and from 35 to 60 cm in 2008 (Table 2-3). The highest variability resulted from cross 5 in 2007 and cross 6 in 2008 (Table 2-3). The average broad sense heritability for plant diameter was 0.8 varying from 0.47 to 0.97 (Table 2-4) Significant variation was also found among families for plant height in 2007 and 2008. Family means for height varied from 32 to 47 cm in 2007 and from 40 to 63 cm in 2008 (Table 2-3). Th e greatest amount of variability was observed for cross 5 in 2007 and for cross 3 in 2008 (Table 2-3). The average broad sense heritability for plant height was 0.79 varying among families from 0.61 to 0.95. Since seed production is considered one importa nt agronomic characteristic of bahiagrass, the total number of inflorescen ces per plant was counted at the end of the 2007 growing season. The mean number of inflorescences produced varied from 13 to 39 among families (Table 2-5). The highest variability was obtained from cross 9. Broad sense heritability varied from 0.19 to 0.82 among families, while the average was 0.48 (Table 2-4). Segregation for Cool-season Regrowth and Freeze Resistance The 12 f amilies included in this studied were determined to be significantly different in terms of plant regrowth during fall 2007, spring 2007 and fall 2008. The largest variability for spring regrowth was contained in family 4, while the lowest was present in family 11 (Table 25). Broad sense heritability estimates varied greatly among families having an average of 0.64 (Table 2-4). Progeny within family 1, 3 and 4 exhi bited the greatest amount of variability for fall

PAGE 30

30 regrowth considering 2007 and 2008 data (Table 2-5). The average broad sense heritability for fall regrowth was 0.63 varying from 0.47 to 0.86 (Tab le 2-4). Significant differences were also observed among these 12 families for freeze resistance in 2008. The greatest amount of variability for this characterist ic resulted from cross 11, while th e lowest variation resulted from cross 12 (Table 2-5). The average broad sense heritability was 0.77, varying from 0.25 to 1.0 (Table 2-4). Discussion The opportunity of fixing superior seed -propagated genotypes and the inform ation generated in the last decade concerning the gene tic control of apomixis have encouraged new attempts for the genetic improvement of apomictic bahiagrass as forage an d turf. This research shows that it is feasible to generate a segregating population by crossing selected F1 sexual and apomictic bahiagrass clones. In fact, F2 hybrids can be created more efficiently than F1 hybrids mainly because of the differences in fertility a nd vigor of the involved fe male parents. While F1 sexual hybrids are superior genotypes selected from an original segregating population and have normal fertility, induced sexual tetraploid clones usually have reduced fertility and low vigor. The ratio between sexual and apomictic progeny was of 4.6:1, which was different from the 1:1 segregation ratio [ 2 = 86.4, P(1 df) 0.001] expected for a monogenic tetrasomic inheritance with apospory as a dominant trait. However, the observed segregation pattern was not different from the 4.3:1 ratio [ 2 = 0.5, P(1 df) = 0.48] observed for the F1 population. These results would indicate that al though there is a strong distortion with an excess of sexual progeny, the segregation patterns for apomixis remain constant through hybridizat ion cycles. However, the proportion of highly apomictic progeny decreased from 11% in the F1 to only 3% in the F2, and the proportion of facultative apom ictic increased from 9% in the F1 to 15% in the F2. These

PAGE 31

31 findings would indicate that th e probability of finding highl y apomictic progeny decreases through hybridization cycles. This variation of gene expressiv ity between generations also suggests that epigenetic inherita nce might be involved in the e xpression of aposporous apomixis. Moreover, Laspina et al. (2008) concluded that the expression of apospory in Paspalum might be triggered by the silencing of a large genomic re gion including the apospory locus. Environmental differences can also be responsible for the observed variable expressivity. Although Burton (1982) showed that the environment has little or no effect on the expression of apomixis in bahiagrass, seasonal variation of apospory was reported for Paspalum cromiorrhyzon Trin. (Quarin, 1985). The most successful forages of southeastern US A have a prostrate grow th habit that allows them to maintain their growing points without being defoliated by grazing animals. Marked variability for spreading and plant hei ght was observed in the generated F2 progeny (Figure 2-1 and Table 2-6). A considerably high proportion of this variability for growth habit was determined, using broad-sense heritability estimat es, to be the result of genetic variation among the progeny (Table 2-6). These resu lts indicate that selection can be efficiently used to develop clones with desirable growth habit. Previously reported broad sens e heritability estimates for the F1 progeny (Acua et al., 2009) were similar to the average estimates now reported for the F2 (Table 2-6). This is another indication that large genetic variation for grow th habit was present in the F2. One of the advantages of bahiagrass as a fo rage and utility turf re lies on seed propagation. Production of inflorescences is one trait that can be recorded ear ly in the evaluation of a large segregating population as an indirect estimation of seed production. The variation observed for production of inflorescences in the F2 was lower than that reported for the F1 (Table 2-6). In

PAGE 32

32 addition, a considerable portion of the variation observed in the F2 was attributed to environmental differences. The decline in variance and heritability estimates indicates that one cycle of selection significantly reduced th e genetic variability for this trait. Warm-season grasses have a delimited growing season in subtropical areas mainly because of photoperiod responses (Sinclair et al., 2001). This physiological re sponse, that seems to be a mechanism of freeze damage avoidance, reduces forage production during spring and fall. Efforts are being made to reduce this photoperiod sensitivity and increase freeze resistance by genetic manipulation of bahiagrass at the diploi d and tetraploid level (Blount and Acua, 2009). A large variation was observed in the F2 for cool-season regrowth and freeze resistance (Table 26). A large part of this variation was attri buted to genetic varia tion based on the obtained heritability estimates. A minimal reduction of ge netic variance and heritability estimates for spring and fall regrowth was observed when comparing estimates for the F1 and F2 (Table 2-6). This small change reflects the effect of one cycle of phenotypic selection on the genetic variability for cool-season regrowth. In contra st, the genetic variance for freeze resistance was larger for the F2 compared with that for the F1. These results might relate to the fact that the F2 was exposed to lower temperatures befo re the data was collected. While the F1 was exposed to a minimal temperature of -2 C, the F2 was exposed to -6 C before freeze resistance was estimated. Heritability estimates obtained in the F1 and F2 for freeze resistance did not change significantly. In conclusion, hybridization between sexual and apomictic bahiagrass clones can be used efficiently to generate a large segregati ng population. Minimal redu ction of the genetic variability for traits of agronomic interest can be expected after one cycl e of selection. The low number of highly apomictic genotypes that can be found in each generation is a major limitation

PAGE 33

33 of this breeding approach. Vari ation of apomixis expressivity can also be expected among generations.

PAGE 34

34 Table 2-1. Seed set, germination and reproductive efficiency resulting from crosses between sexual and apomictic tetraplo ids clones of bahiagrass. Cross Seed Set Germination Reproductive Efficiency Sexual Female Apomictic Male Number --------------------------%--------------------------FL-83 FL-13 1 32 98 32 FL-137 FL-13 2 51 89 46 FL-99 FL-3B 3 35 99 35 FL-14 FL-93 4 44 80 35 FL-16 FL-93 5 26 70 18 FL-41 FL-25 6 23 83 19 FL-3C Argentine 7 39 97 38 FL-7 Tifton 7 8 24 92 22 FL-62 Tifton 7 9 26 72 19 FL-47 Tifton 7 10 23 84 20 FL-16 Argentine 11 13 66 9 FL-20 Argentine 12 20 59 12 Average 30 82 25 Seed set: percentage of obtained seed from the total number of pollinated florets. Reproductive efficiency: percentage of obtained plants from the total number of pollinated florets. Table 2-2. Bahiagrass tetraploid progeny clas sification for method of reproduction based on embryo sacs observations. Cross Analyzed Progeny Apomictic Facultative Sexual 1 22 0 3 19 2 33 0 4 29 3 22 0 3 19 4 43 2 12 29 5 23 3 3 17 6 12 1 0 11 7 4 0 2 2 8 19 1 1 17 9 16 0 0 16 10 3 0 0 3 11 2 0 0 2 12 12 0 3 9 Total 211 7 31 173

PAGE 35

35 Table 2-3. Plant diameter and plant height fo r progeny of 12 combinations of sexual and apomictic clones. Plant Diameter Plant Height 2007 2008 2007 2008 Mean SD Mean SD Mean SD Mean SD Cross --------------------------------------------------------cm --------------------------------------------------------------1 33 8.9 50 13.9 40 8.6 51 8.6 2 42 11.0 59 22.6 34 7.7 48 8.9 3 35 8.0 47 11.2 43 7.9 53 31.6 4 34 8.4 50 12.4 47 8.0 63 9.0 5 40 14.4 60 15.7 40 9.4 52 21.2 6 40 7.7 60 39.2 40 8.2 48 11.1 7 28 8.3 35 15.6 40 7.9 51 15.4 8 40 8.7 59 28.4 36 7.7 48 10.7 9 37 9.4 49 14.6 36 8.4 47 10.2 10 37 10.0 54 20.3 32 7.6 42 11.8 11 38 7.8 59 33.0 36 7.5 40 11.5 12 37 6.4 59 10.6 43 8.1 57 8.8 MSD 2.0 4.2 1.7 3.0 Minimum significant difference, Waller-D uncan means separation procedure. Table 2-4. Broad-sense heritability estimates for several characteristics measured for 12 combinations of sexual and apomictic bahiagrass tetraploid clones. Cross Number of Inflorescences Height Diameter Spring Regrowth Fall Regrowth Freeze Resistance 1 0.40 0.78 0.82 0.46 0.49 0.66 2 0.66 0.61 0.85 0.71 0.49 0.68 3 0.22 0.95 0.59 0.86 0.78 0.52 4 0.65 0.90 0.74 0.63 0.66 0.85 5 0.82 0.93 0.86 0.41 0.65 1.00 6 0.40 0.78 0.97 0.72 0.47 0.68 7 0.50 0.90 0.80 0.75 0.86 0.74 8 0.19 0.71 0.93 1.00 0.79 1.00 9 0.27 0.65 0.80 0.63 0.56 0.25 10 0.36 0.80 0.83 0.56 0.52 1.00 11 0.55 0.84 0.94 0.31 0.49 1.00 12 0.68 0.70 0.47 0.65 0.77 0.81 Mean 0.48 0.79 0.80 0.64 0.63 0.77

PAGE 36

36 Table 2-5. Number of inflorescences, spring and fall regrowth, and freeze resistance of progeny resulting from 12 combinations of sexual a nd apomictic bahiagrass tetraploid clones. Number of Inflorescences Spring Regrowth Fall Regrowth Freeze Resistance Cross Mean SD Mean SD Mean SD Mean SD 1 13 12.4 2.0 0.9 2.6 1.0 1.8 0.8 2 21 15.4 2.8 1.0 2.9 0.9 2.0 0.8 3 26 15.7 2.7 1.1 3.1 1.0 2.0 0.8 4 28 17.3 3.5 1.2 3.3 1.0 2.6 1.0 5 31 17.2 2.2 0.9 2.9 0.9 1.8 0.7 6 22 14.6 2.3 1.0 2.8 0.8 1.7 0.7 7 11 11.6 2.0 0.9 2.6 0.9 2.0 1.0 8 16 18.1 2.7 0.9 3.0 0.7 2.0 0.8 9 18 18.5 2.3 0.9 2.8 0.8 1.7 0.6 10 25 17.7 2.0 0.9 2.2 0.8 1.7 0.6 11 39 14.0 1.7 0.6 2.1 0.6 1.6 0.5 12 36 13.3 3.4 0.9 3.0 0.7 2.8 1.1 MSD 3.2 0.2 0.1 0.2 Minimum significant difference, Waller-D uncan means separation procedure. Table 2-6. Genetic variance and broad sense heritability (H2) estimates for several agronomic characteristics for a F1 population generated by cr ossing induced sexual and apomictic clones (Acua et al., 2009) and a F2 population generated by crossing sexual and apomictic F1 clones. F1 F2 Trait Variance H2 Variance H2 Spreading 102.9 0.7 86.8 0.8 Height 55.5 0.8 68.2 0.8 Inflorescences 495.8 0.9 147.3 0.5 Spring Regrowth 0.88 0.8 0.73 0.6 Fall Regrowth 0.82 0.8 0.61 0.6 Freeze Resistance 0.25 0.7 0.58 0.8

PAGE 37

37 Figure 2-1. Growth habit variation among bahiagrass hybrids generated by crossing F1 sexual and apomictic clones.

PAGE 38

38 CHAPTER 3 SEASONAL GROWTH, NITROGEN AND PHOSPHORUS UPTAKE OF NOVEL APOMICTI C BAHIAGRASS HYBRIDS Introduction Bahiagrass is widely adapted to sub-tropica l and tropical regions around the world and is considered an im portant forage and utility turf for southeastern USA (Blount and Acua, 2009). As forage, it is used as a long-term perennial pa sture or in short-term rotations with row crops, such as cotton and peanut. As turf, it is so wed along road-ways, athl etic fields and other recreational areas. Bahiagrass has potential as a bioenergy crop because of its long-term persistence, and, as other C4 species, high water and n itrogen (N) use efficiency. It is also used as ground cover in different parts of the world to prevent soil erosion and ground water pollution resulting from nutrient runoff and leaching. This is especially important in areas with well drained, sandy soils (Brady and Weil, 2002). Bahiagrass growth is restricted to a relativ ely short period in s ubtropical regions. In southeastern USA, most of its above-ground biom ass is produced between June and August. This growth pattern is mainly the result of its sensitivity to photoperiod and low temperatures (Sinclair et al., 2001; Gates et al., 1999). Extendi ng the growth period can increase its use as forage or turf. Genetic variability for extending the bahiagrass growth period has been observed in diploid and tetraploid bahiagrass germpl asm (Blount and Acua, 2009; Acua et al., 2009). The possibility of fixing bahiagrass genotype s with an extended growing season by using apomixis makes the tetraploid germplasm a good candi date for genetic improvement of this crop. Several apomictic hybrids exhibiting an extende d growing season as individual plants were identified among a segregating population gene rated by crossing induced sexual and apomictic tetraploid clones (Acua et al., 2009). Howeve r, it was unknown if these individual plant hybrids were able to exhibit the same response wh en grown in swards. It was also unknown how

PAGE 39

39 selecting for an extended growing season would re late to root and rhizome mass. These organs are essential for the long-term survival and pr oduction of this crop. H ypothetically, if these hybrids produce more biomass in the cool-season, they will likely remove more total nutrients during this period. An extended growing season ma y also result in greater annual biomass and nutrient uptake. The objective of this study was to determine the seasonal biomass yield, N and P accumulation of novel apomictic bahiagrass hybrids grown in swards. Materials and Methods Plant Material Thirte en tetraploid apomictic bahiagrass cl ones were selected for this study, based on expected variability in seasonal forage producti on. Two of these clones, Argentine and common, are natural ecotypes introduced fr om South America. Argentine is a productive tetraploid clone commonly grown in Florida. Both, Argentine and common were selected for this study because of their relatively short growing season. The ot her eleven clones were novel apomictic hybrids generated by crossing sexual and apomictic tetr aploid clones. One of the better known clones, Tifton 7, was developed by Dr. Gl enn Burton (USDA-ARS, Tifton, Ge orgia), and selected based on superior forage yields when grown in fiel d plot trials. Three ot hers, C-49, C-65 and C-92, were generated by Camilo Quarin (Corrientes, Argentina), and selected based on superior individual plant forage yields. The other se ven clones (FL-3, FL-3B, FL-13, FL-14, FL-21, FL93 and FL-122) were developed at the University of Florida, and sele cted, based on superior vigor, cool-season regrowth, and freeze resistance of spaced plants. Since these 13 clones are classified as highly apomictic, variation among replications was primarily restricted to environmental variation.

PAGE 40

40 Experimental Design and Plot Management Gainesville Seeds were scarified (10 min) using concentrated sulfuric acid and rinsed with tap water. The scarified seed were sown in March 2006, in a greenhouse, using plas tic trays containing a sterile germination mix. After two weeks, seedlings were transplanted to seedling flats containing multiple cells. Seedlings were transp lanted into a field located at Gainesville (29 48'12" N, 8224'47" W), Florida, on 15 May, 2006. The so il classification at this location was a loamy, siliceous, subactive, thermic, Arenic Endoaquult. Forty s eedlings from individual clones were planted (40 cm x 40 cm) into 2-m x 3-m plots. Five blocks of 12 pure-stand plots were planted without alleys in a randomized complete block design. The ecotype common was not planted at this location. Pl ots were irrigated with appr oximately 20 mm of water after transplanting. Weeds were manua lly removed during the establis hment year. In July 2006, plots were fertilized with 500 kg ha-1 of 16-4-8 (N-P2O5-K2O). On 31 October, 2006, plots were harvested using a sickle bar mower leaving a 5-cm stubble height. A 2.35 m x 0.7 m strip was cut in the middle of each plot, the forage collecte d, weighed, and a subsample (approximately 700 g) immediately taken after harves t. The subsample was weighed and the material dried at 60 C for 48 h, prior to reweighing for dry mass. Plot s were harvested again on 4 May, 2007, and every four weeks during the rest of the 2007 growing season (Table A-1). With the exception of the last harvest of each year, plots were fertilized with 286 kg ha-1 of 21-7-14 (N-P2O5-K2O) following each cutting. In 2008, plots were harvested for the first time on 13 May and every four weeks during the growing season (Table A-2). Plots were fertilized with 375 kg ha-1 of 16-4-8 (N-P2O5-K2O) following each cutting.

