CHEMICAL CONTROL OF WILD GARLIC (AIIIum vineale L.) IN WINTER WHEAT (Triticum aestivum L.) AND THE MOVEMENT AND METABOLISM OF AN EFFECTIVE HERBICIDE 2- METHDXY-B. 6-DICHLORDBENZDIC ACID (DICAMBA) Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY LARRY K. BINNING 1959 MlChleU‘: 54' 1:- Uni ersicy This is to certify that the thesis entitled Chemical Control of Wild Garlic (Allium vineale L.) in Winter Wheat (Triticum aestivum L.) and the Movement and Metabolism of an Effective Herbicide 2-Methoxy-3,6-Dichlorobenzoic Acid (chamba) presented by Larry K. Binning has been accepted towards fulfillment of the requirements for _PhD— degree in M1106 Li) limiflfk. meemif Major professonj [3 Date (79144/1 /9é- 9 0469 -‘ ABSTRACT CHEMICAL CONTROL OF WILD GARLIC (Allium vineale L.) IN WINTER WHEAT (Triticum aestivum L.) AND THE MOVEMENT AND METABOLISM OF AN EFFECTIVE HERBICIDE 2-METHOXY-3,6-DICHLOROBENZOIC ACID (DICAMBA) BY Lar ry K. Binning Wild garlic (Allium vineale L.) is a problem in winter wheat. The aerial bulbils, of the same size and weight as the wheat kernel, are difficult to remove from the harvested grain and reduce wheat quality due to the garlicky odor. Studies were undertaken to find effective control measures for wild garlic in winter wheat and to determine the effect of herbicides on the wild garlic plants. Wild garlic was controlled with an application of 0.25 1b/A of 2-methoxy—3,6-dichlorobenzoic acid (dicamba) plus 0.5 lb/A of 2,4-dichlorophenoxyacetic acid low volatile ester (2,4-D LVE). This treatment reduced scape production and caused bulb shrink- age. Application at the resumption of spring growth of the wheat was effective in wild garlic control, but caused wheat injury. The Larry K. Binning fully tillered stage of wheat growth was the safest time to apply the treatment. Laboratory results indicated that the bulbs and bulbils acted as metabolic sinks during development. Application timed to correspond to the development of these structures was most effective. Movement of 14C labeled dicamba in the wild garlic plants was toward these areas. Dicamba moved in the xylem and the phloem with xylem movement being much faster than movement in the phloem. More dicamba moved into the plant from leaf treatment than from treatment on the scape. The maximum amount of labeled material was found in the plant 7 days after treatment. Herbicide concentration in the plant decreased 1 month after treatment which was possibly due to volatilization and/or decarboxylation of dicamba. Pretreatment of the plants with Z-chloroethanephosphonic acid (ethrel) prior to treatment with dicamba caused a reversal of dicamba movement. Ethrel possibly stimulated the lower bulb area and caused it to become an active metabolic sink. Wild garlic does not degrade dicamba rapidly. Seven days after treatment there was no apparent metabolism, however, after 1 month of exposure to dicamba there was some metabolism, but the parent compound accounted for most of the radioactivity. CHEMICAL CONTROL OF WILD GARLIC (Allium vineale L.) IN WINTER WHEAT (Triticum aestivum L.) AND THE MOVEMENT AND METABOLISM OF AN EFFECTIVE HERBICIDE 2-METHOXY-3,6-DICHLOROBENZOIC ACID (DICAMBA) By W a Larry K: Binning A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Crop Science 1969 ACKNOWLEDGMENT The author wishes to express his sincere apprecia- tion to Dr. w. F. Meggitt for his willing assistance through- out this study and in the preparation of the manuscript. Appreciation is also expressed to Dr. C. M. Harrison, Dr. S. K. Ries, Dr. E. Everson, and Dr. D. Penner for their critical review of the manuscript. The advice of Dr. D. Penner in regard to the laboratory phase of the manuscript is also greatly appreciated. Recognition is also given to Robert C. Bond for giving freely of his time to aid in the completion of field “studies for the manuscript. Special gratitude is given to my wife, Dorothy, for her unassuming attitude through four years of study and for her assistance in typing and reviewing the manu- script. ii TABLE OF CONTENTS LIST OF TABLES ..... . ......... . ..... ....... LIST OF FIGURES ..................................... iii 10 20 58 61 Table 10. ll. LIST OF TABLES Wild garlic control and wheat yield from 4 herbicide treatments applied in the spring, 1966 ................................. . ........ Wild garlic control and wheat yield from fall application of herbicides in 1965 ............. Wild garlic control and wheat yields from.spring application of herbicides in 1966 ............. Wild garlic control from fa11 application of herbicides on fallow soil in 1965 ............. Wild garlic control from spring application of herbicides on fallow soil in 1966 ............. Wild garlic control and wheat yield from herbi- cide applications made at 2 stages of wheat growth in the fall of 1966 .................... Wild garlic control and wheat yields from.herbi- cide applications made at 4 stages of wheat growth in the Spring of 1967 ............ ‘ ...... Wild garlic control and wheat yields from herbi- cide applications made at 2 times in the spring of 1968 Quantitative determination of 14C dicamba move- ment in wild garlic pretreated with ethrel 7 days prior to dicamba application ,,,,,,,,,,,,,,,,,, Rf values for extracts and standards chromato- graphed in 2 solvent systems .................. Percent distribution of recovered 14C labeled dicamba from wild garlic ...................... iv Page 21 22 24 25 26 28 29-30 33 50 55 57 Figure II. III. IV. VIM LIST OF FIGURES Upper: Scape treatment with 14C dicamba har- vested after 1 day of exposure. Lower: Scape treatment with 14C dicamba harvested after 2 days of exposure. Treated plants on the left and the autoradiograms on the right ........ Upper: Scape treatment with 14C dicamba har- vested after 7 days of exposure. Lower:' Scape treatment with 14C dicamba harvested after 1 month of exposure. Treated plants on the left and the autoradiograms on the right ........ Upper: Leaf treatment with 14C dicamba har- vested after 1 day of exposure. Lower: Leaf treatment with 14C dicamba harvested after 2 days of exposure. Treated plants on the left and the autoradiograms on the right ........ Upper: Leaf treatment with 14C dicamba har- vested after 7 days of exposure. Lower: Leaf treatment with 14C dicamba harvested after 1 month of exposure. Treated plants on the left and the autoradiograms on the right ........ Upper: Treatment of vegetative plant with 14C dicamba harvested after 1 day of exposure. Lower: Treatment of vegetative plant with 14C dicamba harvested after 2 days of exposure. Treated plants on the left and the autoradio- grams on the right ......................... Upper: Treatment of vegetative plant with 14C dicamba harvested after 7 days of exposure. Lower: Treatment of vegetative plant with 14C dicamba harvested after 30 days of exposure. Treated plants on the left and the autoradio- grams on the right ......................... Page 37 38 39 4O 41 42 F igure Page VII. Treatment of a vegetative plant in an immature stage of growth. Treated with 14C dicamba and harvested after 7 days of exposure. Treated plant on the left and the autoradiogram on the right ...................................... 