Organization, Cooperating Agencies, and Principal Leaders:
 

David H. Teem, Administrative Advisor SAES, AL

Jack M. Barnes, CSREES Representative USDA, CSREES

Robin N. Huettel, CSREES Representative USDA, CSREES

R. Charudattan, Project Chairman SAES, FL

Donald J. Daigle, Secretary USDA-ARS, LA

William L. Bruckart, Annual Meeting Chairman USDA-ARS, MD

C. Doug. Boyette, Chairman, Objective 1 USDA-ARS, MS

David O. TeBeest*, Chairman, Objective 2 SAES, AR

William M. Connick, Chairman, Objective 3 USDA-ARS, LA

Norman W. Schaad, Chairman, Objective 4 USDA-ARS, MD

Paul A. Backman* SAES, AL

Fenny K. Dane SAES, AL

J.A. Shaw SAES, AL

David O. TeBeest* SAES, AR

Gregory J. Weidemann SAES, AR

R. Charudattan* SAES, FL

Abraham H. Epstein* SAES, IA

Mark A. Jackson* USDA-ARS, IL

William M. Connick* USDA-ARS, LA

Donald J. Daigle USDA-ARS, LA

Rex W. Millhollon USDA-ARS, LA

Thomas A. Bewick SAES, MA

Frank L. Caruso* SAES, MA

William L. Bruckart* USDA-ARS, MD

Efstathios Hatziloukas USDA-ARS, MD

Douglas G. Luster USDA-ARS, MD

C. Mischke USDA-ARS, MD

Norman W. Schaad USDA-ARS, MD

Shaw-Ming Yang USDA-ARS, MD

Hamed K. Abbas USDA-ARS, MS

C. Doug. Boyette* USDA-ARS, MS

Robert E. Hoagland USDA-ARS, MS

Robert F. Nyvall* SAES, MN

Joe C. Neal* SAES, NY

Herb J. Hopen* SAES, WI
 

*Principal Leaders/Voting Members

Other Participants in the 1998 Meeting:

James Anderson USDA-ARS, MD

Bryan Arroyo US Fish and Wildlife Service, VA

Bryan Bailey USDA-ARS, MD

Karen Bailey Agric. & Agri-Food, Sask., Canada

Susan M. Boyetchko Agric. & Agri-Food, Sask., Canada

Nancy L. Brooker Pittsburg State Univ., KS

Charlie Brown USDA-APHIS, MD

Jennifer Cook SAES, NC

Farivar Eskandari Gaithersburg, MD

Robin Goodson Dept. of Agric., NC

Doug Gurian-Sherman US EPA, Washington, DC

Robin Huettel USDA-CSREES, DC

Steve Hutcheson SAES, MD

David Johnson SAES, MN

Dennis Johnson USDA-ARS, MD

Robert Kremer USDA-ARS, MO

John Lydon USDA-ARS, MD

Cheryl McRae NSW, Australia

S. Krishna Mohan SAES, ID

Richard Ostrowski United Agric Products, CO

John C. Porter SAES, Cranberry Experiment Station, MA

Mariana Purnell Agric. Res. Council, Embassy of S. Africa

Albert Pye BioLogic Biocontrol Products, PA

Erin N. Rosskopf USDA-ARS, FL

Nicky Staunton Virginia Native Plant Society

Dorothy Wayson USDA-APHIS, MD

Terry Wheeler SAES, TX

Yun Wu USDA-FS, WV

Nina Zidack SAES, MT

Objective 1: Evaluate selected pathogens as bioherbicides. The following selected pathogens were evaluated.
 

COLTRU Project: Evaluation of Colletotrichum truncatum (COLTRU) for control of hemp sesbania in cotton, rice, and soybean was continued and coordinated by C.D. Boyette (South. Weed Science Lab. [SWSL], USDA-ARS, MS), W.J. Connick and D.J. Daigle (South. Reg. Res. Lab. [SRRC], USDA-ARS, LA), M.A. Jackson (Natl. Center for Agricultural Utilization Res. [NCAUR], USDA-ARS, IL), and R. Ostrowski (United Agri Products, CO). Other researchers who contributed to this research are H.K. Abbas and R.E. Hoagland (USDA-ARS, SWSL, MS).
 

Hemp sesbania was effectively controlled in soybean (Stoneville, MS) and rice (Stuttgart, AR) test plots with a 1:3 (v:v) unrefined corn oil /Colletotrichum truncatum (COLTRU) formulation containing 0.2%. Silwet L-77. A dried formulation consisting of dextrose and hydrated silica gel suspended in an experimental oil (Quoil) or in a United Agri Product proprietary oil (M+ oil) were also effective in controlling hemp sesbania. PESTA formulations containing conidia and microsclerotia (produced at SRRC) provided some control when applied either preemergence or preplant incorporated, but were significantly less effective than postemergence application of fungus/oil formulations. Control of hemp sesbania in cotton was significantly reduced by fungicide seed treatments and in-furrow fungicide and insecticide treatments. Rice formulations of COLTRU were tested for viability and virulence to hemp sesbania. The fungus was highly virulent to weed seedlings even after 8 years of storage at -20 C, and only slightly less virulent at 4 C. Survival of COLTRU in formulations stored at room temperature was significantly less than at the lower temperatures.
 

Dodder Project: An Alternaria sp. is undergoing evaluation as a bioherbicide for dodder (Cuscuta spp.) in cranberries. T.A. Bewick (SAES, MA) coordinated this project with the following states cooperating: MA, WI, and CA. United Agri Products, CO; and SRRC also participated. The Alternaria sp. has been identified by Dr. Emory Simmons. The name of this new species will be published in Mycologia. In both field trials and greenhouse studies, there was some concern that whiteflies might be responsible for movement of the pathogen since the pathogen was found in control plots. In Massachusetts cranberry fields, there were fewer weeds in the treated field. Plots that had been treated the previous year with 'Pesta' containing the pathogen experienced no infestation of dodder the following year. Studies on preemergence application of the pathogen in the field were substantiated by greenhouse results. Postemergence trials with inoculum produced by Sylvan Spawn Laboratory, Inc., Cabot, PA in either 10% corn oil or Quoil (a formulation developed by C. Quimby) gave positive results but no differentiation between the two oils.
 

Pseudomonas syringae pv. tagetis (PST) Project: D.R. Johnson (Encore Technologies, MN) coordinated the cooperative tests with PST to control Asteraceae weeds. At Encore Technologies, a microbial products company that manufactures Collego® and other biocontrol products, D.R. Johnson and F. Schendel are conducting research on industrial-scale fermentation and stabilization of PST. In an effort to advance product identity, fatty acid profiles (FAME), Biolog data, and MIDI 16S Ribosomal RNA sequence data were obtained for PST. Data from PST collections from different geographical regions and hosts revealed that diverse isolates were nearly identical in those features measured. The bacterium is amenable to liquid culture, and was
routinely grown to a density of 3 X 10¹⁰ cfu/ml in minimal glucose-nitrate medium.
 

Joseph C. Neal (SAES,NC), conducted greenhouse studies of PST on several important Asteraceae weeds. He also evaluated the relationship of organosilicone surfactant (Silwet L-77) rate to PST effectiveness, as well as the effect of clipping plants as a means of facilitating infection. The isolate used was 1-502a, obtained from D.R. Johnson. Two greenhouse experiments were conducted. The concentration of PST inoculum in both tests was 10⁸ cfu/ml applied with a hand-held spray bottle at 188 ml per m². Test plants were about 6-week old from either seed, or rhizome or stolon pieces. Seed-propagated plants were dandelion (Taraxacum officinale), eclipta (Eclipta prostrata), galinsoga (Galinsoga ciliata), and horseweed (Conyza canadensis). Vegetatively propagated species were English daisy (Bellis perennis), mugwort (Artemisia vulgaris), oxeye daisy (Chrysanthemum leucanthemum), yarrow (Achillea millefolium), and yellow hawkweed (Hieracium pratense). In the second test only seedling plants were tested, including bull thistle (Cirsium vulgare), eclipta, galinsoga, common groundsel (Senecio vulgaris), and western salsify (Tragopogon dubius). No disease symptoms were observed on plants inoculated with PST without Silwet L-77. Symptoms and percent control were increased by increasing Silwet concentration from 0.2% to 0.4%. Clipping plants following inoculation reduced the effectiveness of treatments. Considerable variability in susceptibility was observed among species. In the most susceptible hosts, symptoms developed within one week. Most species demonstrated signs of recovery by 3 weeks after inoculation. Common groundsel was controlled 80% to 100% within 1 week of inoculation and that level of control was maintained for the 5-week duration of the test. Eclipta was controlled 60% to 70% for the duration of the test. Galinsoga was controlled 60 to 80% at 2 weeks but by 5 weeks 0% to 50% control was observed, for 0.2% and 0.4% Silwet L-77, respectively. Bull thistle, mugwort, and western salsify were controlled 50% to 80% for about 2 weeks, but recovery was evident by the 3^rd week. By the 5^th week, control of these three species was between 0 and 35%. Dandelion and horseweed were symptomatic but control never exceeded 32%. Of the vegetatively propagated plants tested, only yellow hawkweed control was greater than 50% after 5 weeks.
 

H.K. Abbas conducted greenhouse studies of PST and tagetitoxin on various hosts, including common cocklebur (Xanthium strumarium) The isolate used was 1-502a, obtained from D.R. Johnson. Ten cocklebur biotypes from MS, IL, OK and TX , including MSMA-resistant and imazaquin-resistant biotypes, were spray-inoculated at the 2- to 3-leaf growth stage with an aqueous suspension of 1 X 10⁸ cfu/ml plus Silwet L-77 (0.2% v/v). All biotypes were susceptible to PST, but disease severity varied with biotype. There was no apparent correlation between herbicide resistance and disease severity. When purified tagetitoxin was tested on cocklebur, a dose-response relationship was observed. Tagetitoxin was delivered to 2- to 3-leaf cocklebur seedlings by wounding the stem and applying an 8 l droplet of 2, 4, 8, 31.5, 62.5, 125, or 250ng on the toxin/l. Apical chlorosis symptoms were observed on all plants after 48 h. Chlorophyll content in leaves of treated plants was reduced 5-98%, depending on dose. In plants treated with 31.5 ng/l tagetitoxin, chloroplast abnormalities in third leaves were observed 24 h after treatment. Responses of other weed seedlings to tagetitoxin were also tested. Sensitive species included Canada thistle (Cirsium arvense), common ragweed (Ambrosia artemisiifolia), horseweed (Conyza canadensis), safflower, wild and cultivated sunflowers (Helianthus annuus), and tropical soda apple (Solanum viarum). Morningglories (Ipomoea sp.), sicklepod (Cassia obtusifolia), and velvetleaf (Abutilon theophrasti) were weakly sensitive.
 