PAGE 41

41 Soil cores (4.7-cm diameter and 15-cm depth) were collected in spring, summer, and fall 2007, and spring and fall 2008. Samples were collected from the center of an original (or mother) plant located near the border of each plot. Soil samples were placed on a 2-mm mesh screen and washed with a gentle stream of tap water to recover rhizomes and roots. Samples where dried at 60C for 48 h and dry mass recorded. Dry forage and rhizome-root samples from each harvest were ground using a Willey mill, and passed through a 1-mm screen. Nitrogen and P concentrations were determined at the University of Florida Forage Evaluation Suppor t Laboratory (Gainesville, FL). For N and P analyses, samples were digested using a modification of the alum inum block digestion procedure of Gallaher et al. (1975). Brie fly, dry sample (0.25 g) was combined with 6 ml of H2SO4, 2 ml H2O2 and catalyst (1.5 g of 9:1 K2SO4:CuSO4) and digestion conducted fo r at least 4 h at 375C. Nitrogen and P in the digestate were determ ined by semiautomated colorimetry (Hambleton, 1977). Live Oak and Quincy Seeds from the 13 apomictic clones were germinated in March 2007, in a greenhouse located at Gainesville, FL. After two weeks, seedlings were transplanted to seedling flats containing multiple cells. Seedlings were transplant ed into a field located at the North Florida Research and Education Center, Live Oak (32 18'03" N, 82 53'53" W), FL, on 9 May. The soil at is classified as a thermic, coated Typic Qu artzipsamment. Four blocks (rows) were planted, containing 13 (1.2 m x 1.2 m) pl ots separated by a 1 m alley, as a randomized complete block design. Each plot contained 36 seedlings of each clone (20 cm x 20 cm), and clones were randomized within each block. Pl ots were irrigated with approximately 20 mm of water after transplanting, and weeds re moved manually. Plots were fertilized with 530 kg ha-1 of 34-0-0 (N-

PAGE 42

42 P2O5-K2O) and 100 kg ha-1 of 0-0-60 (N-P2O5-K2O) during 2007. A 70-cm wide strip was cut across each plot on 15 May, 2008, using a sickle ba r mower, leaving a 5-cm stubble height. Plots were harvested every 4 weeks (Table A-3), and fertilized with 376 kg ha-1 of 16-4-8 (N-P2O5K2O) following each cutting, for the remainder of the growing season. Forage samples were dried at 60 C for 48 h, and dry mass recorded. Seedlings from 13 apomictic clones germinated in March 2007 at Gainesville were also planted at the North Florida Research and Education Center, Quincy (30 32'59" N, 84 36'02" W), Florida, on 10 May, 2007. The soil in this location is classified as fine-loamy, kaolinitic, thermic, Typic Kandiudult. Forty se edlings were transplanted (40 cm x 40 cm) into each plot (2 m x 3.2 m). Five replication of each pure-stand plot were planted in a randomized complete block design. Plots were irrigated with 25 mm of water after transplanting. Weeds were controlled manually and mowed during 2007. On 2008, plots were harvested on 13 May for the first time, and every four weeks (Table A-4). On 22 May, plots were fertilized with 171 kg ha-1of 35-0-0 (N-P2O5-K2O) and 28 kg ha-1of 0-0-60 (N-P2O5-K2O). On 13 June, plots were fertilized with 20 kg P ha-1, 67 kg N ha-1, and 42 kg K ha-1. In 11 July and after each subsequent harvest, plots were fertilized with 376 kg of 16-4-8 (N-P2O5-K2O). A 2.35 m x 0.7-m strip was cut in the middle of each plot, the forage weighed, and a subsample was taken. The subsample fresh weight was obtained immediately after harves t. Subsamples were dried at 60 C for 48 h, and dry mass recorded. Statistical Analyses Gainesville data for biom ass, N concentrati on, P concentration, N content, and P content were analyzed as repeated measures usi ng Proc Mixed procedure (SAS version 9.2, SAS Institute, Cary, NC). Clones and harvest dates were considered fixed, while replicates were

PAGE 43

43 considered random. Biomass, N and P accumulated in spring, summer, and fall were also analyzed as repeated measures. Biomass data collected every four weeks at Gainesville, Live Oak, and Quincy in 2008 were also analyzed as repeated measures usi ng Proc Mixed of PC SAS. Locations, clones and harvest dates were considered fixed, while replic ates were considered random. When significant treatment effects ( P = 0.05) were found, the minimum significant difference (MSD) among means was calculated using the Waller-Duncan test. Results Seasonal Biomass Yields Gainesville Biom ass yields among 12 bahiagrass tetraplo id clones varied signi ficantly among harvests in 2007 and 2008 (Figures 3-1 and 3-2). The highes t biomass yields were observed in June 2007 and July 2008 (Tables A-1 and A-2). A signifi cant interaction between harvests and clones resulted from differences on seasonal biomass pr oduction among clones, as illustrated with 4 of the clones (Figure 3-1). The la rgest genotypic differences were observed in the spring of both years (Table 3-1 and Figure 3-1). With few exceptions, most hybrids produced more biomass than Argentine during the spring of both years (T able 3-1). For example, clone FL-13 yielded 4.3 times more biomass than Argentine in spri ng 2007 and 1.9 times more in spring 2008. In contrast, Argentine was among the most produc tive clones during the summer of both years. Eight hybrids produced as much as Argentine in summer 2007, while only 3 hybrids were able to yield as much as Argentine in summer 2008. Although there were signi ficant differences among clones, the fall biomass yield wa s significantly lower than that observed for spring and summer. Tifton 7 had among the greatest fall yields across years. There were significant differences among tetraploid clones for annual biomass accumulation in 2007 and 2008 (Table 3-1). While

PAGE 44

44 Argentine accumulated less annual biomass th an several clones in 2007, no clones produced more than Argentine in 2008. Multi-location biomass production A significan t location effect and a significant interaction between location and clones were observed when the 2008 biomass data collected at Gainesville, Live Oak and Quincy were compared. There was also a significant harvest effect and an interaction between harvests and clones. Clones were significantly different for biomass production at each harvest for each of the three locations (Tables A-3 and A-4). Spring biomass varied greatly among clones at Live Oak and Quincy (Tables 3-2 and 3-3, and Figures 3-3 and 3-4). FL-13 and FL-93 were among the highest spring yielding clones, while common and Argentine were among the lowest yielding clones at the two locations. However, some clones performed differently between loca tions. For example, C-49 was ranked low at Quincy and high at Live Oak. Regardless, most novel hybrids outperformed Argentine for spring biomass (Tables 3-2 and 3-3; Figures 3-3 and 3-4). Significant differences were also observed among clones for summer biomass production at Live Oak and Quincy (Tables 3-2 and 3-3). However, no hybrid produced more biomass than Argentine. Common, FL-21 and FL-13 produced less biomass than Argentine at Live Oak, while only common produced less than Argentine at Quincy during the summer. Genotypic performance for fall biomass yield was different between Live Oak and Quincy. While seven hybrids produced more fall biomass than Argentine at Live Oak, no hybrid outperformed Argentine at Quincy. Total annual biomass during 2008 differed among clones. While four hybrids accumulated more biomass than Argentine at Live Oak, fi ve hybrids outperformed Argentine at Quincy

PAGE 45

45 (Tables 3-2 and 3-3). FL-93, C-92 and C-49 pr oduced more annual biomass than Argentine during 2008 at both locations. Rhizome+Root Mass and Nutrient Concentration and Acc umulation Significant differences among clones for rhiz ome+root dry mass were observed in spring, 2007 (Figure 3-5 and Table A-5). Clone FL-13 ha d significantly greater mass than the C-65, C49 and Argentine. No genotypic effect was detected for summer or fall 2007, nor spring or fall 2008. Rhizome+root mass decreased continuously from spring 2007 to fall 2008 (Table 3-4). Only small genotypic differences in N concentration were observed in summer 2007 and fall 2008 (Table 3-5). No genotypic effect was observed for N concentration in spring 2007, spring 2008, or fall 2007. The lowest N values were observed in spring of both growing seasons (Table 3-4). Small differences were also obser ved for P concentrations in spring and summer 2007, and fall 2008 with the greatest P concentrat ions observed in fall 2007 and 2008 (Table 36). Genotypic differences for rhizome+root N accumulation were observed in spring 2007 (Figure 3-6). These differences were mainly due to differences in dry mass (Figure 3-5). Nitrogen rhizome+root N accumulation increased during the 2007 growing season. However, N accumulation did not change between spring and fall 2008 (Table 3-4). Differences among genotypes were observed for P accumulation onl y in spring 2007 (Figure 3-7). While P accumulation increased during the 2007 growing s eason, it decreased during 2008 (Table 3-4). Forage Nutrient Composition and Accumulation Nitrogen concentration remained relatively constant during spring and summer of the 2007 and 2008 growing seasons (Figure 3-8 and 3-9). With the exception of the forage harvested at the end of October 2007, N concentration was higher during fall than spring and summer of both years. Minimal genotypic differences for N con centration were observed at the first spring

PAGE 46

46 harvest of 2007 and 2008 and a few additional harv ests during the 2-year study period (Tables A6 and A-7). There was not a consistent patter n of N concentration differences among clones. Phosphorus concentrations increased duri ng spring and summer of 2007 and then decreased during the fall (Figure 3-10). A marked increase on P concentration was observed in spring 2008 (Figure 3-11). It co ntinued increasing during summe r and fall 2008 but at a lower rate. Significant differences were found among clones for P concentration at each harvest in 2007 (Table A-8). Differences were also observed at the summer and fall harvests of 2008 (Table A-9). Argentine consistently ha d the greatest forage P concentr ations throughout all harvests (Figures 3-10 and 3-11). A similar seasonal pattern was observed for N and P uptake in each year. In 2007, N and P uptake were low at the beginning of the spring (Figures 3-12 and 3-14). They increased during spring reaching a maximum at the end of this season, decreased and stayed low during the summer, increased at the end of the summer reaching another maximum, and decreased during fall reaching a minimum at the end of the seas on. In 2008, N and P uptake were also low at the beginning of spring (Figures 3-13 and 3-15). They increased continuously until reaching a maximum at the end of the summer. N and P uptak e decreased during fall reaching a minimum at the end of the season. Genotypic differences in N accumulation were ob served for most harvests (Tables A-10 and A-11). The main differences were observed in spring of both years (T able 3-7). FL-3, FL-13 and FL-93 accumulated significantly more N than Argentine in spring 2007 and 2008. Clones also accumulated different amounts of N in summer 2007 and 2008. Any of the novel hybrids was able to accumulate more N than Argentine in summer. Differences for fall N accumulation were only observed in 2008. Tifton 7 was the only clone able to accumulate more N than

PAGE 47

47 Argentine in fall 2008. There were no genotypic differences for annual N accumulation in 2007 or 2008. Genotypic differences for P uptake were al so observed for most harvests (Tables A-12 and A-13). Several hybrids accumulated more P than Argentine in spring 2007 and 2008 (Table 3-8). In contrast, no other clone accumulated more P than Argentine in summer 2007 or 2008. No differences were detected among clones for P accumulation in fall 2007. However, Tifton 7 accumulated more P than most clones in fall 2008. There were differences among clones for annual P accumulation in 2007 and 2008. However, any of the novel hybrids was able to accumulate more P than Argentine in 2007 or 2008. Discussion Genetic variability for seasona lity of above ground biom ass production is present in the tetraploid bahiagrass germplasm. The main diffe rences can be expected to occur in spring and fall when bahiagrass is grown in subtropical z ones. Hybridization of sexual and tetraploid bahiagrass clones selected for superior cool-seas on re-growth and freeze resistance (Acua et al., 2009) resulted in hybrids with an extended growing season. The heterosis observed for these characteristics of hybrids grown in a space-plant nur sery have been fixed by their high degree of inherent apomixis. The growing season extension is especially important when bahiagrass is used as forage. Cultivars with an extended growing season can provide forage for grazing livestock longer during the year. Increasing the gr owing season of bahiagrass lines can also be exploited for turf. Turf cultivars with an extended growing season will maintain good sod quality longer during each season. Further research is ne eded to determine if these bahiagrass hybrids exhibiting a longer growing season are less photoper iod sensitive, as reporte d by Sinclair et al. (2001). Anatomical or physiologi cal characteristic behind their higher freeze resistance should also be further investigated.

PAGE 48

48 The observed decline in rhizome-root mass throughout the experiment could be the result of frequent defoliation, or the tran sition between individual plants in a plot to a well developed sward. Since samples were always co llected from the center of the original plants, this decline in mass might result from spread ing and independence of new rh izomes. The lack of genotypic differences for rhizome+root mass would support th is hypothesis since the cultivar Argentine is well know for its robust rhizome-root system and tolerance to intense defoliation. The decline in N and P concentrations from fall to spring seems to be the result of utiliza tion of stored nutrients for spring regrowth. Blue (1973) reported that the bahiagrass rhizome-root system can be a source of nutrients for new growth. Although remob ilization of N and P stored in these organs would have little relevance for fo rage yields in intensive produc tion systems, it seems to be an important survival mechanism. Hybrids that produced more biomass in spring also removed more N and P. For example, FL-93 produced consistently more spring biomass than Argentine across years and locations, and it also removed more N and P in spring 2007 and 2008. This indicates that hybrids with an extended growing season would reduce the amount of nutrient losses in runoff water in years with high precipitation in the spring. Hybrids showing this characteristic will be more appropriate for crop rotation with row crops, since remnant fert ilizer would be extracted and converted into forage soone r during the crop cycle. Seasonal biomass yield differences affected total biomass in 2007 at Gainesville. For example, C-92 produced more biomass in spring and fall than Argentine, resulting in higher annual biomass accumulation. FL-13 producing more spring biomass than Argentine resulted in a higher annual biomass yield. Some of these novel hybrids with an extended growing season were also able to accumulate more biomass than Argentine in Live Oak and Quincy in 2008. In

PAGE 49

49 contrast, superior spring or fall biomass yields did not result in higher annual biomass accumulation at Gainesville in 2008. These results indicate that in most years and locations across north Florida hybrids showing superior spring and fall growth will accumulate annually higher biomass. However, long term studies will be needed to determine the consistency of annual forage accumulation of these novel apomictic hybrids across multiple years. Based on our data, hybrids with an extended growing season would be also more appropriate for hay or biomass production systems with one or two harvests per year. Greater spring and fall biomass yield did not result in higher annual N and P accumulation. This is because clones with a shorter growing season tended to have greater N and P tissue concentrations which offset the lower biomass yiel ds. This was particularly true with Argentine, which had higher P concentrations than othe r clones for most of 2007 and 2008. If annual nutrient removal is the considered objective fo r planting bahiagrass on nutrient impacted land, available cultivars should be considered appropri ate. The possibility of transferring genes for superior P and N accumulation present in avai lable cultivars should be further evaluated. In this study we reported seasonal biom ass production, N and P concentrations and accumulation of several apomictic bahiagrass cl ones. Genetic variability for seasonality of biomass yields was reported. Higher cool-seas on biomass production and N and P removal was observed for a few novel hybrids. Greater cool-season biomass resulted in higher annual biomass accumulation of newly developed clones.

PAGE 50

50 Table 3-1. Seasonal and annual biomass accumulati on of 12 apomictic bahiagrass clones grown at Gainesville, FL. 2007 2008 Clone Spring Summer Fall Total Spring Summer Fall Total ---------------------------------------------g m-2----------------------------------------FL-13 483 657 94 1233 353 720 257 1330 C-92 455 835 71 1360 319 953 248 1520 C-65 428 835 85 1347 356 926 292 1574 FL-14 398 952 30 1380 337 887 319 1543 Tifton 7 379 893 79 1350 240 981 401 1622 FL-3 354 887 63 1303 353 886 287 1526 FL-93 353 879 44 1276 366 873 278 1517 FL-3B 331 788 42 1162 298 908 253 1460 FL-122 278 805 50 1133 217 842 231 1291 FL-21 261 775 42 1078 334 728 214 1276 C-49 217 808 44 1069 247 841 289 1377 Argentine 112 940 37 1089 188 1022 253 1464 MSD 107 138 29 183 89 100 75 207 Clones were ordered based on spring 2007 biomass. Minimum significant difference, Waller-D uncan means separation procedure. Table 3-2. Seasonal biomass accumulation of 13 apom ictic bahiagrass clones grown at Live Oak, FL. Clone Spring Summer Fall Total --------------------------------------g m-2-------------------------------------FL-93 260 916 90 1265 C-92 229 984 115 1328 FL-13 226 750 97 1073 C-65 210 835 127 1172 FL-3 195 867 97 1159 FL-14 195 888 144 1226 FL-3B 188 880 136 1204 FL-21 157 763 74 994 Tifton 7 149 870 109 1127 FL-122 137 894 114 1145 C-49 135 930 167 1231 Argentine 121 895 82 1099 Common 0 563 55 618 MSD 38 106 23 123 Clones were ordered based on spring biomass. Minimum significant difference, Waller-D uncan means separation procedure.