44 VIII. Plants used in the exudation study. The upper photograph is of the treated plant and auto- radiogram and the lower photograph is of the control plant grown in the same nutrient solution ................................... 46 IX. Upper: Dicamba 14C treated plant on the left with autoradiogram on the right. Lower: Plant pretreated with ethrel 7 days prior to dicamba 14C treatment. Treated plants on the left and autoradiograms on the right. Both plants harvested 2 days after dicamba treat- ment ....................................... 48 X. Upper: Dicamba 14C treated plant on the left with autoradiogram on the right. Lower: Plant pretreated with ethrel 7 days prior to dicamba 14C treatment. Treated plants on the left and autoradiograms on the right. Both plants harvested 7 days after treatment .... 49 XI. Scan of TLC plate developed in 8:1:1 butanol, ammonia, and water (v/v/v) ................. 53 XII. Scan of TLC plate with dicamba 14C enriched extracts developed in 8:1:1 butanol, ammonia, and water (v/v/v) .......................... 54 vi INTRODUCTION Wild garlic (Allium vineale L.) is an immigrant to the United States from the European Continent where it has been a problem for centuries. It was introduced into the United States in the mid Seventeenth or early Eighteenth Century, probably in crop seed (l4). Extensive areas of the United States, primarily east of the Mississippi River, are infested with the plant which causes major problems in small grain and pasture production. The offensive odor caused by allyl sulfide in wild garlic can be imparted to plant and animal products making them undesirable. Wild garlic is a problem in wheat production in Michigan. Aerial bulbils which are of the same size and weight as wheat kernels are produced on scapes that stand as tall as wheat. The weight and size of the bulbils and height of the scapes make field separation in the threshing ‘process difficult and drying is required to shrink the 'bulbils to facilitate separation. Wild garlic does not present a severe problem in othexr crops, but will reproduce and remain a problem when wheat: is grown again in the rotation. Cultural control is not totally effective. Success- ful herbicide treatments must reduce scape production and cause bulbil shrinkage. Recommended herbicides will not eliminate the problem or approach the 100% control necessary in wheat. The objectives of this study were to find effective control measures for wild garlic, and to determine the translocation, sites of accumulation, and the metabolism of an effective herbicide. LITERATURE REVIEW Wild garlic plants are of two types, scapigerous, or seed stalk producing, and nonscapigerous, or vegetative. Four bulb types are formed by these plants; scapigerous plants producing soft offsets, hard-shelled and aerial bulbils, nonscapigerous plants producing a central bulb, and hard-shelled offsets. Soft offsets or central bulbs are produced 1 per plant, hard—shelled bulbs several, and aerial bulbils 50-200 per plant (10, ll, 12). In wheat, aerial bulbils cause the immediate prob- lem in the production year. Hard-shelled bulbs may remain dormant in soil up to 6 years. Wild garlic seed production in Michigan is not a problem (12, 31). Most bulbs germinate in the fall of the year and complete growth in spring with maturation in July. Some aerial bulbs may germinate in spring. Scapigerous plants arise primarily from central or soft offsets, some from lutrd-shelled bulbs, and rarely from aerial bulbils (23, 25). Wild garlic is not a serious competitor with crop Plants (17) . Lazenby (16) in his work noted that plant competition reduced production and establishment of wild garlic. High fertility and coarse soil tilth may also reduce growth of wild garlic plants (16). Lazenby also found that bulbs placed at the surface of the soil did not produce plants as well as those that were buried to moderate depths, with the exception of hard-shelled bulbs which germinated best at the soil surface. Bulbs have the capac- ity to produce plants when buried deep in the soil. Soft offsets and central bulbs emerge 60% of the time from a 16- inch depth (18). Hard-shelled bulbs buried below 8 inches will remain dormant and aerial bulbs do not consistently emerge from more than 4 inches (18). In a cultivation study by Lazenby (19), fall culti- vation did little to reduce the total production of wild garlic plants. Cultivation buried the vegetative portions of the plant and caused a delay in plant maturity. Spring cultivation caused the greatest reductions in plant numbers and weight. Wild garlic bulbs when uprooted have been shown to redistribute stored reserves from the main bulbs to offsets vflrich.cou1d remain viable for 5-6 months (20)- Lazenby (21) indicated that cutting to remove topgrowth reduced PrOChJction, but did not kill the plants, although it could eLhniniate aerial bulbil production. Cultural control was implemented by deep plowing fall and spring in an effort to bury as many bulbs as pos- sible (32). Scape production decreased with depth of plant- ing or burying, although plants still survived vegetatively. Spring tillage caused serious damage to the garlic population. Row crop production added to vegetative destruction (32). 2,4-Dichlorophenoxyacetic acid (2,4-D) has been the most effective herbicide for garlic control (13). Populations have been reduced, but not eliminated, by using 1 lb/A. 2,3,6-Trichlorobenzoic acid (2,3,6-TBA) was found to be more effective by Davis et a1. (9) in re- ducing bulb production than 2,4-D. Davis et a1. (11) also found that 2,3,6-TBA caused greater morphological effects on wild garlic bulbs, such as necrosis, softening, and bulb shrinkage than 2,4-D. 2-Methoxy-3,6-dichlorobenzoic acid (dicamba) used at relatively high rates has also been effective. Repeated applications of herbicides in spring and fall over a period of years have caused greater re- ductions of stand than single applications (1, 15). Klingman (15) found that a fall application was most ef- fective in arresting bulb production. Hard-shelled bulbs were not affected by repeated treatment unless they were induced to germinate (1). Treatment to cause germination induction may be possible. Beasley (2) found that rhizome buds on Johnsongrass were induced to germinate by a 2- chloroethanephosphonic acid (ethrel) treatment. Herbicide uptake, movement, and metabolism have been affected by the physiological state of the plant and external