J. Gronwald, K. Plaisance, D.R. Johnson, and D. Wyse collaborated on a project to investigate population the dynamics of PST in spray-inoculated host and nonhost plants. Seedlings tested were in the 1- or 2-leaf stage, depending on the host. Soybean and sunflower were grown from seed, while Canada thistle plants were clones from a single rootstock. The isolate used was D.R. Johnson's 1-502a, originally isolated from Canada thistle. Spray applications were made with an air-powered sprayer using Silwet L-77 at 0.3% v/v. Volume was approximately 1.1 ml/seedling. Inundative foliar application of PST in an aqueous suspension containing an organosilicone surfactant provided good control of sunflower, variable control of Canada thistle, but had no effect on soybeans. Population dynamics of endophytic PST following foliar application were evaluated in host (sunflower, Canada thistle) and nonhost (soybean) plants. The endophytic PST population in sunflower leaves, measured 60 min after spraying with 10⁹ cfu/ml, was 10⁷ cfu/g fresh wt. The endophytic population increased rapidly during the subsequent 24 h, reaching a peak of 10⁹ cfu/g fresh wt, which declined slightly during the subsequent 48 h. Chlorophyll content of sunflower leaves that emerged following application of 10⁹ cfu/ml PST was reduced by about 90%. Spraying 10⁹ cfu/ml PST on soybean leaves resulted in a low endophytic population (10⁵ cfu/g fresh wt) due to entrapment of spray droplets by leaf trichomes. When trichomes were removed by rubbing ethanol on the leaves with a Kimwipe, it was possible to establish higher PST populations in soybean. In contrast to sunflower, the endophytic PST population in soybean leaves did not increase during a 72 h interval after application, nor was chlorophyll content of developing leaves reduced. Canada thistle plants sprayed with 10⁹ cfu/ml PST exhibited variable endophytic populations and injury as measured by chlorosis of developing leaves.
 

Nina Zidack studied the effects of PST on houndstongue (Cynoglossum officinale) and discovered that necrotizing agents such as pelargonic acid (Scythe^®) greatly enhanced PST efficacy. The PST isolate used was D.R. Johnson's 1-502a, obtained from Mycogen Corporation. An aerosol sprayer (Crown sprayer) was used for applications in greenhouse experiments, and hand pump, air-pressurized sprayers were used in field experiments. Silwet L-77 surfactant rates were 0.2% v/v in greenhouse studies, 0.25% v/v in the field. Inoculum density was 1 X 10¹⁰ cfu/ml. Greenhouse experiments showed that the optimal rate of Scythe was 4% v/v, applied at 100 GPA.
 

A strong synergistic relationship was shown for PST and the contact herbicide Scythe when applied to houndstongue. When Scythe followed PST application, immediate necrosis resulted and subsequent regrowth was severely chlorotic. The synergy seemed to be the result of the contact herbicide necrotizing the mature leaves of the plant, thereby reducing the plants' ability to produce carbohydrates for regeneration of healthy tissue. Tagetitoxin produced in the necrotized tissue inhibited chloroplast development in the new tissue. Plant reserves were depleted by supporting chlorotic regrowth that produced substantially reduced levels of photosynthate. This synergy is not unique to Scythe. Similar synergy was also demonstrated plants damaged by frost or steam.
 

Field research in 1997 produced very promising results for control of houndstongue with PST in combination with Scythe. In June, a replicated experiment on a natural infestation of houndstongue was initiated near the East Gallatin river in Southwest Montana. Treatments included PST alone, PST plus Scythe, Scythe alone, and an untreated control. Ten plants in each treatment were tagged on the day of application . One month after spraying, plots were rated for necrosis and chlorosis of regrowth on a 0 to 5 scale with 0 being completely healthy and 5 being dead. Data are presented in Table 1.
 

Table 1. Average efficacy rating of PST treatments on houndstongue 1 month after treatment
 

Treatment

Average efficacy

 

PST alone	0.66
PST plus Scythe	3.05
Scythe alone	0.94
Untreated control	0

 

In the PST plus Scythe treatment, 37% of the tagged plants were dead. In the PST alone treatments, no plants were dead and in the Scythe alone treatments 16% were dead. The most promising aspect of this study was observed in the fall of 1997; on September 25^th, many of the plants in the field plots that had appeared to recover earlier in the season had died. Most plots with PST plus Scythe had 0-2 live houndstongue plants. There was also some houndstongue mortality in the PST alone and Scythe alone plots.

T. Wheeler, P. Dotray and T. Sheikh studied the effects of PST on wollyleaf bursage (Ambrosia grayi) an important perennial weed in Texas. The isolate used was D. Johnson's 1-502a obtained form Mycogen Corporation. In 1997 field tests, bacterial inoculum concentrate (10⁷ to 10⁸ cfu/ml) was diluted by 1/40th, mixed with 0.25 % v/v Silwet L-77 and applied to runoff using a Solo gasoline powered backpack sprayer. Plants were rated at 24 and 48 hrs and weekly thereafter. Tests conducted consisted of spraying plots I) once a month starting in April and terminating in October; ii) with a dilution series of the bacteria from concentrate to 1/10,000 dilution; iii) at different times during the day, starting at 8:00 a.m. and terminating at 8:00 p.m.; iv) at 2- week and 1-week intervals, versus no sprays. Another experiment was conducted with field application equipment that had been adjusted to deliver different water pressure and volume as the sprayer was tractor driven over the test area. Volumes varied from 50 to 320 gallons/acre and pressure from 20 to 80 psi. This experiment was repeated twice.
 

The monthly application of PST resulted in good symptom development during April and moderate symptom development during May. The weeds recovered from chlorosis by 4-5 wks after application. There was poor symptom development when weeds were sprayed for the first time during months after May. There was significant attrition of the weed plots which were sprayed in April, beginning in late July and continuing in August. This drop in weed density coincided with an increase in apical chlorosis. By September, 80% of the surviving weeds in the plots sprayed in April showed apical chlorosis.
 

Attempts to apply the bacteria with normal field spraying equipment was unsuccessful. A backpack sprayer (Solo gasoline powered) was used to apply PST (successfully) in fall of 1996 and spring 1997, but after Roundup was used in the sprayer, applications no longer resulted in symptom development. Agricultural pesticides such as Roundup may affect the ability to apply with field equipment.
 

J. Porter and T. Bewick studied effects of organosilicone surfactants and herbicides on PST as a potential biological control agent for Asteraceae weeds in cranberry. The most important of these weeds are narrow-leaved goldenrod (Solidago tenuifolia), pitchforks (Bidens frondosa), ragweed (Ambrosia artemisiifolia), and white and blue asters (Aster ericoides and A. novi-belgii, respectively). The surfactants tested were the Silwet L-77, Silwet-1, and Silwet-408, and LI-700 (Loveland Industries Inc., P.O. Box 1289, Greeley, Co. 80632), all non-ionic surfactants. The herbicides tested were 2,4-D and glyphosate (as Weedar 64 and Roundup Ultra, respectively). These herbicides are routinely used to control Asteraceae weeds on cranberry bogs. Live bacterial cells suspended in sterile distilled water were added to serially diluted solutions of each of the chemicals (each diluted solution being half the strength of the one that preceded it) at rates that produced 100 cells/ml suspensions. Controls consisted of bacteria suspended in distilled water only. Plates containing solid King's Medium B were inoculated with one ml samples of the suspensions and incubated at 24C. The number of bacterial colonies growing on each plate was counted 2 days later. There was a significant range in the lethality of the different adjuvants and herbicides. The organosilicones L-77 and Silwet-1 had very little, if any, impact on bacterial survival, while the adjuvant LI-700 caused complete mortality when applied at rates as low as 0.333%. Silwet-408 was a little less lethal than LI-700, allowing a higher percentage of the bacteria to survive at a particular rate, but the highest rate that allowed 100% of the bacteria to survive was roughly the same as it was for LI-700 (i.e., roughly 0.0476%). As for the herbicides, Roundup Ultra was almost twice as potent as Weedar 64, killing all of the bacteria at rates as low as 0.4167% while Weedar 64 killed all bacteria at rates as low as 0.833%. Certain concentrations of the organosilicones and/or herbicides that did not have an impact on bacterial survival could still have a serious impact on the growth rate of the bacterium. To determine if this was the case with any of these chemicals, bacteria were added to flasks containing liquid King's Medium B and amended with one of the organosilicones or herbicides at a rate sufficient to produce 100 cells/ml suspensions. The amounts of organosilicones or herbicide added to each of the flasks were the highest amounts that still allowed 100% of the bacteria to survive (determined in the previous study). The organosilicones L-77 and Silwet-1, the two chemicals in the previous study that did not seem have an impact on bacterial survival, were applied at a rate of 1%, the highest rate they would probably ever be applied at in commercial situations. Controls for the experiment consisted of bacteria growing in liquid King's Media B only. After flasks were inoculated, they were incubated on a rotary shaker at 200 rpm. Bacteria were allowed to grow in these flasks until the suspension became turbid. Bacterial suspensions were then removed from samples of these suspensions through centrifugation, resuspended in equivalent amounts of distilled water, and counted. This was done by determining the percent absorbance of the solution using a spectrophotometer, different rates of absorbance corresponding to different bacterial concentrations. The length of time it took for inoculated flasks to become turbid was then divided by the number of generations it took for the starting concentration to become the final concentration. This calculation gave a value for the length of time it would take for a bacterium to mature and reproduce in the different suspensions (i.e., a measure of the bacterium's growth rate). Some chemicals had a rather profound effect on the rate of growth of the bacteria while others did not. Bacteria in suspensions containing LI-700 and Weedar 64 grew only half as fast as bacteria in the control did, while bacteria in suspensions containing Silwet-1 and Silwet-408 grew roughly as fast as bacteria in the control did.
 

Other selected pathogens that were evaluated in 1997 are: several unnamed fungal pathogens of Canada thistle, wild oats, and green foxtail (Boyetchko, et al., SASK, Canada); Phomopsis amaranthicola (pigweeds), Dactylaria higginsii (purple nutsedge), Pseudomonas solanacearum (tropical soda apple), and a mixture of three fungi (several grasses) (Charudattan, et al., FL); Colletotrichum gleosporoides (coffee senna and sicklepod) (Boyette, et al., MS); Alternaria helianthii (cocklebur) (Abbas, et al., MS); rose rosette disease (RRD) agent (multiflora rose) (Epstein (IA); and Xanthomonas convolvuli (field bindweed) (Mohan, ID).
 

Objective 2. Enhance the efficacy of bioherbicide candidates. Several processes and materials are being developed and tested to enhance bioherbicide efficacy through inoculum formulations, enhancement of infection and disease development and increased levels of weed control. TeBeest (AR) has developed methods to cross formae spciales of Colletotrichum gleosporoides, including the COLLEGO fungus, but the progeny exhibited low fertility. COLLEGO and various oils were found to be capable of controlling northern jointvetch more effectively by reducing the dew period by 4 hours.
 

Charudattan, et al. (FL) have developed a cocktail of three fungi, Drechslera gigantea, Exserohilium rostratum, and E. longirstratum, which controlled seven grasses: large crabgrass, crowfootgrass, johnsongrass, guineagrass, southern sandbur, Texas panicum, and yellow foxtail in the greenhouse. Field testing on guineagrass with the cocktail mixture was also successful. Advantages of this multi-pathogen strategy are the improved range of weeds controlled, the ability to customized the mixture, overcoming age-related host resistance, and the potential to avoid resistance buildup.
 