PAGE 51

51 Table 3-3. Seasonal biomass accumulation of 13 a pomictic bahiagrass clones grown at Quincy, FL. Clone Spring Summer Fall Total --------------------------------------g m-2-------------------------------------FL-13 215 525 129 776 C-49 212 651 185 972 C-92 206 631 136 882 FL-14 193 577 155 841 FL-93 176 642 144 869 FL-3B 161 638 155 880 FL-3 153 636 148 875 C-65 137 635 153 868 Tifton 7 135 542 169 790 FL-21 129 542 146 747 FL-122 110 642 156 853 Argentine 86 570 160 769 Common 1 460 123 583 MSD 44 80 43 99 Clones were ordered based on spring biomass. Minimum significant difference, Waller-Duncan means separation procedure. Table 3-4. Rhizome+root dry mass, N and P concentrations and accumulation (Gainesville). Season Rhizome+root mass Nitrogen Phosphorus Nitrogen Phosphorus mg cm-3 g kg-1 g kg-1 mg cm-3 mg cm-3 Spring 2007 41 10.2 0.68 0.42 0.028 Summer 2007 36 11.0 0.85 0.38 0.030 Fall 2007 32 15.7 1.20 0.50 0.040 Spring 2008 18 9.3 0.88 0.17 0.020 Fall 2008 14 12.6 0.97 0.18 0.010 MSD 4 1.0 0.09 0.06 0.005 Minimum significant difference, Walle r-Duncan means separation procedure.

PAGE 52

52 Table 3-5. Nitrogen concentrati on in rhizome+roots of 12 bahiag rass clones (Gainesville). 2007 2008 Clone Spring Summer Fall Spring Fall -----------------------------------------g kg-1-----------------------------------FL-21 9.4 14.0 15.1 9.5 16.8 FL-13 11.6 12.7 13.7 12.2 15.2 C-65 11.8 11.4 15.5 9.5 12.6 FL-122 8.3 7.4 12.6 7.4 12.6 FL-3B 11.0 14.0 17.9 9.6 12.3 FL-3 10.0 12.2 14.7 10.7 12.1 C-49 10.0 11.4 14.6 8.8 12.1 FL-93 9.5 9.7 16.2 10.1 12.1 Argentine 10.7 11.3 15.9 9.3 11.7 FL-14 9.5 7.2 15.0 6.6 11.6 C-92 10.9 9.9 18.2 8.4 11.2 Tifton 7 9.1 11.3 18.8 9.6 10.4 MSD ns 4.1 ns ns 4.4 Clones were ordered based on N concentration for fall 2008. Minimum significant difference, Waller-Duncan means separation procedure. Table 3-6. Phosphorus concentration in rhizome+r oots of 12 bahiagrass clones (Gainesville). 2007 2008 Clone Spring Summer Fall Spring Fall ---------------------------------------g kg-1-----------------------------------Argentine 0.91 0.97 1.19 1.02 1.33 FL-122 0.83 1.02 1.12 1.20 1.22 FL-21 0.69 0.83 1.05 0.87 1.04 C-65 0.81 1.02 1.45 0.69 0.97 FL-93 0.57 0.68 1.27 0.79 0.94 FL-3 0.65 0.83 1.34 0.88 0.93 C-49 0.63 0.87 1.10 0.67 0.92 FL-13 0.69 0.75 1.56 0.68 0.91 FL-14 0.47 0.60 1.13 1.20 0.90 Tifton 7 0.66 0.97 1.16 1.06 0.87 C-92 0.70 0.97 1.31 0.82 0.84 FL-3B 0.53 0.68 1.32 0.69 0.73 MSD 0.26 0.33 ns ns 0.26 Clones were ordered based on P concentration for fall 2008. Minimum significant difference, Waller-Duncan means separation procedure.

PAGE 53

53 Table 3-7. Seasonal and annual forage N uptake of 12 apomictic bahiagrass clones (Gainesville). 2007 2008 Clone Spring Summer Fall Total Spring Summer Fall Total ---------------------------------------------g m-2----------------------------------------C-92 7.9 11.8 6.1 25.8 4.6 14.6 5.6 24.9 FL-13 7.6 9.1 6.2 22.9 5.4 11.4 5.8 22.7 C-65 7.6 11.3 6.3 25.2 5.2 14.1 6.2 25.5 FL-3 6.4 11.3 7.1 24.7 5.4 13.4 6.1 25.0 Tifton 7 6.4 11.6 6.9 24.9 3.9 14.3 8.4 26.4 FL-93 5.8 11.5 7.2 24.5 5.4 13.8 6.0 24.7 FL-14 5.6 12.0 7.3 25.0 4.6 14.1 6.5 25.3 FL-3B 5.5 10.2 6.0 21.7 4.3 14.5 5.4 24.2 FL-122 5 11.1 6.3 22.4 3.5 13.4 5.2 22.1 FL-21 4.3 9.7 6.6 20.6 5.3 12.5 4.9 22.7 C-49 4.2 11.7 5.6 21.5 3.5 12.6 6.2 22.2 Argentine 2.0 11.0 7.8 20.7 2.9 16.2 5.6 24.7 MSD 3.0 2.4 ns ns 2.3 2.6 1.8 ns Clones were ordered based on spring 2007 N accumulation. Minimum significant difference, Waller-D uncan means separation procedure. Table 3-8. Seasonal and annual forage P uptake of 12 apomictic bahiagrass clones grown at Gainesville, FL. 2007 2008 Clone Spring Summer Fall Total Spring Summer Fall Total ---------------------------------------------g m-2----------------------------------------C-92 0.7 1.4 0.5 2.6 0.4 2.0 0.6 2.9 FL-13 0.7 1.1 0.6 2.3 0.4 1.5 0.6 2.5 Tifton 7 0.6 1.6 0.7 2.9 0.3 2.2 1.0 3.5 C-65 0.6 1.3 0.6 2.5 0.4 1.9 0.6 2.9 FL-93 0.6 1.4 0.6 2.5 0.4 1.8 0.6 2.8 FL-14 0.5 1.4 0.7 2.7 0.4 1.8 0.8 3.0 FL-3 0.5 1.3 0.7 2.5 0.4 1.8 0.7 2.9 FL-3B 0.5 1.2 0.6 2.3 0.3 1.7 0.5 2.6 FL-122 0.4 1.4 0.6 2.4 0.3 1.9 0.5 2.8 C-49 0.4 1.2 0.5 2.1 0.3 1.7 0.6 2.6 FL-21 0.4 1.2 0.6 2.2 0.4 1.5 0.5 2.4 Argentine 0.2 1.7 0.8 2.7 0.2 2.6 0.7 3.6 MSD 0.2 0.3 ns 0.6 0.1 0.3 0.2 0.4 Clones were ordered based on spring 2007 P accumulation. Minimum significant difference, Waller-D uncan means separation procedure.

PAGE 54

54 0 50 100 150 200 250 300 350 400 9-Mar28-Apr17-Jun6-Aug25-Sep14-Nov3-JanForage mass, g m-2 Fl-13 C-92 Fl-93 Argentine Figure 3-1. Biomass of 4 bahiagrass clones grown at Gainesville in 2007. 0 50 100 150 200 250 300 350 400 450 500 12-Apr1-Jun21-Jul9-Sep29-Oct18-DecForage mass (g m-2) Fl-13 C-92 Fl-93 Argentine Figure 3-2. Biomass of 4 bahiagrass clones grown at Gainesville in 2008.

PAGE 55

55 0 50 100 150 200 250 300 350 400 450 12-Apr1-Jun21-Jul9-Sep29-Oct18-DecForage mass (g m-2) Fl-13 C-92 Fl-93 Argentine Common Figure 3-3. Biomass of 5 bahiagrass clones grown at Live Oak in 2008. 0 50 100 150 200 250 300 12-Apr1-Jun21-Jul9-Sep29-Oct18-DecForage mass (g m-2) Fl-13 Fl-93 C-92 Argentine Common Figure 3-4. Biomass of 5 bahiagrass clones grown at Quincy in 2008.

PAGE 56

56 Argentine C-49 Tifton 7 Fl-3 C-65 Fl-14 Fl-93 Fl-92 Fl-3B Fl-122 Fl-21 Fl-13 0 25 50 75 100abc abc abcab a bc bc bc bc bc c cRhizome+root mass (mg cm-3) Figure 3-5. Spring rhizome-root mass of 12 bahi agrass clones grown at Gainesville, FL. Error bars represent the standard deviations. Cl one means followed by the same letter are not different. C-49 Argentine Tifton 7 Fl-3 Fl-14 Fl-93 Fl-122 C-65 C-92 Fl-3B Fl-21 Fl-13 0.00 0.25 0.50 0.75 1.00 c N (g cm-3)c bc bc bc bc bc ab bc bc bc a Figure 3-6. Spring 2007 rhizome+r oot nitrogen accumulation of 12 bahiagrass clones grown at Gainesville, FL. Error bars represent the standard devi ations. Clone means followed by the same letter are not different.

PAGE 57

57 C-49 Fl-14 Argentine Tifton 7 Fl-93 Fl-3 Fl-3B C-65 C-92 Fl-21 Fl-122 Fl-13 0.000 0.025 0.050 0.075P (mg cm-3)a ab bc ab c bc bc d c c c c Figure 3-7. Spring 2007 rhizome+root phosphorus accumulation of 4 representative bahiagrass clones grown at Gainesville, FL. Error bars represent the standa rd deviations. Clone means followed by the same letter are not different. 0.5 5.5 10.5 15.5 20.5 25.5 30.5 9-Mar28-Apr17-Jun6-Aug25-Sep14-Nov3-JanN (g kg-1) Fl-13 C-92 Fl-93 Argentine Figure 3-8. Nitrogen concentra tion of 4 representative ba hiagrass clones during the 2007 growing season.

PAGE 58

58 0.5 5.5 10.5 15.5 20.5 25.5 30.5 12-Apr1-Jun21-Jul9-Sep29-Oct18-DecN (g kg-1) Fl-13 C-92 Fl-93 Argentine Figure 3-9. Nitrogen concentra tion of 4 bahiagrass clones du ring the 2008 growing season. 0.05 0.55 1.05 1.55 2.05 2.55 3.05 9-Mar28-Apr17-Jun6-Aug25-Sep14-Nov3-JanP (g kg-1) Fl-13 C-92 Fl-93 Argentine Figure 3-10. Phosphorus concentration of 4 bahiagrass clones during the 2007 growing season.

PAGE 59

59 0.05 0.55 1.05 1.55 2.05 2.55 3.05 12-Apr1-Jun21-Jul9-Sep29-Oct18-DecP (g kg-1) Fl-13 Fl-21 C-92 Fl-93 Argentine Figure 3-11. Phosphorus concentration of 4 bahiagrass clones during the 2008 growing season. 0 1 2 3 4 5 6 7 8 9-Mar28-Apr17-Jun6-Aug25-Sep14-Nov3-JanN (g m-2) Fl-13 C-92 Fl-93 Argentine Figure 3-12. Nitrogen accumulation of four bahiagrass clones during the 2007 growing season.

PAGE 60

60 0 1 2 3 4 5 6 7 8 12-Apr1-Jun21-Jul9-Sep29-Oct18-DecN (g m-2) Fl-13 C-92 Fl-93 Argentine Tifton 7 Figure 3-13. Nitrogen accumulation of four bahiagrass clones during the 2008 growing season. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 9-Mar28-Apr17-Jun6-Aug25-Sep14-Nov3-JanP (g m-2) Fl-13 C-92 Fl-93 Argentine Figure 3-14. Phosphorus accumulation of four bahiagrass clones during the 2007 growing season.

PAGE 61

61 0 0.2 0.4 0.6 0.8 1 1.2 12-Apr1-Jun21-Jul9-Sep29-Oct18-DecP (g m-2) Fl-13 C-92 Fl-93 Argentine Tifton 7 Figure 3-15. Phosphorus accumulation of four bahiagrass clones during the 2008 growing season.

PAGE 62

62 CHAPTER 4 ROOT DEPTH DEVELOPMENT IN APOMICTIC BAHIAGRASS Introduction Drought is an occasion al hazard in all types of dry-land agriculture. In wet regions, crops are especially vulnerable to sudden dry spells re sulting in more severe damage than crops grown in dry regions. Success with dry-land farm ing depends upon appropriate crop and cultivar selections (Loomis and Connor, 1992) Higher yielding genotypes coul d be bred more efficiently if attributes that confer yield under water-limited conditions are identified and used as selection criteria. The general principle is that crops yield best when thei r developmental cycle avoids or tolerates periods of water shortage and they ma ke the best use of th e available water supply (Ludlow and Muchow, 1990). One such plant characteristic that ma y optimize productivity under water-limited conditions is rapid vertical root developmen t for recovering soil water (Ludlow and Muchow, 1990; Sinclair, 1999). Rapid vertical root development is expected to result in deeper rooting. Studies with subtropical grasses have indicated benefits of de ep rooting. Burton et al. (1954) observed marked differences for early root pene tration among subtropical forages. Forages with deeper roots were classified as more drought to lerant. In contrast, tota l root dry mass was not correlated with drought toleran ce. In another experiment, Ma rcum et al. (1995) compared maximum root depth of 25 Zoysiagrass ( Zoysia sp.) cultivars and found th at cultivar rooting depth correlated with the drought tolerance of the cultivars. Genot ypic variation for rooting depth was also found within grain crops, such as ri ce (Shen et al., 2001), sunflower and soybean (Dardanelli et al., 1997). Deep rooting was also c onsidered important for deep nitrogen uptake of vegetable crops, especially in areas where nitr ate can be leached out of the soil profile, potentially contaminating ground water (K ristensen and Thorup-Kristensen, 2004).

PAGE 63

63 Bahiagrass, Paspalum notatum Flgg, is a warm-season grass cultivated in southeastern USA as forage and utility turf. It is considered a drought tolerant species because of its deep and dense root system (Burton, 1943). The bahiagrass tetraploid ge rmplasm reproduces asexually by apomixis, which allows the perpetuation of superi or hybrids. The genetic va riability of a variety of agronomically important traits has been report ed for this species (Gates et al., 2004; Acua et al., 2007; Acua et al., 2009). However, it is unk nown if there is genetic variability for root architecture and performance. The evaluation of the bahiagrass germplasm for rate of root development could indicate geno types that can reach stored wa ter in deep soil layers more rapidly. A screening technique would be advantageous to evalua te a large number of genotypes in less space and more timely. Since bahiagrass is frequently defoliated mechanically or by livestock the effect of defoliation on the rate of root depth development also needs to be considered. The aim of this research was to generate a sc reening technique that can be readily used to detect genotypic variation for rate of root depth development (RRDD) among tetraploid cultivars or breeding lines of bahiagrass, analyze th e variability for RRDD among novel tetraploid hybrids, and determine the effect of stress (i.e., defoliation) on RRDD. Materials and Methods Plant Material Thirte en bahiagrass tetraploid clones were used for this research. The cultivar Argentine, an ecotype known as Common and the unreleased clone Tifton 7 (obtaine d from Dr. G. Burton, USDA-ARS, Tifton, GA) were selected for this st udy because they were considered well adapted clones in the southeastern USA (Burton, 1992), a nd because they were classified as highly apomictic, based on embryo sac observations and fi eld progeny tests (Acua et al., 2007). The other ten clones were novel hybrids generated by crossing sexual induced autotetraploid clones

PAGE 64

64 as female parents and apomictic clones as pollen donors. Seven of these hybrids (FL-3, FL-3B, FL-13, FL-14, FL-21, FL-93 and FL-122) were gene rated at the University of Florida, and selected for this study because of their exte nded growing season, cold tolerance, and high apomictic expression (Acua et al., 2009). The other 3 hybrids (C-49, C-65, and C-92) were generated at Universidad Nacional del Nordeste, Argentina, and se lected for their high herbage accumulation (Quarin, personal communication). Development of a Screening Technique The rate of root depth developm ent (RRDD) which is the slope of the linear function between time and depth of the deepest root, was monitored by growing Argentine and Tifton 7 plants in clear acrylic tubes (Figure 4-1). Tubes 100-cm length and either 3.5-cm or 10-cm diameter were tested. These two tube sizes were used to evaluate the effect of soil volume and plant competition on RRDD. All tubes were filled with a sandy soil (Thermic, coated Typic Quartzipsamment) collected at Live Oak, Flor ida, or with commercial potting mix (Jungle Growth, Professional Grower Mix, Piedmont Pacifi c Inc., Statham, GA). These two soils were used to test the potential eff ect of organic and inorganic so ils on RRDD. The field soil was collected from three successive 30.5-cm thick layers, the bulk density of each layer was determined, and columns were filled with material from these three layers maintaining thickness and density. Seeds from the two genotypes were scar ified using concentrated sulfuric acid for 10 min, washed, and sown on the substrate surface of each column. A single plant was grown in 3.5-cm diameter tubes, while 5 plants were gr own in the 10-cm diameter tubes. The growing medium was maintained at field capacity by wa tering every other day. The depth of the deepest visible root was recorded three times per w eek. Aboveand below-ground plant dry mass were determined at the end of each trial. Ten tubes were used for each treatment and these were

PAGE 65

65 considered replicates. Two tria ls in 2006 (from 16 May to 17 Ju ly, and from 7 August to 7 October) were located outdoors in Gainesville, FL. Effect of Defoliation on the Ra te of Root Depth Development Using the sam e procedure described above, Ar gentine and Tifton 7 plants were grown in clear acrylic columns from 7 August to 7 October, 2006. Plants were grown in 3.5-cm or 10-cm diameter tubes filled with inorganic or organi c soil. Plants were manually defoliated with scissors at 3 cm from the soil surface weekly or biweekly starting on 8 September. Ten replications were used for each treatment. Genotypic Variation for Rate of Root Depth Development Ten novel apom ictic hybrids, Argentine, Ti fton 7 and Common were grown as described above in 3.5-cm diameter columns filled with sandy soil from 21 May to 5 July 2007. Eight replications of each clone were used for this tr ial. The depth of the deepest visible root was recorded three times per week. Aboveand below-ground plant dry mass were determined at the end of the experiment. Statistical Analyses Regression analysis was used to estim ate th e slope (RRDD) of the linear function between time and depth of the deepest visible root. The RRDD, above and below ground mass data were analyzed using PROC GLM (PC SAS version 9.2, SAS Institute, Cary, NC ) as a randomized complete block design. When significant differences were detected for one variable, the Fishers least significant difference (LSD) was used for comparing two means and the Duncans Multiple Range Test was used for mean separations when comparing more than two means.