conditions which might contribute to this state (28, 30). Sargent (29), using 2,4-D as an example, found that young tissue was much more susceptible to herbicide entry than old tissue. Water stress enhanced 2,4-D uptake. Wetting agents and high temperature also increased uptake. The data of Sargent (28) showed that low pH increased foliar uptake, indicating that ionic and nonionic molecules of herbicide entered the plant. The pH effect was greater in the light than in the dark, indicating both physical and active up- take of herbicide molecules. Inhibition of plant metabolism caused alterations in the uptake rate, indicating active uptake. Typical 2,4-D foliar entry into susceptible species, as indicated by Blackman (3), proceeded rapidly at first, stopped after about 2 hours, and reverted out of the plant tissue by the end of 24 hours. Saunders (30) showed that in resistant species uptake continued in a linear fashion. If weak auxin-type herbicides were applied to susceptible species, uptake was linear. Wheat is a 2,4-D tolerant riff?! "m' w J .1» I" 4." species and uptake continues in a linear fashion over time. This linear uptake by resistant species is probably related to some form of detoxification of the parent herbicide molecule. Examples would be glucoside formations, con- jugations to some sugar molecule, removal of functional groups from the parent molecule to make it less toxic, or the addition of some functional group to the parent molecule for the same purpose (30). Dicamba is included in the class of auxin-type herbicides where uptake and movement in plants were shown to be somewhat similar to 2,4-D. Dicamba moved freely in the xylem and phloem tissues (8). The physiological state of the plant determined the type of movement (22). Quimby's work (27) indicated that dicamba was effective in wild buckwheat control at 0.25 lb/A applied at the 2 leaf stage. In this trial, wheat was in the 2-4 leaf stage of growth at the time of treatment. Species tolerance or susceptibility was related to uptake and sub- sequent metabolism of a compound (27)- Wheat was found to metabolize dicamba at a significantly higher rate than wild buckwheat. Wheat was found by Broadhurst (4) to metabolize dicamba to 5-OH-dicamba as do many of the resistant plant species. After 18 days, all of the original l4C dicamba was converted to the metabolite. Some metabolism by wild buckwheat to salicylic acid was noted by Quimby (26), but very little, if any, to 5-0H-dicamba. Chang (6) found the major metabolite in tartary buckwheat to be S—OH dicamba, but only 13% of the total label was the metabolite after 40 days of exposure. Metabolism, therefore, proceeds at a relatively slow rate in this susceptible species. Most of lthis metabolite was isolated from the plant as a conjugate with a sugar molecule. -Magalhaes (22), working with purple nutsedge and 14C labeled dicamba, found that movement depended on numerous factors- The light regime under which the.plants were grown greatly altered the distribution in the plant. Under low light, 14C dicamba was uniformly distributed throughout the topgrowth. The physiological stage of the plant also caused various distribution patterns. If the plants had seed heads, the movement of label was:primarily.to.the seed head area; Vegetative and actively growing plants exhibited an even distribution.of label throughout the upper leaves. This indicated that if an active metabolic sink existed, translocation proceeded in that direction. Purple nutsedge is a dicamba-susceptible species and metabolism of dicamba could not be observed in this species. The parent compound was the only material re- covered in the extractions. In exudation studies, the parent compound was observed to be exudated. Chang (5), working with Canada.thistle, found dicamba movement to be primarily toward the leaf tips. Accumulation increased in the younger tissue. Metabolism proceeded at a low rate so that 54 days after treatment, 37% was found in a form other than free dicamba. This may have been a conjugate of the parent compound. Substantial amounts of free dicamba still remained. Decarboxylation did take place, but very slowly. After 54 days of exposure, about 8% of total radioactivity was-collected as 14002. Chang (6) in his work also found a small amount of C02 produced in buckwheat, but it proved to be an.insignificant amount. MATERIALS AND METHODS Field and laboratory studies were initiated in the fall of 1965 and continued into 1969. The field studies were conducted in southwestern Michigan on a silt loam soil. Four hundred and fifty lb/A of a 5-20-20 fertilizer was applied prior to sowing and 30 1b/A of N as ammonium nitrate was topdressed in spring. Wheat (Triticum aestivum L.), variety Genesee, was sown for fall and spring treatment with a six-foot grain drill. »Plots were laid out for 2 experiments. The first was a rotation study using 12 by 50 foot plots with 3 replica- tions in a randomized block design- The treatments as active material in lb/A were: 0.25 of dicamba plus 0.50 of 2,4-D low volatile ester (LVE); 1.0 N-oleyl-l,3- propylenediamine salt of 2,4-dichlorophenoxyacetic acid (2,4-D oil soluble amine); 0.125 4-amino-3,5,6-trichloro- picolinic acid (picloram); and 0.50 dicamba. Treatments were applied to wheat at the fully tillered stage with a tractor-mounted boom-type plot sprayer in 23 gallons of water at 30 psi. The same 10 ll treatments were also applied after corn harvest and prior to soybean planting on the same plots in subsequent years over the rotation period. In addition, the corn and soy- beans were treated with recommended herbicides at the time of planting and cultivated as required for row crop produc— tion. Wheat plots were rated visually for control and injury. Yield data was collected as a measure of crop injury and scape counts were made at harvest to determine the level of control. Yield data and scape counts were obtained from the 5 center rows of each plot. The second experiment with wheat was a timing and screening study. Fall and spring applications were made on separate plots replicated 3 times in a randomized block design. The plots were 7 by 25 feet. Yield and scape counts were taken as above. The treatments as-active material in 1b/A were: 0.125 picloram; 0.50 dicamba, 0.50 2,3,6-trichlorophenylacetic acid (fenac); 1.0 2,4-D oil soluble amine; 1.0 2,4-D LVE; 1.0 2,3,6-TBA; 0.25 dicamba plus 0.50 2,4-D LVE; 0.25 fenaclplus 0.50 2,4-D.LVE; 0.25 2,3,6PTBA.plus 0-50 2,4-D LVE; 0.25 dicamba plus 0.25 fenac; 0.125 picloram plus 0.25 dicamba; and no treatment. 