Unrefined corn oil-Silwet L-77 emulsions containing Alternaria helianthi spores increased pathogenesis to several cocklebur biotypes under moisture-limiting conditions. The bacterial pathogen Pseudomonas syringae pv. tagetis also infected and killed these biotypes when formulated with Silwet L-77 and unrefined corn oil (Boyette, et al., USDA-ARS, MS).
 

Objective 3: Production and Formulation. Connick (USDA-ARS-SRRC, LA) coordinated this objective and the following contributed: Boyette (USDA-ARS-SWSL, MS); Connick and Daigle (USDA-ARS-SRRC, LA); Jackson (USDA-ARS-NCAUR, IL); Zidack and Quimby (USDA-ARS, MT); Charudattan, et al. (SAES, FL), Bewick and Caruso, Cranberry Experiment Station, MA, and Boyetchko,et al. (Agric. and Agri-Food, Canada, Sask).
 

Development of liquid culture production methods for spores and microsclerotia continued for the bioherbicidal fungus Colletotrichum truncatum (COLTRU), a pathogen of hemp sesbania (Sesbania exaltata). A collaborative effort of Boyette, Jackson, Zidack, Quimby, and United Agri Products (Ostrowski) led to field trials conducted in Stoneville, MS, for control of hemp sesbania in cotton and rice using COLTRU spores produced using 20-L and 100-L fermentors. Granular formulations, adjuvants, and "Stabileze" were tested. The "Stabileze" method (Quimby, Zidack et al.) involves stabilization of microbes with sucrose and oil in a starch matrix, which is then granulated with hydrated silica. The product contained 1.35 x 10⁷ spores/g.
 

Studies are underway to evaluate the use of COLTRU microsclerotia as bioherbicidal propagules. In an attempt to reduce fermentation times for the production of microsclerotia, 3-day-old COLTRU precultures were used to inoculate production flasks. Using this method, production time was reduced from 10 days to 6 days. The benefit of using a preculture inoculum for microsclerotia production in fermentors will be determined.
 

Pesta^tm granules containing inoculum of a Fusarium oxysporum pathogenic to coca retained viability throughout storage for at least two years at 25^oC and at least one year at 35^oC, if the water activity was maintained at 0.12 (12% relative humidity).
 

Solid-state fermentation (SSF) of Colletotrichum truncatum and an Alternaria sp. (pathogenic to dodder) on rice flour resulted in about a 100-fold increase in fungal viability when the infested flours were incorporated in a granular matrix by extrusion followed by fluid bed drying at 50C. Apparently, the aerial conidia produced by SSF are hardier than biomass produced in liquid culture (COLTRU and Dodder Projects).
 

A technique for mass production and multiple-harvesting of bioherbicide fungi by SSF was developed and tested with two bioherbicide agents, Drechslera gigantea and Exserohilum rostratum. Shake flask cultures were transferred to trays which were exposed to alternating light and dark cycles. Spores were harvested twice at 24 and 48 hours. In related work, sorghum and oat grains were suitable SSF substrates for these fungi and produced abundant spores when exposed to the light/dark cycles for two weeks.
 

Greenhouse and field trials were conducted with Pesta^tm formulations of Alternaria sp. for control dodder. None of the formulations improved disease levels in the greenhouse. In the field, differences between formulations were not seen, but there was significant infection in dodder treated with Pesta in trials in CA and MA. In each instance, in different plots, the dodder did not reappear the year following application.
 

Principles for selection of adjuvants used in bioherbicide formulations are being developed by Boyetchko, et al. Fungal species from each of the genera Colletotrichum, Phoma, Fusarium, and Alternaria were selected to characterize their compatibility with common laboratory surfactants. Tween 20, Tween 40, Tween 80, Tergitol 9, Tergitol 10, sorbitol, and gelatin were evaluated. Conidial germination varied with the surfactant, surfactant concentration, the fungal pathogen, and the inoculum density. Tween 40 and Tween 80 were compatible with all fungi; gelatin was compatible with Phoma and Fusarium; sorbitol was compatible with Fusarium and Alternaria. Self inhibition of conidia was released in Colletotrichum with Tween 80; gelatin had similar effects on Phoma. Tergitol was detrimental to conidial germination for most fungal species, while Tween 40 and Tween 80 had the mildest effects. In general, fungi belonging to the Coelomycetes were more sensitive to adjuvants than those belonging to the Hyphomycetes.
 

In 1997, field trials were conducted to evaluate several rhizobacteria to control green foxtail and wild oats. Granular formulations, including peat prills and clay formulations of rhizobacteria were tested. Generally, peat prill formulations provided significant reductions in weed emergence (30-35% reduction) and aboveground biomass (30% reduction) while the clay formulations did not reduce weed emergence or biomass. However, one bacterial strain formulated in the clay formulation reduced biomass of green foxtail by approximately 30%, but weed emergence was not affected. Efforts on improving these formulations are continuing and field trials will be conducted in 1998.
 

Technology is being developed to mass-produce fungal and bacterial agents, the former on solid substrates. A key factor in fungal sporulation depends on the amount of aeration (oxygenation) and length of time in the liquid phase prior to spreading on the solid surfaces. Fermentation studies were conducted to determine the nutritional requirements of specific rhizobacterial isolates with weed-suppressive properties. Shake-flask culturing incorporating various nutrients that lead to significant biological control activity in combination with mass production of bacterial cells was undertaken. Specific carbon and amino acid combinations that provide optimum activity of each bacterial isolate are currently being selected. In addition, phase kinetic studies are being conducted to evaluate the growth rate of bacteria for scale-up from shake flask to 5 and 10 L fermentors.
 

Objective 4: Develop genetic characterization and transformation of bioherbicide candidates as a means of enhancing efficacy and assessing environmental risk. The ability to distinguish biotypes of purple nutsedge may be important for developing strategies for successful biological control of this weed. W. Pereira, a visiting scientist from EMBRAPA-CNPH, Brazil, working in Charudattan's laboratory, examined purple nutsedge accessions for variability in susceptibility to the bioherbicide isolate of Dactylaria higginsii. A comparison of morphological and phenological characteristics of 63 purple nutsedge accessions from Brazil, India, Mexico, Puerto Rico, and United States (Florida, California, and Hawaii) indicated the presence of distinctive intraspecific biotypes in this collection. Pathogenicity trials indicated that 90.5% of the nutsedge accessions tested were susceptible to D. higginsii, 7.9% resistant, and 1.6% immune. Molecular variability among the accessions is being characterized by RAPD analysis to determine the genetic make up of purple nutsedge populations and to identify possible genetic markers of susceptibility to D. higginsii.
 

An attempt is being made to understand the possible molecular basis for pathogenic variability in Cercospora piaropi Tharp and C. rodmanii Conway, two pathogens of waterhyacinth (Eichhornia crassipes). Sixty isolates of Cercospora spp. pathogenic to waterhyacinth were collected from the United States, Mexico, Venezuela, Brazil, South Africa, and Zambia and screened in a quarantine greenhouse in Gainesville to assess their variability in virulence toward waterhyacinth. A bioassay was used which consisted of immersing waterhyacinth leaves, attached to plants, in a suspension of inoculum. The inoculum was prepared from mycelium grown in still liquid cultures in V8 medium for 14 days at 25+ 3C. The mycelium was separated from the broth by filtration and blended for 5 sec at a concentration of 80 mg/ml of water. The suspension was amended with 0.5% Metamucil and 0.05% Silwet. Four replicates, each with one 2-leaved plant, were used and the experiment was done twice. Control leaves were immersed in a solution containing only the amendments. The inoculated and control plants were placed in a dark dew chamber at 25 ±2C for 12 h and then held in a greenhouse for 2 wk. Disease severity (percentage of leaf area necrosed) was assessed 4, 7, and 14 days after inoculation using a visual rating scale of 0-7. The isolates differed significantly in virulence (P<0.05), varying from avirulent (rating 0) to highly virulent (rating 7, leaf death). The variation in virulence among the isolates was independent of their geographic origin, spore size, or cultural differences. To determine the extent of similarities and evolutionary divergence among these isolates, DNA sequences of three conserved genes were amplified through polymerase chain reaction (PCR), sequenced, and compared through cladistic analysis. Comparisons were based on PCR fragments of -tublulin, histone H3, and elongation factor 1 genes, corresponding to 377, 288, and 459 nucleotides, respectively. All these sequences included at least one intron region. The number of isolates used in this study were 20, 15, and 9 for -tubulin, histone H3, and elongation factor 1 genes, respectively. The set of primers used for the PCR amplification of the -tubulin and histone H3 genes were obtained from the literature, meanwhile the primers for amplification of the elongation factor 1 gene was specially designed for this research. For the cladistic analysis, the software Phylogenetic Analysis Using Parsimony (PAUP) version 3.1 was used, having the species Cercospora beticola as an outgroup. In a preliminary analysis, the cladograms based on these genes were similar and included isolates of C. piaropi and C. rodmanii in the same clade. Thus, it is suggested that all isolates studied belong to a single species, according to the cladistic concept of species.
 

Abbas, H.K., and Boyette, C.D. 1996. Control of morningglory species using Fusarium solani and its extracts. Int. J. Pest Manag. 42:235-239.

Abbas, H.K., and Duke, S.O. 1997. Plant pathogens and their phytotoxins as herbicides. Pages 1-20 in: Toxins in Plant Disease Development and Evolving Biotechnology, Eds. R.K. Uphadhyay and K.G. Mukerji.

Abbas, H.K., Duke, S.O., Shier, W.T., Dadria, F.A., Woodward, R.P., and Mirocha, C. J. 1997. Comparison of ceramide synthase inhibitors with other phytotoxins produced by Fusarium species. J. Nat. Toxins 6:163-181.

Abbas, H.K., Smeda, R.J., Duke, S.O., and Shier, W.T. 1997. Fumonisin-plant interactions. Bull. Inst. Compr. Agr. Sci., Kinki Univ., 5:63-71.

Bianco, S., Pitelli, R.A., Mullahey, J.J., and Charudattan, R. 1997. Response of tropical soda apple (Solanum viarum Dunal) to soil liming. WSSA Abstracts. 37:29.

Bailey, K.L., Boyetchko, S.M., Mortensen, K., and Wolf, T. 1997. Biological Control of Weeds Using Plant Pathogens. Pages 205-210 in Proc. Soils & Crops Workshop ¡97, February 20-21, 1997. University of Saskatchewan.

Bailey, K.L. and Mortensen, K. 1997. Evaluation of a fungal pathogen for control of Canada thistle. Pages 181-183 in Expert Committee on Weeds, Proc. 1996 National Meeting, Victoria, B.C., December 9-12, 1996. B.C. Ministry of Forest, Victoria, B.C.