PAGE 66

66 Results Development of a Screening Technique A linear relationship between root depth and time was obs erved when Argentine and Tifton 7 were grown in clear acr ylic tubes (Figures 4-2 and 43). The increase in root depth proved to be highly linear under all circum stan ces over the entire observation period. The RRDD values, (i.e., slope) were similar between genot ypes (Figures 4-4 a nd 4-5). There were no interactions between genotypes and the growing medium or tube size. There were no RRDD differences when the experiment was carried out in spring or summer. The RRDD of plants grown in potting mix or soil were similar, indica ting that bulk density a nd other soil attributes (within the range tested) had no measurable effect on RRDD. Therefore, either medium can be used to screen a larger number of bahiagrass genotypes. RRDD were similar when plants were grown in small or large tube diameters indicating th at small tubes can be used efficiently to test a large number of genotypes. Tifton 7 produced more above ground mass (AGM ) than Argentine (Figures 4-4 and 4-5), and field results supported these findings (Cha pter 3). Tifton 7 also produced more below ground mass (BGM) than Argentine during the spring, but they were not significantly different during the summer. In general, plants produced more biomass when grown in potting mix, which might be the result of its higher nitrogen content, compar ed with the sandy soil us ed in this experiment. Effect of Defoliation on the Ra te of Root Depth Development W eekly defoliation significantly reduced th e RRDD of plants grown in small tubes containing soil (Figure 4-6). Argentines RRDD was equally reduced by both defoliation treatments (weekly and biweekly). However, defoliation did not have an effect on RRDD of plants growing in small tubes with potting mix or big tubes with soil. The lack of response for plants grown in the potting mix mi ght again result from readily av ailable nutrients to the roots.

PAGE 67

67 Since the bigger tubes contained more soil, they too offered a greater amount of nutrients than when plants were grown in the small tubes. Genotypic Variation for Rate of Root Depth Development Since there were no app reciable differences among growing medium, tube size or date during the growing season, the novel apomictic hybr ids were grown in small tubes filled with soil and were grown at the end of the spring 2008. Hybrid FL-122 had a greater RRDD than the other 12 apomictic hybrids in this study (Figure 47) including its male parent, Argentine, or other hybrids sharing this same male parent. Unfortunately, there were not any other hybrids sharing the same female parent (Q-4188). The su periority of FL-122 for RRDD might be related to genes present in Q-4188 or it could also be rela ted to heterosis resulting from this combination of parents. No differences were observe d among the other 12 hybrids for RRDD. The low variation among the eight re plications included in this experi ment indicates the consistency of these results. Hybrid FL-122 also produced more above and below ground dry mass compared with the other hybrids (Figure 4-7). Discussion These experim ents showed that clear acrylic tubes can be effectively used to screen bahiagrass germplasm for RRDD. Of particular interest was the lack of variation resulting from the use of two contrasting growing media. This indicates that mineral or organic growing media can be used indifferently to screen bahiagrass genotypes for RRDD. The results also indicated that small tubes (3.5-cm diameter) can be used as efficiently as big tube s (10-cm diameter) to screen genotypes for RRDD differences during earl y growth. These results are important because small tubes are lighter, use less growing medi um, and require less space, allowing for the screening of a larger number of genotypes. The results also indi cate that the screening for RRDD can be carried out at the be ginning or middle of the bahiag rass growing season without a

PAGE 68

68 significant effect on the results. Kramer and Bo yer (1995) stated that a RRDD of 10 to 12 mm per day was considered typical among grasses. In our expe riment, the average RRDD was 18 mm per day. However, it is important to recognize that these values are e xpected to be affected by environmental variables, such as temperature, soil available water, a nd nutrient concentration in the growing medium. Although bahiagrass is able to tolerate severe and frequent grazing or mowing (Gates et al., 2004), our results indicated that defoliation decreases the RRDD. It seems that defoliation limits the plants ability to mobilize ca rbohydrates for root depth devel opment. These results indicate that frequent defoliation would compromise bahiagrass ability for water and nutrient uptake, and probably establishment. Genotypic variability for RRDD was found am ong a group of novel apomictic bahiagrass hybrids with different genetic backgrounds. Although most root studies have been based on measurements of root mass and maximum root de pth instead of root depth development, some indirect comparisons can be made with st udies involving maximum root depth. Genotypic variation for maximum root depth was reported fo r other grasses, such as zoysiagrass (Marcum et al., 1995), fescue and perennial ryegrass (Bonos et al., 2004). Subsequent studies have shown that selection for maximum rooting depth can result in drought tole rant plants, and that selection based on root characteristics can be more eff ective than selection ba sed on field screening (Karcher et al., 2008). In our e xperiment, hybrid FL-122 not only showed greater RRDD but also greater above and below ground mass. This might be the result of its ability for exploring deeper soil layers by extending its roots faster. However, further experiments are needed to examine this hypothesis. These characteristics observed with line FL-122 may result in faster establishment,

PAGE 69

69 which is highly desirable in perennial grasses. The genetic variability for RRDD observed by this experiment indicates the potential use of th is trait for bahiagrass genetic improvement.

PAGE 70

70 Figure 4-1 Bahiagrass plants growing in clear acrylic t ubes. A) 3.5-cm diameter tubes filled with potting mix. B) 10-cm diameter tubes filled with sandy soil. y = 2.5794x 5.5252 R2 = 0.9953 y = 2.4271x 10.357 R2 = 0.9953 0 10 20 30 40 50 60 70 80 90 100 010203040 daysRoot Depth (cm) Argentine Tifton 7 Figure 4-2. Rate of root depth development for Argentine and Ti fton 7 growing in clear acrylic tubes (10-cm diameter) filled with potting mix. A B

PAGE 71

71 y = 2.5396x 10.836 R2 = 0.9986 y = 2.5671x + 1.8863 R2 = 0.9901 0 10 20 30 40 50 60 70 80 90 010203040 daysRoot Depth (cm) Argentine Tifton 7 Figure 4-3. Rate of root depth development for Argentine and Ti fton 7 growing in clear acrylic tubes (3.5-cm diameter) filled with soil.

PAGE 72

72 Argentine Tifton 7 0 1 2 3 soil, 3.5 cm potting mix, 3.5 cm soil, 10 cm potting mix, 10 cm Root elongation, cm d-1 Argentine Tifton 7 0 10 20 30 soil, 3.5 cm potting mix, 3.5 cm soil, 10 cm potting mix, 10 cm Above ground mass, g Argentine Tifton 7 0 5 10 15 soil, 3.5 cm potting mix, 3.5 cm soil, 10 cm potting mix, 10 cm Below ground mass, g Figure 4-4. Rate of root depth development, and above and below ground mass for Argentine and Tifton 7 grown in clear ac rylic tubes during spring 2006.

PAGE 73

73 Argentine Tifton 7 0 1 2 3 soil, 3.5 cm potting mix, 3.5 cm soil, 10 cm potting mix, 10 cm Root elongation, cm d-1 Argentine Tifton 7 0 10 20 30 soil, 3.5 cm potting mix, 3.5 cm soil, 10 cm potting mix, 10 cm Above ground mass, g Argentine Tifton 7 0 5 10 15 soil, 3.5 cm potting mix, 3.5 cm soil, 10 cm potting mix, 10 cm Below ground mass, g Figure 4-5. Rate of root depth development and above and below ground mass for Argentine and Tifton 7 grown in clear acrylic tubes during summer 2006.

PAGE 74

74 Argentine Tifton 7 0 1 2 3Control biweekly weekly a b ab a b bRoot elongation, cm d-1 Argentine Tifton 7 0 1 2 3Control biweekly weekly a a a a a aRoot elongation, cm d-1 Argentine Tifton 7 0 1 2 3Control biweekly weekly a a a a a aRoot elongation, cm d-13.5-cm diameter tubes filled with sandy soil 10-cm diameter tubes filled with sandy soil 3.5-cm diameter tubes filled with potting mix Figure 4-6. Effect of defoliation on root depth de velopment for Argentine and Tifton 7.

PAGE 75

75 Fl-21 Fl-3 Argentine Fl-93 Fl-14 C-92 Common C-49 C-65 Fl-3B Fl-13 Tifton 7 Fl-122 0 1 2 3RRDD, cm d-1b b b b b b b b b a bb b Fl-21 Fl-3 Argentine Fl-93 Fl-14 C-92 Common C-49 C-65 Fl-3B Fl-13 Tifton 7 Fl-122 0.0 0.1 0.2 0.3 0.4Above ground mass, ga b b b b bb b b b b bb Fl-21 Fl-3 Argentine Fl-93 Fl-14 C-92 Common C-49 C-65 Fl-3B Fl-13 Tifton 7 Fl-122 0.0 0.1 0.2 0.3 0.4 0.5Below ground mass, gb b b b b b b b b a b b b Figure 4-7. Rate of root depth development (RR DD), and above and below ground mass for 13 novel apomictic bahiagrass hybrids growing in clear plastic tubes during spring 2007.

PAGE 76

76 CHAPTER 5 IMPORTANCE OF RAPID VERTICAL ROOT DEVELOPMENT FOR RECOVERING OF SOIL NITR OGEN IN TETRAPLOID BAHIAGRASS Introduction Most of the physiological pro cesses contributing to crop yiel d are quantitatively dependent on plant nitrogen (N) availa bility (S inclair and Valdez, 2002). Therefore, the addition of organic or inorganic sources of N is essential for crop production in most environments. The highly weathered soils of the U.S. Southern Coastal Plain require relatively high N inputs for adequate crop production. Therefore, concer ns exist that the regions su btropical environment (warm temperatures and heavy rains) may contribute to significant N contamin ation of surface and groundwaters (Hubbard et al., 2004). Several studies have shown that deep-rooted crops can reduce the potential for N leaching losses by captu ring N from deep soil or subsoil horizons (Peterson et al., 1979; Huang et al., 1996; Thorup-Kristensen, 2001; Kristensen and ThorupKristensen, 2004). Rapid vertical ro ot growth was theoretically propos ed as an important trait for developing deep root systems and for recovering stored wa ter (Ludlow and Muchow, 1990; Sinclair, 1999). Moreover, N catch crops showing more rapid rates of vertical root development were able to deplete more N from deep soil layers than those showing slower rates (ThorupKristensen, 2001). The crucifer catc h crops were faster in devel oping deep rooting and depleting nitrogen from the subsoil than grass crops. Only small differences for rates of root penetration were found within botanical groups. Bahiagrass, Paspalum notatum Flgg, is a warm season perennial grass used as a forage and turf in the southeastern United States. It is also utilized in crop rota tions, particularly with row crops and various vegetable crops (Blount and Acua, 2009). Higher yields of row crops, great reductions in insect, nematode, and diseas e problems associated with peanut, cotton and soybean have been linked to short-term crop rotati on with bahiagrass. In a ddition, bahiagrass is a

PAGE 77

77 good candidate for mitigating excess soil N since it ha s a large, deep fibrous root system (Burton, 1943). Different rates of root depth development have been reported for novel apomictic bahiagrass hybrids (Chapter 4). The objective of this research was to determine if velocity differences in vertical root development are im portant for the uptake of N present in deep soil layers. Materials and Methods Plant Material Four bahiagrass clones were used for this experim ent: Argentine (PI 148996), FL-3, FL122 and C-65. Clones FL-3 and FL-122 were obtained by hybridizing sexual and apomictic tetraploid clones as part of a breeding program conducted at th e University of Florida, and selected based on superior cool-season regrowth, and freeze tolerance when grown as individual plants (Acua et al., 2009). Clone C-65 was also generated by crossing sexual and apomictic clones, but it was selected based on superior biomass yields when grown as individual plant (Quarin, personal communication). These four clones were selected for this experiment because variation for rate of root depth development was previously observed among them (Chapter 4). The four clones were classified as highly a pomictic (Acua et al., 2007; Acua et al., 2009), therefore, plant variation among replications was exp ected to be restricted to the environment. Root Development and Harvest Measurements The rate of root depth developm ent (RRDD) was monitored by gr owing the bahiagrass clones in clear acrylic tubes. Tubes (3.5-cm di ameter and 100-cm length) were filled with a sandy soil (Thermic, coated Typic Quartzipsamments ) collected at Live Oak, Florida. The field soil was collected from three c onsecutive 30.5-cm thick layers, the bulk dens ity of each layer was determined, and columns were filled with material from these three layers maintaining thickness and density. Seeds were scarified using concentrated sulfuric acid for 10 min, rinsed

PAGE 78

78 with tap water, and sown at the top of each column. The growing medium was maintained at field capacity by watering every other day. No fer tilizer was added to the growing media. Clear acrylic tubes where kept in ver tical position inside a PVC pipe. The tops of the acrylic tubes were painted white to exclude light. The depth of the deepest observable root was recorded two times per week. At the end of each trial, plan ts and growing media were placed on a 2-mm mesh and plants were washed with a stream of water. Roots and shoots (rhizomes+leaves) were separated by cutting the roots at the base of th e rhizomes, dried at 60 C for 48 h, and dry weights were determined. Eight tubes (replications) were used fo r each clone and the study was conducted twice in 2008 (from 2 May to 7 June, a nd from 31 July to 16 September), outdoors in Gainesville, Florida. 15N Uptake and Partitioning between Roots and Shoots To determine if the four bahiagrass clones differ in their ability to uptake N with soil depth, 15N label was applied to four replicates. To de monstrate, when at least one clone within each replicate was observed having roots d eeper than 60 cm, a solution containing 15N was prepared by dissolving 1444 mg of 10% enriched 15N as KNO3 in 200 ml of distilled water. A 2mm hole was drilled through the tube wall where th e deepest root of each plant was observed, and 5 ml of labeled solution wa s injected into the soil. The label was also imposed to the remaining fo ur replicates by injecting these tubes at the depth where the deepest root of FL-122 (clone having the fastest ve rtical root development) was observed for that replicate. These data helped to determine if differences in RRDD resulted in differences in N uptake. Three days after the in jection, plants were wash ed and roots and shoots were separated. Samples were dried at 60 C for 48 h and ground with a cyclone mill to 1-mm particle size. Determinations of 15N atom abundance were conducted at the Soil and Water

PAGE 79

79 Science Department Stable Isotope Mass Spectrome try Laboratory at the University of Florida. The atom % 15N/14N ratios were determined using a Costech model 4010 elemental analyzer (Costech Analytical I ndustries, Valencia, Calif.) coupled to a Finnigan MAT DeltaPlusXL mass spectrometer (continuous flow isotope ratio mass spectrometry; Thermo Finnigan, San Jose, Calif.) via a Finnigan Conflo III interface. Results were standardized using Ammonium Sulfate at 0.5, 1.0, 1.5, 2.0 and 5.0 atom percent 15N. Statistical Analyses Regression analysis was used to estim ate th e slope (RRDD) of the linear function between time and depth of the deepest visible root. Data were analyzed using PROC GLM (SAS version 9.2, SAS Institute, Cary NC) as a randomized complete block design. When significant differences were detected, the D uncans Multiple Range Test wa s used for mean separations. Results Root Depth Development As reported for other bahiagrass clones (Chapt er 4), root depth increased linearly with tim e (Figure 5-1). The rates of root depth development (RRDD), which is the slope of the root depth development plots, were significantly differe nt among clones. Clone FL-122 had higher RRDD than the other 3 apomictic clones (Figure 5-2). Although there was a significant season (spring vs. summer) effect for RRDD, the apomictic clones were ranked in a similar order in both seasons (Figure 5-2). Clone C-65 showed greater RRDD than FL-3 and Argentine during spring. This superiority of FL-122 for early RRDD confir med the results reported previously (Chapter 4). Plant Mass and Tiller Number There were no significant differences am ong clones for the number of tillers produced by the end of the experiment in either trial. Significant differences were observed among the 4

PAGE 80

80 bahiagrass clones for above-ground (AGM) a nd below-ground (BGM) mass during spring 2008 and only for BGM during summer 2008 (Figure 53). While FL-65 produced more AGM than FL-3 and Argentine during the spring, it did not differ from FL-122. C-65 and FL-122 produced more BGM during the spring than FL-3, but th ey did not produce significantly more than Argentine. During the summer FL-122 produced significantly more BGM than the other 3 clones. 15N Uptake and Partitioning between Roots and Shoots Atom % 15N abundance in roots and shoots was no t significantly different among clones when 15N-enriched KNO3 was applied where the deepest visible root of each plant was observed (Figure 5-4). Results were similar in spring and summer (Figure 5-4). These results indicate that roots from these four clones have the same ability for N uptake once they reach a particular soil depth. The atom % 15N abundance was significantly greater in shoots than in roots indicating that the N absorbed after the injection of 15N-enriched KNO3 was preferentially mobilized to shoots. Atom % 15N abundance in roots and shoots was significantly different among clones when 15N-enriched KNO3 was applied to all clones at the point of the single deepest visible root of FL122 (Figure 5-5). Roots and shoots fr om clone FL-122 had higher atom % 15N abundance three days after the injection than the other three test clones. Clone C-65 had higher 15N abundance than FL-3 and Argentine, but only for the spring trial. Shoots of FL-122 had higher 15N abundance than roots of the same clone. However, there were no differences for 15N abundance between roots and shoots of C-65, FL-3 or Argentine. Correlations between 15N abundance in roots and RRDD were high for the spring (r = 0.91) and summer (r = 0.80). The correlations between 15N abundance in shoots and RRDD were also high for spring (r = 0.90) and summer (r = 0.80).