12 An experiment was started on fallow soi1 using in- creased rates of herbicides applied spring and fall on separate plots in a randomized block design replicated 3 times. The plots were 6 by 25 feet and scape counts were taken as a measure of control. The treatments as active material in lb/A were: 0.25 picloram; 1.0 dicamba; 0.50 dicamba plus 1.0 2,4-D LVE; 1.0 fenac; 0.50 fenac plus 1.0 2,4-D LVE; 1.0 2,3,6- TBA; 0-50 2,3,6-TBA plus 1.0 2,4-D LVE; 2.0 2,4-D oil soluble amine; 2.0 2,4-D LVE; 3.0 2-chloro-4-ethy1amino-6- isopropylamino-l,3,5—triazine (atrazine); and no treatment. In 1967, the rotation study was continued with corn as the second crop in the rotation. Picloram was discontinued due to lack of control and crop tolerance. The remaining treatments applied after corn harvest were the same as the 66 spring rotation treatments on wheat. Atrazine at 2 lb/A was used to control broadleaf weeds in the corn. A time of application study was initiated on wheat in the fall of 1966. The plots were 6 by 30 feet and arranged in a split plot design with 4 replications. Treat- ment times corresponded to the following stages of wheat growth: , ..— .. . . __-_ m was «a ' - inf-“r - ' 13 Fall treatments: 1. Two leaf stage. 2. Completion of fall growth. Spring treatments: 3. Resumption of spring growth, early April. 4. Fully tillered stage, 6-8 inches in height. 5. Late tiller, early boot stage. 6. Late mature stage - 10 days prior to harvest. The treatments applied at the above times as active material in lb/A were: 1.0 2,4-D LVE; 0.50 dicamba; 0.25 dicamba plus 0.50 2,4-D LVE; 0.25 fenac plus 0.50 2,4-D LVE; 0.50 dicamba plus 0.50 2,4-D LVE; 0.75 dicamba; and no treat- ment. Yield data was taken to measure crop injury and scapes were counted as an indication of control. In addi- tion, bulbil weight per scape was recorded to determine the degree of control. Sterilant plots were established in 1966 and rated for 2 years visually to determine the effectiveness of various herbicides on eradication of garlic- These plots were 12 by 50 feet and replicated twice. Garlic control was determined by a 0-10 rating scale, 0 being no control and 10 being complete kill. The treatments as active material in lb/A were: 10.0 2,3,6-TBA; 1.0 picloram; 10.0 atrazine; 10.0 Zechloro- 4,6-bis(ethylamino)-s-triazine (simazine); 20.0 3-(3,4— dichlorophenyl)-l,l—dimethylurea (diuron); 6.0 dicamba; 2.0 3-amino-l,2,4-triazole (amitrole) plus 10.0 simazine; 10.0 5-bromo—3-sec-butyl—6-methyluracil (bromacil); 0.50 picloram; 10.0 fenac; 8.0 2,6-dichlorobenzonitrile (dichlo- benil); 2.0 3-tert-butyl-5-chloro-6-methyluracil (terbacil); 4.0 terbacil; 8.0 dichlobenil incorporated; and no treat- ment. In 1967 wheat was planted in the fall for an evalua- tion of spring treatments and timinglof spring treatments. Fall applications were dropped because of extensive crop injury. The plots were 6 by 30 feet in a randomized complete block design with 4 replications. The times of treatment vvere as follows: 1. Resumption of active spring growth. 2. Fully tillered stage. Tflne treatments as active material in lb/A were: 1.0 2,4—D 'LAIE at the fully tillered stage; 0-50 dicamba plus 0.50 2el4-D LVE at growth resumption; 0.50 dicamba.plus 0.50 2,14-D at the fully tillered stage; 0.25 dicamba plus 0.50 2:¢L-D.LVE at growth resumption; 0.25 dicamba plus 0.50 I 15 2,4-D LVE at the fully tillered stage; 0.25 dicamba at growth resumption plus 0.50 2,4-D LVE at the fully tillered stage; and no treatment. Yield data was recorded for crop injury and bulbil weight per scape recorded as degree of control. The rotation study was continued in 1968. Rotation treatments were applied prior to preparing the soil for soybean planting. These treatments were the same as those applied after corn. The soybean variety Chippewa 64 was used. Two pounds of 3-amino-2,5-dichlorobenzoic acid (amiben) plus 1.0 pound of 3-(3,4-dichlorophenyl)-1-methoxy- l-methylurea (linuron) were applied per acre for broadleaf weed control. Wheat was planted in the fall of 1968 on the rotation area again to complete the rotation. Harvest data will be taken at maturity. Protein content of the wheat treated with various lierbicides.was determined by the micro-kjeldahl method (24). This method was later correlated with the Udy Protein Iknalyzerl. Further testing of protein content in subsequent yVBars used the latter method. Baking.quality tests-were made 'b37 the Soft Wheat Quality Laboratory at Wooster, Ohio. 1 [hdy Analyzer Company, Boulder, Colorado. 16 Experiments using 14C labeled dicamba were begun in 1968 with wild garlic plants and bulbs dug from the field. The 140 labeled dicamba (2-methoxy-3,6-dichloro- benzoic acid) was labeled in the carboxy carbon and obtained from the Velsicol Chemical Corporation. The specific activ- ity was 1.8 no per um. The plants used were grown in the greenhouse. Those plants that produced scapes were sep- arated and used in the first experiment. The experiment was arranged so that one-half of the plants would be treated on the youngest fully-developed leaf and one-half would be treated on the scape. All plants were treated on the same day, and about at the time of bursting of the spathes, or umbel enclosure. The treatment rate was 0.1 no of 14C dicamba per plant applied in 10 ul of solution. The solution contained 0.1% of X77 as a spreader since wild garlic has a waxy cuticle. This was applied to an area approximately 1.5 cm by 0.2 cm which was bounded by a lan- olin enclosure. The plants in each series were allowed to continue growth for 1, 2, 7, and 30 days after treat- nnent. At these respective times, duplicate plants were Ilarvested from each series. These plants were washed free fdrom the soil, frozen immediately in dry ice, and freeze dltied, a procedure outlined by Crafts and Yamaguchi (7). 17 In the second experiment, only vegetative plants with germinated offsets were used. Rate of treatment was the same as above. Only 1 series was involved since the plants were all vegetative. They were harvested and handled as the above series. These plants were grown in a growth chamber with a 16 hr day at 21 C. A plant that was in an immature stage of growth was treated and prepared as above to determine movement at this stage. The plant was grown for 7 days and freeze dried. A study was initiated to determine whether root exudation occurred. Two plants were placed in each of two 300 m1 erlenmeyer flasks and 250 ml of Hoagland's No. II nutrient solution was added. The solution was aerated with a porous clay aerator and the air pumped by a small diaphragm (zirculating pump. The experiment was done in the growth <2hamber with a 16 hr day at 21 C. Treatment was made on 1. of the plants in each flask. The remaining plant was tine check plant. These were allowed to grow for 10 days Ennd were then harvested and freeze dried. One remaining labeling study was conducted with Plants treated with ethrel to determine if herbicide move- Ineult patterns could be changed by the chemical alteration of plant metabolism. 