Boyetchko, S.M. 1996. Formulating bacteria for use as biological control agents. Pages 85-88 IN: Proc. 1996 National Meeting of Expert Committee on Weeds, Victoria, B.C., December 9-12, 1996.

Boyetchko, S.M. 1997. Efficacy of rhizobacteria as biological control agents of grassy weeds. Pages 460-465 In: Proc. Soils & Crops Workshop ¡97, Saskatoon, Sask.

Boyetchko, S.M. 1997. Principles of biological weed control with microorganisms. HortScience 32:201-205.

Boyetchko, S.M., Bailey, K.L., Mortensen, K., Wolf, T.M. and Zhang, W.M. 1997. Survey and evaluation of fungal pathogens for biological control of grass weeds. Phytopathology 87(6):S11 [Abstr]

Boyetchko, S.M., Wolf, T.M., Bailey, K.L., Mortensen, K. and Zhang, W.M. 1997. Survey and evaluation of fungal pathogens for biological control of grass weeds. Expert Committee on Weeds, November 29 - December 4, 1997, Charlottetown, P.E.I. [Abstr]

Chandramohan, S. and Charudattan, R. 1997. Bioherbicidal control of grassy weeds with a pathogen mixture. WSSA Abstracts. 37:56.

Chandramohan, S., and Charudattan, R. 1997. Use of pathogen mixtures in a "Multiple Pathogen Strategy" for bioherbicidal control of several weeds. A patent disclosure, approved by UF for submission to the U.S. Patent Office. Patent application in preparation by a patent attorney.

Chandramohan, S., Charudattan, R., and Megh Singh. 1997. Field-testing of a "multiple-pathogen strategy" for bioherbicidal control of grassy weeds. Phytopathology 87 (Suppl.): S17 (Abstract).

Chandramohan, S., Charudattan, R., and Sonoda, R.M. 1997. Field-testing of a pathogen mixture for control of guineagrass. Phytopathology 87 (Suppl.): S18 (Abstract).

Charudattan, R., Y.M. Shabana, J.T. DeValerio, and E.N. Rosskopf. 1996. Phomopsis species fungus useful as a broad-spectrum bioherbicide to control several species of pigweeds. U.S. Patent No. 5,510,316. April 23, 1996.

Connick, W.J., Jr., Daigle, D.J., Williams, K.S., Vinyard, B., Boyette, C.D., and Quimby, P.C., Jr. 1996. Shelf life of a bioherbicide product. Amer. Biotechnol. Lab. 14(10):34-36.

Duke, S.O., Abbas, H.K., Amagasa, T., and Tanaka, T. 1996. Phytotoxins of microbial origin with the potential for use as herbicides. Pages 82-113 in Natural Products and their Potential in Agriculture, Critical Reviews in Applied Chemistry Series, Ed. L. Copping.

Duke, S.O., Abbas, H.K., Duke, M.V., Lee, H.J., Vaughn, K.C., Amagasa, T., and Tanaka, T. 1996. Microbial toxins as potential herbicides. J. Environ. Sci. Health, B31 (3), 427-434.

Epstein, A. H. and Hill, J.H.. 1995. The biology of rose rosette disease: a mite transmitted disease of uncertain aetiology. J. of Phytopathology 143 : 353 - 360.

Epstein, A.H., Hill, J.H. and Nutter, F.W. 1997 Augmentation of rose rosette disease for biocontrol of multiflora rose (Rosa multiflora). Weed Science 45 : 172 - 178.

Green, S., Stewart-Wade, S.M., Boland, G.J., Teshler, M.P., and Liu, S.H. 1998. Formulating microorganisms for biological control of weeds. Pages 249-281 In: G.J. Boland and L.D. Kuykendall (eds.), Plant-Microbe Interactions and Biological Control, Marcel Dekker, New York.

Kadir, J. and Charudattan, R. 1997. Control of Cyperus spp. with a fungal pathogen. U.S. Patent No. 5,698,491. December 16, 1997.

Kadir, J.B., Charudattan, R., Berger, R.D., Stall, W.M., and Brecke, B.J. 1997. Field efficacy of Dactylaria higginsii for control of purple nutsedge. Phytopathology 87 (Suppl.): S49 (Abstract).

Kadir, J.B., Charudattan, R., Stall, W.M., and Bewick, T.A. 1997. Effect of Dactylaria higginsii on the interference of purple nutsedge with tomato and pepper. Phytopathology 87 (Suppl.): S50 (Abstract).

Isaacson, D.L. and Charudattan, R. 1998. Biological Control of Weeds. Chapter 23 In: J.R. Ruberson, ed. Handbook of Pest Management. Marcel Dekker, New York. In Press.

Mortensen, K. 1998. Biological Control of weeds using microorganisms. Pages 223-248 In: G.J. Boland, and L.D. Kuykendall (Eds.), Plant-Microbe Interactions and Biological Control. Marcel Dekker Inc., New York..

Mortensen, K. and Makowski, R.M.D. 1997. Effects of Colletotrichum gloeosporioides f. sp. malvae on plant development and biomass of non-target field crops under controlled and field conditions. Weed Research 37: 351-360.

Pereira, A., Pitelli, R.A., Nemoto, P., Mullahey, J.J., and Charudattan, R. 1997. Seed production by tropical soda apple (Solanum viarum Dunal) in Brazil. WSSA Abstracts. 37:29.

Pitelli, R.A., Charudattan, R., and DeValerio, J.T. 1998. Effect of Alternaria cassiae, Pseudocercospora nigricans, and soybean (Glycine max) planting density on the biological control of sicklepod (Senna obtusifolia). Weed Technol. 12:000-000. In press.

Rosskopf, E.N., Gaffney, J.F., and Charudattan, R. 1997. The effect of spray propellant on the efficacy of bioherbicide candidates. WSSA Abstracts. 37:62.

Semer, C.R., IV and Charudattan, R. 1997. First report of Rhizoctonia solani causing a foliar leaf spot on Brazilian peppertree (Schinus terebinthifolius) in Florida. Plant Dis. 81:424.

Semer, C.R, IV, Charudattan, R., and Yandoc, C.B. 1997. Evaluation of a new species of Botryosphaeria (Anamorph: Fusicoccum) as a biocontrol agent for melaleuca [Melaleuca quinquenervia (Cav.) S.T. Blake]. WSSA Abstracts. 37:60.

Shabana, Y.M. and Charudattan, R. 1997. Preparation and regeneration of mycelial protoplasts of Alternaria eichhorniae. Europ. J. Plant Pathol. 145:335-338.

Shabana, Y.M., Charudattan, R., and DeValerio, J.T. 1997. Herbicidal activity of microorganisms against hydrilla [Hydrilla verticillata (L.f.) Royle]. WSSA Abstracts. 37:57.

Shabana, Y.M., Charudattan, R., DeValerio, J.T., and Elwakil, M.A. 1997. An evaluation of hydrophilic polymers for formulating the bioherbicide agents Alternaria cassiae and A. eichhorniae. Weed Technol. 11:212-220.

Simmons, E.G. and Mortensen, K. 1997. 218. Alternaria cirsinoxia Simmons & Mortensen. Pages 72-76 in: Alternaria themes & variations (151-223). Mycotaxon 53:1-91.

Smither-Kopperl, M.L., Charudattan, R., and Berger, R.D. 1997. Epidemiology of an underwater disease II: Settlement viability and germination of spores of Fusarium culmorum, a potential biocontrol agent of hydrilla [Hydrilla verticillata (L.f.) Royle], in aqueous solutions. WSSA Abstracts. 37:59.

Smither-Kopperl, M.L., Charudattan, R., and Berger, R.D. 1997. Plectosporium tabacinum, a pathogen of the aquatic weed Hydrilla verticillata in Florida. Phytopathology 87 (Suppl.): S92 (Abstract).

Smither-Kopperl, M.L., Charudattan, R., and Berger, R.D. 1998. Epidemiology of an underwater disease: III dispersal of Fusarium culmorum spores in aquatic systems. Phytopathology. 88:000-000. In Press.

Zhang, W.M., Wolf, T.M., Bailey, K.L., Mortensen, K., and Boyetchko, S.M. 1998. Principles for selection of adjuvants used in bioherbicide formulations. WSSA Abstracts 38:44 [Abstr.].
 
 

PREPARED BY:

__________________________________________ _________________________

Project Chairman Date
 
 

APPROVED:

__________________________________________ _________________________

Administrative Advisor Date

Objective 1: Evaluate selected pathogens as bioherbicides. The following selected pathogens were evaluated.
 

A five-year study of Phomopsis amaranthicola, a bioherbicide agent for pigweeds has reached a stage where further regional trials appear warranted. The pathogen, discovered in Florida, has been described as a new species, Phomopsis amaranthicola Rosskopf, Charudattan, & Shabana, based on its morphological and molecular characteristics. The efficacy of this organism as a biological control agent to manage pigweeds and amaranths has been evaluated. Conidial suspensions of P. amaranthicola, compared to mycelial suspensions, were most effective in causing high levels of plant mortality in greenhouse and field trials. A dew period of 24 h promoted plant mortality regardless of the type of inoculum suspension or the amendment, although 8 h of dew was adequate for the onset of severe infection. Fungal suspensions amended with psyllium mucilloid (Metamucil; Procter and Gamble) were effective in causing plant mortality even in the absence of a dew period. Conidial suspensions of 1.5 x 10⁶ to 1.5 x 10⁷ conidia/ml were most effective in causing high levels of mortality of pigweeds at the two- to four-leaf stage. Dew temperatures ranging from 25-35^oC were most conducive for disease development and plant mortality. Thirty-three pigweed/amaranth accessions belonging to 22 known and two unknown species of Amaranthus were tested for susceptibility to the fungus. All accessions were susceptible to the fungus, but susceptibility did not lead to mortality in all cases. Species in which there was a minimum of one biotype succumbing to 80-100% mortality included A. acutilobus L., A. lividus L., A. powellii S. Wats., A. retroflexus L., and A. viridus L. Plants within the family Amaranthaceae, but outside the genus Amaranthus, as well as crops in which pigweeds are a problem, were also tested for susceptibility to the fungus. A substantial number of plants that are reported to have a host-pathogen association with another member of the genus Phomopsis were also tested. No plants outside the genus Amaranthus showed any symptoms, nor was P. amaranthicola found to be present in their tissues as evidenced microscopically or by isolation techniques. The fungus was field tested during the summers of 1993, 1994, and 1995. The species A. hybridus, A. lividus, A. spinosus L., A. retroflexus, and A. viridus were included. In addition, a triazine-resistant accession of A. hybridus was also used. Field treatments consisted of single or double applications of mycelium and two concentrations of conidia. As in greenhouse trials, conidial suspensions were most effective in causing high levels of plant mortality, although A. lividus and A. viridus were effectively controlled with all treatments. These results indicate that P. amaranthicola could be an effective bioherbicide to control pigweeds and amaranths. The bioherbicidal use of this fungus is protected under two U.S. patents. (Erin Rosskopf and R. Charudattan).
 

