<%BANNER%>

Biology, Ecology and Management of Natalgrass (Melinis repens)

Permanent Link: http://ufdc.ufl.edu/UFE0042608/00001

Material Information

Title: Biology, Ecology and Management of Natalgrass (Melinis repens)
Physical Description: 1 online resource (108 p.)
Language: english
Creator: Stokes, Courtney
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: biology, ecology, invasive, management, melinis, natalgrass
Agronomy -- Dissertations, Academic -- UF
Genre: Agronomy thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: BIOLOGY, ECOLOGY AND MANAGEMENT OF NATALGRASS (Melinis repens) By Courtney Ann Stokes December 2010 Chair: Gregory E. MacDonald Major: Agronomy Natalgrass (Melinis repens (Willd.) Zizka) is a species native to Africa. Introduced to the U.S. in the 1800s, natalgrass was grown as a hay crop in central Florida in the early 1900s. Although no longer cultivated, natalgrass is still common throughout much of the state. As the restoration of native plant communities becomes increasingly important, land managers are struggling to control this species. To develop an effective management plan, more information is needed about seed biology and ecology as well as the chemical management of natalgrass. Natalgrass seeds were exposed to various light, temperature, pH and osmotic potential treatments to better characterize the conditions in which these seeds germinate. Natalgrass does not require light for germination. Maximum germination occurs at 20 C and greater and at pH levels ranging from 6 to 8. Germination did not occur at water potentials less than -0.2 MPa. Natalgrass seeds were buried at different depths to determine maximum depth of emergence. Natalgrass seeds emerged from a depth of 5 cm, the greatest depth tested in the study. Preliminary tests show that natalgrass likely requires an afterripening period after seed shed to reach maximum potential for germination. Natalgrass seed longevity was studied under field conditions. Seed burial tubes were constructed, buried and exhumed after periods ranging from 0 to 15 mo. An initial decline in germination was observed after 3 mo, with no further decline. These results indicate the onset of dormancy in natalgrass seeds after burial. This finding will be useful to land managers who plan to utilize tillage for natalgrass control, a practice that buries seeds. Seed longevity was also studied on the ground surface, where dense layers of seeds form in infested areas. Exclusion frames were placed over seed deposits to prevent further seed rain and germination under the frames was monitored for 12 mo. After 1 mo, high levels of germination occurred, but levels declined to 0 seedlings/m2 within several months. This finding indicates that surface seed deposits are quickly depleted if land managers can prevent further seed production. Natalgrass seeds were exposed to wind at varying speeds and the distance traveled was measured. Seeds traveled less than 2 m at the highest wind speed, 32 km/h. A number of herbicides were tested in the greenhouse and in the field to determine potential for natalgrass control pre- and postemergence. Metsulfuron and fluazifop offered little or no control. Glyphosate provided excellent control, but the lack of residual activity resulted in immediate reinfestation. Pendimethalin and metolachlor offered good control preemergence, but were detrimental to native species. Hexazinone and sulfometuron provided good control pre- and postemergence, but were also detrimental to native species. Imazamox, imazapyr and imazapic offered less control but stunted growth and delayed flowering. These herbicides were also less harmful to native species. Many native plants are tolerant to imazapic and gained a competitive advantage when this herbicide was applied.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Courtney Stokes.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: MacDonald, Greg.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042608:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042608/00001

Material Information

Title: Biology, Ecology and Management of Natalgrass (Melinis repens)
Physical Description: 1 online resource (108 p.)
Language: english
Creator: Stokes, Courtney
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: biology, ecology, invasive, management, melinis, natalgrass
Agronomy -- Dissertations, Academic -- UF
Genre: Agronomy thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: BIOLOGY, ECOLOGY AND MANAGEMENT OF NATALGRASS (Melinis repens) By Courtney Ann Stokes December 2010 Chair: Gregory E. MacDonald Major: Agronomy Natalgrass (Melinis repens (Willd.) Zizka) is a species native to Africa. Introduced to the U.S. in the 1800s, natalgrass was grown as a hay crop in central Florida in the early 1900s. Although no longer cultivated, natalgrass is still common throughout much of the state. As the restoration of native plant communities becomes increasingly important, land managers are struggling to control this species. To develop an effective management plan, more information is needed about seed biology and ecology as well as the chemical management of natalgrass. Natalgrass seeds were exposed to various light, temperature, pH and osmotic potential treatments to better characterize the conditions in which these seeds germinate. Natalgrass does not require light for germination. Maximum germination occurs at 20 C and greater and at pH levels ranging from 6 to 8. Germination did not occur at water potentials less than -0.2 MPa. Natalgrass seeds were buried at different depths to determine maximum depth of emergence. Natalgrass seeds emerged from a depth of 5 cm, the greatest depth tested in the study. Preliminary tests show that natalgrass likely requires an afterripening period after seed shed to reach maximum potential for germination. Natalgrass seed longevity was studied under field conditions. Seed burial tubes were constructed, buried and exhumed after periods ranging from 0 to 15 mo. An initial decline in germination was observed after 3 mo, with no further decline. These results indicate the onset of dormancy in natalgrass seeds after burial. This finding will be useful to land managers who plan to utilize tillage for natalgrass control, a practice that buries seeds. Seed longevity was also studied on the ground surface, where dense layers of seeds form in infested areas. Exclusion frames were placed over seed deposits to prevent further seed rain and germination under the frames was monitored for 12 mo. After 1 mo, high levels of germination occurred, but levels declined to 0 seedlings/m2 within several months. This finding indicates that surface seed deposits are quickly depleted if land managers can prevent further seed production. Natalgrass seeds were exposed to wind at varying speeds and the distance traveled was measured. Seeds traveled less than 2 m at the highest wind speed, 32 km/h. A number of herbicides were tested in the greenhouse and in the field to determine potential for natalgrass control pre- and postemergence. Metsulfuron and fluazifop offered little or no control. Glyphosate provided excellent control, but the lack of residual activity resulted in immediate reinfestation. Pendimethalin and metolachlor offered good control preemergence, but were detrimental to native species. Hexazinone and sulfometuron provided good control pre- and postemergence, but were also detrimental to native species. Imazamox, imazapyr and imazapic offered less control but stunted growth and delayed flowering. These herbicides were also less harmful to native species. Many native plants are tolerant to imazapic and gained a competitive advantage when this herbicide was applied.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Courtney Stokes.
Thesis: Thesis (M.S.)--University of Florida, 2010.
Local: Adviser: MacDonald, Greg.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042608:00001


This item has the following downloads:


Full Text

PAGE 1

1 BIOLOGY, ECOLOGY AND MANAGEMENT OF NATALGRASS ( Melinis repens) By COURTNEY A NN STOKES A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MA STER OF SCIENCE UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Courtney A nn Stokes

PAGE 3

3 To my family and friends

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my supervisory committee chair, Dr. Greg MacDonald, for allowing me this opportunity to further my educat ion. Dr. MacDonald provided invaluable guidance as I worked toward the completion of my degree program. In addition to guiding my research, he offered me the chance to become involved in teaching, extension and international development work. I would also like to thank my supervisory committee members, Dr. Ken Langeland, Dr. Carrie Reinhardt Adams and Dr. Debbie Miller for their advice. Dr. Jason Ferrell and Dr. William Haller also provided me with a great deal of guidance throughout the course of my study. In addition, I thank Bob Querns for his help in the laboratory and the students in Weed Science for their assistance Without the help of a number of my fellow students, it would have been a great deal more difficult to conduct my research. I than k Danon Moxley, Tim King and the others at the Tenoroc Fish Management Area for providing me with a research site. Likewise, I thank Mike Green and the staff at Mitigation Resources, Inc. for providing me with a site at the Lake Louisa Mitigation Bank. T he Florida Fish and Wildlife Conservation Commission and the UF/IFAS Center for Aquatic and Invasive Plants provided funding, which I greatly appreciate. Finally, I thank my family and friends for their advice, pep talks and willingness to listen when I ne eded it most. Your encouragement helped me a great deal. I am grateful to each and every one of you for your love and support.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Biology ................................ ................................ ................................ .................... 13 Management ................................ ................................ ................................ ........... 15 Research Objectives ................................ ................................ ............................... 15 2 SEED GERMINATION CHARACTERISTICS OF NATALGRASS ( Melinis repens) ................................ ................................ ................................ .................... 16 Introduction ................................ ................................ ................................ ............. 16 Materials and Methods ................................ ................................ ............................ 18 Seed Source ................................ ................................ ................................ ..... 18 General Germination Test Protocol ................................ ................................ .. 18 P reliminary Germination Test ................................ ................................ ........... 19 Light ................................ ................................ ................................ ................. 20 Temperature ................................ ................................ ................................ ..... 20 pH ................................ ................................ ................................ ..................... 20 Water Stress ................................ ................................ ................................ ..... 21 Depth of Burial ................................ ................................ ................................ .. 21 Statistical Analysis ................................ ................................ ............................ 21 Results and Discussion ................................ ................................ ........................... 22 Preliminary Germination Test ................................ ................................ ........... 22 Light ................................ ................................ ................................ ................. 23 Temperature ................................ ................................ ................................ ..... 23 pH ................................ ................................ ................................ ..................... 23 Water Stress ................................ ................................ ................................ ..... 24 Depth of Burial ................................ ................................ ................................ .. 25 Conclusions ................................ ................................ ................................ ............ 25 3 SEED ECOLOGY OF NATALGRASS ( Melinis repens ) ................................ .......... 28 Introduction ................................ ................................ ................................ ............. 28 Materials and Methods ................................ ................................ ............................ 29 Seed Burial ................................ ................................ ................................ ....... 29

PAGE 6

6 Seed Exclusion ................................ ................................ ................................ 31 Wind ................................ ................................ ................................ ................. 32 Statistical Analysis ................................ ................................ ............................ 32 Results and Discussion ................................ ................................ ........................... 33 Seed Burial ................................ ................................ ................................ ....... 33 Seed Exclusion ................................ ................................ ................................ 34 Wind ................................ ................................ ................................ ................. 35 Conclusions ................................ ................................ ................................ ............ 35 4 CHEMICAL CONTROL OF NATALGRASS ( Melinis repens ) ................................ 42 Introduction ................................ ................................ ................................ ............. 42 Materials and Methods ................................ ................................ ............................ 43 Greenhouse Preemergence Study ................................ ................................ ... 43 Greenhouse Postemergence Study ................................ ................................ 45 Field Preemergence Study ................................ ................................ ............... 46 Field Postemergence Study ................................ ................................ ............. 47 Results and Discussion ................................ ................................ ........................... 48 Greenhouse Preemergence Study ................................ ................................ ... 48 Greenhouse Postemergence Study ................................ ................................ 51 Field Preemergence Study ................................ ................................ ............... 55 Field Postemergence Study ................................ ................................ ............. 61 Conclusions ................................ ................................ ................................ ............ 64 5 CONCLUSION S ................................ ................................ ................................ ..... 97 APPENDIX: AFTERRIPENING STUDY ................................ ................................ ...... 100 Introduction ................................ ................................ ................................ ........... 100 Materials and Methods ................................ ................................ .......................... 100 Results and Discussion ................................ ................................ ......................... 102 LIST OF REFERENCES ................................ ................................ ............................. 104 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 108

PAGE 7

7 LIST OF TABLES Table page 4 1 I 50 and I 90 values for various herbicides applied preemergence to natalgrass in the greenhouse. ................................ ................................ .............................. 67 4 2 I 50 and I 90 values for various herbicides applied postemergence to natalgrass in the greenhouse (Trial 1). ................................ ................................ ................. 67 4 3 I 50 and I 90 values for various herbicides applied postemergence to natalgrass in the greenhouse (Trial 2). ................................ ................................ ................. 6 8 4 4 Influence of herbicide treatments applied preemergence on natalgrass cover at the Tenoroc Fish Management Area in 2009. ................................ ................. 68 4 5 Influence of herbicide treatments applied preemergence on natalgrass cover at the Lake Louisa Mitigation Bank in 2009. ................................ ....................... 69 4 6 Influence of herbicide treatments applied postemergence on natalgrass cover at the Tenoroc Fish Management Area in 2008. ................................ ................. 70 4 7 Influence of herbicide treatments applied postemergence on natalgrass cover at the Lake Louisa Mitigation Bank in 2009. ................................ ....................... 70 4 8 Disturbance to plots caused by wild hogs at the Lake Louisa Mitigation Bank in 2009. ................................ ................................ ................................ ............... 71 4 9 Presence of species within treatment areas at t he Tenoroc Fish Management Area ................................ ................................ ................................ .................. 71 4 10 Presence of species within treatment areas at the Lake Louisa Mitigation Bank. ................................ ................................ ................................ .................. 72 A 1 Saturated salt solutions and the corresp onding relative humidity levels at 25 C 1 ................................ ................................ ................................ .................... 103

PAGE 8

8 LIST OF FIGURES Figure page 2 1 The effect of storage length on natalgrass germination.. ................................ .... 26 2 2 The effect of temperature on natalgrass germination.. ................................ ....... 27 3 1 Seed burial tube opened and placed in the greenhouse. ................................ ... 37 3 2 Seed exclusion frame at the Lake Louisa Mitigation Bank. ................................ 38 3 3 Effects of burial on natalgrass seedling emergence.. ................................ ......... 39 3 4 Effects of length of seed exclusion on the number of natalgrass seedlings per m 2 .. ................................ ................................ ................................ ..................... 40 3 5 Effects of wind speed on distance traveled by natalg rass seeds.. ...................... 41 4 1 The effect of metsulfuron applied preemergence on natalgrass biomass.. ......... 73 4 2 The effect of hexazinone appli ed preemergence on natalgrass biomass.. ......... 74 4 3 The effect of imazapyr applied preemergence on natalgrass biomass.. ............. 75 4 4 The ef fect of imazapic applied preemergence on natalgrass biomass.. ............. 76 4 5 The effect of imazamox applied preemergence on natalgrass biomass.. ........... 77 4 6 The effect of sulfometuron applied preemergence on natalgrass biomass.. ....... 78 4 7 The effect of metolachlor applied preemergence on natalgrass biomass.. ......... 79 4 8 The effect of pendimethalin applied preemergence on natalgrass biomass.. ..... 80 4 9 The effect of metsulfuron applied postemergence on natalgrass bioma ss. Trial 1.. ................................ ................................ ................................ ............... 81 4 10 The effect of metsulfuron applied postemergence on natalgrass biomass. Trial 2.. ................................ ................................ ................................ ............... 82 4 11 The effect of s ulfometuron applied postemergence on natalgrass biomass. Trial 1.. ................................ ................................ ................................ ............... 83 4 12 The effect of sulfometuron applied postemergence on natalgrass biomass. Trial 2.. ................................ ................................ ................................ ............... 84 4 13 The effect of hexazinone applied postemergence on natalgrass biomass. Trial 1.. ................................ ................................ ................................ ............... 85

PAGE 9

9 4 14 The effect of hexazinone applied postemergence on natalgrass bio mass Trial 2.. ................................ ................................ ................................ ............... 86 4 15 The effect of fluazifop applied postemergence on natalgrass biomass. Trial 1.. ................................ ................................ ................................ ....................... 87 4 16 The effect of f luazifop applied postemergence on natalgrass biomass. Trial 2.. ................................ ................................ ................................ ....................... 88 4 17 The effect of imazamox applied postemergence on natalgrass biomass. Trial 1.. ................................ ................................ ................................ ....................... 89 4 18 The effect of imazamox applied postemergence on natalgrass biomass. Trial 2.. ................................ ................................ ................................ ....................... 90 4 19 The effect of imazapyr applied postemergence on natalgrass biomass. Tria l 1.. ................................ ................................ ................................ ....................... 91 4 20 The effect of imazapyr applied postemergence on natalgrass biomass. Trial 2.. ................................ ................................ ................................ ....................... 92 4 21 The effect of imazapic appl ied postemergence on natalgrass biomass. Trial 1.. ................................ ................................ ................................ ....................... 93 4 22 The effect of imazapic applied postemergence on natalgrass biomass. Trial 2.. ................................ ................................ ................................ ....................... 94 4 23 The effect of glyphosate applied postemergence on natalgrass biomass. Trial 1.. ................................ ................................ ................................ ............... 95 4 24 The effect of glyphosate applied postemergence on natalgrass biomass. Trial 2.. ................................ ................................ ................................ ............... 96

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science BIOLOGY, ECOLOGY AND MANAGEMENT OF NATALGR ASS ( Melinis repens) By Courtney Ann Stokes December 2010 Chair: Gregory E. MacDonald Major: Agronomy Natalgrass ( Melinis repens (Willd.) Zizka) is a species native to Africa. Introduced to the U.S. in the 1800s, natalgrass was grown as a hay crop in c entral Florida in the early 1900s. Although no longer cultivated, natalgrass is still common throughout much of the state. As the restoration of native plant communities becomes increasingly important, land managers are struggling to control this species To develop an effective management plan, m ore information is needed about seed biology and ecology as well as the chemical management of natalgrass. Natalgrass seeds were exposed to various light, temperature, pH and osmotic potential treatments to bett er characterize the conditions in which these seeds germinate. Natalgrass does not require light for germination. Maximum germination occurs at 20 C and greater and at pH levels ranging from 6 to 8. Germination did not occur at water potentials less tha n 0.2 MPa. N atalgrass seeds were buried at different depths to determine maximum depth of emergence. Natalgrass seeds emerged from a depth of 5 cm, the greatest depth tested in the study. P reliminary tests show that natalgrass likely requires an afterr ipening period after seed shed to reach maximum potential for germination.