PAGE 81

81 Discussion Genetic va riability for RRDD exists within the bahiagrass tetraploid germplasm. Our results confirm the findings of our previous expe riments (chapter 4) s howing that clone FL-122 had greater RRDD as compared to other apomictic bahiagrass clones. It indicates that genetic improvements for rapid root penetration by tetr aploid bahiagrass are possible. Lehman and Englke (1991) reported relatively high narrow-sense heritability for root extension of creeping bentagrass ( Agrostis palustris Huds.). However, further studies are needed to evaluate the possibility of using this characteristic in a bahiagrass breeding program. The presence of apomixis in tetraploid bahiagrass cytotypes (Gates et al., 2004) woul d make feasible the perpetuation of new hybrids show ing this characteristic. No genotypic differences were observed for 15N abundance in shoots or roots when 15N was injected at rooting depth (Fig ure 5-4). These results indicate th at the rate of nitrogen uptake of actively growing roots was not different among genotypes. Thes e results also indicate that once N is taken up by bahiagrass plants, it is mob ilized to shoots at a constant rate. Since the N uptake capacity of roots is consta nt, the amount of N that a bahiagrass clone can accumulate at establishment mainly depends on soil e xploration by actively growing roots. When 15N was injected at the rooting depth, higher 15N abundance was observed in shoots of bahiagrass plants compared to roots. This result indicates that ni trogen is preferentially mobilized to shoots after it is taken up by bahiag rass plants. Shoots are the predominant site of nitrate reduction in grasses (Scheurwater et al., 2002). The observed rapid accumulation of 15N in bahiagrass shoots could be the result of absorbed nitrate being preferentially translocated to shoots for reduction and incorporat ion in organic forms. This ne w hypothesis needs to be tested by analyzing xylem sap after ni trate absorption, or by measuri ng the activity of nitrogen reductase in roots and shoots.

PAGE 82

82 Our results showed that rapid root penetra tion was determinant for rapid access and uptake of N present in deep soil layers (Figure 5-5). Ba hiagrass is typically characterized as having slow establishment due, in part, to weak seedlings th at may be outcompeted by weeds (Gates et al, 2004). Early rapid root penetration would result in a faster establishment due to rapid access to soil N and probably soil water. Uti lization of bahiagrass cultivars with rapid root penetration in crop rotations would lead to better soil N mitigation by reducing the chance for N leaching. Further research is needed to de termine the interaction of this trait with soil water and nutrient stresses. The 15N injection technique used for this experiment can be utilized with other species to discriminate between N uptake capacity of roots and consequences of deeper rooting. In this experiment, the technique was successfully used to determine that the N uptake capacity of several bahiagrass clones did not differ, and that higher RRDD was critical for rapid access to N in deep soil layers.

PAGE 83

83 y = 3.8039x 29.408 R2 = 0.9906 y = 2.5051x 23.756 R2 = 0.9926 0 10 20 30 40 50 60 70 80 90 100 01 02 03 04 0days after germinationroot depth (cm) Fl-3 Fl-122 Figure 5-1. Root depth development for FL-3 and FL -122 bahiagrass growing in clear acrylic tubes filled with sandy soil.

PAGE 84

84 ArgentineFl-3C-65Fl-122 0 1 2 3 4 5a b c cRRDD, cm d-1 ArgentineFl-3C-65Fl-122 0 1 2 3 4 5a b b bRRDD, cm d-1Spring Summer Figure 5-2. Rate of root depth development for Ar gentine, FL-3, C-65 and FL-122 bahiagrass grown in clear acrylic tubes filled with soil during the spring and summer 2008. Different letters indicate si gnificant differences between means. Error bars represent standard deviations.

PAGE 85

85 ArgentineFl-3C-65Fl-122 0.0 0.1 0.2 0.3Roots Shoots aba a a b b babMass, g ArgentineFl-3C-65Fl-122 0.0 0.1 0.2 0.3Roots Shoots a a a a a b b bMass, gSpring Summer Figure 5-3. Root and shoot mass for Argentine, FL -3, C-65 and FL-122 bahiagrass grown in clear acrylic tubes filled with soil dur ing the spring and summer 2008. Different letters represent significan t differences among clones. The error bars represent standard deviation.

PAGE 86

86 ArgentineFl-3C-65Fl-122 0 1 2 3 4 5 6Roots Shoots a a a a a a a aAtom %15N abundance ArgentineFl-3C-65Fl-122 0 1 2 3 4 5 6Roots Shoots a a a a a a a aAtom %15N abundanceSpring Summer Figure 5-4. Atom % 15N abundance in roots and shoots of Argentine, FL-3, C-65 and FL-122 bahiagrass grown in acrylic tubes during the spring and summer 2008. 15N-enriched KNO3 was injected where the deepest root of each clone was observed. Different letters represent significan t differences among clones. The error bars represent standard deviations.

PAGE 87

87 ArgentineFl-3C-65Fl-122 0 1 2 3 4 5 6Roots Shoots Atom %15N abundancea a b c c c b c ArgentineFl-3C-65Fl-122 0 1 2 3 4 5 6Roots Shoots Atom %15N abundancea a b b b b b bSpring Summer Figure 5-5. Atom % 15N abundance in roots and shoots of Argentine, FL-3, C-65 and FL-122 bahiagrass grown in acrylic tubes during the spring and summer 2008. 15N-enriched KNO3 was injected where the deepest root of FL-122 was observed. Different letters represent significant differences among clones. The error bars represent standard deviations.

PAGE 88

88 CHAPTER 6 IMPORTANCE OF ROOT MASS AND ROOT LENGTH DENSITY ON FORAGE PRODUCTI ON OF NOVEL APOM ICTIC BAHIAGRASS HYBRIDS Introduction Bahiagrass, Paspalum notatum Flgg, is a warm -season, pe rennial grass extensively cultivated as forage in Florida and the southe rn Coastal plain region of USA. Recent breeding efforts have resulted in high yielding experimental lines and cultivars of this species (Blount and Acua, 2009). The identification of traits responsible for the observed differences in forage yields could be utilized to e nhance the genetic improvement of bahiagrass and other subtropical grasses. Although rainfall is relatively high in the region where bahiagrass is cultivated, water is considered one of the main f actors limiting its growth (Gates et al., 2004). Regular water shortages, high vapor pressure gradients dur ing the growing season, and low water holding capacity of light textured soils often result in growth bei ng limited by water availability. Availability of essential minerals in the soil also limits bahiag rass forage production. This crop is mainly cultivated in soils with inherent low fer tility and low cation and anion exchange capacity. The study of root mass and root length density (RLD) in superficial soil layers might indicate the capacity of a genotype for uptake of nu trients with low mobility in the soil, such as phosphorus. Sinclair and Valdez (200 2) indicated that an extensiv e root system is especially advantageous for plant production in phosphoruslimited environments. The study of root mass and RLD in deep soil layers is expected to indicate the ability of a genotype for water and nitrogen uptake under limited condi tions. Plant species with deep roots are expected to yield more in water-limited environments due to thei r capacity to recover st ored soil water (Ludlow and Muchow, 1990). Selection for higher deep-root to shoot mass ratio in tall fescue (Bonos et al., 2004) resulted in genotypes with higher drou ght tolerance (Karcher et al., 2008). Deep

PAGE 89

89 rooting has been also been associated with the capacity of certain crops to deplete nitrogen (N) from N-enriched subsoils (Thorup-Kristensen, 20 01). In this study, crucifer crops removed more nitrogen from the subsoil than grass crops be cause they developed a deeper root system. There are a limited number of reports about root characteristics of fo rage grasses that can be grown in subtropical regi ons. Burton (1943) reported marked differences among warm-season grass species during the first year of root production. Variation among warm-season grasses was also reported for root penetrati on, root yields, and root activity (Burton et al., 1954). Intraspecific variability for root characteristics was reported for Zoysiagras s (Marcum et al., 1995). Genetic variability for maximum root depth, root mass and root number at different soil depths was reported for this species. Bahiagrass is a diverse species that contains races with different ploidy levels and linked reproductive characteristics (Acua et al., 2007 ). Asexual reproduction through seeds (apomixis) is characteristic of the tetraploid germplasm, and it can be manipulated to fix genotypes with superior agronomic characteristics. Several hundr ed bahiagrass hybrids were recently generated by crossing induced sexual and apom ictic tetraploid clones (Acua et al, 2009). Several of these progeny were selected, based on their high a pomixis expression, improved vigor, and freeze resistance when grown as individu al plants. Since they were cla ssified as highly apomictic, it should be possible to establish pure-stand plots of each hybrid by seed propagation. However, it is unknown if selection based on ph enotypic characteristics of i ndividual plants will result in higher forage production when these novel hybrids are grown in swards. If genetic variability exists for forage production among novel tetraploid hybrid s and tetraploid ec otypes, variability for root mass and RLD might also be present among them. Moreover, variation in root mass and RLD among tetraploid clones might explain variation in forage production.

PAGE 90

90 Our objectives were to determine the geneti c variability for forage and root production among apomictic bahiagrass clones grown under different fertilization rates, and to evaluate the relationship between root mass a nd RLD, and forage production. Materials and Methods Plant Material Thirte en tetraploid apomictic bahiagrass cl ones were selected for this study, based on expected variability in forage production. Two of these clones, Argentine (PI 148996) and Common, are natural ecotypes introduced from South America. Argentine is a productive tetraploid clone commonly grown in Florida. The other eleven clones were novel apomictic hybrids generated by crossing sexual and apomictic tetraploid clones. One of the better known clones, Tifton 7, was generated by Dr. Glenn Bu rton (Tifton, Georgia). Selection was based on superior forage yields, based on field plot trials. Three others, C-49, C-65 and C-92, were generated by Camilo Quarin (Corrientes, Argentina), and selected, based on superior individual plant forage yields. The othe r seven clones (FL-3, FL-3B, FL-13, FL-14, FL-21, FL-93 and FL122) were developed by University of Florida forage breeders, and selected, based on superior vigor, cool-season regrowth, and cold tolerance of spaced plants. Since these 13 clones are classified as highly apomictic, variation among replications was primarily restricted to environmental variation. Experimental Design and Plot Management Seeds were scarified using concentrated su lfuric acid for ten m inutes, and sown on 3 March, 2007, in a greenhouse, using plastic flats containing a sterile germination mix. After two weeks, seedlings were transplanted to seedling flats containing multiple cells. Seedlings were transplanted into a field located at the North Florida Research and Education Center, Live Oak, FL, on 9 May. Eight rows were planted, containi ng 13 (1.2 m x 1.2 m) pure-stand plots separated

PAGE 91

91 by a 1 m alley. The clones were randomized with in each row. Pure-stand plots were established by planting 36 plants (20-cm apart) of each cl one in each plot. Plots were irrigated with approximately 25 mm of water immediately after planting. Weeds were removed manually on 15 June and plots were fertilized with 60 or 20 kg N ha-1 (34-0-0) and 20 kg P ha-1 (0-0-60). The fertilizer treatment was randomly assigned to each of four blocks containing 2 rows. The experiment was a 2 (fertilizer ra tes) by 13 (clones) factorial in a completely randomized splitplot design, where fertilizer rate was the whole plot and clones the subplot. A 70 cm wide strip was cut across each plot on 30 July using a sickle bar mower, leaving a 5 cm stubble height. The harvested material was dried at 60 C for 48 h and weighed. After harvest, plots were fertilized with 120 or 40 kg N ha-1 and 40 kg P ha-1. Plots were harvested again on 13 November 2007. Plots were harvested on 15 May for the first tim e in 2008, and every four weeks. Plots were harvested seven times during 2008. At the first an d fourth harvest date for 2008, all plots were fertilized with 60 kg N ha-1 (16-4-8). At the second, third, fift h and sixth harvest dates only plots originally assigned for the high fertilizati on treatment were fertilized with 60 kg N ha-1. The resulting rates of fertiliz ation were 180 or 60 kg N ha-1 year-1 in 2007, and 360 or 120 kg N ha-1 year-1 in 2008. Collection and Analyses of Root Samples Between 19 July and 21 July 2007, single soil co res were taken from each of the 104 plots using a trailer-mounted, hydraulic soil coring m achine (Giddings, model HDGSRPST) (Figure 61). Soil cores were obtained by inserting a 5.1-cm-diam steel tube wi th plastic liner to a depth of 120 cm. On 12 and 13 August 2008, two additional so il cores were taken from plots containing clones FL-3, FL-122, C-65, and Argentine. After sa mple collection, the internal plastic liners were removed from the steel tubes, capped and the samples transported to Gainesville, FL, and

PAGE 92

92 kept in a cold room overnight. Plastic liners we re then marked and the sample divided in 3 sections: 0-40 cm, 40-80 cm, and 80-120 cm. Each section was placed on a 1.5-mm mesh screen and washed with a gentle stream of water to re cover roots. Samples were placed in plastic bags and kept refrigerated (4 C). Roots were spread on clear plas tic trays and images generated and analyzed using WinRHYZO (Regent Instruments Inc., Quebec, Canada) scanner and software (WinRHIZO Pro v2002c). Root length density wa s calculated by dividing sample root length (cm) by sample soil volume (cm-3). After the images were generated, roots were dried at 60 C for 48 h, and weighed. Statistical Analysis Data was an alyzed separately by year because plants were establishing in year one and fertilization practice differed between years. Root data and annual forage accumulation were analyzed as a split-plot design where fertilizer rate was considered th e whole plot and clones were subplots. Seasonal biomass data was anal yzed as a split-plot design with repeated measures. Fertilizer rate, clone, and harvest dates were considered fixed, while replicates where considered random. The data was analyzed usi ng Proc mixed of SAS (SAS version 9.2, SAS Institute, Cary NC). When signifi cant differences were detected ( P = 0.05), the least significant difference (LSD) was computed. Correlations between root variables and fo rage production were calculated using the Pearson productmoment correlation coefficient ( r ). Results Forage Production Significant differences were observed am ong hybrids for forage annual accumulation (FAA) in 2007 (Table 6-1). As expected, there was a significant N effect on forage production. No significant interaction between hybrids and N occurred in 2007, so hybrids were ranked in similar order for FAA (Table 6-1). With the ex ception of FL-21, artificially generated hybrids