Another fungal pathogen, Dactylaria higginsii (Luttrell) M.B. Ellis, was isolated from diseased purple nutsedge plants collected in Gainesville and shown to be an effective bioherbicide for control of purple nutsedge (Cyperus rotundus). In greenhouse trials, nearly complete control of purple nutsedge was achieved when D. higginsii was applied at 10⁶ conidia per ml in 100 ml of water per m² (= inoculum concentration of 10 ¹² conidia in 1000 liter per ha). The fungus was highly pathogenic to younger plants (4- to 6-leaf stage compared to > 6-leaf stage) at a temperature range from 20^oC to 30^oC and with a minimum of 12 h of exposure to dew period. At this temperature range and dew period, D. higginsii was capable of providing excellent control of young purple nutsedge plants. The host range of D. higginsii was restricted to the genus Cyperus only. Important crop plants and other economic plants tested were immune, but several weedy members of Cyperus, including C. brevifolius (Rottb.) Endl. ex Hassk. (= Kyllinga brevifolia Rottb.), C. compressus L., C. difformis L., C. esculentus L., and C. iria L. were susceptible. The ability of D. higginsii to suppress the growth of purple nutsedge plants when grown in competition with a crop plant was determined in the greenhouse using tomato and pepper as crops. The fungus, applied at 10⁶ conidia/ml under simulated cropping situation, was able to reduce purple nutsedge growth components by almost 90%, effectively suppressing purple nutsedge interference. The efficacy of D. higginsii in purple nutsedge-infested, crop-free fields was also demonstrated. Up to three applications of D. higginsii at an inoculum dose of 10⁶ conidia per ml in 100 ml/m² application volume (= 10¹² conidia per 1000 liter per ha) was required to suppress the growth of purple nutsedge. An inoculum concentrations lower than this level or a single application 10⁶ conidia per ml did not provide effective control. The high application volume used in this study may not be a requirement but it was used because it was convenient. A U.S. patent has been issued, protecting the bioherbicidal use of this fungus. Further development of this bioherbicide agent is underway. (Jugah Kadir and R. Charudattan).
 

The potential to control several grasses with a mixture of pathogens was field-tested using three fungi native to Florida: Drechslera gigantea (Heald & F.A. Wolf) Ito, Exserohilum rostratum (Drechs.) K.J. Leonard & E.G. Suggs), and E. longirostratum (Subramanian) Sivanesan, isolated from large crabgrass (Digitaria sanguinalis [L.] Scop.), crowfootgrass (Dactyloctenium aegyptium [L.] Willd.), and johnsongrass (Sorghum halepense [L.] Pers.), respectively. Twenty five 2-wk-old seedlings of seven grasses: large crabgrass, crowfootgrass, johnsongrass, guineagrass (Panicum maximum Jacq.), southern sandbur (Cenchrus echinatus L.), Texas panicum (Panicum texanum Buckl.), and yellow foxtail (Setaria glauca [L.] Beauv.) were transplanted randomly within each 1 m² plot and inoculated when 4-wk old with spore suspensions of each pathogen alone or a mixture of the three pathogens (1:1:1 by vol). The fungi were applied as foliar sprays (5 x 10⁵ spores per ml) in one of three carriers: water, 0.5% aqueous Metamucil, or an emulsion. Appropriate controls were included. During the next 14 wk, two more applications of treatments were made at 2 or 3 wk after initial inoculation (WAI). Disease severity (DS; percentage of foliar necrosis) was recorded weekly for 4-6 WAI. At 4 WAI, DS in the emulsion treatments ranged from 50.8-100% (crabgrass isolate), 78.1-99.6% (crowfootgrass isolate), 43.7-100% (johnsongrass isolate), and 63.3-100% (mix). It was possible to control the grasses using the emulsion-based inoculum preparation of each pathogen as well as the pathogen mixture. (S. Chandramohan & R. Charudattan).
 

Control of a natural population of guineagrass with the pathogen mixture (multiple-pathogen strategy) was field-tested using the three fungi mentioned above. The guineagrass plants within each 1 m² plot were inoculated with spore suspensions of each pathogen or a mixture of the three pathogens (1:1:1 by vol). The fungi were applied as foliar sprays (5 x 10⁵ spores per ml) in one of three carriers: water, 0.5% aqueous Metamucil, or an emulsion. Appropriate controls were included. During the next 10 wk, a second application of all treatments was done at 2 WAI. Disease severity was recorded weekly for 4-6 WAI. At 2 WAI, DS on guineagrass ranged from 23.4-56.3% in the emulsion treatments. At 4 WAI, DS ranged from 93.8-98.5%. Individual pathogens as well as the mixture of three pathogens were effective. It was possible to control the natural population of guineagrass using the emulsion-based inoculum preparation. (S. Chandramohan & R. Charudattan)
 

A new disease of hydrilla (Hydrilla verticillata [L.] Royle) caused by Plectosporium tabacinum was discovered in Florida by Dr. Margaret Smither-Kopperl, a Research Associate. Excised shoots of the submerged aquatic weed hydrilla, maintained in test tubes containing 5% Hoaglands solution, became chlorotic and died within two weeks after symptom appearance. Isolations from leaf and stem tissue onto Komadas medium and potato dextrose agar (PDA) yielded 100% recovery of a fungus that produced copious, short, single-septate conidia (10-12 x 3-4 µm). Colony characters were variable on PDA: mycelium was from whitish, creamy colored, to pale pink, often with a slimy appearance. Brown ascomata were produced infrequently. Asci were persistent at the base with two-celled ascospores. The fungus was identified as Plectosporium tabacinum (Beyma) Palm et al. which is synonymous with Plectosphaerella cucumerina, the teleomorph of Fusarium tabacinum. Hydrilla shoots maintained in 5% Hoaglands solution in test tubes were inoculated with a suspension of 10⁶ to 10⁸ conidia. After two weeks, noninoculated control plants remained symptomless and 100% of the inoculated plants were chlorotic or dead. The fungus was reisolated from the inoculated but not from the control plants. Disease was most severe on plants in 5% Hoaglands solution and progressively less severe on plants maintained in sterile river water, spring water, and tap water. (Margaret Smither-Kopperl, R. Charudattan, and R.D. Berger).
 

Objective 2: Enhance efficacy of bioherbicide candidates.
 

Ten polymers capable of forming aqueous gels, two surfactants, and an invert emulsion were compared for their suitability as materials for formulation and capacity to enhance the efficacy of different bioherbicide candidates. The characteristics of eight polymeric gels were determined with respect to their capacity for hydration, to remain hydrated over time, to promote spore germination, and to prolong viability of germinated spores (= germlings) of Alternaria cassiae Jurair & Kahn, a bioherbicide agent for sicklepod (Senna [= Cassia] obtusifolia L.). The polymers were also evaluated for their ability to enhance the pathogenicity of mycelial inoculum of A. eichhorniae Nag Raj & Ponnappa, a bioherbicide agent for waterhyacinth (Eichhorniae crassipes [Mart.] Solms.). The best concentration of each polymer that yielded 95-100% viable germinating spores within 6 h after hydration was chosen. The proportion of alive germlings versus the spores that germinated was used as the measure of longevity of effectiveness of the polymers. Based on these two measures, the most effective concentrations of the gels were: Gellan gum, 4.3%; Kelzan xanthan gum, 0.4%; Evergreen 500 polyacrylamide, 0.04%; Kelgin-HV, 0.04%; Kelgin-MV, 0.04%; Kelgin-LV, 0.004%; Metamucil, 0.4%; and N-Gel, 0.04%. When compared at a standard 0.1% w/w (gel/water) concentration, the eight gels retained hydration for 6.5 to 7.5 days with no significant differences among them. Thus, the addition of any of these polymers to the inoculum suspension should enable the fungal propagules to remain moist for a prolonged period, to benefit from the high ambient moisture to improve germination, and thereby promote disease on the host plants. Seven of the gels were tested for their ability to promote pathogenicity of a mycelial inoculum of A. eichhorniae. Gellan gum and Kelgin-HV were most effective in promoting disease, followed by Evergreen 500 polyacrylamide, Kelgin-LV, Metamucil, Kelzan xanthan gum, and N-gel. (Y.M. Shabana, R. Charudattan, and J.T. DeValerio).
 

Silwet L-77 (0.02% v/v), Triton X-100 (0.02% v/v), Metamucil (0.5% w/v), N-Gel (0.5% w/v), Kelzan S (0.5% w/v), Natrosol (0.5% w/v), and combinations of Silwet+Metamucil, Silwet+N-Gel, and Silwet+Kelzan S were tested as amendments to promote the germination of conidia of Dactylaria higginsii. Conidia suspended in water was used as the control. Metamucil, N-Gel, and combinations of Metamucil or N-Gel with Silwet promoted better germination of conidia than the control, but the percentage of spore germination was not significantly higher in Metamucil, N-Gel, or the combination of Metamucil or N-Gel with Silwet. Germination of spores in the suspensions containing surfactants (Silwet L-77 and Triton X-100) were quite low compared to the gels, but significantly higher than in the control. The gels and the combination of a gel and Silwet promoted more lesion development on the leaf. Significantly higher numbers of lesions developed when sprayed with inoculum containing the gels (P = 0.05). Fewer lesions developed on leaves sprayed with the inoculum plus surfactants, and much fewer lesions developed on leaves sprayed with spores suspended in water. Metamucil, N-Gel, and Kelzan were better materials than others for formulating the conidia of D. higginsii. Plants sprayed with conidia suspended in .05% Metamucil and 0.5% N-Gel developed severe disease symptoms (100% disease incidence), and all leaves on these plants were severely diseased (95% and 87% disease severity). Kelzan S, a xanthan gum, did not perform as well as Metamucil and N-Gel, although it was slightly better than Silwet L-77 and Triton X-100. Very low levels of disease developed on plants inoculated with conidia suspended in water only. Thus, Metamucil and N-Gel appear to be suitable materials to use as amendments-humectants for conidial inoculum of D. higginsii. (Jugah Kadir and R. Charudattan).
 

Two compositions of invert emulsions were tested to enhance the efficacy of three fungal pathogens used against grass weeds in citrus The pathogens were: Drechslera gigantea, Exserohilum rostratum, and E. longirostratum, used alone or as a mixture (see Objective 1 for details). The emulsions consisted of Sunspray 65 oil, mineral oil, and aqueous conidial suspension (at 10⁶ conidia per ml) combined at 4:10:86 or 80:20:100 proportions. They were prepared according to the procedures recommended by Dr. Sam Yang (USDA-ARS, Ft. Detrick). The emulsified inocula were compared with conidia+water only and conidia+0.5% Metamucil in replicated, repeated field trials. All preparations contained the same inoculum concentrations. The results confirmed that the emulsion-based inoculum consistently out-performed the other two inoculum formulations. Complete control of the grasses was possible with the emulsion-based inocula, whereas the other two types of inocula did not provide full control. For details, see Objective 1. (S. Chandramohan and R. Charudattan).
 

Objective 3: Develop systems for mass-production of stable bioherbicide formulations.
 