PAGE 11

11 Natalgrass seed longevity was studied under field conditions. Seed burial tubes were constructed, buried and exhumed after per iods ranging from 0 to 15 mo An initial decline in g ermination was observed after 3 mo with no further decline. These results indicate the onset of dormancy in natalgrass seeds after burial. This finding will be useful to land managers who plan to utilize tillage for natalgrass control a practice that b uries seeds. Seed longevity was also studied on the ground surface, where dense lay ers of seeds form in infested areas. Exclusion frames were placed over seed deposits to prevent further seed rain and germination under the frames was monitor ed for 12 mo. After 1 mo high levels of germination occurred, but levels declined to 0 seedlings/m 2 within several months. This finding indicates that surface seed deposits are quickl y depleted if land managers can prevent further seed producti on. Natalgrass seeds were exposed to wind at varying speeds and the distance traveled was measured. Seeds traveled less than 2 m at the highest wind speed, 32 km/h. A number of herbicides were tested in the greenhouse and in the field to determine potential for natalgrass con trol pre and postemergence. Metsulfuron and fluazifop offered little or no control. Glyphosate provided excellent control, but the lack of residual activity resulted in immediate reinfestation. Pendimethalin and metolachlor offered good control preemer gence, but were detrimental to native species. Hexazinone and sulfometuron provided good control pre and postemergence, but were also detrimental to native species. Imazam ox, imazapyr and imazapic offe red less control but stunt ed growth and delay ed flow ering. These herbicides were also less harmful to native species. Many native plants are tolerant to imazapic and gained a competitive advantage when this herbicide was applied.

PAGE 12

12 CHAPTER 1 INTRODUCTION Natalgrass ( Melinis repens (Willd.) Zizka) is a sp ecies native to Africa. This species was f irst described as Saccharum repens by Willdenow in Ghana in 1797 and in the last 100 years has also been recognized by the names Tricholaena rosea Nees Tricholaena repens Nees, Rhynchelytrum roseum (Nees) Stapf & Hubb. and Rhynchelytrum repens (Willd.) C.E. Hubb. ( Wunderlin and Hansen 2009). Natalgrass is widespread in many tropical and subtropical regions of the world, including much of Africa, southeast Asia, Central and South America and parts of the United St ates (Haselwood and Motter 1966; Kleinschmidt and Johnson 1977 ; Hfliger and Scholz 1980 ). In the U.S., specimens have been collected from Florida, Georgia, Louisiana, Texas, New Mexico, Arizona, California, North Carolina, Maryland and Hawaii ( USDA 2010 ) In Florida, vouchered specimens have been collected from 50 of 67 counties ( Wunderlin and Hansen 2009 ; FLMNH 2010 ). It is unknown when natalgrass was introduced to the U.S., but it was cultivated as an ornamental as early as 1866 (Tracy 1916). Tracy (1916) describes several examples of natalgrass spread from introductions in Florida and notes that this species was tested as a forage grass in the Department of Agriculture trial gardens as early as 1878. Between 1884 and 1894, seed sources of natalgras s grown for forage included Brazil, So uth Africa, Australia, India, Hawaii and Florida In 1892, the Florida Agricultural Experiment Station released natalgrass as its first forage grass cultivar (Mislevy and Quesenberry 1999). The U.S. Department of Agri culture (USDA) reported over 30,000 acres of natalgrass cultivated for hay in central Florida in 1915 (Tracy 1916). Scott (1913) also

PAGE 13

13 natalgrass was commonly grown as a war m season crop in rotation with various winter crops. Natalgrass was also commonly cultivated as a hay crop in the row middles of citrus groves in Florida (Tracy 1916). In addition to discussing the spread of natalgrass from points of introduction in the state of Florida, Tracy (1916) also notes the value of natalgrass as a smother crop: sand spurs ( Cenchrus spp.), a common pest in citrus groves, in areas where natalgrass was seeded. Finally, Tracy notes that if a field of natalgrass is established, it will continue to produce a crop for many years, particularly if cultivated occasionally. Although natalgrass has not been cultivated in Florida in recent memory, this species is still w idespread throughout the state (Wunderlin and Hansen 2009) The same qualities which made natalgrass a good forage crop in the early 1900s have also caused na talgrass to become a problematic weed. The Florida Exotic Pest Plant Council (FLEPPC) now lists natalgrass as a Category I invasive species in the state, indicating that research has shown that this species is considered to have caused significant ecologi cal harm (FLEPPC 2009). Biology Natalgrass is alter nately described as an annual, perennial or short lived perennial species ( Small 1933 ; Haselwood and Motter 1966; Kleinschmidt and Johnson 1977 ; Hfliger and Scholz 1980 ) In Florida, this species will perenniate if freezing temperatures do not occur but acts as an annual if freezing temperatures do occur (C. A. Stokes, unpublished data). In its native range, temperatures remain warm

PAGE 14

14 throughout the year but periodic dry seasons occur (Klages 1947). In addition, heavy grazing pressure occurs seasonally in these areas, contributing to an annual growth habit. Natalgrass grows erect to about 1 m in height and possesses slender culms that are often geniculate and root at the nodes. This species has fibrous roots and does not produce rhizomes. The leaf blades are flat and linear, usually reaching up to 20 cm in length and ranging from 3 to 10 mm wide (Hitchcock 1950; Hfliger and Scholz 1980) The lig ule is a rosy fringe of hairs 1 to 2 mm in length. Pube scen ce may occur on the leaf sheath s and both surfaces of the leaf blades. The inflorescenc e of natalgrass is a panicle 10 to 15 cm long and 5 to 10 cm wide ( Hfliger and Scholz 1980) The spikelets are 3 to 6 mm long and approximately 2 mm wide. The sp ikelets are pedicelled and covered with dense silky hairs that are initially rosy pink in color but fade to white with age Natalgrass plants sometimes accumulate anthocyanins in response to stress ( Small 1933; Hitchcock 1950; Hfliger and Scholz 1980). Natalgrass is a prolific seed producer and flowers nearly year round in Florida if it is not killed by freezing temperatures. In areas infested with natalgrass, dense seed deposits up to 5 cm deep have been observed on the ground (C. A. Stokes, unpublish ed data). Tracy (1916) suggests that 45.5 kg (100 lb) of seeds per 0.4 ha (1 acre) could be collected from the initial growth of a stand of natalgrass. Natalgrass is most commonly found growing in dry, sandy soils and does not tolerate wet conditions ( Hit chcock 1950; Haselwood and Motter 1966). This species is most commonly found gr owing at low elevations but has also been observed at an elevation of approximately 1500 m (C. A. Stokes, unpublished data).

PAGE 15

15 Management There is little available information r egarding natalgrass management. Tracy (1916) states that plants are controlled by tillage H owever, it is noted that this act buries seeds and new growth will occur if another tillage operation uncovers the seeds. This was considered a desirable trait w hen natalgrass was used as a forage but suggests that natalgrass may be difficult to remove from a site when considered an undesired species. Hernndez Quiroz (2010) found that natalgrass seedling emergence in Chihuahua, Mexico grasslands was unaffected by prescribed fire. Fire is often used as adapted ecosystems to promote desired native species (Provencher et al. 2001). Based on the findings of Hernndez Quiroz it does not appear that this management strategy is an effective choice for the control of natalgrass. Nat algrass can be controlled by spot treatments of glyphosate or by imazapyr (MacDonald et al. 2008). Imazapic is also reported to offer some control of natalgrass (Kluson et. al 200 0; Richardson et al. 2003). However, many of the other herbicides commonly used in natural areas have not been tested for natalgrass control. Research Objectives Some information is available regarding the cultivation of natalgrass as a forage, but litt le information is available regarding seed biology, ecology or management. The objectives of this research are therefore to better characterize the biology, ecology and management of natalgrass with the goal of contributing to a more comprehensive managem ent plan for the control of this species.

PAGE 16

16 CHAPTER 2 SEED GERMINATION CHA RACTERISTICS OF NATA LGRASS ( MELINIS REPENS) Introduction Natalgrass ( Melinis repens (Willd.) Zizka, formerly Rhynchelytrum repens (Willd.) C.E. Hubb.) is a grass native to Africa that has become a problematic weed in many tropical and subtropical regions around the world, including Florida, Mexico, the Caribbean, Central America, Brazil and many Pacific islands ( Haselwood and Motter 1966; Kleinschmidt and Johnson 1977 ; Hfliger and Scho lz 1980 ). In Florida, natalgrass can be found in many areas but is particularly widespread along the central Florida ridge in citrus groves and reclaimed phosphate mining areas (Kluson et al. 2000) The Florida Exotic Pest Plant Council (FLEPPC) consider s natalgrass a Category I invasive in Florida, indicating that this species is considered to have caused significant ecological harm (FLEPPC 2009 ). For instance, research show s that natalgrass invade s undisturbed ecosystems such as pine rocklands in Flori da (Possley and Maschinski 2006). Natalgrass is sometimes grown as an ornamental and was grown for this purpose in the United States as early as 1866 (Tracy 1916). Tracy (1916) also states that natalgrass was grown as a forage plant in the U.S. Department of Agriculture (USDA) trial gardens in 1878. Between 1891 and 1894, the USDA received natalgrass seeds from Natal, South Africa; Queensland, Australia; India and Hawaii. In 1892, natalgrass was the first forage grass cultivar released by the Florida Agr icultural Experiment Station (Mislevy and Quesenberry 1999). Over 30,000 acres of cultivated natalgrass were reported in central Florida in 1915. Natalgrass was often grown between rows of

PAGE 17

17 in citrus groves (Tracy 1916). Natalgrass is an annual species that sometimes perenniates in warmer climates. Although its native range in south and east Africa has a warm climate, these regions experience dry conditions and heavy grazing pressure from m igrating animals during certain times of the year, resulting in plant die back (Klages 1949). In Florida, natalgrass will sometimes perenniate if temperatures do not reach freezing (C A Stokes, unpublished data ). Natalgrass forms tussocks that grow up to 1 m in height. While this species does not produce rhizomes, it is capable of rooting at the nodes and sometimes develops a sprawling appearance (Haselwood and Motter 1966). Natalgrass inflorescences are panicles up to 20 cm long; initially rosy pink the inflorescences fade to silver with age. Natalgrass produces pedicelled spikelets that are covered with dense, silky hairs, giving the plant a feathery or fluffy appearance (Hfliger 1980). Natalgrass is a prolific seed producer, and these seeds are windborne. Tr acy (1916) suggests that 45.4 kg (100 lb) of seeds per 0.4 ha (1 acre ) could be expected from the initial growth of a natalgrass crop. In areas where severe natalgrass infestations occur, dense layers of seeds up to 5 cm thick have been obse rved on the ground (C. A. Stokes, unpublished data ). Natalgrass seeds appear to be key to the rapid spread of this species and extensive seed deposits are likely a reason for the persistance of natalgrass in a given area. Little research has been conduct ed concerning the biology of natalgrass and no published research is available that addresses seed biology. If effective management plans for this species are to be

PAGE 18

18 developed, there is a need for a better understanding of the environmental factors that af fect natalgrass seed germination. Therefore, the objectives of this research were to determine whether seed dormancy was present and to examine the effects of light, temperature, pH, water stress and depth of burial on natalgrass seed germination. Materia ls and Methods Seed Source Natalgrass seeds were collected in November 2007 in Polk County, FL from both the duff layer on the ground and from the seedheads of mature plants. Duff layer seeds were likely deposited over the previous summer and fall seasons Additional seeds were collected from the Lake Louisa Mitigation Bank in Lake County, FL in November 2008 from both the duff layer and the seedheads of mature plants and again in December 2009 from only the duff layer. Both groups of seeds collected in 2008 were used for preliminary germination tests, while seeds from the duff layer were used for the first run of all further experiments. Duff layer seeds collected in 2009 were used when each experiment was repeated. Unless otherwise stated, seeds were stored in a paper bag at room temperature. General Germination Test Protocol Unless otherwise stated, natalgrass seed germination was tested by placing thirty seeds with intact husks evenly in a 9 cm Petri dish 1 containing 1 piece of filter paper 2 The fi lter paper was moistened with 4 mL of deionized water (pH = 6) or test solution. Each Petri dish was sealed with parafilm and placed in a growth chamber at 30 1 C under constant light (200 mol/m 2 / s 1 photosynthetic flux density [PPFD]). Germination 1 Fisherbrand Petri dishes, Fisher Scientific, 2000 Park Lane Drive, Pittsburgh, PA 15275. 2 Fisherbrand P8 filter paper, Fisher Scientific, 2000 Park Lane Drive, Pi ttsburgh, PA 15275.

PAGE 19

19 was visually determined after 14 d. Seeds were considered germinated when the radicle emerged from the seed coat. Any ungerminated seeds were tested for viability according to procedures described in the Handbook for Tetrazolium Testing (Moore 1985). In th is procedure, seeds were removed from the husk and seed coat and placed in a 0.25% tetrazolium solution for 24 h in the dark. Seeds were examined under a dissecting microscope and were counted as viable if the entire embryo was stained red or pink. Perce ntage of seed germination was calculated by dividing the number of germinated seeds by the number of total viable seeds in each Petri dish, then multiplying by 100. Each Petri dish was considered a replication, each treatment was replicated 4 times and ea ch experiment was conducted twice. Preliminary Germination Test Preliminary germination tests (data not shown) indicated that seeds collected from the duff layer had a much higher germination rate than seeds collected directly from seedheads, suggesting that natalgrass seeds require an afterripening period after the seeds are shed. To test this hypothesis, seeds were collected with a sweep net from mature seedheads at the Lake Louisa Mitigation Bank. A sample of these seeds was tested for germination im mediately, and the remaining seeds were divided into 2 groups. The first group was stored in a paper bag at 4 C and the second group was stored in a paper bag at 25 C. Seeds from each group were tested for germination at intervals of 2, 4, 6, 8, 10 and 1 5 weeks. For dormancy tests, percent germination was calculated by dividing the number of germinated seeds by the total number of seeds tested. All environmental conditions were the same as described in the general germination protocol.