PAGE 93

93 produced more forage than the natural ecotype s, Argentine and Common in 2007 and the trend continued in 2008 (Table 6-2). There was a significant interact ion between hybrids and N in 2008. Some hybrids, such as FL-14 and FL-13, were ranked higher under the low fertilization treatment (Table 6-3), while others, such as C92, were ranked higher und er the high fertilization treatment (Table 6-2). When the data were analyzed to consider forage production from each harvest date, fertilizer rate had a significan t effect on forage production in both years. Genotypic differences were observed for forage production in both year s. There was a significa nt interaction between fertilizer rate and hybrid only in 2008. In addition, the effect of harvest date on forage yield indicated the marked seasonal grow habit of bahiagrass. Since ther e was a significant interaction between harvest date and cult ivar in 2007 and 2008, the data we re further analyzed at each harvest. Although significant differences were de tected among hybrids for forage mass at each harvest (Tables 6-1, 6-2 and 6-3), the chief differences were observe d at the beginning and at the end of the growing season (Tables 6-2 and 6-3; Figures 6-2 and 6-3). Most of the artificially generated hybrids included in this study had a more extended growing season than the natural ecotypes (Figures 6-2 and 6-3). At the peak of the growing season hybrids did not out-produce Argentine, however most of them produced more forage than Argentine in May and October (Tables 6-2 and 6-3). Root Mass and Root Length Density No significant differences were observed am ong hybrids for root m ass or RLD from any of the three analyzed soil layers in 2007. Nitrogen fertilization did not have an effect on root mass or RLD from any of the analyzed soil layers in 2007. The averag e root mass (across hybrids and fertilization rates) was 0.50 mg cm-3 at the 0to 40-cm depth, 0.13 mg cm-3 at the 40to 80-cm depth, and 0.08 mg cm-3 at the 80to 120cm depth. The av erage root length density was 1.1 cm

PAGE 94

94 cm-3 at the 0and 40-cm depth, 0.15 cm cm-3 at the 40and 80-cm depth, and 0.08 cm cm-3 at the 80to 120-cm depth. There were no significant diffe rences between hybrids for root mass from any of the analyzed soil layers in 2008. Additionally, the fer tilizer rate did not have a significant effect on root mass from any soil layer. The average root mass (across hybrids and fertilizer rates) was 0.87 mg cm-3 at the 0to 40-cm depth, 0.24 mg cm-3 at the 40to 80-cm depth, and 0.18 mg cm-3 at the 80to 120-cm depth. Significant differe nces for RLD were observed among hybrids in 2008 (Figure 6-4). However, genotypic differences for RLD at the 0to 40-cm and 40to 80-cm depths were only observed for the high fertilizer treatment. In c ontrast, genotypic differences for RLD at the 80to 120-cm depth were only observed for the low fertilizer treatment (Figure 6-4). A significant positive correlation was found between RLD and root mass at the 0to 40-cm depth in 2007 (Figure 6-5) and 2008. Positive correlations were also found between RLD and root mass at the 40to 80-cm depth (r2 = 0.7), and for the 80to 120-cm depth (r2 = 0.6) in 2007. Results were similar for the 2008 growing seas on. Positive correlations were found between RLD and root mass at the 0to 40-cm depth (r2 = 0.4), 40to 80-cm depth (r2 = 0.7), and 80to 120-cm depth (r2 = 0.7). There were no significant corr elations between root mass at the 0to 40cm depth and forage annual accumulation (Figure 6-6) or for the forage harvested at the time of root collection in 2007 or 2008. Although significant, the correlation between root mass at the 40to 80-cm and 80to 120-cm depths and forage annual accumu lation were low (Figures 6-7 and 6-8). No correlations were found between root mass at these two depths and forage annual accumulation nor forage harvested at the time of root collection in 2008.

PAGE 95

95 No significant correlations we re found between RLD at the 0to 40-cm depth and forage annual accumulation (Figure 6-9) nor forage harves ted at the time of root collection for either year. Significant but low correla tions were found between RLD at the 40to 80-cm and 80to 120-cm depths and forage annual accumulation (Figures 6-10 and 6-11) only in 2007. There were not significant correlations between RLD at the 40to 80-cm and 80to 120-cm depths and forage harvested at the time of root collection for either year. Discussion Most of the hybrids included in this study produced m ore forage than the cultivar Argentine, which is the most commonly used bahiag rass tetraploid cultivar in the southeastern USA. Differences in forage production were mainly observed during spring and fall. These findings indicate that phenotypic selection of individual progeny for cool-season regrowth and freeze resistance (Acua et al., 2009) can successfully result in hybrids with extended growing season when grown in swards. A lthough the physiological reasons fo r these differences were not part of this research, previous reports indica te that the two major factors influencing the extension of the bahiagrass growing season were photoperiod (Sincl air et al., 2001) and temperature (Gates et al., 1999). The results al so indicate that the ge netic variability for photoperiod sensitivity and freeze resistance can be successfully fixed through apomixis. However, it is important to mention that apomic tic bahiagrass plants are occasionally able to sexually reproduce resulting in a low degree of segregation for these characteristics. Once established, the apomictic hybrids included in this study responded differently to the application of different amounts of fertilizer. This respons e indicates that some lines, such as FL13, may be more appropriate for systems with mini mal fertilizer inputs, while others, such as C92, may be more appropriate for intensive produc tion systems receiving high fertilizer inputs.

PAGE 96

96 Further research is need ed to investigate the tolerance of these hybrids to grazing stress, and associated nutritive value. The lack of genetic variability for root ma ss in 2007 and 2008, in addi tion to the lack of significant (or low) correlations between root mass and forage mass would indicate that root mass can not explain the genotypic differences ob served for forage production. These results indicate that root activity of perennial grasses might play an important role for plant production. Genotypic differences for forage production may be more greatly affected by differences in above-ground growth habits (i.e., rhizome produc tion, prostate vs. upright). These attributes were not evaluated in this study. In addition, furt her research is needed to analyze the seasonal (year-long) root production, since the major differences for forage production were observed in spring and fall, while, root sampling (mid July ) best represented the mid-season production. Although genotypic variation for RLD was obs erved in 2008, this variation was not significantly correlated w ith the genotypic variation observe d for forage production. This would indicate that RLD was not lim iting those hybrids producing less forage, even under lower soil fertility. Passioura (1983) indicated that crop RLD appears much larger than what is required to extract water at reasonable rates, and that selec ting for less massive root systems (particularly in the topsoil) may be beneficial in water-limited environments. Sinc e water can be expected to be the most limiting factor to many production systems in southeastern USA, selection of a smaller root system in the topsoil should be considered for bahiagrass breeding es pecially for intensive production systems. Increasing the fertilizer rate resulted in greater forage production but it did not result in greater root mass or root length density. This lack of root response to the 2 fertilizer rates is further evidence for the presence of an excessively big root system in bahiagrass. However, comparisons with lower fertility soils need to be evaluated. Additionally, a more

PAGE 97

97 massive root system might be an advantage fo r bahiagrass long-term survival under very low input systems, such as rangela nds in Florida, southern Braz il, Uruguay, northeastern Argentina and other parts of the world.

PAGE 98

98 Table 6-1. Average biomass produced by of 13 apomictic bahiagrass clones in 2007. Biomass Yield Low Fertilization High Fertilization Clone 30 July 13 November Annual 30 July 13 November Annual ----------------------------------------------g m-2------------------------------------------C-92 245 325 570 345 348 693 C-49 225 360 585 319 372 692 C-65 182 273 456 351 336 688 FL-13 262 283 545 338 343 681 FL-93 241 249 491 327 331 658 FL-3B 257 320 576 328 310 638 Tifton 7 202 344 546 261 363 624 FL-122 198 285 482 310 306 616 FL-3 241 290 531 315 299 615 FL-14 236 342 578 287 310 597 FL-21 202 237 439 247 230 477 Argentine 141 190 331 237 207 443 Common 154 128 283 201 113 314 LSD 70 62 113 75 87 119 Clones were ranked based on annual bioma ss accumulation with high fertilization. Least significant difference. Table 6-2. Average biomass produced by 13 bahiag rass clones and a high fertilizer rate in 2008. Biomass Yield Hybrid 15 May 12 June 10 July 7 August 4 September 2 October 31 October Annual ------------------------------------------------------g m-2--------------------------------------------------C-92 82 147 373 396 216 85 30 1328 FL-93 97 162 304 377 235 72 18 1266 C-49 41 94 298 376 255 115 52 1231 FL-14 88 107 305 346 237 106 38 1226 FL-3B 66 121 323 338 219 98 38 1204 C-65 85 125 274 317 244 97 30 1172 FL-3 80 115 276 377 213 69 28 1159 FL-122 37 100 258 395 241 80 34 1145 Tifton 7 52 96 273 369 229 76 33 1127 Argentine 21 101 258 383 254 71 12 1099 FL-13 103 124 264 294 191 67 30 1073 FL-21 70 87 276 279 209 58 16 994 Common 0 0 116 253 194 46 8 618 LSD 28 21 46 80 43 19 10 133 Clones were ranked based on annual biomass accumulation. Least significant difference.

PAGE 99

99 Table 6-3. Average biomass produced by 13 bahiag rass clones and low fertilizer rate in 2008. Biomass Yield Hybrid 15 May 12 June 10 July 7 August 4 September 2 October 31 October Annual --------------------------------------------------g m-2----------------------------------------------------FL-14 73 117 252 227 229 82 14 993 FL-13 96 138 235 216 217 49 9 960 FL-93 68 126 249 204 225 45 3 919 C-92 54 124 263 196 213 58 5 913 C-65 67 118 223 192 205 75 14 893 FL-3 64 112 207 212 197 63 9 863 FL-122 44 108 204 190 219 80 14 859 FL-3B 46 111 218 202 206 65 7 855 C-49 34 94 215 183 221 81 16 843 FL-21 54 108 214 193 183 55 9 816 Tifton 7 69 100 160 209 207 60 6 810 Argentine 16 80 178 201 201 50 6 732 Common 0 0 82 152 157 40 2 432 LSD 31 21 47 28 41 17 6 117 Clones were ranked based on annual biomass accumulation. Least significant difference.

PAGE 100

100 Figure 6-1. Root sample collection on bahiagrass plots. A hydraulic soil sampling equipment can be seen at front. A person removing an intern al plastic liner can be seen in the back. 0 50 100 150 200 250 300 12-Apr1-Jun21-Jul9-Sep29-Oct18-DecHerbage Mass, g m-2 Fl-13 Fl-14 Argentine Common Figure 6-2. Seasonal biomass yields of the cult ivar Argentine, the ecotype Common and hybrids FL-13 and FL-14 during 2008. Plots we re fertilized with 120 kg N ha-1 year-1.

PAGE 101

101 0 50 100 150 200 250 300 350 400 450 12-Apr1-Jun21-Jul9-Sep29-Oct18-DecHerbage Mass, g m-2 Fl-93 C-49 Argentine Common Figure 6-3. Seasonal biomass yields of the cult ivar Argentine, the ecotype Common and hybrids C-49 and C-93 during 2008. Plots we re fertilized with 360 kg N ha-1 year-1.

PAGE 102

102 Fl-3C-65Fl-122Argentine 0 1 2 30to 40-cm 40to 80-cm 80to 120-cm RLD (cm cm-3)a a a a a ab a b b b b b Fl-3C-65Fl-122Argentine 0 1 20to 40-cm 40to 80-cm 80to 120-cm RLD (cm cm-3)a a a a a a a a a ab b bHigh N Low N Figure 6-4. Root length density of 4 bahiagrass hybrids fe rtilized with 360 kg N ha-1 yr-1 (high N) or 360 kg N ha-1 year-1(low N). Bars at specific depths having different letters have significance at alpha = 0.05.

PAGE 103

103 y = 0.2537x + 0.0765 R2 = 0.5042 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 00.511.522.5RLD (cm cm-3)Root mass (g) Figure 6-5. Relationship between ro ot mass and RLD at the 0to 40-cm soil layer of bahiagrass hybrids in 2007. 0 100 200 300 400 500 600 700 800 900 00.10.20.30.40.50.60.70.8Root mass (g)Herbage mass (g m-2) Figure 6-6. Relationship between root mass at th e 0to 40-cm soil layer and forage annual accumulation of bahiagrass hybrids in 2007.

PAGE 104

104 y = 363.39x + 510.17 R2 = 0.0673 0 100 200 300 400 500 600 700 800 900 00.10.20.30.40.50.60.7Root mass (g)Herbage mass (g m-2) Figure 6-7. Relationship between root mass at th e 40to 80-cm soil layer and forage annual accumulation of bahiagrass hybrids in 2007. y = 375.64x + 523.96 R2 = 0.0825 0 100 200 300 400 500 600 700 800 900 00.10.20.30.40.50.60.70.8Root mass (g)Herbage mass (g m-2) Figure 6-8. Relationship between root mass the 80to 120-cm soil layer and forage annual accumulation of bahiagrass hybrids in 2007.

PAGE 105

105 0 100 200 300 400 500 600 700 800 900 00.5 1 1.5 22.5RLD (cm cm-3)Herbage mass (g m-2) Figure 6-9. Relationship between RLD at the 0to 40-cm soil layer and forage annual accumulation of bahiagrass hybrids in 2007. y = 284.17x + 500.86 R2 = 0.0541 0 100 200 300 400 500 600 700 800 900 00.10.20.30.40.50.60.7RLD (cm cm-3)Herbage mass (g m-2) Figure 6-10. Relationship between RLD at the 40to 80-cm soil layer and forage annual accumulation of bahiagrass hybrids in 2007.

PAGE 106

106 y = 318.04x + 519.62 R2 = 0.0571 0 100 200 300 400 500 600 700 800 900 00.10.20.30.40.50.6RLD (cm cm-3)Herbage mass (g m-2) Figure 6-11. Relationship between RLD at the 80to 120-cm soil layer and forage annual accumulation of bahiagrass hybrids in 2007.

PAGE 107

107 CHAPTER 7 CONCLUSIONS Genetics and Breeding Naturally occurring tetraploid bah iagra ss races reproduce ase xually by aposporous apomixis. Crosses between sexual female parent s and apomictic male pollinators allowed the generation of segregating populations for the genetic improvement of this species. Two populations (F1 and F2) were generated in the last 4 years and evaluated for mode of reproduction, growth habit, cool-season re growth, freeze resistance, and production of inflorescences. The F1 was generated by crossing sexual induc ed tetraploid clones and apomictic tetraploid ecotypes, while the F2 was generated by crossing sele cted and sexual and apomictic F1 progeny. The proportion of progeny able to form aposporous embryo sacs remained constant between the F1 and F2 (Chapter 2). Approximately 20% of the progeny from a sexual x apomict cross can be expected to inher it the apospory locus. The propor tion of highly apomictic progeny was reduced from 11% in the F1 to only 3% in the F2. This variation in expr essivity suggests that epigenes are involved in the expr ession of aposporous apomixis in bahiagrass. Since the fixation of superior hybrids is the ultimate goal of manipulating apomixis for breeding purposes, this low proportion of obligate apomicts within the progeny seriously compromise the practicability of this breeding approach. More research is needed to investigate possible methods to increase the proportion of obligate apomicts in the progeny. The possibility of using sexual clones that show a minimum expression of apomixis as female parents should be evaluated as a potential breeding approach. The genetic variability and heritability estimates for grow th habit, cool-season regrowth, and freeze resistance remained rela tively constant between the F1 and F2 indicating that

PAGE 108

108 phenotypic mass selection could be efficiently used for continuous improvement of this species (Chapter 2). Seven apomictic F1 hybrids were selected based on superior spreading, cool-season regrowth and freeze tolerance. These hybrids were cultivated in purestand plots in three locations across north Florida. Five hybrids produced higher cool-season biomass yields compared with commercial cultivars when grown in swards (Chapter 3). The main differences for cool-season biomass producti on were observed at the begi nning of the growing season. Higher cool-season growth resulted in most cas es in higher annual biomass accumulation. These F1 hybrids were also able to remove more n itrogen and phosphorus from the soil during the spring than commercial tetraploid cultivars (Cha pter 3). The genetic variability observed for nitrogen and phosphorus concentration in foliage of these novel hybrids was very low. Rhizomes and roots of these novel hybrids were not compro mised by frequent defoliation during the three years of evaluation. It is necessary to further i nvestigate the physiological and anatomical characteristics that might be responsible for the extended growing season of these novel tetr aploid hybrids. Best management practices need to be determined for the use of these hybrids for hay production or as a bioenergy crop. The potent ial use of these hybrids for grazing needs to be determined in larger trials where they will be evaluated for pers istence under grazing by livestock. Performance of these novel lines will need further evaluation in re gards to animal performance, palatability and preference. Ecological Determinants of Growth Clear acrylic tubes were used to screen several bahiagrass tetraploid clones for rate of root depth developm ent during the first 6 to 8 weeks after germination (C hapter 4). This trait can be evaluated using different growing media and tube si zes with minimal variati on in the results. The rate of root depth development was also shown to be independent of the growing season when

PAGE 109

109 plants are frequently irrigated. Th e technique was determined to be adequate for the screening of a large number of plants in a reduced spaced (3.5-cm diameter tubes). The rate of root depth development in bahiagrass is not significantl y affected by close and frequent defoliation. Bahiagrass is typically known to withstand cl ose defoliation, and the speed of root depth development and recovery following defoliation may be related to this plants success when cultivated as grazed forage or used for hay production. The bahiagrass germplasm contains genetic vari ability for rate of root depth development (Chapter 4). Faster r oot development was observed for clone FL-122. This clone is one of the novel F1 apomictic hybrids generated by crossing i nduced sexual and apomictic tetraploid clones. Since the female parent of this clone was derived from an induced autotetraploid, the genes behind this superiority could be present in the diploid or tetraploid germplasm. This trait always resulted in higher aboveand belowground biomass when grown in tubes. We hypothesized that higher biomass results from rapid exploration of water and nutrients present in deep soil layers. Field observations confirmed the rates of root depth development observed for plants grown in acrylic tubes. Pl ants growing in the field were able to explore a depth of 100 cm after 60 days which reflects a rate of depth in crease approximately the same as observed for plants grown in clear acrylic tubes (1.8 cm d-1). Higher rate of root depth devel opment resulted in faster acce ss and uptake of label nitrogen deposited in deep soil layers (Chapter 5). Th is observation indicated that high biomass yield observed for FL-122 was associated with faster soil depth exploration and water and nutrient uptake. Although FL-122 was not one of the highes t yielding apomictic clones, it could be used in a breeding program as a source of genes for early vigor and quick es tablishment. Clone F-122 was generated by crossing Q-4188 as female pare nt and Argentine as male parent. Since

PAGE 110

110 Argentine showed lower rates of root depth de velopment than FL-122, the source of superior genes for this trait might also be present in Q-4188. Other apomictic male pollinators could be crossed with Q-4188 to test this hypothesis. Genetic variability for root length density is present am ong the bahiagrass tetraploid germplasm (Chapter 6). This variability can be detected in the second growing season once a sward is established. Genotypic variation for r oot length density varies depending on the amount of fertilizer applied. Clones like Argentine are able to produce a prolific root system at the soil surface when high rates of fertilizer are applied. In contrast, othe r clones like FL-3 are able to produce a more extensive root system in the subsoil when low amounts of fertilizer are applied. The potential advantages of the observed genetic va riability need to be further investigated. The prolific root system observed for Argentine is probably related with its superior ability for phosphorus uptake when high amounts of fertiliz er were applied. The relationship between higher root length density in the subsoil under the low fertilization treatment of FL-3 should be investigated further for a potenti al superior drought tolerance a nd deep soil mineral recovery. The observed genetic variability for root length density was no t related to differences in above-ground biomass yields (Chapt er 6). This result indicates th at lower root length density observed for some clones did not limit their prod uctivity. Although the extensive root systems of these clones should have contributed to the success of this specie s in native grasslands, it might be excessively high for a semi-intensive production system. It would be essential to complement these results by studying the seasonal variation for root length density since tetraploid clones have shown a marked variability for seasonal ity of above-ground biomass production. It would be particularly important to dete rmine the root length density of these clones in spring and fall since the main differences for biomass produc tion were detected in these two seasons.