A technique for mass production and multiple-harvesting of bioherbicide fungi by solid-substrate culturing was developed and tested with two bioherbicide agents, Drechslera gigantea and Exserohilum rostratum. Mycelial plugs (1-wk old) were used to inoculate 1000 ml V8 broth in 2 liter flasks. The inoculated flasks were shake-cultured (100 rpm) for 2-3 days at 25^oC. The contents of each flask plus 10 ml of an antibiotic solution (3.7 mg/ml streptomycin and 2.5 mg/ml chloramphenicol) were blended in a Waring blender at low speed for 30-60 sec and 500 ml of this suspension was poured onto a layer of V8 agar (500 ml) containing antibiotics (as above) in trays (37.5 x 30 x 1.25 cm) lined with aluminum foil. The trays were exposed to alternating light and dark cycles (12 h per cycle) at room temperature (25-28ºC). The initial crop of spores appeared within 24 h. These spores were collected in two steps: first, the spores were gently scraped-off with a rubber spatula into sterile water. The remaining spores were then rinsed-off the agar surface with sterile water. The spore suspensions were pooled and the spores were allowed to settle down. The excess supernatant was decanted and the spores were resuspended in 250 ml sterile water. The trays were reincubated under light as before and the spores were harvested twice at 24 h and 48 h. The spore yields for D. gigantea and E. rostratum averaged 2.05 x 10⁵, 3.43 x 10⁵, and 1.89 x 10⁵, and 2.44 x 10⁵, 2.16 x 10⁵, and 1.00 x 10⁵ spores/ml/harvest, respectively. Thus, it is feasible to mass produce these fungi by solid-substrate culturing and multiple spore harvests. (S. Chandramohan and R. Charudattan).
 

The suitability of natural substrates to support vigorous growth and abundant sporulation of fungi was evaluated in the laboratory. Sorghum and oat grains in flasks were autoclaved twice and inoculated with blended mycelia of two Drechslera spp. isolated from cogongrass (Imperata cylindrica (L.) Beauv.) and crabgrass (Digitaria sanguinalis). Cultures were incubated at 26ºC with alternating light and dark periods (12 h each). Both substrates yielded spores after 2 wk. Twenty five grams of grains yielded 100 ml of suspension with 10⁵ to 10⁶ spores/ml. The isolates produced more spores in oats (10⁶ spores/ml) than in sorghum (10⁵ spores/ml), (P=0.05). Spores harvested from the grains were as infective as the spores grown on V8 agar plates. Mass production of inoculum may be achieved by using larger amounts of grains and inoculating with an appropriate volume of blended mycelia as seed inoculum.
 

Objective 4: Develop genetic characterization and transformation of bioherbicide candidates as a means of enhancing efficacy and assessing environmental risk.
 

The ability to distinguish biotypes of purple nutsedge may be important for developing strategies for successful biological control of this weed. Dr. Welington Pereira, a visiting scientist from EMBRAPA-CNPH, Brazil, examined purple nutsedge accessions for variability in susceptibility to the bioherbicide isolate of Dactylaria higginsii (see Objective 1 for details of this pathosystem). Tubers of 63 purple nutsedge accessions from Brazil, India, Mexico, Puerto Rico, and United States (Florida, California, and Hawaii) were maintained in a quarantine greenhouse in Gainesville during these studies. A comparison of morphological and phenological characteristics indicated the presence of distinctive intraspecific biotypes in this collection. A suspension of 10⁶ conidia/ml was used to inoculate 21-day-old plants raised from tubers. The conidial suspension was amended with 0.5% Metamucil. Metamucil, without the fungus, was used as a control. Plants were evaluated 15 days after inoculation for susceptibility based on a visual-disease assessment scale (0=immune, 1 and 2=resistant, 3 and 4=susceptible). The experiment was done twice and each time Koch's postulates were fulfilled. Results indicated that 90.5% of the nutsedge accessions tested were susceptible, 7.9% resistant, and 1.6% immune. Molecular variability among the accessions is being characterized by RAPD analysis to determine the genetic make up of purple nutsedge populations and to identify possible genetic markers of susceptibility to D. higginsii.
 

An attempt is being made to understand the possible molecular basis for pathogenic variability in Cercospora piaropi Tharp and C. rodmanii Conway, two pathogens of waterhyacinth (Eichhornia crassipes). Sixty isolates of Cercospora spp. pathogenic to waterhyacinth were collected from the United States, Mexico, Venezuela, Brazil, South Africa, and Zambia and screened in a quarantine greenhouse in Gainesville to assess their variability in virulence toward waterhyacinth. A bioassay was used which consisted of immersing waterhyacinth leaves, attached to plants, in a suspension of inoculum. The inoculum was prepared from mycelium grown in still liquid cultures in V8 medium for 14 days at 25+ 3^oC. The mycelium was separated from the broth by filtration and blended for 5 sec at a concentration of 80 mg/ml of water. The suspension was amended with 0.5% Metamucil and 0.05% Silwet. Four replicates, each with one 2-leaved plant, were used and the experiment was done twice. Control leaves were immersed in a solution containing only the amendments. The inoculated and control plants were placed in a dark dew chamber at 25 ±2ºC for 12 h and then held in a greenhouse for 2 wk. Disease severity (percentage of leaf area necrosed) was assessed 4, 7, and 14 days after inoculation using a visual rating scale of 0-7. The isolates differed significantly in virulence (P<0.05), varying from avirulent (rating 0) to highly virulent (rating 7, leaf death). The variation in virulence among the isolates was independent of their geographic origin, spore size, or cultural differences. To determine the extent of similarities and evolutionary divergence among these isolates, DNA sequences of three conserved genes were amplified through polymerase chain reaction (PCR), sequenced, and compared through cladistic analysis. Comparisons were based on PCR fragments of -tublulin, histone H3, and elongation factor 1 genes, corresponding to 377, 288, and 459 nucleotides, respectively. All these sequences included at least one intron region. The number of isolates used in this study were 20, 15, and 9 for -tubulin, histone H3, and elongation factor 1 genes, respectively. The set of primers used for the PCR amplification of the -tubulin and histone H3 genes were obtained from the literature, meanwhile the primers for amplification of the elongation factor 1 gene was specially designed for this research. For the cladistic analysis, the software Phylogenetic Analysis Using Parsimony (PAUP) version 3.1 was used, having the species Cercospora beticola as an outgroup. In a preliminary analysis, the cladograms based on these genes were similar and included isolates of C. piaropi and C. rodmanii in the same clade. Thus, it is suggested that all isolates studied belong to a single species, according to the cladistic concept of species.
 
 
 
 

Introduction
 

1997 marks the 12th year since our research on rose rosette disease (RRD) commenced at Iowa State University. Along the way we have been able to clear up a number of issues resulting from conflicting observations reported in the earlier literature, and at the same time, we have been able to develop significant new information about the disease, its biology and epidemiology, and how it affects the plant. The data indicates that implementing RRD as a biological control to reduce the present extent of multiflora rose infestation poses no serious risk to plantings of ornamental rose if some simple precautions are observed. An important added benefit of reducing the present infestation of multiflora rose will be a reduction of this weedy plant as a source of inoculum for many of the diseases attacking ornamental roses.
 

1997 Results and Observations
 

The augmentation plot initiated in 1993 at the McNey Research Farm in Lucas County is now virtually multiflora rose - free. Only 6 of the 586 plants present in the stand at the outset are still uninfected. These will be graft-inoculated to be sure that they are not resistant to RRD.
 

The risk assessment plots containing 10 plants each of the cultivars "Color Magic, "Chrysler Imperial, "Bonica", and seedlings of multiflora rose also established in 1993 had no new infections in 1997. "Bonica" seems to be resistant to mite infection, but is susceptible to graft inoculation (That glossy leaf results from an unusually thick cuticle, and the mites may not be able to feed through it).
 

Three ranks of additional risk assessment plots spaced at 20 meter intervals (5 plots in each rank) from the augmentation site were added in 1996. These sustained no RRD infections in either 1996 or 1997 although the mite populations were still very high in the augmentation plot in 1996. The data strongly suggests that RRD has a very high lapse rate.
 

The annual disease survey (since 1992) was conducted again in the major rose plantings in Iowa and Missouri in 1997. No new diseased plants were found in the Iowa Gardens (State Center and Des Moines). A total of six newly infected plants were found in the rose collection at the Missouri Botanic Gardens (ca the same number as in 1996). The City of St Louis' massive plantings of the cultivar "Red Meidiland" (very susceptible to RRD) were found to be riddled with RRD in 1996. A large concentration of these were located along Highway 44 just north of the Botanic gardens and may have been the principle source of initial infection at these Gardens. These have been destroyed. The infected plants at the "Garden" have also been destroyed. Unless something was missed in those removals by the City, this should be the end of their RRD problem. I have yet to find a rose planting with RRD infections that does not have a source (usually infected multiflora rose plants) nearby.
 

We have isolated disease - specific nucleic acids, and several disease - specific proteins. These are now being analyzed and we are hopeful that we will soon have specific molecular probes which will enable us to identify and characterize the causal agent. These probes will also be useful as rapid tests for detecting the presence of the RRD causal agent. Once this is accomplished we will have a tool that will enable us to determine the exact manner in which the disease agent is acquired, retained, and transmitted by the vector. It will also be of great value to rose breeders in developing RRD - resistant rose
 
 
 
 

John Porter and Thomas Bewick studied effects of organosilicone surfactants and herbicides on Pseudomonas syringae pv. tagetis as a potential biological control agent for Asteraceae weeds in cranberry. The most important of these weeds are Narrow-leaved Goldenrod (Solidago tenuifolia), Pitchforks (Bidens frondosa), Ragweed (Ambrosia artemisiifolia) and White and Blue Asters (Aster ericoides and A. novi-belgii, respectively). The surfactants tested were the organosilicones Silwet L-77, Silwet-1, and Silwet-408, and LI-700 (Loveland Industries Inc. P.O. Box 1289, Greeley, Co. 80632) a non-ionic surfactant. The herbicides tested were 2,4-D and glyphosate (in the forms of Weedar 64 and Roundup Ultra, respectively). These herbicides are routinely used to control Asteraceae weeds on cranberry bogs.
 

These studies were conducted to determine the effects of surfactants and herbicides on Pseudomonas syringae pv. tagetis. Live bacterial cells suspended in sterile distilled water were added to serially diluted solutions of each of the chemicals (each diluted solution being half the strength of the one that preceded it) at rates that produced 100 cells/ml suspensions. Controls consisted of bacteria suspended in distilled water only. Plates containing solid King's Media B were inoculated with one ml samples of the suspensions and incubated at 24C. The number of bacterial colonies growing on each plate was counted 2 days later.
 

Results from this study are summarized in Figure 1. There was a significant range in the lethality of the different adjuvants and herbicides. The organosilicones L-77 and Silwet-1 had very little, if any, impact on bacterial survival, while the adjuvant LI-700 caused complete mortality when applied at rates as low as 0.333%. Silwet-408 was a little less lethal than LI-700 was, allowing a higher percentage of the bacteria to survive at a particular rate, but the highest rate that allowed 100% of the bacteria to survive was roughly the same as it was for LI-700 (i.e. roughly 0.0476%). As for the herbicides, Roundup Ultra was almost twice as potent as Weedar 64, killing all of the bacteria at rates as low as 0.4167% while Weedar 64 killed all bacteria at rates as low as 0.833%.
 