PAGE 20

20 Light To determ ine the effect of light on natalgrass seed germination, dry seeds were placed in Petri dishes containing water in a dark room with only green light present to ensure that all hydration occurred in the absence of light. The dishes were immediately wrapped in 2 layers of aluminum foil to prevent light penetration. Seed germination was then compared to control seeds exposed to continous light. All other environmental conditions were the same as described in the general germination protocol. Temperature The effects of temperature on natalgrass seed germination were determined by placing seeds in Petri dishes Dishes were incubat ed at constant temperatures of 10, 20, 25, 30 or 35 C. All other environmental conditions were the same as described in the general germination protocol. pH To evaluate the effects of varying pH levels on natalgrass seed germination, seeds were placed in Petri dishes containing buffer solutions at pH values of 4, 6, 8 and 10. Seeds serving as the control were placed in Petri dishes c ontaining deionized water. Buffer solutions were prepared at 25 M and included potassium hydrogen phthalate, 2 (4 morpholino)ethanesulfonic acid (MES), N 2(2 hydroxyethyl) piperazine N 2 ethanesulfonic acid (HEPES) and Tris(hydroxymethyl) aminomethane ( TRIS) for pH levels of 4, 6, 8 and 10, respectively. Buffer solutions were titrated with HCl or NaOH to achieve the desired pH. All other environmental conditions were the same as described in the general germination protocol.

PAGE 21

21 Water Stress To determine t he effects of water stress on natalgrass seed germination, seeds were placed in Petri dishes with aqueous solutions of polyethylene glycol (PEG 600) with osmotic potentials of 0.2, 0.4, 0.6, 0.8 and 1.0 MPa. Seeds serving as the control were placed i n Petri dishes with deionized water. A vapor pressure osmometer 3 calibrated with aqueous solutions of sodium chloride was used to confirm water potential. All other environmental conditions were the same as described in the general germination protocol. Depth of Burial To determine the effects of burial depth on natalgrass seedling emergence, seeds were planted in a wooden box containing field soil (Apopka sand : loamy, siliceous, subactive, hyperthermic Grossarenic Paleudults ) collected from the Lake Loui sa Mitigation Bank in Clermont, FL. Treatments were separated by wooden dividers. Ten seeds each were planted at depths of 0, 0.25, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 4.0 or 5.0 cm. The box was placed in the greenhouse and received irrigation as needed to ma intain adequate soil moisture. Seedlings were counted 21 d after planting. After the number of seedlings was counted, each seedling was exhumed to confirm that the soil did not shift and that the seed husk was present at the proper depth. Statistical Ana lysis Unless otherwise stated, all experiments were conducted twice using a completely randomized design with 4 replications of each treatment. Data were subjected to analysis of variance. There was no significant (P > 0.05) trial by treatment interactio n 3 Wescor vapor pressure osmometer, Model 5500, Wescor, Inc., 459 S. Main Street, Logan, UT 84321.

PAGE 22

22 for each experiment; data were therefore pooled for analysis. Means were separated using 95% confidence intervals. Germination is reported as percent germination the 95% confidence interval. Regression analysis was u s ed to determine the effects of t emperature and depth of burial. Results and Discussion Preliminary Germination Test S eed collected from the duff layer had a n initial germination rate of 49% 3.8%, while seed collected directly from seedheads had a germination rate of 6% 5.1%. There w as no difference (P > 0.05) between the germination rates of seed stored at 4 C and seed stored at 25 C, so data were pooled for analysis. Germination increased over time to 25% 4.6% after 15 weeks of storage (Figure 2 1) These results indicate that n atalgrass may require an afterripening period after seed shed to reach maximum potential for germination. Afterripening is defined as the loss of a dormant state over a period of time through exposure of seeds to a set of environmental conditions after ma turation and separation from the parent plant (Simpson 1990). Afterripening has been observed in a number of grass species and is well documented in wild oat ( Avena fatua L.) and red rice ( Oryza punctata L.) ( Quail and Carter 1969 ; Leopold et al. 1988; Fo ley 1994). Afterripening can be influenced by environmental conditions such as moisture status and temperature; the conditions that best facilitate afterripening vary with species (Foley 2001). While the results from this experiment suggest that afterrip ening does occur in natalgrass, further research is required to better characterize the afterripening mechanism. This could provide useful information to land managers attempting to control natalgrass infestations. Because dense seed deposits can form on the ground surface in areas where infestations occur, management should not end

PAGE 23

23 when all plants have been eliminated. A better understanding of natalgrass afterripening could result in a prediction of how long germination should be expected from duff laye r deposits after seed rain has ended. Light No difference (P > 0.05) was observed between the rate of natalgrass seed germination in the light and in the dark The rate of germination was 90% 5.8% in light and 74% 17.3% in dark. These results suggest that light is not a requirement for natalgrass seed germination. Based on these results, experiments to determine the effects of varying light quality (phytochrome based studies) were not performed. Temperature Natalgrass seed germination did not occur a t a constant temperature of 10 C, and only 4% 5.4% germination occurred at 15 C (Figure 2 2) However, germination increased as temperature increased The highest level of germination observed was 89% 10.2% at 30 C, although there was no significant difference among germination rates at 20 C or higher. This outcome was not surprising, because many tropical species have optimum germination rates above 20 C (Teuton et al. 2004; Wilder 2009). These results could explain why natalgrass has not become in vasive in U.S. states other than Florida and Hawaii and why natalgrass has not spread to areas outside the southern U.S. pH Natalgrass seed did not germinate at pH 4 or 10. The rate of natalgrass seed germination at pH 6 was 96% 6.3% and at pH 8 was 91% 5.3%. Germination rates at pH 6 and 8 were not significantly different (P > 0.05). These results indicate that natalgrass seeds do not successfully germinate at acidic or basic pH levels. Natalgrass

PAGE 24

24 is most often a problem in disturbed areas such as reclaimed phosphate mining areas and newly cultivated soils; however, populations appear to decline over time. This decline could possibly be a result of soils growing more acidic as they revert back to n atural pH levels closer to 5 (Adjei and Rechcigl 20 04 ). Water Stress Natalgrass seed germination was greatly affected by water stress. Germination was 85% 15.4% in deionized water, while seeds placed in the 0.2 MPa test solution had a germination rate of 98% 23.4%. At osmotic potentials less than 0 .2 MPa no germination was observed. These results indicate that natalgrass seed germination is dependent on adequate soil moisture. Bradford (1990) describes seed germination as the process of initiating growth of a previously quiescent or dormant embryo a process that usually begins with the imbibition of water. The rate of water imbibition and seed germination is generally very sensitive to changes in soil water potential. Evans and Etherington (1990) found that some species adapted to dry or well dr ained habitats germinate well even in soil with osmotic potentials as low as 1.5 MPa. Other research has also shown that some species adapted to dry conditions, such as yankeeweed ( Eupatorium compositifolium Walt.), germinate at low osmotic potentials (M acDonald et al. 1992). However, Evans and Etherington (1990) found that a number of species typically found in dry or well drained areas did not germinate in dry soils. They suggested that this may be a response that confers an ecological advantage on a species in dry conditions. If seeds of these species germinate in very low water potentials, seedling establishment failures may occur if dry conditions continue. However, if the seeds do not germinate until higher levels of moisture are present in the s oil, there is a greater chance that seedlings will survive. This response may allow

PAGE 25

25 natalgrass to successfully germinate and reach maturity in the dry, sandy areas in which it is typically found. Because natalgrass germination is so dependent on availabl e soil moisture, it may also be possible to predict, based on current rainfall patterns in an area, when large numbers of seeds will germinate in the field. Depth of Burial There was no significant difference among treatments; emergence was fairly uniform across all the depths tested. These results indicate that natalgrass can emerge from depths of up to 5 cm. F urther testing is required to determine the depth at which natalgrass emergence is impeded Conclusions Natalgrass seeds do not require light for germination. Although germination occurs at 15 C, high levels of germination occur at temperatures of 20 C and greater. Natalgrass germinates within a fairly neutral pH range of 6 to 8. Germination also appears to be dependent on soil moisture. Natalgr ass seedlings can emerge from depths of at least 5 cm. Finally, natalgrass appears to require an afterripening period after seed shed to reach maximum germination potential. These results indicate that land managers should expect most natalgrass germinat ion to occur as soil temperatures reach 20 C and above and rainfall becomes more consistent. A preemergence herbicide application at this time may inhibit seedling growth and effectively reduce natalgrass populations.

PAGE 26

26 Figure 2 1 The effect of storage length on natalgrass germination. Values represent the mean of 8 replications with standard error y = 5.53 + 19.03(1 e 0.18 x ) R 2 = 0. 90

PAGE 27

27 Figure 2 2 The effect of temperature on natalgrass germination. Values re present the mean of 8 replications with standard error y = 138.19 + 15.17x + 0.25x 2 R 2 = 0. 87

PAGE 28

28 CHAPTER 3 SEED ECOLOGY OF NATA LGRASS ( MELINIS REPENS ) Introduction Natalgrass ( Melinis repens (Willd.) Zizka ) is a species native to the grasslands of south and east Africa. Natalgrass has become an increasing problem in many tropical and subtropical areas around the world, including Mexico, the Caribbean and numerous countries in Asia and Central and South America, along with Florida ( Haselwood and Motter 1966; Kleinschmidt and Johnson 1977 ; Hfli ger and Scholz 1980 ) Natalgrass has been present in the U.S. since at least 1866. When natalgrass was observed grow ing ing easily by seed, it was tested as a forage grass (Tracy 1916). Eventually, natalgrass was released as the Quesenberry 1999). In Florida, natalgrass has been observed to quickly invade open, disturbed areas such as roadsides, citrus groves and phosphate m ining areas (Kluson et al. 2000; Possley and Maschinski 2006) Natalgrass often becomes a problem after another invas ive species such as cogongrass ( Imperata cylindrica (L.) Beauv. ) has been removed (G. E. MacDonald, personal communication ). Natalgrass h as also been shown to invade undisturbed ecosystems such as the pine rocklands of south Florida (Possley and Maschinski 2006). Natalgrass produces large numbers of seeds which are dispersed readily by wind Rapid dispersal of these seeds is presumed to be the method by which natalgrass quickly invades. A 37.6 ha (93 acre) site at the Lake Louisa Mitigation Bank in Lake County, FL offers a good example of natalgrass spread. According to the land

PAGE 29

29 manager s the area was formerly bahiagrass ( Paspalum nota tum Fluegg ). In preparation for native plant establishment the bahiagrass was removed with glyphosate, resulting in nearly bare ground Almost immediately, the land manager s noticed natalgrass spreading from a small patch at the corner of the property. Within 6 mo natalgrass had spread across the entire 37.6 ha (M Green, personal communication) The spread occurred in the direction of prevailing winds; this conclusion was supported by the observation of older plants near the initial infestation whi le younger plants were observed at a greater distance from this area Once a natalgrass infestation has occurred, dense duff layer seed deposits often form particularly in the fall and early winter months Little is known about how long these seeds persi st on the soil surface. In addition, little is known about how long these seeds persist if buried. Natalgrass plants are easily controlled by mechanical cultivation, but this action buries seeds. More information is needed about natalgrass seed persista nce under field conditions so that land managers can act in the most effective manner possible to control this species. The objectives of this study we re to determine the length of natalgrass seed persistence on the soil surface, to determine the length of natalgrass seed persistence when buried, and to better characterize the dispersal of natalgrass seeds by wind. This information will be important to land managers when developing a management plan for this species. Materials and Methods Seed Burial Duff layer material, including large numbers of seeds, was collected from a single field site within the Lake Louisa Mitigation Bank in Lake County, FL in November 2008.

PAGE 30

30 Field soil (Apopka sand : loamy, siliceous, subactive, hyperthermic Grossarenic Paleudults ) was collected from the same site after the duff layer was completely removed. To determine if natalgrass seeds were present in the sample, the soil seed bank was assessed by placing trays of soil in the greenhouse and monitoring seedling emergence. The seed bank was determined to b e negligible because of the complete lack of natalgrass seedlings and the low levels of seedling emergence for all other species (data not shown). Material collected from the duff layer was comprised of seeds (approximately 5 0% of the total volume), soil (approximately 10% of the total volume) and other debris such as small twigs and stems (about 40% of the total volume). It proved to be difficult to separate the seeds from the other duff layer contents because of the fluffy hairs present on the husk, which clung to the other materials present. As a result, seeds for each replication of this study were not counted. Instead, duff layer material was mixed evenly and then divided by weight. PVC pipe with a diameter of 4 in was cut into segments 30 cm in length and used to create seed burial tubes. Each 30 cm segment of pipe was cut in half lengthwise. Five 0.5 cm holes were drilled in each half, evenly spaced and running in a line lengthwise down the center of each half. Nylo n mesh fabric was folded in half, forming a packet, and filled with a mixture of 10 g field soil and 1.2 g of the duff layer material. The 2 halves of the PVC pipe were filled with field soil and placed back together with the mesh packet centered between t hem. The 2 halves were secured together with duct tape 4 Window screen 5 was placed over each end to prevent the field soil from 4 All Purpose Duct Tape, Shurtape Technologies, LLC, 1506 Highland Ave NE, Hickory, NC, 28601.

PAGE 31

31 escaping the tube. The end product was a 4 in diameter tube filled with field soil, with a mesh fabric packet containing both field soil and seeds suspended within the field soil in the center of the tube and stretching the 30 cm length of the tube (Figure 3 1) The holes drilled in the sides and the screen placed over the ends were present to allow the maximum moisture and gas exchange possible between the soil inside the tube and the outside environment. The contents of the tubes were kept dry at all times to prevent water imbibition and possible germination before the tubes were placed in the field. One hundred tubes were bu ried in 10 groups of 10 tubes each within a 37.6 ha study area at the Lake Louisa Mitigation Bank in Lake County, FL in June 2009 The tubes were buried on end, with the top end level with the ground surface. Each location was marked with a flag and the GPS coordinates noted. One tube from each group was exhumed at 0, 3, 6, 9, 12 and 15 mo after burial. The tubes were split in half lengthwise by removing the duct tape holding the halves together. The mesh packets were then opened without disturbing th e contents. Each tube was then placed in the greenhouse and irrigated as needed to maintain adequate soil moisture. After 2 weeks, the number of natalgrass seedlings present was counted. The total number of seedlings was noted as well as the number of s eedlings in each area of the tube corresponding to depths of 0 to 5, 5 to 10, 10 to 15, 15 to 20, 20 to 25 and 25 to 30 cm when the tubes were buried on end. Seed Exclusion Ten areas within the study site at the Lake Louisa Mitigation Bank were chosen base d on 3 criteria: the presence of a duff layer natalgrass seed deposit, an area of at 5 Fiberglass screen wire, PHIFER Inc., P.O. Box 1700, Tuscaloosa, AL, 35403.