18 Six plants were pretreated with ethrel at a con- centration of 1000 ppm. This solution was sprayed on the plant to the point of runoff. Seven days after ethrel treatment, they were treated with the standard l4C dicamba treatment of 0.1 no in 10 ul of solution. Duplicates were then harvested at 2 and 7 days after treatment. At each time 1 plant not treated with ethrel, but treated with 14C dicamba, was harvested as a check. These plants were then frozen in dry ice and later freeze dried. In all of the above l4C labeling experiments after the plants were freeze dried they were prepared for auto- radiography. The plants were remoistened in a humidity chamber for flexibility. They were then mounted on blotter paper and held in place with a casein glue. The mounts were air dried and pressed against masonite to get a flat surface for good contact with X-ray film. After being pressed, the mounts were placed in cassettes with no screen X-ray film and exposed for 15 days. After exposure, the film was developed by standard developmental procedures (7). Qualitative data on the plant metabolism of dicamba Vmas obtained by hot ethanol extraction for chromatographic éinalysis, the remainder hydrolyzed with 4.0 M HCL to break énny glucoside linkages that might have been formed as done 19 by Broadhurst et al. (4). The hydrolysate was then extracted with chloroform and the aqueous phase discarded. These extracts were then chromatographed on microcrystalline cellulose thin layer plates in solvent systems of butanol, ammonia, and water in the proportions of 8:1:1 (v/v/v) and in butanol, ethanol, and water in proportions of 2:1:1 (v/v/v). Autoradiograms of these plates were made in the first experiments and in subsequent experiments the plates were scraped into vials in 1 cm segments per vial and counted in a liquid scintillation counter. The scintilla- tion solution contained 0.3 g of dimethyl POPOP, 5.0 g of PFC per liter of toluene. Quantitative data was obtained by separating the plants into 3 portions, foliage, treated area, or area of application of 14C labeled dicamba, and the root system. The portions were then ground in a Wiley mill through a 40 mesh screen. A sample of this ground material teas then combusted in an oxygen atmosphere (Schoeninger (combustion technique) and the carbon dioxide collected in El solution of ethanol amine plus absolute ethanol in a lsatio of 1:2 (v/v). An aliquot of this was removed and C Ounted (33) . RESULTS AND DISCUSSION The rotation study indicated that picloram used at .125 lb/A did not control garlic and was injurious to wheat (Table l). Dicamba at 0.5 lb/A did not cause a wheat yield reduction, although changes in plant structure were observed. Garlic control was not adequate to avoid dockage. 2,4rD oil soluble amine at 1 lb/A did not reduce yield or cause any noticeable effects on wheat, but control was again not adequate. Dicamba at 0-25 lb/A plus 2,4-D LVE at 0.5 lb/A gave the best results of these treatments. Bulbil weights were not taken in this study. From observa- tions made at the time of harvest, the dicamba-plus 2,4-D combination caused a greater amount of bulbil shrinkage than all other treatments. There was an 83% reduction in scape production. This, coupled with the shrinkage, Inade-the treatment adequate. 20 21 Table l.--Wild garlic control and wheat yield from 4 herbi- cide treatments applied in the spring, 1966. Rate Yield Scape2 Herbicide Lb/A Bu/A Reduction, % Dicamba + 0.23 + 48 a1 83 2,4-D LVE 0.50 Dicamba 0.50 46 ab 67 2,4-D Oil 301- 1.0 43 be 54 uble Amine Picloram 0.125 32 d 39 No Treatment - 43 bc - 1Means followed by different letters are significantly different at the 5% level. 2The control contained 22,400 scapes/A. There was only a slight yield reduction of wheat for some fall treatments (Table 2). This was mainly due to cold weather which stopped any further growth and resulted in reduced injury to the crop. The fall application of dicamba plus 2,4-D LVE was effective, but not as effective as spring application. Bulbils from plants that escaped fa11 treatment were observed to be plumper than those that escaped spring treatment. Table 2.--Wild garlic control and wheat yield from fa11 application of herbicides in 1965. 22 Rate Yield1 Scape Herbicide Lb/A Bu/A Reduction, % Picloram 0.25 36 e 100 Dicamba 0.50 49 she 98 Picloram + 0.125 + 43 bed 98 Dicamba 0.25 Picloram 0.125 43 bcd 97 2,3,6-TBA + 0.25 + 41 de 95 2,4-D LVE 0.50 Dicamba + 0.25 + 46 abcd 91 2,4-D LVE 0.50 2,3,6-TBA 0.50 46 abcd 89 2,4-D LVE 1.0 40 de 86 Dicamba + 0.25 + 52 a 86 Fenac 0.25 2,4-D Oil 801- 1.0 42 cde 70 uble Amine Fenac + 0.25 + 44 abcd 59 2,4-D LVE 0.50 Fenac 0.50 50 ab 42 Pic Treatment 43 bcd - m 1 Ddeans followed by different letters are significantly different at the 5% level. 2 TFhe control contained 23,000 scapes/A. 23 The most effective spring treatment was the combina- tion of dicamba plus 2,4-D LVE (Table 3). This treatment resulted in a 98% reduction in scape production and the bulbils that remained were shrunken. Without shrinkage of bulbils, control was not considered adequate. The stand of wild garlic in all of the plot areas was a natural stand. In some cases, the check value from fall application of herbicides on fallow soil (Table 4) is lower than some treated plots. This was due to stand variation. Stimulation by herbicide treatment did not seem probable. Dicamba plus 2,4—D LVE was the most effective treatment. 2,4-D LVE was almost as effective as the dicamba plus 2,4-D LVE combination, as indicated by scape number, but those plants that escaped 2,4-D injury produced well formed bulbils. Atrazine had no effect on wild garlic. Spring application of atrazine, as fall applica- tion, had no effect on garlic (Table 5). Dicamba plus 2,4-D LVE gave the best control of wild garlic. Dicamba alone at 1 pound also gave adequate control. The results from this series of experiments in- dicated that the combination of dicamba plus 2,4-D LVE was as effective or more effective than either herbicide alone. The combination was better than all other rTable 3.--Wi1d garlic control and wheat yields from spring application of herbicides in 1966. 24 Rate Yield1 Scape2 Herbicide Lb/A Bu/A Reduction, % Dicamba + 0.25 + 44 ab 98 2,4-D LVE 0.50 Fenac + 0.25 + 47 a 93 2,4-D LVE 0.50 2,4-D Oil 801- 1.0 37 bc 91 uble Amine 2,4-D LVE 1.0 41 ab 90 Picloram 0.125 23 d 86 Picloram 0.25 17 d 83 2,3,6-TBA + 0.25 + 42 ab 80 2,4-D LVE 0.50 Dicamba 0.50 40 abc 78 Picloram.+ 0.125 + 33 c 74 Dicamba 0.25 Dicamba + 0.25 + 41 ab 72 Fenac 0.25 2,3,6-TBA 0.50 45 a 44 Fenac 0.50 45 a 40 No Treatment 42 ab - ‘— lMeans followed by different letters are significantly different at the 5% level. 2The control contained 22,500 scapes/A. 25 Table 4.