PAGE 111

111 Genetic variability for root mass was not observed among the bahiagrass tetraploid germplasm at any of the evaluated soil depths (Chapter 6). Variation in above ground biomass was not related with variation on below-ground biomass. Selection based on root mass should not be considered appropriate for bahiagrass if the breeding objective is to increase aboveground biomass production. Root length density and root mass were not affected by the amount of fertilizer applied (Chapter 6). Although increasing th e amount of fertilizer increased the amount of foliage that was harvested, it did not have a significant effect on the root system. These results are further evidence that the bahiagrass root sy stem might be in excess of what is needed to uptake nutrients in a semi-intensive production system. Our results would also indicate that root activity instead of root architecture is probably involved in the uptake of nutrients and biomass yields. Further research will be needed to test how root activit y is correlated to biomass production, and if this is a major genotypic trait responsible for differences found in biomass yields. Perspective on Future Research The genetic variability cont ained in the bahiagrass germ plasm should be further investigated to enhance future breeding. Molecular markers can be used to identify relationships among tetraploid populations growing in the wild or contained in a germplasm bank. The identification of heterotic groups using molecula r techniques can be tested by hybridization of individuals from related and unrel ated populations. One of the most important aspects to improve upon is the proportion of highly apomictic progeny that can be generated by hybridization. Since the apomixis locus is inherited as a dominant Mendelian factor, id entification of selfincompatible sexual mother plants will enhance future breeding. The relationship between seasonal herbage pr oduction and root development and activity should also be further investigat ed. These root parameters can be estimated at beginning and end

PAGE 112

112 of the growing season by quantifying the uptake of phosphorus isotopes (32P) placed at different soil depths. Genotypes exhibiting contrasting growi ng patterns may be utilized for this purpose. Apomictic hybrids exhibiting an extended growing season should be further evaluated for tolerance to defoliation by livestock. The util ization of mob grazing on small pure-stand plots can be considered for this purpose. Additionall y, the potential of using these novel hybrids in crop rotation with row crops should be investigate d. Field trials evaluating the seasonal growing patterns of novel hybrids with current cultivars should determine a best fit for crop rotation systems. From the efforts of this project we be lieve there exists great potential for the genetic improvement of bahiagrass through manipulation a nd exploitation of apomixis in this genus.

PAGE 113

113 APPENDIX SEASONAL FORAGE AND RHIZOME+RO OT MASS, AND NITR OGEN AND PHOSPHORUS CONCENTRATIONS A ND ACCUMULATIONS IN HERBAGE Table A-1. Biomass of 12 bahiagrass clone s grown at Gainesville in 2006 and 2007. Biomass Hybrid 10/31/06 05/04/07 06/04/07 07/01/07 07/31/07 08/27/07 09/24/07 10/23/07 12/04/07 --------------------------------------------------------g m-2------------------------------------------------FL-3 375 153 234 341 239 177 214 52 16 FL-13 383 246 282 275 170 126 147 80 22 FL-14 287 171 265 375 244 174 249 30 3 FL-21 173 129 157 312 182 152 202 39 7 C-49 512 66 172 296 229 174 184 42 6 C-65 499 187 281 384 209 157 163 72 20 C-92 370 175 323 376 215 157 166 59 19 FL-93 348 157 229 349 217 177 218 39 9 FL-122 339 114 191 318 232 137 194 48 7 FL-3B 298 127 236 283 214 173 192 36 11 Argentine 239 37 85 280 268 217 263 41 0 Tifton 7 359 214 201 342 248 185 202 71 16 MSD 118 64 72 77 55 40 137 25 11 Minimum significant difference, Walle r-Duncan means separation procedure. Table A-2. Biomass of 12 bahiagrass clones grown at Gainesville in 2008. Biomass Hybrid 13 May 10 June 8 July 5 August 2 September 1 October 28 October ----------------------------------------g m-2---------------------------------------FL-3 176 177 338 361 187 200 87 FL-13 179 174 267 295 159 178 79 FL-14 243 94 343 360 184 217 102 FL-21 143 191 270 310 148 151 63 C-49 171 76 313 350 179 197 92 C-65 214 142 381 343 202 204 88 C-92 207 112 349 392 213 177 71 FL-93 186 180 329 357 187 202 76 FL-122 77 140 328 351 163 161 71 FL-3B 158 140 372 362 174 156 98 Argentine 66 122 321 447 254 195 58 Tifton 7 124 116 362 398 221 288 113 MSD 79 58 62 59 53 88 9 Minimum significant difference, Walle r-Duncan means separation procedure.

PAGE 114

114 Table A-3. Biomass of 13 bahiagrass clones grown at Live Oak in 2008. Biomass Hybrid 15 May 12 June 10 July7 August4 September 2 October 31 October ----------------------------------------------g m-2-----------------------------------------FL-3 80 115 276 377 213 69 28 FL-13 103 124 264 294 191 67 30 FL-14 88 107 305 346 237 106 38 FL-21 70 87 276 279 209 58 16 FL-93 97 162 304 377 235 72 18 FL-122 37 100 258 395 241 80 34 FL-3B 66 121 323 338 219 98 38 C-49 41 94 298 376 255 115 52 C-65 85 125 274 317 244 97 30 C-92 82 147 373 396 216 85 30 Argentine 21 101 258 383 254 71 12 Common 0 0 116 253 194 46 8 Tifton 7 52 96 273 369 229 76 33 MSD 28 21 46 80 43 19 10 Minimum significant difference, Walle r-Duncan means separation procedure. Table A-4. Biomass of 13 bahiagra ss clones grown at Quincy in 2008. Biomass Hybrid 13 May 10 June 8 July 5 August 2 September 1 October 28 October ----------------------------------------g m-2--------------------------------------FL-3 90 63 156 221 259 110 39 FL-13 122 92 131 188 206 103 26 FL-14 108 85 151 179 247 116 39 FL-21 59 70 114 174 255 119 27 C-49 136 76 159 224 268 141 44 C-65 80 56 145 209 280 120 33 C-92 114 92 178 189 264 114 22 FL-93 84 92 157 209 275 120 24 FL-122 55 55 137 204 301 123 32 FL-3B 87 73 173 203 262 129 26 Argentine 39 47 120 179 271 137 23 Common 0 1 58 176 227 101 21 Tifton 7 78 56 124 176 242 134 36 MSD 29 31 28 57 46 42 11 Minimum significant difference, Walle r-Duncan means separation procedure.

PAGE 115

115 Table A-5. Rhizome+root mass of 12 bahiagrass clones (Gainesville). 2007 2008 Clone Spring Summer Fall Spring Fall -----------------------------------------mg cm-3----------------------------------FL-13 72.2 49.0 25.7 23.9 16.4 FL-21 54.9 30.7 36.1 14.1 11.4 FL-122 48.4 40.3 31.4 13.2 10.8 FL-3B 46.1 33.1 25.0 16.6 15.6 C-92 45.8 44.2 32.8 18.7 13.7 FL-93 42.1 39.9 37.7 23.6 13.7 FL-14 40.7 39.4 28.5 18.7 13.6 C-65 34.0 27.9 32.4 17.8 16.2 FL-3 33.1 30.7 26.6 21.5 15.3 Tifton 7 32.5 43.2 41.3 14.8 15.5 C-49 23.4 21.6 30.3 13.9 11.3 Argentine 22.8 25.5 30.5 17.3 15.7 MSD 28.0 ns ns ns ns Clones were ordered based on rhizome+root mass for Spring 2007. Minimum significant difference, Waller-Duncan means separation procedure. Table A-6. Nitrogen concentration in forage of 12 bahiagrass clones grown at Gainesville in 2006 and 2007. Nitrogen Hybrid 10/31/06 05/04/07 06/04/07 07/01/07 07/31/07 08/27/07 09/24/07 10/23/07 12/04/07 ----------------------------------------------------g kg-1---------------------------------------------------FL-3 11.30 15.63 16.41 14.70 13.79 16.51 26.14 17.93 27.83 FL-13 12.24 13.06 15.36 15.04 15.39 18.74 27.01 19.07 28.52 FL-14 11.81 13.4 12.79 14.02 13.88 19.45 27.08 18.1 25.48 FL-21 12.04 13.6 16.35 14.19 15.15 16.64 28.34 17.81 28.41 C-49 9.50 16.76 18.92 16.54 15.49 18.97 25.05 19.48 27.08 C-65 10.26 15.62 16.08 14.70 13.96 17.18 25.23 19.25 31.58 C-92 10.79 16.4 15.01 16.01 14.93 16.94 27.28 19.46 27.21 FL-93 12.05 15.28 14.83 15.44 14.56 16.56 28.44 18.83 28.11 FL-122 11.56 17.18 16.1 16.90 14.30 17.32 26.71 18.35 27.57 FL-3B 10.03 13.79 15.91 14.47 15.09 17.15 26.05 19.32 27.83 Argentine 12.20 17.01 15.64 15.09 13.02 15.01 26.84 17.64 Tifton 7 11.16 14.49 15.88 15.04 13.51 16.54 25.71 18.39 25.74 MSD 1.9 3.3 ns ns ns 3 ns ns ns Minimum significant difference, Walle r-Duncan means separation procedure.

PAGE 116

116 Table A-7. Nitrogen concentration in forage of 12 bahiagrass clones grown at Gainesville in 2008. Nitrogen Hybrid 13 May 10 June 8 July 5 August 2 September 1 October 28 October ---------------------------------------g kg-1--------------------------------------------------FL-3 14.59 15.99 14.27 15.22 16.49 20.02 24.26 FL-13 12.62 18.15 15.49 15.88 16.75 21.62 25.61 FL-14 12.99 16.23 17.67 14.68 15.11 18.48 24.48 FL-21 15.05 16.97 17.96 16.37 17.68 21.92 24.72 C-49 12.44 17.28 15.74 14.45 14.41 19.78 24.91 C-65 13.46 16.87 16.20 14.25 15.18 19.83 24.05 C-92 13.99 16.50 16.32 14.37 15.63 21.53 25.57 FL-93 15.15 14.77 15.55 15.25 17.58 20.37 26.29 FL-122 14.48 16.95 16.54 15.30 17.19 20.76 26.82 FL-3B 12.28 17.32 17.64 14.75 15.02 19.30 24.22 Argentine 16.53 15.29 15.93 14.90 17.22 20.47 27.00 Tifton 7 14.88 17.52 15.23 14.28 15.21 19.57 24.30 MSD 1.6 ns 2.3 ns 0.9 1.7 2.1 Minimum significant difference, Walle r-Duncan means separation procedure. Table A-8. Phosphorus concentratio n in forage of 12 bahiagrass clones grown at Gainesville in 2007 Phosphorus Hybrid 05/04/07 06/04/07 07/01/0707/31/0708/27/0709/24/0710/23/07 12/04/07 --------------------------------------------g kg-1-------------------------------------------FL-3 1.30 1.38 1.72 1.66 1.94 2.55 2.14 2.15 FL-13 1.04 1.55 1.85 1.86 1.98 2.54 2.10 1.97 FL-14 1.25 1.21 1.60 2.01 2.01 2.49 2.05 2.36 FL-21 1.41 1.42 1.55 1.97 2.02 2.59 2.00 1.92 C-49 1.70 1.64 1.69 1.77 1.98 2.36 2.00 1.98 C-65 1.24 1.41 1.66 1.81 1.90 2.26 2.05 1.71 C-92 1.35 1.43 1.80 1.82 1.87 2.32 1.97 1.69 FL-93 1.34 1.54 1.79 1.85 1.83 2.36 2.09 1.88 FL-122 1.16 1.24 2.08 2.01 2.20 2.45 2.18 1.81 FL-3B 1.13 1.67 1.72 1.84 1.90 2.46 2.02 2.16 Argentine 1.54 1.61 2.11 2.10 2.40 2.73 2.57 Tifton 7 1.34 1.68 1.90 2.01 2.21 2.58 2.28 2.51 MSD 0.20 0.20 0.30 ns 0.30 0.20 0.20 0.50 Minimum significant difference, Walle r-Duncan means separation procedure.

PAGE 117

117 Table A-9. Phosphorus concentratio n in forage of 12 bahiagrass clones grown at Gainesville in 2008 Phosphorus Hybrid 13 May 10 June 8 July 5 August 2 September 1 October 28 October ---------------------------------------g kg-1---------------------------------------FL-3 1.12 1.28 1.98 1.90 2.34 2.33 2.53 FL-13 0.97 1.51 2.06 2.13 2.23 2.22 2.38 FL-14 1.07 1.53 2.02 1.98 2.07 2.32 2.75 FL-21 0.98 1.43 2.02 2.00 2.20 2.21 2.37 C-49 1.06 1.66 2.21 1.80 2.01 2.03 2.26 C-65 1.03 1.48 2.16 1.95 2.15 2.02 2.05 C-92 0.96 1.41 2.28 1.90 1.98 2.05 2.36 FL-93 1.11 1.29 2.06 2.05 2.17 2.19 2.34 FL-122 1.05 1.48 2.33 2.23 2.36 2.26 2.34 FL-3B 0.97 1.36 1.98 1.81 1.95 1.94 2.39 Argentine 1.25 1.34 2.43 2.52 2.73 2.86 2.77 Tifton 7 1.14 1.58 2.21 2.22 2.47 2.46 2.77 MSD 0.60 0.80 0.60 0.20 0.20 0.30 0.30 Minimum significant difference, Walle r-Duncan means separation procedure. Table A-10. Nitrogen accumulation in forage of 12 bahiagrass clones grown in Gainesville in 2006 and 2007. Nitrogen Hybrid 10/31/06 05/04/07 06/04/07 07/01/07 07/31/07 08/27/07 09/24/07 10/23/07 12/04/07 ------------------------------------------------------g m-2--------------------------------------------------FL-3 4.2 2.4 4.0 5.0 3.3 2.9 5.7 1.0 0.5 FL-13 4.6 3.3 4.3 4.2 2.6 2.3 4.0 1.5 0.6 FL-14 3.4 2.3 3.4 5.3 3.4 3.4 6.7 0.5 0.1 FL-21 2.0 1.8 2.6 4.4 2.8 2.5 5.7 0.7 0.2 C-49 4.9 1.1 3.2 4.8 3.5 3.3 4.7 0.8 0.2 C-65 5.1 3.0 4.6 5.6 3.0 2.7 4.2 1.4 0.6 C-92 3.9 3.0 4.9 6.0 3.1 2.7 4.5 1.1 0.5 FL-93 4.1 2.4 3.4 5.4 3.2 2.9 6.2 0.7 0.3 FL-122 3.9 2.0 3.1 5.4 3.3 2.4 5.2 0.9 0.2 FL-3B 3.0 1.8 3.7 4.1 3.2 2.9 5.0 0.7 0.3 Argentine 2.8 0.6 1.3 4.2 3.5 3.3 7.0 0.7 0.0 Tifton 7 4.0 3.2 3.3 5.2 3.4 3.0 5.2 1.3 0.4 MSD 1.2 2.3 1.9 1.6 ns 0.8 ns 0.5 0.4 Minimum significant difference, Walle r-Duncan means separation procedure.