PAGE 32

32 least 1 m 2 and a lack of plants of any species growing within the area. In April 2009, each area was covered with a square wooden frame (1 m by 1 m) secured to the ground and covered with screen 6 to exclude any additional seeds from entering the plot (Figure 3 2) Hardware cloth 7 (3 squares/in) was placed over the screen to prevent damage from birds or other animals. The number of natalgrass seedlings growing in each plot was counted 1, 2, 3, 4, 5, 6, 8, 10 and 12 mo after the frames were put into place. After 6 mo the soil and duff layer of half of the plots were disturbed to a depth of 3 cm. After each seedling count, a 1% solution of glyphosate was applied as needed to remove any plants growing in the plot. Wind Wind was generated using a seed blower and wind speed measured with a pocket wind meter 8 Seeds were released into the airstream 46 cm from the floor and 50 cm from the air source. Air speeds included 4, 8, 16, 24 and 32 km/h At each speed, 5 seeds were released; each seed was considered a replication. The study was repeated 4 times for a total of 5 trials. Statistical Analysis The seed burial study and seed exclusion study were conducted using a completel y randomized design. The se studies were each conducted once and included 10 replications. T he wind study was conducted a total of 5 times and included 5 replications per treatment Data were subjected to analysis of variance and m eans were separated usi ng 95% confidence intervals. 6 Fiberglass Screen Wire, PHIFER Inc., P.O. Box 1700, Tuscaloosa, AL, 35403. 7 3 Mesh Galvanized Hardware Cloth, TWP Inc., 2831 Tenth Street, Berkeley, CA, 94710. 8 Kestrel 3000 Pocket Wind Meter, Nielsen Kellerman, 21 Cre ek Circle, Boothwyn, PA 19061.

PAGE 33

33 Results and Discussion Seed Burial There was no significant difference (P > 0.05) in the number of seedlings counted within the areas corresponding to different burial depths; therefore, only the total number of seedlings per b urial tube is discussed. The control group (tubes buried for 0 mo ) had a mean of 50 8 see d lings per tube. There was no significant difference (P > 0.05) among the mean number of seedlings from tube s buried for 3, 6, 9, 12 and 15 mo ; however, these valu es were significantly lower than the mean at 0 mo indicating a decline in natalgrass germination after burial. The lowest mean number of seedlings observed was 13 5 after 9 mo of burial. However, after the initial decline germination levels have remai ned relatively consistent, indicating possible dormancy. There are a number of causes for increased dormancy during burial and decreased seed germination after burial. Several studies have shown that seeds can become light sensitive after burial, even whe n germination of the same seeds was not previously light dependent (Wesson and Wareing 1969, Mandoli and Briggs 1981). In addition, several studies have suggested that the increase in dormancy and decrease in overall germination observed after burial coul d be the result of toxic metabolites produced by anaerobic respiration occurring in hypoxic conditions (Holm 1972, Benvenuti and Macchia 1995). Finally, decreased seed germination can occur because of seed mortality. This mortality can be a result of pre dation, failed germination, aging or attack from patho gens (Baskin and Baskin 1998). Seed burial studies are often criticized for creating conditions that can result in accelerated seed mortality. Van Mourik et al. (2005) found that when high densities o f seeds are buried in mesh bags during seed burial studies, attack from pathogenic fungi often causes an overestimation

PAGE 34

34 of the rate of soil seed bank depletion. However, when lower densities of seeds were mixed with soil inside the mesh bags, mortality wa s significantly lower. Based on the findings of Van Mourik et al. (2005), the decision was made to mix field soil with natalgrass seed samples before burial to more closely mimic natural conditions. Because the study methods were designed in such a way as to limit artificial acceleration of seed mortality, decreased germination over the course of the study can more positively be attributed to an increase in dormancy. Natalgrass plants are easily controlled by cultivation (Tracy 1916), but the effects of se ed burial as a result of cultivation or other disturbance were unknown prior to this research study. The results from this study suggest that, if a subsequent cultivation or other disturbance brings natalgrass seeds back to the soil surface, these seeds h ave the potential to germinate at least 1 year after the initial burial. Land managers utilizing cultivation as a control method for natalgrass should expect germination to occur if they repeat tillage operations within this time period. Seed Exclusion Th e mean number of natalgrass seedlings per m 2 was 521 354 after 1 mo of seed exc lusion from the plots. T his number decreased to 6 7 seedlings after 2 mo and 0 after 4 mo of seed exclusion. No further germination was observed in any of the plots, even after disturbance. These results suggest that seeds deposited in the duff layer over the summer and fall months have largely concluded afterripening by April, when the study began. Afterripening is the loss of a dormant state over a period of time throug h exposure of seeds to a set of environmental conditions after maturation and separation from the parent plant (Simpson 1990). Afterripening commonly occurs in a number of grass species, and has been studied extensively in wild oat ( Avena fatua L. )

PAGE 35

35 and re d rice ( Oryza punctata L.) ( Quail and Carter 1969 ; Leopold et al. 1988; Foley 1994). With rainfall and warm temperatures during the month of April, most of the seeds present in the seed bank during this study germinated at once. These results are consis tent with results from the seed biology studies that show that most natalgrass germination occurs at temperatures higher than 15 C and in conditions with adequate moisture available. These conditions existed at the study site in April 2009 ( FAWN 2010 ). T hese results suggest that if land managers can prevent further seed deposition during the spring, active management of emerging seedlings for several months should result in a significant reduction of new natalgrass growth. Wind There was a large amount of variability in the distance traveled by natalgrass seeds at different wind speeds. At 4 km/h, the mean distance traveled by natalgrass seeds was 67 33 cm. This distance was similar to that observed at 16 km/h (65 26 cm). Both of these distances wer e significantly different from the 158 40 cm the seeds traveled at 32 km/h. No other statistical differences were observed. Natalgrass seeds do not appear to travel great distances when exposed to wind in this manner. However, the methods used in this study may not be the most accurate method of characterizing wind dispersal of seeds. Further research utilizing alternate methods would be advisable. Conclusion s Natalgrass seed does appear to enter a state of increased dormancy when buried. This is con sistent with the observations of Tracy (1916), who instructed growers to plow under natalgrass fields in the winter and then cultivate again in the spring to induce new

PAGE 36

36 growth. Land managers should be mindful of this potential for new growth if they utili ze tillage in natural areas to control natalgrass or other species when natalgrass is present. Although natalgrass can form dense seed deposits in infested areas, the seed bank appears to quickly become depleted when conditions are favorable for germinatio n and further seed rain is prevented. If a land manager can prevent seed production through mechanical or chemical means while effectively controlling germinating seedlings for several months, n atalgrass can be greatly reduced

PAGE 37

37 Figure 3 1 Seed buri al tube open ed and placed in the greenhouse.

PAGE 38

38 Figure 3 2 Seed exclusion frame at the Lake Louisa Mitigation Bank.

PAGE 39

39 Figure 3 3 Effects of burial on natalgrass seedling emergence. Values reflect the mean of ten re plications with 95% confidence intervals.

PAGE 40

40 Figure 3 4 Effects of length of seed exclusion on the number of natalgrass seedlings per m 2 Values reflect the mean of 10 replications with 95% confidence intervals.

PAGE 41

41 Figure 3 5 Effects of wind speed on distance traveled by natalgrass seeds. Values reflect the mean of 25 replications with 95% confidence intervals.

PAGE 42

42 CHAPTER 4 CHEMICAL CONTROL OF NATALGRASS ( M ELINIS REPENS ) Introduc tion Natalgrass ( Melinis repens (Willd.) Zizka ) is native to the savanna regions of south and east Africa. This species has become a weed problem in many tropical and subtropical areas around the world, including Central America and the Caribbean, Florida and Hawaii in the U.S., Brazil and areas in the Pacific (Haselwood and Motter 1966; Kleinschmidt and Johnson 1977 ; Hfliger and Scholz 1980 ). In Florida, natalgrass has become a common problem in dry, sandy areas along the central ridge. This species is especially prevalent in citrus groves, abandoned fields and areas formerly mined for phosphates. The Florida Exotic Pest Plant Council (FLEPPC) considers natalgrass to be a Category I invasive species (FLEPPC 2009), meaning that research has shown that th is species is capable of invading undisturbed ecosystems such as the pine rocklands of south Florida (Possley and Maschinski 2006). Natalgrass was introduced to the U.S. in the 1800s and was grown as an ornamental as early as 1866. Natalgrass was also gro wn as a forage plant by the U.S. Department of Agriculture as well as the Florida Agricultural Experiment Station, which released natalgrass as its first forage grass cultivar in 1892 (Mislevy and Quesenberry 1999) Over 30,000 acres of natalgrass were gr own for hay in central Florida in 1915 (Tracy 1916). Natalgrass is no longer cultivated, but remains widespread in Florida. As restoration becomes an increasing priority for land managers, there is an increased need for natalgrass management. There is li ttle literature available addressing chemical control strategies for this species. Recommendations usually include spot treatments of glyphosate ( MacDonald et al. 2008 ), but the effects of

PAGE 43

43 glyphosate for large scale treatments are unknown. Several studie s report that imazapic shows potential for natalgrass control (Kluson et al. 2000; Richardson et al. 2003), but imazapic control has not been compared to the other herbicides available for use in natural areas The objective of this study wa s to thoroughl y explore the potential of various herbicides for natalgrass control in natural areas in Florida. Materials and Methods Greenhouse Preemergence Study A greenhouse study was ini tiated in May 2010 to determine the potential of various rates of 8 different he rbicides applied preemergence for natalgrass control. The study was repeated. Natalgrass plants were gro wn in the greenhouse in 8.9 cm square pots 1 filled with 360.5 cm 3 of field soil (Apopka sand : loamy, siliceous, subactive, hyperthermic Grossarenic Pa leudults ) collected from the Lake Louisa Mitigation Bank in Lake County, FL This soil was used to mimic field conditions as closely as possible Seeds were collected from the same site in Lake County, FL in December 2009 and stored at room temperature u ntil planting. Approximately 50 s eeds were scattered on the soil surface and lightly pressed into the soil. Pots were irrigated daily in the greenhouse to maintain adequate soil moisture. Herbicides were assigned a 1x rate based on label recommendations, personal observations of the authors and comments from land managers. These rates were estimated to be that which would provide adequate control of the treated plants. Herbicides were also applied at 0.0625, 0.125, 0.25, 0.5, 2 and 4x rates; these were calculated based on the chosen 1x rates. The herbicides used and the 1x rates chosen 1 Traditional Square Pot, Kord U.S.A., Inc., 103 Lachicotte Rd, Lugoff, SC

PAGE 44

44 included imazapic 2 [ 1.4 kilograms active ingredient/hectare ( kg ai/ha) ] imazapyr 3 (1.12 kg ai/ha), imazamox 4 (0.56 kg ai/ha), hexazinone 5 (0.56 kg ai/ha), sulfometuron 6 (0.22 kg ai/ha), metsulfuron 7 (1.12 kg ai/ha), pendimethalin 8 (1.12 kg ai/ha) and metolachlor 9 (1.12 kg ai/ha). Untreated plants served as the control. One pot was considered a replication and each treatment included 4 replications. Treatments were applied within 24 h of seed sowing with a backpack sprayer calibrated to deliver 187 L/ha (20 gal lons /acre) spray solution. Pots were then placed back in the greenhouse and received irrigation as necessary to maintain adequate soil moisture. Two weeks after treatment, seedling density in each pot wa s visually rated as a percentage of the untreated control. Ten weeks after treatment, the number of seedlings in each pot was determined and all above ground biomass was harvested, dried at 60 C for 4 d, and weighed. The total dry weight of harvested mat erial from each pot was converted to a percentage of the average biomass per pot for the untreated controls. Data was analyzed using analysis of variance and means separated at the 0.05 e (LSD) test. Nonlinear regression was used to describe the response of natalgrass to each herbicide 2 Plateau herbicide, BASF Corporation, 26 Davis Dr, Research Triangle Park, NC 27709 3 Habitat herbicide, BASF Corporation, 26 Davis Dr, Research Tria ngle Park, NC 27709 4 Clearcast herbicide, BASF Corporation, 26 Davis Dr, Research Triangle Park, NC 27709 5 Velpar L herbicide, I.E. duPont de Nemours and Company, 1007 Market St, Wilmington, DE 19898 6 Oust XP herbicide, I.E. duPont de Nemours and Compan y, 1007 Market St, Wilmington, DE 19898 7 Escort XP herbicide, I.E. duPont de Nemours and Company, 1007 Market St, Wilmington, DE 19898 8 Prowl H2O herbicide, BASF Corporation, 26 Davis Dr, Research Triangle Park, NC 27709 9 Dual Magnum herbicide, Syngenta Crop Protection, P.O. Box 18300, Greensboro, NC 27419

PAGE 45

45 and I 50 and I 90 values were calculated to determine the herbicide rates necessary to reduce natalgrass biomass by 50 percent and 90 percent, respectively Greenhouse Postemergence Study A greenhouse study was initiated in May 2010 to determine the potential of various rates of 8 different herbicides applied postemergence for natalgrass co ntrol. The study was repeated 2 mo after the initial study. Natalgr ass plants were grown in the greenhouse in 8.9 cm square pots filled with 360.5 cm 3 of Fafard 4 potting mix 10 Seeds were coll ected from Lake County, FL in 2008 and stored at 4 C until planting. Seeds were scattered on the soil surface and plants were thi nned to 1 plant per pot after seedling emergence Pots were irrigated daily in the greenhouse to maintain adequate soil moisture. Herbicides were assigned a 1x rate based on label recommendations, personal observations of the authors and comments from lan d managers. These rates were estimated to be that which would provide adequate control of the treated plants. Herbicides were also applied at 0.0625, 0.125, 0.25, 0.5, 2 and 4x rates; these were calculated based on the chosen 1x rates. The herbicides us ed and the 1x rates chosen included imazapic (1.4 kg ai/ha), imazapyr (1.12 kg ai/ha), imazamox (0.56 kg ai/ha), glyphosate 11 (0.84 kg ai/ha), hexazinone (0.56 kg ai/ha), sulfometuron (0.22 kg ai/ha), metsulfuron (1.12 kg ai/ha) and fluazifop 12 (1.12 kg ai/h a). Treatments included non ionic surfactant (NIS) at 0.25% v/v as required by the herbicide labels. An additional treatment consisting of water plus NIS was included and untreated plants served as the 10 Conrad Fafard, Inc., 770 Silver St, Agawam, MA 01001 11 Roundup WeatherMax herbicide, Monsanto Company, 800 N. Lindbergh Blvd, St. Louis, MO 63167 12 Fusilade herbicide, Syngenta Crop Protection, P.O. Box 18300, Greensboro, NC 27419

PAGE 46

46 control. One pot was considered a replication and e ach treatment included 4 replications. Treatments were applied 8 to 10 weeks after planting with a backpack sprayer calibrated to deliver 187 L/ha (20 gallons /acre) spray solution. For the first trial, plants were approximately 35 cm tall. For the seco nd tri al, plants averaged 45 cm tall. Plants were placed back in the greenhouse and continued to receive irrigation as necessary. Two weeks after treatment, plants were visually rated for injury as a percent of the untreated control Plants were then harveste d at 3 cm above soil level, dried at 60 C for 4 d, and weighed. Plants were allowed to regrow for 8 additional weeks. At this time, plants were visually rated for percentage of regrowth compared to untreated controls and were again harvested, dried and w eighed. The biomass harvested after regrowth was converted to a percentage of the biomass harvested from the regrowth of the untreated control plants. Data was analyzed using analysis of variance and means separated at the 0.05 probability level using Fis describe the response of natalgrass to each herbicide and I 50 and I 90 values were calculated to determine the herbicide rates necessary to reduce natalgrass biomass by 50 percent and 90 percent, r espectively. Field Preemergence Study Field studies to evaluate the potential of several herbicides for preemergence natalgrass control in natural areas were initiated in April 2009 at the Lake Louisa Mitigation Bank in Lake County, FL and in June 2009 at the Tenoroc Fish Management Area in Polk County, FL. The Lake Louisa site was an Apopka sand (loamy, siliceous, subactive, hyperthermic Grossarenic Paleudults) and the Tenoroc site was a Neilhurst