--Wild garlic control from fa11 application of herbicides on fallow soil in 1965. Rate Scape1 Herbicide Lb/A Reduction, % Dicamba + 0.50 + 98 2,4-D LVE 1.0 2,4-D LVE 2.0 96 Dicamba 1.0 71 2,3,6-TBA + 0.50 + 61 2,4-D LVE 1.0 2,3,6-TBA 1.0 55 2,4-D Oil 801- 2.0 45 uble Amine Fenac 1.0 0 Fenac + 0.50 + 0 2,4-D LVE 1.0 Picloram 0.25 0 Atrazine 3.0 0 No Treatment 1The control contained 12,200 scapes/A. 26 Table 5.--Wi1d garlic control from spring application of herbicides on fallow soil in 1966. Rate Scape1 Herbicide Lb/A Control, % Dicamba + 0.50 + 98 2,4-D LVE 1.0 Dicamba 1.0 96 Fenac + 0.50 + 94 2,4-D LVE 1.0 2,4-D LVE 2.0 85 Picloram 0.25 79 2,4-D Oil 801- 2.0 68 uble Amine 2,3,6-TBA 1.0 42 2,3,6-TBA + 0.50 + 17 2,4-D LVE 1.0 Atrazine 3.0 4 Fenac 1.0 0 No Treatment - - 1The control contained 27,100 scapes/A. 27 herbicides used, if the visual observation on bulbil shrinkage was included in the evaluation. These results led to time of application studies with the most effective herbicides from the screening studies, and various rates of dicamba combined with 2,4-D LVE (Table 6). The 2-4 leaf stage of wheat growth was chosen because of wheat tolerance to dicamba at this time. Results from the previous year indicated that control with- out wheat injury could be attained by late fall treatment. The second application was made at the completion of fall growth. Neither fall application was satisfactory. At the 2-1eaf stage, the wheat plants were tolerant to the herbi- cides, but control was poor. The reverse was true of late fall treatment. The late fall treatment corresponded to the time of application in the previous year which was successful. The variation in results may have been due to weather conditions after treatment. Growing conditions were favorable after treatment in 1966 and crop injury was high. Fall treatments were not considered feasible since weather conditions after late treatment were too critical for con- sistent results and early fall treatment was ineffective. Treatments applied at the resumption of spring growth did not give adequate control (Table 7). Wheat .<\mmnmom ooo.m pmdwwudoo HouucOo anew .Hm>ma Nm mnu um ucmuommwp haucmowmficwfim mam mumuuma udouommwp an posoafiom GEDHoo oamm mnu cw mucosa m ow.H . m mm m om.H u m am I udofiumcuH 02 c oe.H cm 0 mm m em.H o one mm ma.o mesmean om.o m>a o-e.N c es.H ow 0 cm e o~.a he econ mm + om.o + ensecan om.o m>a n-e.m e HN.H oe c em e sm.H o cone cm + mN.o + emcee n om.o m>q n-e.~ c me.a as o em e me.a 0 cc mm + mN.o + eesecam e em.H he a mm c mm.H o am an om.o eeeecan e me. am e ea e em.H o e on o.H m>a m-e.~ seem gem N macaw. acme Hem N macaw «\ea ceaoeeuem manhu Houucoo <\Dm macho Houufiou <\Dm oumm .uB macaw ummnz .u3 Nmamom Huwo£3 ammum £u3ouw Hash mo Gowuwdmfioo nuBouw mo wMMUm mama quN .ooma mo Hamm can :« SHBOHw umosa mo mowmum N um meme mdowumowamam upwownuos scum pHmflh ummna can Houucoo owawww pHHBII.o mHan .<\wmmmom ooo.¢ poawmuaoo Houudoo o£H¢ .Hm>ca Nm osu um ucmummmww hauamofimwawwm one muouuoa uamuommwp an poaoaaom mamm2m .usmanoaw>op mo mwmum coca mum esp um mm? nu309w mo ammum pmuoaawu aaasm mnam .nuaouw o>wuow headway uses? dmnz weaken mnu aw mega man ou pmpcoamouuoo nuzouw mo coaunasmmmH ce o~.a - e on e oN.H .- e an . - eecsuecue 02 es mo.o me e mm m mm.o me e mm ms.o seaweed om.o m>a a-e.~ e Ho.o mm m mm e mm.o me e am + om.o + eeaccao om.o m>a n-e.~ a, one sw.o he e oe e sm.a a e am + mN.o + emcee 2 om.o m>a m-e.~ es oe.o am e on e.¢w.o as e um + m~.o + eeaeoen em He.o me e an e me.a mm e an om.o eeeeoan c HN.N . mm e am e Nw.o am e em o.a ., m>n n-e.~ eeem sea a ease» eeem sea a eases <\ee eeeeeeeem manna Houuaoo <\5m macho Houuaou <\Dm ouwm .uB madam umofiz .uB «venom mucosz NtonHHwH NHHDm anusouo mo aowumfismom .noma mo wdflumm msu aw susouw puma? mo mowmum e um meme muowumowanam mtwownumn Scum mpaowk uses? was Houufioo owaumw UHHBII.N manna e o~.H - e em es om.a - e an - eccaeeeea ez m oo.H .o 9 mm A mm.o mm a mm mn.o mnsmewm om.o m>a n-e.~ m mm.o o a mm a om.o mm as on + om.o + unawoaa om.o m>d nn¢.N m mm.o mm 3 mm no mN.H mm a mm + mN.o + awash om.o m>a o-e.~ mw m mw.o o n em as mn.H ow pm mm + mN.o + unawown m mo.a o as on as mw.o mm A an om.o spamofln a oo.H ma 3 mm m on.o mm mm mm o.H m>A n:¢.~ use: Hem N macaw use: new N macaw «\nq ovaownhom macho Houudoo <\am madam HouucOo <\Dm oumm .u3 madam uwosz .uB endow ummnz umm>Hmm ou Howum when 0H uoom haumm ecueaaae saasm cued Aeeeeeueoov--.s capes 31 'yields were reduced on the average from check yields. The best control occurred with dicamba plus 2,4-D LVE at 0.5 lb/A of each. Treatments applied to wheat at the fully tillered stage resulted in the highest wheat yield and best wild garlic control. Some changes in the wheat plant caused by rates of dicamba above 0.25 lb/A were observed, but these did not affect yield. Dicamba plus 2,4-D LVE combination at 0.5 pound of each per acre gave the best control. Dicamba at 0.75 lb/A effectively controlled garlic, but damaged wheat plants. These treatments caused only slight yield reduction. The dicamba plus 2,4-D LVE combination at the 0.25 plus 0.50 lb/A rate was the best treatment for consistent control with minimum crop injury. 2,4-D LVE applied at the fully tillered stage did not adequately control garlic when considering bulbil weight per scape. The garlic scapes that were formed after 2,4-D LVE treatment appeared to be normal. This was not true when dicamba was included in the treatment. When the treatments were applied at the late tiller- early boot stage, more injury to wheat was noticeable and control was not adequate. The herbicide treatments per- formed as they had at the resumption of spring growth. 32 The last time of treatment in this experiment was 10 days prior to harvest. At this time, shrinkage of the aerial bulbs occurred, but was not apparent from the data. The bulbs had been formed at the time of treatment. The herbicides caused a fusion of the bulbs at the base of the umbel and a softening of the bulb. The shrinkage was severe enough so that field separation of the aerial bulbils in the threshing process was possible. It was not a feasible treatment unless emergency conditions prevailed since physical damage to the crop during application caused a substantial yield loss. These results indicated that spring application gave the most consistent wild garlic control. The resump- tion of spring growth and the fully tillered stage of wheat were the most favorable times for treatment. The 0.25 pound of dicamba plus 0.50 pound 2,4-D LVE combination gave the most consistent results when considering yield and control. In 1968, only 2,4-D LVE and dicamba were used because of the degree of effectiveness observed in previous experi- ments (Table 8). Time (E) in the table corresponds to the stage at which wheat is most tolerant to dicamba and time 1 4 Hilll lilurlfi Ilwlvrwuruura EOHIH WDHWHK UNWCSD Drum. HOMUEOU U.H.HHMH.M DHHgllimw WHQWH i .Hm>ma Nm map on udwumumwp haucmowmwcmflm mum muouuoa ucwummmwp an pm3oaaom mammze .ummts mo £u3ouw wdfiumm w>auom mo wcwadwmmp Ho £u3ouw wcaumm mo cowumEfimmu mtu mmumfiwwmmm m IIII. .nuaouw mo owmum on no pmumaawu haasm can mmumawwmmnm .