Certain concentrations of the organosilicones and/or herbicides that did not have an impact on bacterial survival could still have a serious impact on the growth rate of the bacterium. To determine if this was the case with any of these chemicals, bacteria were added to flasks containing liquid King's Media B and amended with one of the organosilicones or herbicides at a rate sufficient to produce 100 cells/ ml suspensions. The amounts of organosilicones or herbicide added to each of the flasks were the highest amounts that still allowed 100% of the bacteria to survive (determined in the previous study). The organosilicones L-77 and Silwet-1, the two chemicals in the previous study that didn't seem have an impact on bacterial survival, were applied at a rate of 1%, the highest rate they would probably ever be applied at in commercial situations. Controls for the experiment consisted of bacteria growing in liquid King's Media B only. After flasks were inoculated, they were incubated on a rotary shaker at 200 rpm. Bacteria were allowed to grow in these flasks until the suspension became turbid. Bacterial suspensions were then removed from samples of these suspensions through centrifugation, resuspended in equivalent amounts of distilled water, and then counted. This was done by determining percent absorbance of the solution using a spectrophotometer, different rates of absorbance corresponding to different bacterial concentrations. The length of time it took for inoculated flasks to become turbid was then divided by the number of generations it took for the starting concentration to become the final concentration. This calculation gave a value for the length of time it would take for a bacterium to mature and reproduce in the different suspensions (i.e. a measure of the bacterium's growth rate).
 

Results of this study are shown in Figure 2. As this figure shows, some chemicals had a rather profound effect on the rate of growth of the bacteria while others did not. Bacteria in suspensions containing LI-700 and Weedar 64 grew only half as fast as bacteria in the control did, while bacteria in suspensions containing Silwet-1 and Silwet-408 grew roughly as fast as bacteria in the control did.
 

From these studies we can conclude that some adjuvants and herbicides can have much more of an impact on bacterial growth and survival than others will. Through running studies such as these, we can quickly and easily screen out adjuvant and herbicide concentrations that will be too damaging to the bacterium to be able to do much in the way of increasing the amount of damage the bacterium will do to Asteraceae weeds.
 
 
 
 

John Gronwald, Kate Plaisance, David Johnson, and Don Wyse collaborated on a project to investigate population dynamics of Pseudomonas syringae pv. tagetis (PST) in spray-inoculated host and nonhost plants. Seedlings tested were in the first or second leaf stage, depending on host. Soybean and sunflower were grown from seed, while Canada thistle were clones from a single rootstock. The isolate used was David Johnson's 1-502a, originally isolated from Canada thistle. Spray applications were made with an air-powered sprayer using Silwet L-77 at 0.3% v/v. Volume was approximately 1.1 ml/seedling.
 

Inundative foliar application of PST in an aqueous suspension containing an organosilicone surfactant provides good control of sunflower, variable control of Canada thistle, but has no effect on soybeans. Population dynamics of endophytic PST following foliar application were evaluated in host (sunflower, Canada thistle) and nonhost (soybean) plants. The endophytic PST population in sunflower leaves, measured 60 min after spraying with 10⁹ cfu/ml, was 10⁷ cfu/g fresh wt. The endophytic population increased rapidly during the subsequent 24 h, reaching a peak of 10⁹ cfu/g fresh wt, which declined slightly during the subsequent 48 h. Chlorophyll content of sunflower leaves that emerged following application of 10⁹ cfu/ml PST was reduced by about 90%. Spraying 10⁹ cfu/ml PST on soybean leaves resulted in a low endophytic population (10⁵ cfu/g fresh wt) due to entrapment of spray droplets by leaf trichomes. When trichomes were removed by rubbing ethanol on the leaves with a Kimwipe, it was possible to establish higher PST populations in soybean. In contrast to sunflower, the endophytic PST population in soybean leaves did not increase during a 72 h interval after application, nor was chlorophyll content of developing leaves reduced. Canada thistle plants sprayed with 10⁹ cfu/ml PST exhibited variable endophytic populations and injury as measured by chlorosis of developing leaves.
 

Usefulness of These Findings: Population densities of PST in host plants were correlated to observed herbicidal efficacy. Studies of population dynamics in soybean demonstrated that trichomes limit foliar entry of sprayed PST. The cell density also does not increase greatly after inoculation in this nonhost plant. Host specificity may be due to a combination of factors. The nonhost status of soybean does not preclude survival of the bacteria in vivo.
 
 
 
 

David R. Johnson recently departed from the University of Minnesota to join Encore Technologies, a microbial products company that manufactures Collego® and other biocontrol products. At Encore, D. R. Johnson and Fredrick Schendel are conducting research on industrial-scale fermentation and stabilization of Pseudomonas syringae pv. tagetis. In an effort to advance product identity, fatty acid profiles (FAME), Biolog data, and MIDI 16S Ribosomal RNA sequence data were obtained for PST. Data from PST collections from different geographical regions and hosts revealed that diverse isolates were nearly identical in those features we can measure. The bacterium is amenable to liquid culture, and was routinely grown to a density of 3 X 10¹⁰ cfu/ml in minimal glucose-nitrate medium.
 

Usefulness of These Findings: These studies have advanced the knowledge needed for mass-production of a PST product.
 

Work Planned for Next Year: Efforts to optimize yield from culture and stabilize the organism are ongoing. We are also examining application technology and in vivo PST population dynamics in projects with our University of Minnesota cooperators Roger Becker, John Gronwald, and Kate Plaisance. Gronwald et. Al gave some information on this project elsewhere in this report. We will continue to provide PST cultures and technical expertise to anyone wishing to collaborate on PST research.
 
 
 
 

1. Evaluate selected pathogens as potential bioherbicides.
 

Hemp sesbania was effectively controlled in soybean (Stoneville, MS) and rice (Stuttgart, AR) test plots with a 1:3 (v:v) unrefined corn oil /Colletotrichum truncatum (COLTRU) formulation containing .2%. Silwet L-77. A dried formulation consisting of dextrose and hydrated silica gel suspended in an experimental oil (Quoil) or in a United Agri Product proprietary oil (M+ oil) were also effective in controlling hemp sesbania. PESTA formulations containing conidia and microsclerotia (produced at SRRC) provided some control when applied either PE or PPI, but were significantly less effective postemergence fungus/oil formulations. Control of hemp sesbania in cotton was significantly reduced by fungicide seed treatments and in-furrow fungicide and insecticide treatments. Sicklepod was effectively controlled for the second year in soybean test plots when treated with a split application of C. gloeosporioides formulated with Silwet L-77 and unrefined corn oil. Common cocklebur (Xanthium strumarium) was effectively controlled in field test plots with Alternaria helianthi suspended in 1:1 unrefined corn oil and 2% Silwet formulations.
 

2. Overcoming environmental constraints to infection and disease development.
 

Unrefined corn oil-Silwet L-77 emulsions containing Alternaria helianthi spores increased pathogenesis to several cocklebur biotypes under moisture-limiting conditions. The bacterial pathogen Pseudomonas syringae pv. tagetis also infected and killed these biotypes when formulated with Silwet L-77 and unrefined corn oil.
 

Objective 3. Develop systems for mass production of stable bioherbicides.
 

Rice formulations of COLTRU were tested for viability and virulence to hemp sesbania. The fungus was highly virulent to weed seedlings even after 8 years storage at -20 C, and only slightly less virulent at 4 C. Survival of COLTRU in formulations stored at room temperature was significantly less than at the lower temperatures.
 
 
 
 

Hamed K. Abbas, conducted greenhouse studies of PST and tagetitoxin on various hosts, including common cocklebur (Xanthium strumarium L.) The isolate used was 1-502a, obtained from David R. Johnson. Ten cocklebur biotypes from MS, IL, OK and TX , including MSMA-resistant and imazaquin-resistant biotypes, were spray inoculated at the 2-3 leaf growth stage with an aqueous suspension of 1 X 10⁸ cfu/ml plus Silwet L-77 (0.2% v/v). All biotypes were susceptible to PST, but disease severity varied with biotype. There was no apparent correlation between herbicide resistance and disease severity.
 

When purified tagetitoxin was tested in cocklebur, a dose-response relationship was observed. Tagetitoxin was delivered to 2-3 leaf cocklebur seedlings by wounding the stem and applying an 8 ul droplet of 2, 4, 8, 31.5, 62.5, 125, or 250ng/ul. Apical chlorosis symptoms were observed on all plants after 48 h. Chlorophyll content in leaves of treated plants was reduced 5-98%, depending on dose. In plants treated with 31.5 ng/ul tagetitoxin, chloroplast abnormalities in third leaves were observed 24 h after treatment.
 

Responses of other weed seedlings to tagetitoxin were also tested. Sensitive species included Canada thistle (Cirsium arvense), common ragweed (Ambrosia artemisiifolia), horseweed (Conyza canadensis), safflower, wild and cultivated sunflowers (Helianthus annuus) and tropical soda apple (Solanum viarum). Morningglories (Ipomoea sp.), sicklepod (Cassia obtusifolia), and velvetleaf (Abutilon theophrasti) were weakly sensitive.
 

Usefulness of These Findings: These studies demonstrated efficacy of PST against common cocklebur, an important agricultural weed. Imazaquin and MSMA-resistance in cocklebur was not correlated to PST susceptibility, indicating PST might be a valuable tool for controlling herbicide-resistant cocklebur. Studies with tagetitoxin demonstrate that plant response to toxin is variable, and plants outside the organism's natural host range are affected.
 
 
 
 

Nina Zidack studied effects of Pseudomonas syringae pv. tagetis (PST) on houndstongue (Cynoglossum officinale), and discovered that necrotizing agents such as pelargonic acid (Scythe®) greatly enhance PST efficacy. The PST isolate used was David R. Johnson's 1-502a, obtained from. Mycogen Corporation. An aerosol sprayer (Crown sprayer) was used for applications in greenhouse experiments, and hand pump, air-pressurized sprayers were used in field experiments. Silwet L-77 surfactant rates were 0.2% v/v in greenhouse studies, 0.25% v/v in the field. Inoculum density was 1 X 10¹⁰ cfu/ml. Greenhouse experiments showed that the optimal rate of Scythe® was 4% v/v, applied at 100 GPA.
 

A strong synergistic relationship was shown for Pseudomonas syringae pv. tagetis (PST) and the contact herbicide Scythe® (pelargonic acid) when applied to houndstongue. When Scythe® followed PST application, immediate necrosis resulted, and subsequent regrowth was severely chlorotic. The synergy seemed to be the result of the contact herbicide necrotizing the mature leaves of the plant, thereby reducing the plants' ability to produce carbohydrates for regeneration of healthy tissue. Tagetitoxin produced in the necrotized tissue inhibited chloroplast development in the new tissue. Plant reserves were depleted by supporting chlorotic regrowth that produced substantially reduced levels of photosynthate. This synergy is not unique to Scythe®. Similar synergy was also demonstrated plants damaged by frost or steam.
 