PAGE 47

47 sand (thermic, uncoated Typic Quartzipsamments) Plots we re established in areas where natalgrass infestations had been present for at least 1 year to ensure the presence of seed deposits. Plots measured 3.1 m by 7.6 m (10 ft by 25 ft) and were arranged in a randomized complete block design with 4 replications ; each plot was considered a replication Treatments were applied with a CO 2 pressurized backpack sprayer delivering 187 L/ha (20 gallons /acre) spray solution. Treatments included imazapyr (0.14 and 0.28 kg ai/ha), imazapic (0.14 and 0.28 kg ai/ha), metsulfuron (0.042 and 0.126 kg ai/ha), sulfometur on (0.158 and 0.263 kg ai/ha), hexazinone (0.28 and 0.56 kg ai/ha), pendimethalin (1.12 and 2.24 kg ai/ai) and imazamox (0.14 and 0.28 kg ai/ha). All treatments also included glyphosate (3.08 kg ai/ha) to remove any existing plants within the plot area. Plots treated only with glyphosate served as controls. Plots were visually rated for percentage of natalgrass co ver using the following scale: 0 = no natalgrass cover; 100 = complete natalgrass cover Plots at Lake Louisa were rated 1, 2, 4 and 6 mo aft er treatment (MAT). Plots at Tenoroc were rated 2, 3, 5 and 7 MAT. Data were analyzed using analysis of variance and means were separated Field Postemergence Study Field studies to evaluate the potential of several herb icides for natalgrass control in natural areas were initiated in July 2008 at the Tenoroc Fish Management Area and in July 2009 at the Lake Louisa Mitigation Bank. Plots were established in areas with heavy natalgrass pressure but also some native plants Plots were arranged in a randomized complete block design with 4 replications and measured 7.3 m by 15.2 m (24 ft by 50 ft). Each plot served as a replication. Treatments were applied with an ATV

PAGE 48

48 sprayer delivering 140 L/ha (15 gallons /acre) spra y solution per acre. Treatments at both sites included glyphosate (1.12 and 3.36 kg ai/ha), fluazifop (0.28 kg ai/ha), imazapic (0.28 and 0.56 kg ai/ha), imazapyr (0.14 and 0.28 kg ai/ha) and hexazinone (0.56 and 1.12 kg ai/ha). Also included was a metsulfuron treatment (at Tenoroc, 0.28 kg ai/ha; a t Lake Louisa, 0.11 kg ai/ha). All treatments included a non ioni c surfactant (NIS) at 0.25% v/v Untreated plots served as controls. Plots were visually rated for percentage of natalgrass cover using the following scale: 0 = no natalgrass cover ; 100 = c omplete natalgrass cover Plots at Tenoroc were rated 3, 9, 12 and 24 MAT. Plots at Lake Louisa were rated 3, 6 and 12 MAT. At Lake Louisa, significant damage from wild hogs was observed in some plots, so the percentage of each plot that had been distur bed was also rated 3, 6 and 12 MAT. Data were analyzed using analysis of variance and means were separated at the 0.05 probability level Results and Discussion Greenhouse Preemergence Study There was no significant exper iment by treatment interaction D ata were therefore pooled across experiments. Metsulfuron (Figure 4 1 ) reduced natalgrass biomass by 31% at 0.07 kg ai/ha, but at 0.14 kg ai/ha biomass was higher than the untreated control. Biomass declined with each f urther rate increase, but was only reduced by 60% at the highest rate tested, 4.48 kg ai/ha. The I 50 value generated from the regression curve was 2.1 kg ai/ha (Table 4 1) much higher than the maximum labeled use rate for met sulfuron in natural areas (0. 168 kg ai/ha). No I 90 value was generated because 90% contro l was never

PAGE 49

49 observed These results indicate that metsulfuron has limited potential for natalgrass control. Hexazinone (Figure 4 2 ) showed the most inconsistent results of the compounds tested i n this study. At 0.07 kg ai/ha, biomass was reduced by 70%; however, at 0.14 kg ai/ha, biomass was higher than the untreated controls. A 90% reduction in biomass was not observed in this experiment, but the I 50 value generated from the regression curve w as 0.315 kg ai/ha. This value is well within the maximum labeled rate. Hexazinone provided some control, but control was among the lowest of the herbicides tested. Imazapyr (Figure 4 3 ) also provided somewhat inconsistent control of natalgrass seedlings in the greenhouse. A 41% reduction in biomass was observed at 0.07 kg ai/ha, but biomass reduction compared to the untreated control was only 7% at 0.14 kg ai/ha. However, all other rates of imazapyr reduced natalgrass biomass by at least 65%. The I 50 v alue was 0.21 kg ai/ha, well below the maximum labeled use rate for this compound. No I 90 value could be calculated using the regression curve generated. Although imazapyr treatment did not result in the largest reduction in biomass observed, this compou nd did appear to stunt the natalgrass seedlings that did emerge at higher rates. Imazapic (Figure 4 4 ) application resulted in a 48% reduction of biomass at 0.0875 kg ai/ha, but only a 26% reduction at 0.175 kg ai/ha. The I 50 value generated from the regr ession equation was 0.219 kg ai/ha, a rate slightly higher than the maximum labeled rate for natural areas. The I 90 value was 0.92 kg ai/ha, significantly higher than the maximum labeled rate. Imazapic severely stunted natalgrass seedlings

PAGE 50

50 at high rates, and these seedlings also appeared to be more susceptible to attack from fungal pathogens. When imazamox (Figure 4 5 ) was applied at 0.035 kg ai/ha, only a slight (8%) reduction in natalgrass was observed. Between 0.07 and 0.28 kg ai/ha, biomass was red uced by 51% to 68%. At least a 90% reduction in biomass was observed for all rates higher than 0.28 kg ai/ha. The I 50 value was 0.105 kg ai/ha, below the maximum labeled use rate. However, the I 90 value was 1.04 kg ai/ha, higher than the maximum labeled rate. Imazamox also appeared to stunt natalgrass seedlings at high rates, although the high rates required may prevent effective use of this product in the field Sulfometuron (Figure 4 6 ) reduced natalgrass biomass more than 50% at even the lowest rates tested. At 0.0138 kg ai/ha, biomass was reduced by 52%. At 0.22 kg ai/ha, biomass was reduced by 85%. A 99% biomass reduction was observed at 0.44 kg ai/ha and a 98% reduction at 0.88 kg ai/ha. The I 50 value was 0.02, well within the maximum labeled u se rate, but the I 90 value was much higher at 0.9 kg ai/ha, higher than the maximum labeled rate. These results indicate that low rates of sulfometuron may reduce natalgrass biomass, but sulfometuron may not provide desirable levels of control at the maxi mum labeled rate. Metolachlor (Figure 4 7 ) reduced natalgrass biomass by approximately 31% at 0.07 kg ai/ha and 62% at 0.14 kg ai/ha. The I 50 value was 0.11 kg ai/ha and the I 90 value was 1.65 kg ai/ha, well within the labeled use rate for this compound. Metolachlor may be a good option for natalgrass control preemergence. At the lowest rates tested (0.07 and 0.14 kg ai/ha), pendimethalin (Figure 4 8 ) did not provide control of natalgrass; at these rates, biomass was higher than the untreated

PAGE 51

51 controls. H owever, at 0.28 kg ai/ha, pendimethalin resulted in a 28% reduction in biomass. At 0.56 kg ai/ha, a n 85% reduction in biomass was observed, and at all other rates no biomass was measured. The I 50 value was 0.48 kg ai/ha and the I 90 was 1.07 kg ai/ha. Bo th rates are well within the labeled use rate for pendimethalin, suggesting that this compound may be a good option for natalgrass control applied preemergence. Greenhouse Postemergence Study There was a significant trial by treatment interaction, so data were analyzed separately. For many of the herbicides tested, greater control was observed in the first trial than in the second. This was probably a result of the smaller size of the plants in the first trial. At the time of the first harvest, the avera ge dry weight harvested from 8 untreated plants from the first trial was 5.71 g. For the second trial, the average of 8 untreated plants at the time of the first harvest was 12.01 g, slightly more than double the weight of the plants from the first trial. The larger size of the plants used in the second trial is probably the reason for the slightly reduced control. In the first trial, metsulfuron provided little control of natalgrass (Figure 4 9 ) At the lowest rate tested, 0.07 kg ai/ha, biomass wa s about the same as the control. Metsulfuron applied at 0.14 kg ai/ha resulted in an increase in biomass to 134% of the control. At the highest rate tested, 4.48 kg ai/ha, biomass was only reduced by 24%. I 50 and I 90 val ues were not generated (Table 4 2 ) because 50% and 90% reduction s in biomass w ere not observed during this experiment, even though rates tested reached more than 25 times the maximum labeled use rate. The rate of metsulfuron required to reduce natalgrass biomass by 50% would be extremely high. In the second trial, biomass reduction compared to the control was never observed and metsulfuron

PAGE 52

52 application at all rates resulted in higher biomass (Figure 4 10 ) At the highest rate tested biomass was still 137% of the control. I 50 and I 90 val ues were not calculated because no reduction in growth was observed during this trial (Table 4 3) In both trials, plants were flowering at the time the regrowth was harvested. Based on these results, metsulfuron appears to have little activity on natalg rass plants and can even increase growth, possibly by causing bud break and increased tillering. Metsulfuron should not be recommended for the control of natalgrass. Sulfometuron provided fairly good control of natalgrass at high rates in the first trial, but did not provide as much control in the second trial. In the first trial, sulfometuron at 0.01375 kg ai/ha resulted in a 41% reduction in biomass (Figure 4 11 ) At 0.11 kg ai/ha, biomass was reduced by 57%. Biomass was reduced by 95% at 0.44 kg ai/h a and 98% at 0.88 kg ai/ha. The I 50 value was 0.068 and the I 90 was 0.30 kg ai/ha, both below the maximum labeled use rate. In the second trial, sulfometuron at low rates resulted in an increase in biomass compared to the control (Figure 4 1 2 ) At 0.027 5 kg ai/ha, biomass was the highest at 129% of the control. B iomass was only reduced by 19 % at the highest rate tested No I 50 or I 90 values were generated because the regression curve does not reach these values The maximum labeled rate for sulfometur on is 0.42 kg ai/ha. Sulfometuron may be an option for natalgrass control if plants are small but based on these results, control is somewhat inconsistent and sulfometuron may not be the best option. Hexazinone reduced biomass in the first trial by 38% a t 0.035 kg ai/ha (Figure 4 13 ) At 0.28 kg ai/ha, biomass was reduced by 76%. At all higher rates, biomass was 0% of the untreated control and the plants were killed. The I 50 value was 0.09 kg ai/ha

PAGE 53

53 and the I 90 value was 0.35 kg ai/ha. In the second tr ial, biomass was reduced by 71% at 0.035 kg ai/ha (Figure 4 14 ) At the highest rate biomass was 8% of the untreated. The I 50 was 0.07 kg ai/ha, similar to that of the first trial, but the I 90 value was more than double that of the first trial at 0.74 kg ai/ha. However, these rates are still within the maximum labeled rate, suggesting that hexazinone may be a good option for natalgrass control. In the first trial, fluazifop at all but the lowest rates provided 100% reduction in biomass (Figure 4 15 ) At 0.07 kg ai/ha, biomass was 165% higher than the control. At 0.14 kg ai/ha and rates higher biomass was 0% of the untreated control. For the first trial, the I 50 value was 0.14 kg ai/ha and the I 90 value was 0.38 kg ai/ha. Both these rates are less than the maximum labeled rate. In the second trial, biomass was also higher than the control at 0.07 kg ai/ha ( Figure 4 16 ). At 1.12 kg ai/ha biomass was reduced by 98% and was 0% of the untreated control at all higher rates. The I 50 value for the second tr ial was 0.28 kg ai/ha and the I 90 value was 0.73 kg ai/ha. The I 50 value is less than the maximum labeled rate, but the I 90 value is well over the maximum rate. This compound provided good control in the first trial, but less reliable control in the seco nd trial. Fluazifop may provide some control of smaller plants, but is probably not a good option when plants are large. In the first trial, imazamox at 0.035 kg ai/ha reduced natalgrass biomass by 16% (Figure 4 1 7 ) At all rates higher than 0.035, bioma ss was 0% of the untreated control or no more than 1%. The I 50 value was 0.04 kg ai/ha, and the I 90 was 0.12 kg ai/ha. In the second trial, biomass at the lowest rate (0.035 kg ai/ha) was reduced by 57% (Figure 4 1 8 ) Biomass at 2.24 kg ai/ha was 2% of the control The I 50 was

PAGE 54

54 0.02 kg ai/ha and the I 90 was 0.63 kg ai/ha, which exceeds the maximum labeled rate for imazamox (0.56 kg ai/ha). In both trials, none of the plants at rates higher than 0.035 kg ai/ha were killed, but all were severely stunted. Any regrowth t hat occurred at these rates was abnormal; clusters of small, twisted leaves sprouted at some of the nodes and the tissue was very brittle. These plants also appeared to be more susceptible to attack from fungal pathogens. Natalgrass showed a similar response to imazapyr in both trials. In the first trial, imazapyr at 0.07 kg ai/ha reduced natalgrass biomass by 90% (Figure 4 19 ) At 0.14 kg ai/ha, biomass was less than 1% of the untreated control, and at all higher rates biomass was 0%. Th e I 50 value was 0.02 kg ai/ha and the I 90 was 0.07 kg ai/ha. In the second trial, biomass was reduced by 97% when imazapyr was applied at 0.07 kg ai/ha (Figure 4 20 ) The highest rate, 4.48 kg ai/ha, resulted in 0% biomass compared to the control. The I 50 for the second trial was 0.017 kg ai/ha, and the I 90 was 0.05 kg ai/ha. All I 50 and I 90 values are well below the maximum labeled use rate for imazapyr, which is higher than 1 kg ai/ha. No plants in the imazapyr study were killed, but plants were seve rely stunted. Except at the lowest 2 rates, all regrowth was similar to that observed in plants treated with i mazamox; leaves were small and twisted and broke free from the plant easily. Imazapic provided good control of natalgrass in both trials. In the first trial, biomass at 0.0875 kg ai/ha was reduced by 57% (Figure 4 21 ) At 0.175 kg ai/ha, biomass was less than 1% of the untreated control. All higher rates had a biomass of 0% of the control. The I 50 was 0.06 kg ai/ha and the I 90 was 0.19 kg ai/ha In the second trial, biomass was reduced by 90% at 0.0875 kg ai/ha (Figure 4 22 ) Biomass

PAGE 55

55 for all other rates was no more than 2% of the untreated control, although none were 0%. The I 50 value was 0.03 kg ai/ha and the I 90 was 0.08 kg ai/ha. The val ues from both trials are within the labeled use rates for imazapic in natural areas. None of the plants in either trial were killed, but plants at all but the lowest rates were severely stunted. Like imazamox and imazapyr, imazapic caused abnormal regrow th in plants at all but the lowest rates. These plants were not killed, but did not appear likely to recover from the treatments. The plants also appeared more susceptible to attack from fungal pathogens In both trials, glyphosate provided good control of natalgrass. In the first trial, biomass was reduced more at lower rates (Figure 4 23 ) than in the second trial (Figure 4 24 ) At 0.0525 kg ai/ha, biomass was reduced by 30%. B iomass was reduced by 95% at 0.42 kg ai/ha. All higher rates were reduced to 0% of the untreated controls, and these plants were all killed. The I 50 value for the first trial was 0.17 kg ai/ha and the I 90 was 0.56 kg ai/ha, both well below the maximum labeled rate. In the second trial, biomass increased at the lowest rate, bu t was reduced by 90% at 0.42 kg ai/ha. At all higher rates, biomass was 0% of the untreated control. The I 50 value was 0.19 kg ai/ha and the I 90 value was 0.52 kg ai/ha, very similar to the rates generated from regression in the first trial. These rates are all well within the maximum labeled rate, suggesting that glyphosate provides excellent control of natalgrass. Field Preemergence Study Data from the Tenoroc Fish Management Area and the Lake Louisa Mitigation Bank were analyzed separately. Overall, treatments at Tenoroc (Table 4 4 ) showed more control of natalgrass than treatments at Lake Louisa (Table 4 5 ). However,

PAGE 56

56 results at Tenoroc were also more variable and, as a result, treatments were not as different statistically. At both sites, a reductio n in natalgrass was observed in the untreated controls over the course of the study At Tenoroc, cover in the untreated plots declined from 100% 2 MAT to 72% 7 MAT. At Lake Louisa, cover decreased from 100% 1 MAT to 72% 6 MAT. Natalgrass cover also appe ared to be declining in the area surrounding the study plots. Metsulfuron provided very little control of natalgrass at either research site. The only exception was the 0.126 kg ai/ha treatment 1 MAT at the Lake Louisa site, where 40% control was observ ed. In many cases, natalgrass appeared to be more prevalent in plots treated with metsulfuron than in the untreated controls. In the greenhouse studies, metsulfuron application increased natalgrass biomass at some rates. In the field, metsulfuron may ha ve had the same effect. In addition, metsulfuron may have controlled other species that were competing with natalgrass, allowing natalgrass to grow more prolifically in these plots. Metsulfuron appears to be a poor option for natalgrass control. At Tenor oc, imazapyr at 0.28 kg ai/ha provided 62% control 2 MAT while at 0.14 kg ai/ha 48% control was observed. All other imazapyr treatments were not statistically different from the control, with one exception Imazapyr at 0.28 kg ai/ha 5 MAT provided 40% contr ol. At the Lake Louisa site, the only statistical difference between imazapyr and the control was observed at 0.28 kg ai/ha 1 MAT, where imazapyr provided 24% control.