<\mmmmom oom.m pmdwmuaoo Houuaoo maHH m o.m¢ I HH.H I . I ucmaummHH 02 one o.wm I mo.a m om.o caseeam . eon o.mm mm ee.o m ms.o eeseeam om.o m>a aIe.N cee o.wm ms oe.o m + mm.o + eeseeao om.o m>a n-e.~ e o.Nm mm mm.a a + om.o + eeeecao i. a om.o m>a m-e.~ q. es o.oe ooa 00.0 m + mN.o + eeseeao om.o m>a nIe.N ec o.mm ooH oo.o a + mN.o + seaweed om.o m>a oIe.N eee o.em ooH oo.o mm + om.o + eeeecao eee o.em OOH oo.o Na o.H m>a o-e.~ - «\sm a cameo e: ceaeeeaaeaa «\ea ceacaeucm mama» Houucou owaumo mo oEaH Qumm ummcz Hmamom \uB .woma mo wfiwunm can aw mmaHu um mme muowumowaaam mpfiownumn Scum mpamfim umwta cam Houuaoo ofianmw pHMBII.w oHQMH 34 (L) most tolerant to 2,4-D LVE. One combination treat- ment was split, the dicamba applied early and the 2,4-D late. This treatment gave the best wild garlic control with minimum wheat injury. It involved two spray applica- tions and for this reason was less desirable than single applications. The late application of this combination gave excellent control, but wheat injury was observed. If the same treatment was applied early, control was not as good, but yield of wheat was not reduced severely. Wheat had more tolerance to 2,4-D LVE applied early than it had to dicamba applied late. A single application was con- sidered most favorable due to convenience, cost of applica- tion, and low crop value. It was, therefore, better to apply the combination at either the resumption of spring growth of wheat or at the fully tillered stage. Results from Sterilant plots to determine if eradication in non-crop areas was feasible indicated that of the 13 treatments applied, bromacil at 10 1b/A was the only compound giving complete control after 2 years of observation. Picloram at 1 lb/A and fenac at 10 lb/A gave 85% control and 2,3,6-TBA at 10 lb/A gave 50% control. Terbacil and dichlobenil produced changes in plant form, but did not remove the plants. All other treatments failed to produce a response that was apparent through visual ratings. 35 The results of laboratory analysis for protein content of wheat by the micro-kjeldahl method and the Udy method indicated that herbicide treatment had only a slight effect on wheat quality. From 2 years of analysis, only the wheat from the picloram treatment was significantly higher in protein content. Picloram treated wheat was of the poorest baking quality, but was still acceptable. Wheat quality was not significantly affected by herbicide treatment. Yield was the most important criteria for selection of herbicides to be used on wheat. Changes in wild garlic morphology from dicamba were visibly greater than where other treatments were used. It has been shown that wheat is capable of metab- olizing dicamba (4, 26, 27). Since dicamba was found to be effective in wild garlic control and could be metab- olized by wheat, it was of interest to determine the move- ment and metabolism of this herbicide in wild garlic. The plants used here had produced scapes and bulbils and were in the final stages of development. At this stage, the bulbils provided an excellent sink for photosynthates. Dicamba 14C movement was in the direction of these aerial bulbils. Both xylem and phloem.movement was noted from autoradiograms of these plants. Xylem bI 1111 En 36 ‘movement appeared to be much faster than phloem movement. The basal bulbs were not labeled. Growth of these bulbs was completed at the time of treatment; therefore, they did not act as metabolic sinks. Two series of plants are shown. The first series was treated on the scape, Figures I and II, and the second on the youngest fully extended leaf, Figures III and IV. Observations on these plants indicated that after 1 day, treatment on the scape was more effective in translocation of dicamba to the aerial bulbils. After 7 days of treat- ment, little difference was noted. This occurred because both xylem and phloem movement carried dicamba to the aerial bulbil area. When the leaf was treated, phloem movement was primarily responsible for the initial movement out of the treated leaf. Once out of the treated leaf, dicamba could be swept up the xylem to the aerial bulbils. When treatment was made directly on the scape, the xylem trans- port was in a direct manner, from the treatment spot to the bulbil area. The difference between scape and leaf treat- ment could not be detected on the autoradiograms after 1 month of exposure. Results from the next experiment were obtained using vegetative plants that had germinated offset plants growing attached to the basal bulb area (Figures V and VI). SCAPE TREATMENT 1 DAY EXPOSURE ‘c\\ 10‘ SCAPE TREATMENT 2 DAY EXPOSURE Figure I. Upper: Scape treatment with 14C dicamba harvested after 1 day of exposure. with C dicamba harvested after 2 days of exposure. Treated plants on the left and the autoradiograms on the right. Scape treatment 38 SCAPE TNLATMEVT 7 DAY EXPLSURE \ ails” : SCAPE TREATMENT 30 DAY ‘,..... /\ “rum: /' \ .' \ \.~\\\“~K \éé \ a‘ i .. :9 *‘ Figure 11. Upper: Scape treatment with 14C dicamba har- vested after 7 days of exposure. Lower: Scape treatment with 14C dicamba harvested after 1 month of exposure. Treated plants on the left and the autoradiograms on the right. $§GQ§T Figure III. LEAF TREATVEVT 1 DAY EXPOSURE LEAF TREATMENT ‘ 2 0 AT EXPOSURE / 5 (f C \- . 3 Upper: Leaf treatment with 14C dicamba har- vested after 1 day of exposure. Lower: Leaf treatment with C dicamba harvested after 2 days of exposure. Treated plants on the left and the autoradiograms on the right. 4O LEAF TREATVEVT 7 DAY EXPCSURE 'v'A'It Figure IV . LEAF TREATVEVT ,-—r. 30 DAY EXPOSURE Upper: Leaf treatment with 14C dicamba har- vested after 7 dzys of exposure. Lower: Leaf treatment with C dicamba harvested after 1 month of exposure. Treated plants on the left and the autoradiograms on the right. 41 LEAF TREATMENT _ 1 on r EXPOSURE LEAF TREATMENT I " - 2 DAY EXPOSURE Figure V. Upper: Treatment of vegetative plant with 14C dicamba harvested after 1 day of ex osure. Lower: Treatment of vegetative plant with 4C dicamba harvested after 2 days of exposure. Treated plants on the left and the autoradiograms on the right. 42 LEAF TREATVEVT 7 DAY EXPOSURE LEAF TREATVENT 30 DAY EXrLLURE Figure VI. Upper: Treatment of vegetative plant with 14C dicamba harvested after 7 days of exposure. Lower: Treatment of vegetative plant with 1 C dicamba harvested after 30 days of exposure. Treated plants on the left and the autoradiograms on the right. 