PAGE 118

118 Table A-11. Nitrogen accumulation in forage of 12 bahiagrass clones grown in Gainesville in 2008. Nitrogen Hybrid 13 May 10 June 8 July 5 August 2 September 1 October 28 October ----------------------------------------------g m-2-----------------------------------------FL-3 2.6 2.8 4.9 5.5 3.1 4.0 2.1 FL-13 2.2 3.2 4.1 4.7 2.7 3.8 2.0 FL-14 3.2 1.5 6.1 5.3 2.8 4.0 2.5 FL-21 2.1 3.3 4.9 5.1 2.6 3.3 1.6 C-49 2.1 1.3 4.9 5.1 2.6 3.9 2.3 C-65 2.9 2.3 6.2 4.9 3.1 4.0 2.1 C-92 2.9 1.8 5.7 5.6 3.3 3.8 1.8 FL-93 2.8 2.6 5.1 5.4 3.3 4.1 2.0 FL-122 1.1 2.4 5.3 5.3 2.8 3.3 1.9 FL-3B 1.9 2.5 6.6 5.3 2.6 3.0 2.4 Argentine 1.1 1.8 5.1 6.7 4.4 4.0 1.6 Tifton 7 1.9 2.1 5.5 5.7 3.4 5.6 2.7 MSD 1.2 1.4 1.3 ns 1.0 ns 0.3 Minimum significant difference, Walle r-Duncan means separation procedure. Table A-12. Phosphorus accumulation in forage of 12 bahiagrass clones grown in Gainesville in 2007. Phosphorus Hybrid 4 May 4 June 1 July 31 July 27 Au gust 24 September 23 October 4 December -------------------------------------------------g m-2------------------------------------------------------FL-3 0.20 0.33 0.59 0.39 0.34 0.55 0.11 0.04 FL-13 0.25 0.44 0.51 0.32 0.25 0.37 0.17 0.04 FL-14 0.21 0.32 0.60 0.49 0.35 0.62 0.06 0.01 FL-21 0.18 0.22 0.48 0.36 0.31 0.52 0.08 0.01 C-49 0.11 0.27 0.48 0.40 0.34 0.44 0.08 0.01 C-65 0.23 0.40 0.64 0.37 0.30 0.37 0.15 0.03 C-92 0.24 0.46 0.68 0.39 0.29 0.38 0.11 0.03 FL-93 0.21 0.36 0.62 0.41 0.32 0.51 0.08 0.02 FL-122 0.13 0.23 0.66 0.47 0.30 0.48 0.10 0.01 FL-3B 0.14 0.39 0.49 0.39 0.33 0.47 0.07 0.02 Argentine 0.06 0.14 0.59 0.56 0.52 0.73 0.10 0.00 Tifton 7 0.29 0.33 0.66 0.50 0.41 0.52 0.16 0.04 MSD 0.10 0.10 0.20 0.10 0.10 0.30 0.10 0.03 Minimum significant difference, Walle r-Duncan means separation procedure.

PAGE 119

119 Table A-13. Phosphorus accumulation in forage of 12 bahiagrass clones grown in Gainesville in 2008. Phosphorus Hybrid 13 May 10 June 8 July 5 August 2 September 1 October 28 October ----------------------------------------------g m-2------------------------------------------FL-3 0.19 0.22 0.67 0.69 0.44 0.47 0.22 FL-13 0.15 0.26 0.54 0.62 0.35 0.39 0.19 FL-14 0.27 0.14 0.69 0.71 0.38 0.50 0.28 FL-21 0.13 0.27 0.54 0.62 0.33 0.34 0.15 C-49 0.18 0.13 0.69 0.63 0.36 0.40 0.21 C-65 0.21 0.20 0.83 0.66 0.44 0.41 0.18 C-92 0.20 0.15 0.78 0.74 0.42 0.37 0.17 FL-93 0.21 0.22 0.68 0.73 0.41 0.44 0.18 FL-122 0.09 0.21 0.76 0.78 0.38 0.36 0.16 FL-3B 0.14 0.19 0.74 0.65 0.34 0.30 0.23 Argentine 0.09 0.16 0.78 1.13 0.69 0.55 0.16 Tifton 7 0.14 0.18 0.80 0.88 0.56 0.71 0.31 MSD 0.09 0.15 0.21 0.12 0.10 0.18 0.04 Minimum significant difference, Walle r-Duncan means separation procedure.

PAGE 120

120 LIST OF REFERENCES Acua, C.A., A.R. Blount, K.H. Quesenbe rry, W .W. Hanna, and K.E. Kenworthy. 2007. Reproductive characterization of bahiag rass germplasm. Crop Sci. 47:1711-1717. Acua, C.A., A.R. Blount, K.H. Quesenbe rry, K.E. Kenworthy, and W.W. Hanna. 2009. Bahiagrass tetraploid germplasm: reproduc tive and agronomic characterization of segregating progeny. Crop Sci. 49: 581-588. Bashaw, E.C., and W.W. Hanna. 1 990. Apomictic reproduction. p. 100-130. In G.P. Chapman (ed.) Reproduction versatility in grasses. Cambridge Univ. Press, Cambridge, UK. Beaty, E.R., R.A. McCreery, and J.D. Powell. 1960. Response of Pensacola bahiagrass to nitrogen fertilization. Agron. J. 52:453-455. Blount, A.R., and C.A. Acua. 2009. Bahiagrass In : R.J. Singh (ed.) Genetic Resources, Chromosome Engineering, and Crop Improvement Series: Forage Crops Volume 5. CRC press Boca Raton, FL. Blue, W.G. 1973. Role of Pensacola bahiagrass stolon-root systems in fertilizer nitrogen utilization on Leon fine sand. Agron. J. 65:88-91. Bonos, S.A., D. Rush, K. Hignight, and W.A. Me yer. 2004. Selection for deep root production in tall fescue and perennial ryegrass. Crop Sci. 44:1770. Brady, N.C., and R.R. Weil. 2002. The nature an d properties of soils. Prentice Hall, Upper Saddle River, NJ. Burton, G.W. 1943. A comparison of the first years root production of se ven southern grasses established from seed. Jour. Amer. Soc. Agron. 35:192-196. Burton, G.W. 1948. The method of re production in common bahiagrass, Paspalum notatum J. Am. Soc. Agron. 40:443-452. Burton, G.W. 1955. Breeding Pensacola bahiagrass, Paspalum notatum : I. Method of reproduction. Agron. J. 47:311-314. Burton, G.W. 1967. A search for the origin of Pensacola bahiagrass. Econ. Bot. 21:319-382. Burton, G.W. 1974. Recurrent restricted phenotypic selection increases forage yields of Pensacola bahiagrass. Crop Sci. 14:831-835. Burton, G.W. 1982. Effect of environment on apomixis in bahiagrass. Crop Sci. 22:109-111. Burton, G.W. 1992. Manipulating apomixis in Paspalum Proceedings of the Apomixis Workshop, February 11-12, A tlanta, Georgia. p. 16-19.

PAGE 121

121 Burton, G.W., E.H. DeVane, and R.L. Carter. 1954. Root penetration, distri bution and activity in southern grasses measured by yields, dr ought symptoms and P32 uptake. Agron. J. 46:229. Cidade, F.W., M. DallAgnol, F. Bered, and T. Te ixeira de Souza-Chies. 2008. Genetic diversity of the complex Paspalum notatum Flgge (P anicea: Panicoidea). Genet. Resour. Crop Evol. 55:235 Dahmer, N., M.T. Schifino-Wittmann, M. DallAgnol, and B. de Castro. 2008. Cytogenetic data for Paspalum notatum Flgge accessions. Sci. Agric. 65:381-388. Dardanelli, J.L., O.A. Bachmeier, R. Sereno, and R. Gil. 1997. Rooting depth and soil water extraction patterns of different crops in a si lty loam Haplustoll. Fi eld Crops Res. 54:29. Daurelio, L.D., F. Espinoza, C.L. Quarin, and S.C. Pessino. 2004. Genetic diversity in sexual diploid and apom ictic tetraploid populations of Paspalum notatum situated in sympatry and allopatry. Plant Syst. Evol. 244:189-199. Forbes, I., Jr., and G.W. Burton. 1961. Cytology of di ploids, natural and induced tetraploids, and intraspecific hybrids of bahiagrass, Pa spalum notatum Flgge. Crop Sci. 1:402-406. Gallaher, R.N., C.O. Weldon and J.G. Futral. 197 5. An aluminum block digester for plant and soil analysis. Soil Sci. Soc. Amer. Proc. 39:803-806. Gates, R.N., G.M. Hill, and G.W. Burton. 1999. Re sponse of selected and unselected bahiagrass populations to defoliatio n. Agron. J. 91:787-795. Gates, R.N., P. Mislevy, and F.G. Martin. 2001. Herbage accumulation of three bahiagrass populations during the cool season. Agron. J. 93:112. Gates, R.N., C.L. Quarin, and C.G.S. Pedrei ra. 2004. Bahiagrass. p. 651-680. In: L.E. Moser, B.L. Burson, and L.E. Sollenberger (eds.) Warm-Season (C4) Grasses, ASA, CSSA, SSSA, Madison, WI, USA. Gould, F.W. 1966. Chromosome numbers of some Mexican grasses. Can. J. Bot. 44:1683-1696. Hanna, W.W., and E.C. Bashaw. 1987. Apomixis: it s identification and use in plant breeding. Crop Sci. 27:1136-1139. Hambleton, L.G. 1977. Semiautomated method for simultaneous determination of phosphorus, calcium and crude protein in anim al feeds. J.A.O.A.C. 60:845-852. Huang, Y., D.H. Rickerl, and K.D. Kephart. 1996. Recovery of deep-point injected soil nitrogen by switchgrass, alfa lfa, ineffective alfalfa, and corn. J. Environ. Qual. 25:1394. Hubbard, R.K., J.M. Sheridan, R. Lowrace, D.D. Bosh, and G. Vellidis. 2004. Fate of nitrogen from agriculture in the southeastern Coasta l Plain. J. Soil Wat. Conserv. 59(2):72-86.

PAGE 122

122 Karcher, E.D., M.D. Richardson, K. Hignight and D. Rush. 2008. Drought tolerance of tall fescue populations selected for high root/shoot ratios and summer survival. Crop Sci. 48:771-777. Kramer, P.J., and J.S. Boyer. 1995. Water relations of plants and soils. Academic Press, Inc., California, USA. Kristensen, H.L., and K. Thor up-Kristensen. 2004. Root growth and nitrate uptake of three different catch crops in deep soil layers. Soil Sci. Soc. Am. J. 68, 529. Laspina N.V., T. Vega, G. Seijo, A.M. Gonzlez, L.G. Martelotto, J. Stein, M. Podio, J.P.A. Ortiz, V.C. Echenique, C.L. Quarin, S.C. Pe ssino. 2008. Gene expression analysis at the onset of aposporous apomixis in Paspalum notatum Plant Mol. Biol.67: 615-628. Lehman, V.G., and M.C. Engelke. 1991. Heritab ility estimates of cr eeping bentgrass root systems grown in flexible tubes. Crop Sci. 31:1680-1684. Loomis, R.S., and D.J. Connor. 1992. Crop ecology. Cambridge University press, New York. 349 p. Ludlow, M.M., and R.C. Muchow. 1990. A critical evaluation of traits for improving crop yields in water-limited environm ents. Adv. Agron. 43: 107-153. Marcum, K.B., M.C. Engelke, S.J. Morton, an d R.H. White. 1995. Rooting characteristics and associated drought resistance of zoysiagrasses. Agron. J. 87:534. Martnez, E.J, M.H. Urbani, C.L. Quarin, and J.P.A. Ortiz. 2001. Inheritance of apospory in bahiagrass, Paspalum notatum Hereditas 135:19-25. Martnez, E.J., H.E. Hopp, J. Stein, J.P.A. Or tiz, and C.L. Quarin. 2003. Genetic characterization of apospory in tetraploid Paspalum notatum based on the identification of linked molecular markers. Mol. Breed. 12:319-327. Passioura, J.B. 1983. Roots a nd drought resistance. Agri c. Water Manage. 7: 265-280. Peterson, G.A., F.N. Anderson, G.E. Varv el, and R.A. Olson. 1979. Uptake of 15Nlabeled nitrate by sugarbeets from depths gr eater than 180 cm. Agron. J. 71:371. Poehlman, J.M., and D.A. Sleper. 1995. Breeding field crops. Iowa State University Press, Ames. Pozzobon, M.T. and J.F.M.Valls. 1997. Chromosome number in germplasm accessions of Paspalum notatum (Gramineae). Brazilian J. Gen. 20:29-34. Quarin, C.L. 1985. Seasonal changes in the incide nce of apomixis of diploid, triploid, and tetraploid plants of Paspalum cromyorrhizon. Euphytica 35: 515-522.

PAGE 123

123 Quarin, C.L., G.A. Norrmann, and M.H. Ur bani. 1989. Polyploidization in aposporous Paspalum Apomixis Newsl. 1:28-29. Quarin, C.L. 1992. The nature of apomixis and its origin in Panicoid grasses. Apomixis Newsletter 5:8-15. Quarin, C.L., F. Espinoza, E.J. Martnez, S. C. Pessino, and O.A. Bovo. 2001. A rise of ploidy level induces the expression of apomixis in Paspalum notatum Sex. Plant Reprod. 13:243249. Quesenberry, K.H. and R. Smith. 2003. Production of sexual tetraploid bahiagrass using in vitro chromosome doubling agents. Molecular breeding of forage and turf, Third International Symposium, May 18-22, Dallas Texas, USA. Scheurwater, I., M. Corens, H. Lambers, a nd O.K. Atkin. 2002. The contribution of roots and shoots to whole plant nitrate reduction in fast and slow-growing grass species. J. Exp. Bot. 53:1635-1642. Shen, L., B. Courtois, K. McNally, S. Robin & Z. Li. 2001. Evaluation of near-isogenic lines of rice introgresed with QTLs for root depth th rough marker-aided se lection. Theor. Appl. Genet. 103: 75-83. Sinclair, T.R. 1999. Options for sustaining and increasing the limiting yield-plateau of grain crops. Jpn. J. Crop Sci. 67 (extra issue 2), 65. Sinclair, T.R., P. Mislevy, and J.D. Ray. 2001. S hort photoperiod inhibits winter growth of subtropical grasses. Planta 213:488. Sinclair, T. R., and V. Vadez. 2002. Physiologica l traits for crop yield improvement in low N and P environment. Plant Soil 245:1-15. Stein, J., C.L. Quarin, E.J. Martnez, S.C. Pessino, and J.P.A. Ortiz, 2004. Tetraploid races of Paspalum notatum show polysomic inheritance and preferential chromosome pairing around the apospory-controlling locus. Theor. Appl. Genet. 109:186-191. Stein, J., S.C. Pessino, E.J. Martnez, M.P. Rodrigu ez, L.E. Siena, C.L. Quarin, and J.P.A. Ortiz. 2007. A genetic map of tetraploid Paspalum notatum Flgge (bahiagrass) based on singledose molecular markers. Mol. Breed. 20:153-166. Tischler C.R. and B.L. Burson. 1995. Evaluati ng different bahiagrass cytotypes for heat tolerance and leaf epicuticular wax content. Euphytica 84:229-235. Thorup-Kristensen, K. 2001. Are differences in root growth of n itrogen catch crops important for their ability to reduce soil nitrate-N conten t, and how can this be measured? Plant Soil 230:185-195. Young, B.A., R.T. Sherwood, and E.C. Bashaw. 1979. Cleared pistil and thick sectioning techniques for detecting aposporous apomix is in grasses. Can. J. Bot. 57:1668-1672.

PAGE 124

124 BIOGRAPHICAL SKETCH Carlos Alberto Acuna was born in C orpus Christy, northeastern Argentina in 1977. In 1995, he moved to Corrientes, Argentina, and bega n his undergraduate studies at the University of the Northeast. In 2001, he received his Agrono my Engineer degree, after defending his thesis entitle: The relevance of a triploid in the evolution of Paspalum He worked for 4 years, including the last 2 years of his undergraduate studies, with reproductive systems of warmseason grasses at Botany Institute of the Nort heast (IBONE), Corrientes, Argentina. In May 2004 he began his graduate studies at the University of Florida under the dire ction of Dr. Ann Blount and Dr. Kenneth Quesenberry. In May 2006, he wa s awarded a Master of Science degree in agronomy with an emphasis in genetics and pl ant breeding. Immediatel y after graduation, he began his Ph.D. studies in agr onomy working with genetics and plant breeding also under the direction of Dr. Ann Blount. He received his Ph.D. from the University of Florida in May 2009. He plans to return to Argentina, working in the area of plant breed ing, and teaching at the Universidad Nacional del Nordeste.