PAGE 57

57 Imazapyr did not result in a large decrease in natalgrass cover in the field when appl ied preemergence However, imazapyr did severely stunt natalgrass plants within the plots, delaying seed production and allowing many natives to more effectively compete with natalgrass. In addition, the rates tested in the field were fairly low. Based on results from the greenhouse studies, higher rates of imazapyr may provide acceptable levels of control in the field. Imazapyr may be a viable option for land treat cogongrass ( Imperata cylindrica ), a common invader in these areas ( MacDonald 2004 ). At Tenoroc, imazamox at 0.28 kg ai/ha provided 88% control, one of the better treatments, while 61% control was observed at 0.14 kg ai/ha 2 MAT. Three MAT, 60% control w as observed at 0.28 kg ai/ha while no statistical difference was observed between the lower rate and the untreated control. Five MAT, control at 0.28 kg ai/ha was 66% while control at 0.14 kg ai/ha was 50%. Seven MAT, no statistical difference was observ ed between the imazamox plots and the control plots. The higher rate of imazamox ranked among the best treatments for natalgrass control throughout the course of the study, while the lower rate initially was among the middle treatments. At Lake Louisa, co ntrol for both rates of imazamox 1 MAT was 43%. There was no statistical difference between imazamox plots and the untreated control for the latter 3 ratings. At this site, imazamox initially provided some measure of control, but this herbicide did not p rovide much control later in the study. Imazamox showed inconsistent results in the field when applied as a preemergence herbicide. However, plants present in the imazamox plots did appear to

PAGE 58

58 be stunted, an observation that is consistent with the effects of imazamox on natalgrass in the greenhouse. An increase in rate to the maximum labeled rate of 0.56 kg ai/ha would probably result in better control, based on the results of the greenhouse studies. Imazamox does appear to reduce natalgrass populations a nd perhaps delay seed production in plants that survive the application, although natalgrass will probably continue to require management in a reas where imazamox is applied. At Tenoroc, hexazinone at 0.56 kg ai/ha provided 98% control of natalgrass 2 MAT, one of the best treatments. Control declined at 5 MAT to 60% for 0.56 kg ai/ha and 50% for 0.28 kg ai/ha. Seven MAT, control was not statistically different from the control. At 3 and 5 MAT, values for hexazinone at Tenoroc were not statistically differ ent from many of the other treatments; however, these treatments were among the more successful. At Lake Louisa, hexazinone provided 70% control at 0.56 kg ai/ha and 75% control at 0.28 kg ai/ha 1 MAT, among the most successful treatments. Control decli ned to 43% for 0.56 kg ai/ha and 45% for 0.28 kg ai/ha 2 MAT. For the final 2 ratings at this site, hexazinone did not provide a measure of control that was statistically different from the control. Hexazinone initially offered some of the most natalgras s control of the treatments at Lake Louisa, but by the 6 mo rating, hexazinone ranked toward the bottom of the treatments. At both sites, hexazinone offered at least some measure of natalgrass control. Hexazinone did not appear to be the most effective he rbicide for natalgrass control of the compounds tested, but depending upon the desired species composition at a particular site this compound may be a good choice for land managers. A higher rate

PAGE 59

59 would most likely prove more effective at controlling natal grass, but could also be detrimental to native plant populations. At Tenoroc, imazapic provided 100% control at 0.28 kg ai/ha and 95% control at 0.14 kg ai/ha 2 MAT. Three MAT, control was 94% for the higher rate and 88% for the lower rate. Control decli ned slightly 5 MAT to 89% for the higher rate and 84% for the lower. Finally, at 7 MAT, control was 80% for the higher rate and 74% for the lower. At this site, imazapic ranked among the best treatments for natalgrass control throughout the course of the study. At Lake Louisa, imazapic provided 76% control at 0.26 kg ai/ha and 30% control at 0.14 kg ai/ha 1 MAT. Control was 35% for the higher rate 2 MAT, but no statistical difference was observed between the plots for the lower rate of imazapic and the c ontrol. No difference was observed between any of the imazapic plots and the control at 4 and 6 MAT at this site. Although imazapic provided somewhat inconsistent results it did severely stunt natalgrass plants present in the plots. Natalgrass plants in imazapic plots were not flowering. In addition, many native species in Florida are tolerant to imazapic (Kluson et al. 2000, Richardson et al. 2003). However, the maximum labeled rate for imazapic in natural areas is currently 0.21 kg ai/ha, lower tha n the higher rate tested in this study. Imazapic did not provide the most effective control of natalgrass, but it may be one of the best choices for land managers attempting to foster native plant communities. At Tenoroc, pendimethalin provided 69% contro l at 2.24 kg ai/ha and 45% control at 1.12 kg ai/ha 2 MAT. Three MAT, 45% control was observed at the 2.24 kg ai/ha rate. No other pendimethalin treatments were statistically significant at this site.

PAGE 60

60 At Lake Louisa, pendimethalin provided some of the be st results observed in the study. One MAT, 91% control was observed at 2.24 kg ai/ha and 84% control at 1.12 kg ai/ha. At the higher rate, 91% control was also observed 2 MAT, while control declined to 69% at the lower rate. Four MAT, control was 84% at the higher rate and 86% at the lower. Finally, 6 MAT, control was 81% for the higher rate and 83% for the lower. Pendimethalin provided somewhat inconsistent results, but did provide good control of natalgrass at the Lake Louisa site. The rates tested w ere also lower than the maximum labeled rate; an increased rate may provide better control. This herbicide may be a good option for land managers attempting to control natalgrass. At Tenoroc, sulfometuron provided 95% control at 0.263 kg ai/ha and 99% con trol at 0.158 kg ai/ha 2 MAT. Control declined to 81% at the higher rate and 78% at the lower rate 3 MAT. Five MAT, control was 80% at the higher rate and 94% at the lower. Seven MAT, control for the higher rate was not statistically different from the control, but 76% control was observed at the lower rate. At Lake Louisa, both 0.263 kg ai/ha and 0.158 kg ai/ha provided 91% control 1 MAT. Likewise, both rates provided 100% control 2 MAT. Four MAT, control was 89% for the higher rate and 91% for the lo wer. At 6 MAT, both rates provided 85% control. Sulfometuron did appear to provide good control of natalgrass and may be a good option for land managers. However, the treatments appeared to be detrimental to many of the other plant species present in the study area, something that should be

PAGE 61

61 considered before using this herbicide for natalgrass control. In areas where a seed bank exists for native plant species, another herbicide might be a better choice. Field Postemergence Study Data from Tenoroc and La ke Louisa were analyzed separately. At the Tenoroc site, there was no significant difference among any treatments at 12 or 24 MAT. Twelve MAT, natalgrass cover in all plots at Tenoroc averaged 17%, while cover declined to an average of 1% 24 MAT. At bot h sites, n atalgrass cover in the untreated controls declined over time At Tenoroc, natalgrass cover in the control plots declined from 100% 3 MAT to 70% 9 MAT. Twelve MAT, cover in the control plots was 19%, and 24 MAT was less than 1%. At Lake Louisa, cover declined from 85% at 3 MAT to 84% 6 MAT and finally 60% 12 MAT. An overall decline in natalgrass cover was also observed in the areas surrounding the research plots at both sites. Plots treated with metsulfuron did not have natalgrass cover that wa s statistically different from the untreated control plots at either research location. These results are similar to the results observed in other studies. Metsulfuron appears to have little impact on natalgrass in the field and should not be used by land managers to control natalgrass. Fluazifop did not provide any signi ficant measure of control at either research location. These resu lts were also similar to the results observed in other studies. Based on th is information fluazifop does not appear to be a viable option for land managers seeking to control natalgrass. At Tenoroc, glyphosate application at 3.36 kg ai/ha resulted in 3% natalgrass c over in plots 3 MAT, while glyphosate at 1.12 kg ai/ha resulted in 20% cover. Nine MAT, the higher rate resulted in 25% cover while the lower rate resulted in 29% cover.

PAGE 62

62 At Lake Louisa, cover in plots treated with glyphosate was not statistically differe nt from the control plots at any of the rating times. In glyphosate plots at both sites 3 MAT, dead plants were observed among younger plants that were already flowering. Results from the greenhouse studies indicate that glyphosate has very good activity on natalgrass; however, this compound does not provide residual activity. The application appeared to successfully control plants present in the plot at the time of treatment, but natalgrass grew back quickly from seed. In addition, other plants that we re competing with natalgrass were also controlled by the glyphosate, and did not appear as prevalent in the month s after treatment. Based on these results, glyphosate could be a good option if immediate control is required, but land managers should expect natalgrass populations to rebound quickly and perhaps be more dense than before because of the elimination of competing native plants At Tenoroc, imazapyr applied at 0.28 kg ai/ha resulted in 53% natalgrass cover, while 0.14 resulted in 56% cover 3 MAT. Plots treated with the higher rate had 26% cover and plots treated with the lower rate had 16% cover 9 MAT. At Lake Louisa, imazapyr resulted in 34% cover at 0.28 kg ai/ha and 44% cover at 0.14 kg ai/ha 3 MAT. Six MAT, natalgrass cover was not statistica lly different from cover in the untreated control plots. Twelve MAT, cover in plots treated with 0.28 kg ai/ha was 18%. Cover in plots treated with 0.14 kg ai/ha was not statistically different from the untreated control 12 MAT. Imazapyr caused apparent stunting of natalgrass and a reduction in flowering, an observation suppported by results from the greenhouse studies Imazapyr did seem to

PAGE 63

63 pr ovide some native plant species a competitive advantage by reducing natalgrass growth and seed production, but ma ny native species are injured by this product A higher rate of imazapyr might control natalgrass more effectively, but likewise would further injure native plant populations. At Tenoroc, plots treated with hexazinone 1.12 kg ai/ha had 0% natalgrass cover 3 MAT, while plots treated with 0.56 kg ai/ha had 3% cover. Nine MAT, the 2 treatments again showed similar results P lots treated with 1.12 kg ai/ha had 14% cover, while plots treated with 0.56 kg ai/ha had 13% cover. At Lake Louisa, hexazinone at 1.12 kg ai/ha resulted in 4% cover in plots, while 0.56 kg ai/ha resulted in 9% cover 3 MAT. These treatments reduced natalgrass cover by the greatest amount 3 MAT. 6 MAT, hexazinone also reduced natalgrass cover by the greatest amount, resulting in 8% cover at the higher rate and 11% cover at the lower. 12 MAT, these plots still had some of the least natalgrass cover in the study area. Plots treated with 1.12 kg ai/ha had 14% cover, while plots treated with 0.56 kg ai/ha had 20% cover. Hexazinone appeared t o offer some of the longest lasting control of natalgrass under field conditions. However, plots treated with hexazinone had very few other species present 3 MAT, and species composition was not very diverse 12 MAT. Although hexazinone offers good contro l of natalgrass, it may not be the best option if a land manager is attempting to restore native plants. At Tenoroc, imazapic at 0.56 kg ai/ha resulted in 30% natalgrass cover 3 MAT, while 0.28 kg ai/ha resulted in 56% cover. Nine MAT, plots treated with the higher rate had 26% cover, while plots treated with the lower rate had 14% cover. At Lake Louisa,

PAGE 64

64 results were less consistent. Three MAT, plots treated with 0.56 kg ai/ha had 45% cover, while plots treated with 0.28 kg ai/ha had 50% cover. Six and 12 MAT, cover in plots treated with imazapic was not statistically different from cover in the untreated control plots. Although imazapic produced inconsistent results, visual observations indicate that imazapic may be a very good option for natalgrass co ntrol. The amount of natalgrass cover was not significantly different beyond 3 MAT at the Lake Louisa site, but natalgrass in all plots treated with imazapic at both sites appeared to be severely stunted. This observation is supported by results from the greenhouse studies, which show that imazapic does severely stunt natalgrass. Natalgrass was not flowering 3 MAT in imazapic plots, while all other mature natalgrass plants in the areas surrounding the study area were flowering. These plots also appeared to have some of the highest concentrations of native plants. Previous research has shown that many of Aristida beyrichiana Andropogon spp., Eragrostis spp., Liatris spp., Pityopsis gr aminifolia and Solidago stricta (Kluson et al. 2000, Richardson et al. 2003). Andropogon spp. and Eragrostis spp. were present in imazapic plots 3 MAT and were flowering. Although imazapic does not appear t o kill larger natalgrass plants it does cause st unting, allowing native plants to gain a competitive advantage. As a result, natalgrass may be a very good option for land managers trying to control natalgrass and promote native plant communities. Conclusions In all field trials, a decrease in natalgra ss cover over time was observed in the untreated control plots. In addition, natalgrass cover appeared to be declining throughout surrounding areas. This suggests that natalgrass may decline naturally as

PAGE 65

65 plant communities grow more complex. Natalgrass d id not persist as a monoculture at any of the study sites for more than 3 years after disturbance. However, many land managers cannot wait several years for natalgrass to decline, but must actively manage sites. In this case, one of several herbicides mi ght be useful. Metsulfuron and fluazifop do not appear to provide acceptable levels of natalgrass control in the field Likewise, these compounds provided either no control in the greenhouse or control only at high rates. Neither of these compounds would be useful to land man agers attempting to control natalgrass. Glyphosate provides excellent control of natalgrass, but does not provide residual activity. As a result, natalgrass populations are quickly reestablished from seed, easily outcompeting native plants. Glyphosate pr ovides immediate control, but other methods should be employed long term. Glyphosate may be best utilized as a spot treatment. Pendimethalin and metolachlor both offer good control of natalgrass when applied preemergence. However, these herbicides were d etrimental to many of the native species present at the research sites. These herbicides would be best utilized at a site with little to no seed bank for native species. Hexazinone and sulfometuron both provided fairly good control of natalgrass pre and postemergence. However, both compounds were also detrimental to native plants. These compounds would also be good options at sites with few native species present. Imazamox, imazapyr and imazapic provided less control of natalgrass on average than hexazi none and sulfometuron, but were less harmful to native plant populations. Of the three herbicides, imazapic is the best choice when many native species are

PAGE 66

66 present. Many native plants in Florida are tolerant to this compound and gain a competitive advant age when imazapic is used.