Field research in 1997 produced very promising results for control of houndstongue with PST in combination with Scythe ®. In June, a replicated experiment on a natural infestation of houndstongue was initiated near the East Gallatin river in Southwest Montana. Treatments included PST alone, PST plus Scythe®, Scythe® alone, and an untreated control. Ten plants in each treatment were tagged on the day of application . One month after spraying, plots were rated for necrosis and chlorosis of regrowth on a 0 to 5 scale with 0 being completely healthy and 5 being dead. Data are presented in Table 1.
 

Table 1. Average efficacy rating of PST treatments on houndstongue 1 month after treatment
 

Treatment

Average efficacy

 

PST alone	0.66
PST plus Scythe	3.05
Scythe alone	0.94
Untreated control	0

 

In the PST plus Scythe treatment, 37% of the tagged plants were dead. In the PST alone treatments, no plants were dead and in the Scythe alone treatments 16% were dead. The most promising aspect of this study was observed in fall of 1997. On September 25^th, many of the plants in the field plots that had appeared to recover earlier in the season had died. Most plots with PST plus Scythe had 0-2 live houndstongue plants. There was also some houndstongue mortality in the PST alone and Scythe alone plots.
 

Work Planned for Next Year: Whether or not the PST treatments caused early fall senescence or actual plant death will be determined in spring 1998, when the 2-year-old plants emerge.

Usefulness of These Findings: Efficacy of PST was demonstrated on houndstongue, an important rangeland weed. The synergy observed with necrotizing agents may increase herbicidal efficacy of PST and/or expand the host range.
 
 
 
 

Joseph C. Neal, conducted greenhouse studies of PST on several important Asteraceae weeds. He also evaluated the relationship of organosilicone adjuvant (Silwet L-77) rate to PST effectiveness, as well as the effect of clipping plants as a means of facilitating infection. The isolate used was 1-502a, obtained from David R. Johnson.
 

Two greenhouse experiments were conducted. The concentration of PST inoculum in both tests was 10⁸ cfu/ml applied with a hand held spray bottle at 188 ml per m². Test plants were about 6 weeks old from either seed, or rhizome or stolon pieces. Seed propagated plants were dandelion (Taraxacum officinale), eclipta (Eclipta prostrata), galinsoga (Galinsoga ciliata), and horseweed (Conyza canadensis). Vegetatively propagated species were English daisy (Bellis perennis), mugwort (Artemisia vulgaris), oxeye daisy (Chrysanthemum leucanthemum), yarrow (Achillea millefolium), and yellow hawkweed (Hieracium pratense). In the second test only seedling plants were tested, including bull thistle (Cirsium vulgare), eclipta, galinsoga, common groundsel (Senecio vulgaris), and western salsify (Tragopogon dubius).
 

No disease symptoms were observed on plants inoculated with PST without Silwet L-77. Symptoms and percent control were increased by increasing Silwet concentration from 0.2% to 0.4%. Clipping plants following the inoculation reduced the effectiveness of treatments.
 

Considerable variability in susceptibility was observed among species. In the most susceptible hosts, symptoms developed within one week. Most species demonstrated signs of recovery by 3 weeks after inoculation. Common groundsel was controlled 80% to 100% within 1 week of inoculation and that level of control was maintained for the 5-week duration of the test. Eclipta was controlled 60% to 70% for the duration of the test. Galinsoga was controlled 60 to 80% at 2 weeks but by 5 weeks 0% to 50% control was observed, for 0.2% and 0.4% Silwet L-77, respectively. Bull thistle, mugwort, and western salsify were controlled 50% to 80% for about 2 weeks, but recovery was evident by the 3^rd week. By the 5^th week, control of these three species was between 0 and 35%. Dandelion and horseweed were symptomatic but control never exceeded 32%. Of the vegetatively propagated plants tested, only yellow hawkweed control was greater than 50% after 5 weeks.
 

Usefulness of These Findings: Results from these tests are encouraging for potential use of PST for controlling some common weeds of horticultural crops including common groundsel, eclipta, galinsoga, bull thistle, mugwort, and yellow hawkweed. Knowledge about the host range and efficacy of PST was gained. Continued evaluations under field conditions and with repeated applications are warranted.
 
 
 
 

Terry Wheeler, Peter Dotray and Taminah Sheikh studied the effects of Pseudomonas syringae pv. tagetis (PST) on wollyleaf bursage (Ambrosia grayi) an important perennial weed in Texas. The isolate used was David Johnson's 1-502a obtained form Mycogen Corporation. In 1997 field tests, bacterial inoculum concentrate (10⁷ to 10⁸ cfu/ml) was diluted by 1/40th, mixed with 0.25 % v/v Silwet L-77 and applied to runoff using a Solo gasoline engine powered backpack sprayer. Plants were rated at 24 and 48 hrs and weekly. Tests conducted consisted of spraying plots I) once a month starting in April and terminating in October; ii) with a dilution series of the bacteria from concentrate to 1/10,000 dilution; iii) at different times during the day, starting at 8:00 AM and terminating at 8:00 PM; iv) at two week and one week intervals, versus no sprays. Another experiment was conducted with field application equipment that had been adjusted to deliver different water pressure and volume as the sprayer was tractor driven over the test area. Volumes varied from 50 to 320 gallons/acre and pressure from 20 to 80 psi. This experiment was repeated twice. The monthly application of PST resulted in good symptom development during April and moderate symptom development during May. The weeds recovered from chlorosis by 4-5 wks after application. There was poor symptom development when weeds were sprayed for the first time during months after May. There was significant attrition of the weed plots which were sprayed in April, beginning in late July and continuing in August. This drop in weed density coincided with an increase in apical chlorosis. By September, 80% of the surviving weeds in the plots sprayed in April showed apical chlorosis.
 

Attempts to apply the bacteria with normal field spraying equipment was unsuccessful. A backpack sprayer (Solo gasoline powered) was used to apply PST (successfully) in fall of 1996 and spring 1997, but after Roundup was used in the sprayer, applications no longer resulted in symptom development. Agricultural pesticides such as Roundup may affect the ability to apply with field equipment.
 
 
 
 

Objective 1: Evaluation of selected pathogens as bioherbicides
 

Survey for fungi and bacteria
 

Surveys for fungal pathogens of Canada thistle, wild oats and green foxtail are continuing. Both foliar and soilborne pathogens are being evaluated. In addition, selection criteria and screening methods for fungal pathogens was developed for the weed biocontrol group. Choice of survey and evaluation methods were established and a new screening and evaluation process was developed. The screening system considers the growth characteristics in culture, infection site on host, inoculum level and severity of disease, and consistency of pathogen-host response.
 

Research efforts for isolating and screening rhizobacterial strains for biological control of grassy weeds from weed suppressive soils is continuing. A culture collection of over 200 bacteria has been established; three bacterial strains are being extensively evaluated under controlled environment and field conditions.
 

Objective 3: Development of systems for mass production and stabilization of herbicides
 

1. Formulation - foliar fungal agents
 

Principles for selection of adjuvants used in bioherbicide formulations are being developed. Fungal species from each of the genera Colletotrichum, Phoma, Fusarium, and Alternaria were selected to characterize their compatibility with common laboratory surfactants. Tween 20, Tween 40, Tween 80, Tergitol 9, Tergitol 10, sorbitol, and gelatin were evaluated. Conidial germination varied with the surfactant, surfactant concentration, the fungal pathogen, and the inoculum density. Tween 40 and Tween 80 were compatible with all fungi; gelatin was compatible with Phoma and Fusarium; sorbitol was compatible with Fusarium and Alternaria. Self-inhibition of conidia was released in Colletotrichum with Tween 80; gelatin had similar effects on Phoma. Tergitol was detrimental to conidial germination for most fungal species, while Tween 40 and Tween 80 had the mildest effects. In general, fungi belonging to the Coelomycetes were more sensitive to adjuvants than those belonging to the Hyphomycetes.
 

2. Formulation - rhizobacteria
 

In 1997, field trials were conducted to evaluate several granular formulations, including peat prills and clay formulations for control of green foxtail and wild oats. Generally, formulation of bacteria in peat prills provided significant reductions in weed emergence (30-35% reduction) and aboveground biomass (30% reduction) while the clay formulations did not reduce weed emergence or biomass. However, one bacterial strain formulated in the clay formulation reduced biomass of green foxtail by approximately 30%, not weed emergence was not affected. Efforts on improving these formulations are continuing and field trials will be conducted in 1998.
 

3. Mass production of fungi
 

As most of the fungi we are working with do not grow well in liquid culture, technology is being developed to produce these agents on solid substrate culture. A key factor in sporulation of these agents depends on the amount of aeration (oxygenation) and length of time in the liquid phase prior to spreading on the solid surfaces.
 

4. Mass production of bacteria
 

Fermentation studies were conducted to determine the nutritional requirements of specific rhizobacterial isolates with weed suppressive properties. Shake flask-culture tests incorporating various nutrients that lead to significant biological control activity in combination with mass production of bacterial cells was undertaken. Specific carbon and amino acid combinations that provide optimum activity of each bacterial isolate are currently being selected. In addition, phase kinetic studies are being conducted to evaluate the growth rate of bacteria for scale-up from shake flask to 5 and 10 L fermentors.
 

Additional staff:
 

Dr. Chang Yong Chen received his Master¢s degree (1990) and Ph.D. (1994) in plant pathology from the University of Toronto under the direction of Dr. M.C. Heath. From January 1994 to May 1995, he worked with Dr. Barbara Howlett as a PDF in the Department of Botany, University of Melbourne, Australia where he studied the genetics of Leptosphaeria maculans, the blackleg fungus of Brassica spp. and investigated the molecular basis of resistance of different Brassica spp. From June 1995 to August 1997, Dr. Chen worked with Dr. Seguin-Swartz (Agriculture and Agri-Food Canada, Saskatoon) as an NSERC Visiting Fellow on the molecular interactions between the blackleg fungus and oilseed plant. In September 1997, he joined the weed biocontrol group where he has been working as a term research scientist to develop genetically improved fungal pathogens as bioherbicides.
 

Dr. Sarah Green, undertook her Ph.D at Lincoln University, New Zealand where she investigated the potential of a bioherbicide for control of the giant buttercup (Ranunculus acris) in dairy pastures. After successful completion of her Ph.D in 1995 she took up a post-doctoral fellowship at the Nova Scotia Agricultural College with Prof. Glen Sampson to work on the development of a bioherbicide for control of dandelion and other weeds in turfgrass. This project involved collaboration with the universities of Guelph and McGill and industrial partners Dow AgroSciences and Saskatchewan Wheat Pool. Sarah arrived at Agriculture and Agri-Food Canada Saskatoon Research Centre in April 1998 as an NSERC Government Laboratory Visiting Fellow to work with Karen Bailey, Sue Boyetchko and Knud Mortensen on biological control of Canada thistle and grassy weed species.