PAGE 67

67 Table 4 1 I 50 and I 90 values for various herbicides applied preemergence to natalgrass in the greenhouse. Herbicide 1x Rate Treatment (kg ai/ha) I 50 1 I 90 2 Hexazinone 0.56 0.315 3 Imazamox 0.56 0.105 1.04 4 Imazapic 1.40 0.219 0.92 Imazapyr 1.12 0.21 Metolachlor 1.12 0.11 1.65 Metsulfuron 1.12 2.10 Pendimethalin 1.12 0.48 1.07 Sulfometuron 0.22 0.02 0.90 1 I 50 = herbicide rate required to reduce biomass by 50% compared to the untreated control. 2 I 90 = herbicide rate required to reduce biomass by 90% compared to the untreated control. 3 Missing values could not be calculated with the regression equations generated during statistical analysis. 4 Values marked with indicate that the calculated rate is higher than the maximum labeled use rate for the compound. Table 4 2 I 50 and I 90 values for various herbicides applied postemergence to natalgrass in the greenhouse (Trial 1). Herbicide 1x Rate Treatment (kg ai/ha) I 50 1 I 90 2 Fluazifop 1.12 0.14 0.38 Glyphosate 0.84 0.17 0.56 Hexazinone 0.56 0.09 0.35 Imazamox 0.56 0.04 0.12 Imazapic 1.40 0.06 0.19 Imazapyr 1.12 0.02 0.07 Metsulfuron 1.12 3 Sulfometuron 0.22 0.068 0.30 1 I 50 = herbicide rate required to reduce biomass by 50% compared to the untreated control. 2 I 90 = herbicide rate required to reduce biomass by 90% compared to the untreated control. 3 Missing values could not be calculated with the regression equations generated during statistical analysis.

PAGE 68

68 Table 4 3. I 50 a nd I 90 values for various herbicides applied postemergence to natalgrass in the greenhouse (Trial 2). Herbicide 1x Rate Treatment (kg ai/ha) I 50 1 I 90 2 Fluazifop 1.12 0.28 0.73 3 Glyphosate 0.84 0.19 0.52 Hexazinone 0.56 0.07 0.74 Imazamox 0.56 0.02 0.63 Imazapic 1.40 0.03 0.08 Imazapyr 1.12 0.017 0.05 Metsulfuron 1.12 4 Sulfometuron 0.22 1 I 50 = herbicide rate required to reduce biomass by 50% compared to the untreated control. 2 I 90 = herbicide rate required to reduce biomass by 90% c ompared to the untreated control. 3 Values marked with indicate that the calculated rate is higher than the maximum labeled use rate for the compound. 4 Missing values could not be calculated with the regression equations generated during statistical a nalysis. Table 4 4 Influence of herbicide treatments applied preemergence on natalgrass co ver at the Tenoroc Fish Management Area in 2009. Herbicide R ate ------------------------% Co ver 1 ------------------------Treatment (kg ai/ha) 2 MAT 2 3 M AT 5 MAT 7 MAT Untreated 10 0 f 3 10 0g 97 f 72 b d H exazinone 0.28 1 0abc 37 a e 50b d 32 ab 0.56 2 a 28 a d 4 0 a c 47 a c I mazamox 0.14 39 de 62 c g 50b d 57 a c 0.28 12 a d 40 a e 34 a c 52 a c I mazapic 0.14 5 ab 12 a 16 ab 26 a 0.28 0a 6 a 11 a 2 0 a I mazapyr 0.14 52 e 72 e g 67 c f 6 0 a d 0.28 38 c e 65 d g 6 0 c e 6 0a d M etsulfuron 0.042 10 0f 92 fg 97 f 8 5cd 0.126 92 f 10 0 g 9 5ef 10 0 d P endimethalin 1.12 55 e 82 fg 8 5 d f 77 b d 2.24 31 b e 5 5b f 67 c f 5 5a c S ulfometuron 0.158 1 a 22 a c 6 a 24 a 0.263 5ab 19 ab 2 0ab 42 ab 1 Vi sual assessment of natalgrass co ver based on the following scale: 0 = no co ver ; 100 = complete co ver 2 MAT = months after treatment. 3 Values reflect the mean of 4 replications. Means within a column followed by different letters are significantly diff (LSD) test.

PAGE 69

69 Table 4 5 Influence of herbicide treatments applied preemergence on natalgrass co ver at the Lake Louisa Mitigation Bank in 2009. Herbicide R ate -----------------------% Co ver 1 ------------------------Treatment (kg ai/ha) 1 MAT 2 2 MAT 4 MAT 6 MAT Untreated 10 0d 3 10 0f 9 0b 72 b H exazinone 0.28 2 5a 55 cd 7 5b 72 b 0.56 3 0a 57 cd 9 0b 8 5b I mazamox 0.14 57 b 97 f 97 b 97 b 0.28 57 b 9 5f 7 5b 77 b I mazapic 0.14 70 bc 82 d f 76 b 7 5b 0.28 24 a 6 5e 71 b 72 b I mazapyr 0.14 82 cd 9 5f 10 0b 10 0b 0.28 76 bc 9 0ef 97 b 92 b M etsulfuron 0.042 81 cd 10 0f 10 0b 10 0b 0.126 60 bc 10 0f 9 0b 9 5b P endimethalin 1.12 16 a 31 bc 14 a 17 a 2.24 9 a 9ab 16 a 19 a S ulfometuron 0.158 9a 0a 9 a 1 5 a 0.263 9a 0a 11 a 15 a 1 Visual assessment of natalgrass co ver based on the following scale: 0 = no co ver ; 100 = complete co ver 2 MAT = months after treatment. 3 Values reflect the mean of 4 replications. Means within a column followed by different letters are significantly (LSD) test.

PAGE 70

70 Table 4 6 Influence of herbicide treatments applied postemergence on natalgrass co ver at the Tenoroc Fis h Management Area in 2008. Herbicide R ate ------------------% Co ver 1 ------------------Treatment (kg ai/ha) 3 MAT 2 9 MAT Untreated 100a 3 70a Fluazifop 0.28 98a 43a c Glyphosate 1.12 20cd 29b d 3.36 3cd 25b d Hexazinone 0.56 3cd 13d 1.12 0d 14cd Imazapic 0.28 56b 14cd 0.56 30bc 26b d Imazapyr 0.14 56b 16cd 0.28 53b 26b d Metsulfuron 0.28 95a 50ab 1 Visual assessment of natalgrass co ver based on the following scale: 0 = no cover; 100 = complete co ver 2 MAT = months after treatment. 3 Values reflect the mean of 4 replications. Means within a column followed by different letters are significantly (LSD) test. Table 4 7 Influence of herbicide treatments applied postemergence on natalgrass co ver at the Lake Louisa Mitigation Bank in 2009. Herbicide R ate ----------------------% Co ver 1 ----------------------Treatment (kg ai/ha) 3 MAT 2 6 MAT 12 MAT Untreated 85ab 3 84ab 60ab Fluazifop 0.2 8 63bc 71ab 55ab Glyphosate 1.12 91a 90a 71a 3.36 61bc 85a 71a Hexazinone 0.56 9ef 11c 20c 1.12 4f 8c 14c Imazapic 0.28 50cd 75ab 39a c 0.56 45cd 80ab 30bc Imazapyr 0.14 44cd 69ab 41a c 0.28 34de 53b 18c Metsulfuron 0.11 80ab 78ab 55ab 1 Visu al assessment of natalgrass co ver based on the following scale: 0 = no co ver ; 100 = complete co ver 2 MAT = months after treatment. 3 Values reflect the mean of 4 replications. Means within a column followed by different letters are significantly differ (LSD) test.

PAGE 71

71 Table 4 8. Disturbance to plots caused by wild hogs at the Lake Louisa Mitigation Bank in 2009. Herbicide R ate ---------------% Disturbance 1 --------------Treatment (kg ai/ha) 6 MAT 2 12 MAT Untreated 10c 3 0b Fluazifop 0.28 14bc 0b Glyphosate 1.12 10c 0b 3.36 14bc 14a Hexazinone 0.56 0c 0b 1.12 4c 0b Imazapic 0.28 5c 0b 0.56 0c 0b Imazapyr 0.14 23bc 0b 0.28 48ab 3b Metsulfuron 0.11 58a 4b 1 Visual assessment of disturbance to plots based on the following scale: 0 = no disturbance; 100 = complete disturbance. 2 MAT = months after treatment. 3 Values reflect the mean of 4 replications. Means within a column followed by different letter s Significant Difference (LSD) test. Table 4 9. Presence of species within treatment areas at t he Tenoroc Fish Management Area Species 0 MAT 1 3 MAT 6 MAT 9 MAT 12 MAT 24 MAT Aeschynomene spp. 2 X 3 Andropogon spp. X X X X X X Conyza canadensis X Eragrostis spp. X X X X X X Eustachys petraea X X X X X Froelichia floridana X Heterotheca subaxillaris X X X X X X Indigofera hirsuta X X Melinis repens X X X X X X Panicum repens X X X X X Panicum virgatum X X X Paspalum notatum X X X X X X Passiflora incarnata X X Scoparia dulcis X Sporobolus indicus X X X X 1 MAT = months after treatment. 2 Indicates absence of species. 3 Indicates presence of species.

PAGE 72

72 Table 4 10 Presence o f species within treatment areas at the Lake Louisa Mitigation Bank. Species 0 MAT 1 3 MAT 6 MAT 9 MAT 12 MAT Ambrosia artemisiifolia X 2 X X X X Aristida spp. X X X X Conyza canadensis 3 X X X Cyperus rotundus X X X X X Eragrostis spp. X X X X Eupatorium capillifolium X X X X Froelichia floridana X X Heterotheca subaxillaris X X X X X Indigofera hirsuta X Lantana spp. X X X Melinis repens X X X X X Nuttallanthus canadensis X Paspalum notatum X X X X Richardi a scabra X X X X Rumex hastatulum X Urochloa maxima X X X X 1 MAT = months after treatment. 2 Indicates presence of species. 3 Indicates absence of species.

PAGE 73

73 Figure 4 1 The effect of metsulfuron applied preemergence on natal grass biomass (expressed as a percentage of the dry weight of the untreated control) Values represent the mean of 8 replications with standard error y = 33.38 + 66.53 e 0.66 x R 2 = 0. 72

PAGE 74

74 Figure 4 2 The effect of hexazinone applied preemergence on n atalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Values represent the mean of 8 replications with standard error y = 26.13 + 17. 20e 59.35x + 57.33e 2.77x R 2 = 0.52

PAGE 75

75 Figure 4 3 The effect of imazapyr applied preemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Values represent the mean of 8 replications with standard error y = 17.64 + 72.30 e 3.81 x R 2 = 0. 65

PAGE 76

76 Figure 4 4 The effect of imazapic applied preemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Values represent the mean of 8 replications with standard error y = 4.76 + 89.43 e 3.10 x R 2 = 0. 91

PAGE 77

77 Figure 4 5 The effect of imazamox applied preemergenc e on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Values represent the mean of 8 replications with standard error y = 5.52 + 44.81e 26.31x + 53.34e 2.38x R 2 = 0.85

PAGE 78

78 Figure 4 6 The effect of sulfometuron applied pree mergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Values represent the mean of 8 replications with standard error y = 9.00 e (0.13/(x + 0.05)) R 2 = 0. 88

PAGE 79

79 Figure 4 7 The effect of metolachlor applie d preemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Values represent the mean of 8 replications with standard error y = 81.58 e 8.74 x + 19.80 e 0.41 x R 2 = 0.99

PAGE 80

80 Figure 4 8 The effect of pendimethali n applied preemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Values represent the mean of 8 replications with standard error y = 7.40 + 150.0 8 e 2.00 x R 2 = 0.8 0

PAGE 81

81 Figure 4 9 The effect of metsu lfuron applied postemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 1. Values represent the mean of 4 replications with standard error y = 70.94 + 44.57 e 1.39 x R 2 = 0. 54

PAGE 82

82 Figure 4 10 T he effect of metsulfuron applied postemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 2. Values represent the mean of 4 replications with standard error y = 98.74 + 50.59 e 18.62 x R 2 = 0. 74

PAGE 83

83 Figure 4 11 The effect of sulfometuron applied postemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 1. Values represent the mean of 4 replications with standard error y = 80.30 e 6.91 x R 2 = 0. 83

PAGE 84

84 Figure 4 12 The effect of sulfometuron applied postemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 2. Values represent the mean of 4 replications with standard error y = 64.84 + 61.07 e 6.30 x R 2 = 0. 58

PAGE 85

85 Figure 4 13 The effect of hexazinone applied postemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 1. Values represent the mean of 4 replications with standard error y = 0.61 + 91.59 e 6.39 x R 2 = 0. 97

PAGE 86

86 Figure 4 14 The effect of hexazinone applied postemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 2. Values represent the mean of 4 r eplications with standard error y = 57.14 e 22.27 x + 43.17 e 1.98 x R 2 = 0. 94

PAGE 87

87 Figure 4 15 The effect of fluazifop applied postemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 1. Values represent the mean of 4 replications with standard error y = 129.67 e 6.68 x R 2 = 0. 62

PAGE 88

88 Figure 4 1 6 The effect of fluazifop applied postemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 2. Values represent the mean of 4 replications with standard error y = 1.87 + 133.53 e 3.31 x R 2 = 0.8 0

PAGE 89

89 Figure 4 17 The effect of imazamox applied postemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 1. Values represent the mean of 4 replications with standard error y = 108.01 e 19.78 x R 2 = 0. 86

PAGE 90

90 Figure 4 18 The effect of imazamox applied postemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untr eated control) Trial 2. Values represent the mean of 4 replications with standard error y = 87.47 e 40.53 x + 13.49 e 0.47 x R 2 = 0. 94

PAGE 91

91 Figure 4 19 The effect of imazapyr applied postemergence on natalgrass biomass (expressed as a percentage of the dry weigh t of the untreated control) Trial 1. Values represent the mean of 4 replications with standard error y = 100.00 e 32.53 x R 2 = 0. 99

PAGE 92

92 Figure 4 20 The effect of imazapyr applied postemergence on natalgrass biomass (expressed as a percentage o f the dry weight of the untreated control) Trial 2. Values represent the mean of 4 replications with standard error y = 99.96 e 44.01 x R 2 = 0. 96

PAGE 93

93 Figure 4 21 The effect of imazapic applied postemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 1. Values represent the mean of 4 replications with standard error y = 101.40 e 12.16 x R 2 = 0. 98

PAGE 94

94 Figure 4 22 The effect of imazapic applied postemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 2. Values represent the mean of 4 replications with standard error y = 99.98 e 25.56 x R 2 = 0. 99

PAGE 95

95 Figure 4 23 The effect of glyphosate applied postemergence on na talgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 1. Values represent the mean of 4 replications with standard error y = 100.55 e 4.10 x R 2 = 0. 96

PAGE 96

96 Figure 4 24 The effect of glyphosate applied po stemergence on natalgrass biomass (expressed as a percentage of the dry weight of the untreated control) Trial 2. Values represent the mean of 4 replications with standard error y = 2.93 + 120.50 e 4.30 x R 2 = 0. 93

PAGE 97

97 CHAPTER 5 CONCLUSION S Natalgrass seeds do not require light for g ermination. Some germination occurs at 15 C, but most germination occurs at temperatures of 20 C and greater. Natalgrass germinates at pH levels of 6 and 8. Germination was highly dependent on available moisture, only occurring at osmotic potentials of 0.2 MPa and greater. Natalgrass seedlings are capable of emerging from depths of at least 5 cm. Natalgrass also appears to require an afterripening period after seed shed to reach maximum germination potential. Land managers should expect most natalgr ass germination to occur as temperatures in the field reach 20 C and rainfall becomes regular. A preemergence herbicide application prior to this time may result in successful reduction of natalgrass at the site. When buried, natalgrass seed appears to de velop increased dormancy. Land managers can successfully utilize tillage to control natalgrass plants, but should be aware that this act will most likely result in seed burial and, therefore, seed dormancy. If tillage is utilized a second time and seeds are returned to the soil surface, managers should be prepared for germination of these seeds to occur. Natalgrass forms dense seed deposits in infested areas, but these seed deposits appear to become quickly exhausted when conditions are favorable for germ ination. If land managers can control germinating seedlings for several months with a preemergence herbicide while preventing further seed rain at the site, a reduction in natalgrass will likely be achieved.

PAGE 98

98 During monitoring of study sites, an overall d ecrease in natalgrass was observed in untreated plots as well as surrounding areas. This may indicate that natalgrass populations decline naturally as plant communities gain more structure. Natalgrass appears able to quickly invade disturbed areas, but d id not persist as a monoculture for more than several years after disturbance at any of the research sites. Metsulfuron and fluazifop offer little to no control of natalgrass at labeled use rates. The rates required to significantly reduce natalgrass populations are above the maximum labeled us e rates. These herbicides offer little utility to land managers a ttempting to control natalgrass. Glyphosate provides excellent control of natalgrass at fairly low rates. However, glyphosate does not have residual activity. As a result, natalgrass populations quickly became reestablished from seed. Glyphosate may off er good control short term, but may cause an overall increase in natalgrass density long term by eliminating competing plants. Pendimethalin and metolachlor both offer good control of natalgrass when applied preemergence. However, both compounds were detr imental to native plant populations. These herbicides would be best utilized at a site with little to no seed bank present for native species. Hexazinone and sulfometuron provided good control of natalgrass both pre and postemergence. These herbicides w ere also harmful to most native plants present at the study sites. As a result, these herbicides would also be best utilized when few native plant species are present.

PAGE 99

99 Imazamox, imazapyr and imazapic provided less control of natalgrass on average than hex azinone and sulfometuron, but were less harmful to native plant populations. These compounds did not result in a large decrease in natalgrass cover, but did result in severe stunting of natalgrass and delay of flowering and seed set. Of the three herbici des, imazapic appears to be t he best choice when many native species are present. Many native plants in Florida are tolerant to this compound and gain a competitive advantage when imazapic is used.

PAGE 100

100 APPENDIX AFTERRIPENING STUDY Introduction Afterripening is the loss of a dormant state over a period of time through exposure of the seeds to a set of environmental conditions after maturation and separation from the parent plant (Simpson 1990). Common in many grass species such as wild oat ( Avena fatua L.) an d red rice ( Oryza punctata L.) ( Quail and Carter 1969 ; Leopold et al. 1988; Foley 1994) afterripening is influenced by environmental conditions such as temperature and moisture status (Foley 2001). The results of the ex periment discussed in Chapter 2 exa mining natalgrass germination over time suggest that natalgrass likely undergoes an afterripening period after seeds are shed from the parent plant. Based on this conclusion, an experiment was designed to investigate the effects of seed moisture status ov er time on the afterripening process of natalgrass seeds Materials and Methods Natalgrass seeds were collected from the Lake Louisa Mitigation Bank in Lake County, FL in December 2009. Seeds were collected with a sweep net directly from mature seedheads Seeds were left in the husk to mimic natural conditions as closely as possible. The same day, seeds were sterilized by immersion in a 1% bleach solution with a non ionic surfactant (0.25% v/v) to ensure that the bleach solution came in contact with the entire surface. Seeds were air dried under a sterile hood. Seeds were placed into 6 sealed containers 21 each containing a different saturated salt solution. Salt solutions were placed in open Petri dishes in the bottom of the 21 Thermo Scientific Nalgene Autoclavable Plastic Dessicators, Nalge Nunc International, 75 Panorama Creek Drive, Rochester, NY 14625

PAGE 101

101 containers; seeds were place d in open Petri dishes on a metal rack in the upper portion of the chamber. Salts used and the corresponding relative humidity levels at 25 C can be found in Table A 1 (Winston and Bates 1960; Rockland 1960). The containers were placed in a growth chambe r under constant light and a constant temperature of 25 C to maintain the desired relative humidity levels. At 3, 6, 9, 12, 15 and 18 weeks after the beginning of the experiment, each container was opened and approximately 150 seeds quickly removed. Forty seeds from each container were weighed, dried at 50 C for 3 days, and weighed a second time. This data was used to determine seed moisture content at the time of removal from the containers. The remaining seeds were tested to determine germination level s. Germination tests were performed by placing thirty seeds evenly in a 9 cm Petri dish 22 containing 1 piece of filter paper 23 The filter paper was moistened with 4 mL of deionized water (pH = 6). Each Petri dish was sealed with parafilm and placed in a g rowth chamber at 30 1 C under constant light. Germination was visually determined after 14 d. Any ungerminated seeds were tested for viability using a 0.25% tetrazolium solution. Seeds were removed from the husk and seed coat and placed in the tetrazo lium solution for 24 h in the dark. Seeds were examined under a dissecting microscope and were counted as viable if the entire embryo was stained red or pink. Percentage of seed germination was calculated by dividing the number of germinated seeds by the number of total viable seeds in each Petri dish, then multiplying by 100. Each treatment was replicated 4 times. 22 Fisherbrand Petri dishes, Fisher Scientific 2000 Park Lane Drive, Pittsburgh, PA 15275 23 Fisherbrand P8 filter paper, Fisher Scientific 2000 Park Lane Drive, Pittsburgh, PA 15275.

PAGE 102

102 Results and Discussion No r eportable r esults were obtained from this experiment. At 3, 6 and 9 weeks after the beginning of the experiment seeds were removed, weighed and tested for germination. However, accurate weights were not obtained and it was not possible to determine the moisture content of the seeds. If this experiment was performed a second time, a larger quantity of seeds shoul d be weighed to ensure a more accurate measurement. Because natalgrass seeds are so light in weight, 40 seeds was not an adequate amount to measure the difference between dry and fresh weight (the fresh weight of 40 seeds was as low as 0.00426 g in this e xperiment). At such low weights, even small vibrations or tiny bits of debris can greatly influence measurements.

PAGE 103

103 Table A 1 Saturated salt solutions and the corresponding relative humidity levels at 25 C 1 Salt Approximate Relative Humidity (%) Ph osphorus pentoxide [P 2 O 5 ] 0 Lithium chloride [LiCl] 12 Magnesium chloride [MgCl 2 ] 33 Calcium nitrate [Ca(NO 3 ) 2 ] 51 Potassium chloride [KCl] 85 Potassium nitrate [KNO 3 ] 93 1 From Winston and Bates 1960; Rockland 1960

PAGE 104

104 LIST OF REFERENCES Adjei, M. B. and J. E. Rechcigl. 2004. Interactive effect of lime and nitrogen on bahiagrass pasture. Pages 52 56 in Proceedings of the Soil and Crop Science Society of Florida. Gainesville, FL: Soil and Crop Science Society of Florida. Baskin C. C. and J. M. B askin 1998. Seeds: Ecology, b iogeography and e volution of d ormancy and g ermination. San Diego, CA: Academic Press. 666 p. Benvenuti, S. and M. Macchia. 1995. Effect of hypoxia on buried weed seed germination. Weed Res. 35:343 351. Bradford, K. J. 1990. A water relations analysis of seed germination rates. Plant Physiol. 94:840 849. Connell, J. H. and R. O. Slatyer. 1977. Mechanisms of succession in natural communities and their role in community stability and organization. Amer. Natur. 111: 1119 1144. Evans, C. E. and J. R. Etherington. 1990. The effect of soil water potential on seed germination of some British plants. New Phytol. 115:539 548. [FAWN] Florida Automated Weather Network. 2010. Archived Weather Data. http://fawn.ifas.ufl.edu/ Accessed October 13, 2010. 2009 List of Invasive Plant Species. http://www.fleppc.org/list/List WW F09 final.pdf Accessed October 1, 2010. [FLMNH] Florida Museum of Natural History/University of Florida Herbarium. 2010. Collections Catalog. http: //www.flmnh.ufl.edu/herbarium/cat/ Accessed October 8, 2010. Foley, M. E. 1994. Temperature and water status of seed affect afterripening in wild oat ( Avena fatua ). Weed Sci. 42:200 204 Foley, M. E. 2001. Seed dormancy: an update on terminology physiological genetics, and quantitative trait loci regulating germinability. Weed Sci. 49:305 317. Hfliger, E. and H. Scholz. 1980. Rhynch elytrum repens (Willd.) Hubb Page 118 in Grass w eeds I. Basle, Switzerland: Ciba Geigy, Ltd. 142 p. Haselwood, E. L. and G. G. Motter. 1966. Natal redtop. Page 75 in Handbook o f Hawaiian w eeds. Honolulu,

PAGE 105

105 Hernndez Quiroz, N., C. Ortega Ochoa, C. Pinedo Alvarez, O. Viramontes Olivas, and J. A. Ortega Gutierrez. 2010. Fire effect over soil seed bank seedling emergency of n ata l grass ( Melinis repens (Willd.) Zizka) in Chihuahua grasslands. Proceedings of the Weed Science Society of America In press. Hitchcock, A. S. 1950. Rhynchelytrum Nees Pages 716 717 in A. Chase, ed. Manual of the g rasses of the United States 2 nd ed. Washington, DC: United States Department of Agriculture Miscellaneous Publica tion 200. 1051 p. Klages, K. H. W. 1947. Ecological c rop g eography. New York City, NY : The Macmillan Company. 615 p. Kleinschmidt, H. E. and R. W. Johnson. 1977. Red natal grass ( Rhynchelytrum repens ) Page 106 in Weeds of Queensland. Queensland, Australia: S.R. Hampson, Government Printer. 469 p Kluson, R. A., S. G. Richardson, D. B. Shibles and D. B. Corley. 2000. Response of two native and two non native grasses to imazapic herbicide on phosphate mined lands in Florida. Pages 49 57 in Proceedings of the Annual Meeting of the American Soci ety for Surface Mining and Reclamation. Lexington, KY: American Society for Surface Mining and Reclamation. Leopold, A. C., R. Glenister, and M. A. Cohn. 1988. Relationship between water content and afterripening in red rice. Physiol. Plant. 74:659 6 62. Lonsdale, W. M. 1993. Losses from the seed bank of Mimosa pigra : soil micro organisms vs. temperature fluctuations. J. Appl. Ecol. 30:654 660. MacDonald, G. E. 2004. Cogongrass ( Imperata cylindrica ) Biology, ecology and management. Crit. Rev. Plant Sci. 23:367 380. MacDonald, G. E., B. J. Brecke, and D. G. Shilling. 1992. Factors affecting germination of dogfennel ( Eupatorium capillifolium ) and yankeeweed ( Eupatorium compositifolium ). Weed Sci. 40:424 428. MacDonald, G. E., J. A. Ferrell B. Sellers, K. A. Langeland, T. Duperron Bond, and E. Ketterer Guest. 2008. In Invasive s pecies m anagement p lans for Florida. Gainesville, FL: University of Florida Institute of Food and Agricultural Sciences Extension Circular 1529. Mislevy, P. and K. H. Quesenberry. 1999. Development and short description of grass cultivars released by the University of Florida (1892 1995). Pages 12 19 in Proceedings of the Soil and Crop Science Society of Florida. Gainesville, FL: Soil and Crop Science Society of Florida.

PAGE 106

106 Possley, J. and J. Maschinski. 2006. Competitive effects of the invasive grass Rhynchelytrum repens (Willd.) C.E. Hubb. on pine rockland vegetation. Nat. Areas J. 26:391 395. Provencher, L., B. J. Herring, D. R. Gordon, H. L. Rodgers, K. E. M. Galley, G. W. Tanner, J. L. Hardesty, and L. A. Brennan. 2001. Effects of hardwood reduction techniques on longleaf pine sandhill vegetation in northwest Florida. Restor Ecol. 9:13 27. Quail, P. H. and O. G. Carter. 1969. Dormancy in seeds of Av ena ludoviciana and A. fatua Aust. J. Agric. Res. 20:1 11. Richardson, S. G., N. Bissett, C. Knott and K. Himel. 2003. Weed control and upland native plant establishment on phosphate mined lands in Florida. In Proceedings of the Annual Conference on E cosystems Restoration and Creation Tampa, FL: Hillsborough Community College. Rockland, L. B. 1960. Saturated salt solutions for static control of relative humidity between 5 and 40 C. Anal. Chem. 32:1375 1376. Scott, J. M. 1913. Natal grass. Gainesville, FL: University of Florida Agricultu ral Experiment Station Press Bulletin 208. 2 p. Segal, D. S., V. D. Nair, D. A. Graetz, N. J. Bissett, and R. A. Garren. 2001. Post mine reclamation of native upland communities. Bartow, FL: Florida Institute of Phosphate Research Publication 03 122 159. 287 p. Simpson, G. M. 1990. Seed d ormancy in g rasses. New York City, NY : Cambridge University Press. 297 p. Small, J. K. 1933. Tricholaena Schrad Pages 83 84 in Manual of the s outheastern f lora. Lancaster, PA: The Science Press Printing Company. 1554 p. Tracy, S. M. 1916. Natal grass: a southern perennial hay crop. Washington, DC : U.S. Department of A [USDA] United States Department of Agriculture Natural Resources Conservation Service. 2010. The PLANTS Database Plant Profile for Melinis repens (Willd.) Zizka. http://plants.usda.gov/java/profile?symbol=MERE9 Accessed October 11, 2010. Van Mourik, T. A., T. J. Stomph, and A. J. Murdoch. 2005. Why high seed densities within buried mesh bags may overestimate depletion rates of soil seed banks. J. Appl. Ecol. 42:299 305.

PAGE 107

107 Wesson, G. and P. F. Wareing. 1969. The induction of light sensitivity in weed seeds by burial. J. Exp. Bot. 20:414 425. Wilder, B. J. 2009. Seed b iology and c hemical c ontrol of g iant and s mall s mutgrass. s. Gainesville, FL: University of Florida. 66 p. Winston, P. W. and D. H. Bates. 1960. Saturated solutions for the control of humidity in biological research. Ecology 41:232 237. Wunderlin, R. P. and B. F. Hansen. 2008. Atlas of Florida Vascular P lants. http://www.plantatlas.usf.edu/ Accessed October 9, 2010.

PAGE 108

108 BIOGRAPHICAL SKETCH Courtney Ann Stokes was raised in Leesburg, F lorida After graduating as Salutatorian from Leesburg High School in 2004 she att ended the University of Florida, where she earned a Bachelor of Science Cum Laude, in p lant s cience with an emphasis in s ustainable c rop p rodu ction and m anagement While an undergraduate, Courtney was involved in the Agronomy Soils Club and was a captain on the U niversity of F lorida Fencing team. In May 2008, Courtney began graduate studies in a gronomy with a concentration in w eed s cience. As a graduate student, Courtney was inducted into Alpha Zeta and Gamma Sigma Delta and was department repre sentative to the U niversity of Florida Graduate Student Council for 2.5 years. Courtney also served as a Teaching Assistant for PLS4601, Principles of Weed Science. of Florida in the fall of 2010 She plans to pursue a doctorate degree She would like to continue research involving invasive plants and further her knowledge of ecology