43 If the basal bulbs were being set at the time of treatment, there would have been a metabolic sink similar to the aerial bulbils in the previous experiment. These plants had matured beyond the point of basal bulb development and no accumula- tion of herbicide was noted in the basal area. It was noted that dicamba moved from the main plant to germinated offsets. This movement was proposed to be through remain- ing xylem elements between the parent plant and the offset. Movement was slow as indicated by the elapsed time between treatment and appearance of dicamba in the offset plant. Movement toward the base of the plant was probably in the phloem and most of the upward movement in the xylem. The upward movement was much faster and, therefore, no accumula- tion occurred in the lower portion. The basal bulbs of the offsets were not acting as metabolic sinks at this time since the new growth of the offset was translocating stored reserves from the bulb to the plant. Figure VII indicates the type of translocation of the herbicide which occurred if a plant was setting basal bulbs. An offset had also germinated from this plant. At that physiological stage, the basal storage area aeted as a metabolic sink. The connection between the parent plant and the offset was still functional. The offset was actively 44 it» nunm . -m. 1 DAY EXPOSURE Y» Figure VII. Treatment of a vegetative plant in an immature stage of growth. Treated with 14C dicamba and harvested after 7 days of exposure. Treated plant on the left and the autoradiogram on the right. 45 growing and the bulb was not being filled at the time of treatment so movement was toward the leaf tip. The results from an exudation study to determine if the label that was observed in the offset was trans- ported through direct xylem connection or if it was exudated and taken up through the root system of the offset are shown in Figure VIII. Exudation of dicamba by wild garlic apparently was not significant. No label could be detected in the nutrient solution in which the plants were grown. These solutions were extracted and assayed for radioactivity 140 dicamba in a liquid scintillation counter 10 days after treatment. The treated plant was heavily labeled and the test plant grown in the same nutrient solution was not labeled. The lower photograph is of the test plant, and the autoradiogram did not show any label. This experiment indicated that dicamba was moving through remaining xylem connections between the main plant and the offset and was not exuded and taken up through the root system of the offset. The physiological stage of growth was the major influence on the movement of dicamba in the plant. The following experiment was done to determine if chemical induction of physiological change would cause alterations lll Figure VIII. 46 EIUDATION Exp. Him; EXL‘L-ATIOI EXP. '- u CNLCK "K Plants used in the exudation study. The upper photograph is of the treated plant and auto- radiogram and the lower photograph is of the control plant grown in the same nutrient solution. 47 in translocation patterns. Vegetative wild garlic plants were used. They were pretreated with ethrel 7 days before dicamba application. Ethrel treatment caused an overall reversal of dicamba movement (Figures IX and X). More of the label moved downward from the dicamba treated area in the ethrel treated plants. In the untreated check, dicamba movement was toward the leaf tip. The ratio of upward to downward movement shows that with ethrel treatment, dicamba movement is about equal in both directions from the treatment spot 2 days after treat? ment (Table 9). In the check without ethrel, upward move- ment is 8 times greater than downward movement. The same relationship holds 7 days after treatment with a 2:1 ratio in the ethrel treated plant and a 10:1 ratio in the check plant. This increase in downward movement was significant in that once the herbicide accumulates in the bulb area, the bulbs will degenerate. The most probable explanation for this reversal of movement was that ethrel caused an induction of basal axillary bud development. The develop- ing buds constitute a metabolic sink and movement of dicamba was toward this sink. Autoradiograms of the ethrel treated plants were observed to have dicamba 14C in the root system. These were the only plants in which movement into the root system was observed. Figure IX. 48 x t-E'" I ~EATLL 2'.” §I' LP ETHul T :AT U .’ cAT tAN‘. 1‘. Upper: Dicamba 14C treated plant on the left with autoradiogram on the right. Lower: Plant pretreated with ethrel 7 days prior to dicamba 4C treatment. Treated plants on the left and autoradiograms on the right. Both plants harvested 2 days after dicamba treatment. Figure X. 49 NCN‘EI‘ REL TEEAT 7 DAY CPECK 1U {THREL TREATED 1 DAY «in. b “A Upper: Dicamba 14C treated plant on the left with autoradiogram on the right. Lower: Plant pretreated with ethrel 7 days prior to dicamba 4C treatment. Treated plants on the left and autoradiograms on the right. Both plants har- vested 7 days after treatment. 50 Table 9.--Quantitative determination of 14C dicamba move- ment in wild garlic pretreated with ethrel 7 days prior to dicamba application. Days After DPM/mg above DPM/mg below Dicamba Treatment Treatment Ratio of Treatment Treatment Spot (A) Spot (B) A—B Dicamba 2 770 1,041 1:1 after ethrel Dicamba 2 1,966 242 8:1 alone Dicamba 7 3,304 1,617 2:1 after ethrel Dicamba 7 1,248 123 10:1 alone 1 Determination made above the treatment spot. 2 Determination made below the treatment spot. 51 In support of the movement studies, qualitative results on the metabolism of dicamba by wild garlic were obtained. Plants from the various harvest dates were extracted with ethanol and the extract chromatographed against standard compounds. The 7 day and 1 month old plants were of greatest interest. After 7 days, no dicamba metabolites could be detected. This was confirmed by auto- radiography and liquid scintillation counting of chromato- graphed plant extracts. The extract lagged somewhat behind the dicamba standard on the chromatogram. By sequentially increasing extract concentration, the lag was found to be due to concentration of the extract. The spot appearing at a somewhat lower Rf than the dicamba standard was dicamba. Further confirmation was made by mixing standard dicamba with the extract. Only 1 spot appeared and it lagged behind standard alone. After 7 days of exposure to 14C labeled dicamba, only free dicamba was observed to be present. This indicated that glucosides were not present after 7 days and that metabolites in appreciable quantities had not been formed. Extractions were made of plants exposed to 14C dicamba for 1 month. Variable results were obtained depending on the plants which were used. The greatest 52 amount of label extracted in all cases was still present as dicamba. A metabolite was observed at Rf .93, when chromatographed in 8:1:1, butanol, ammonia, and water (v/v/v), from the first plants that were extracted. This peak decreased in intensity after hydrolysis, but was still present. The identity was unknown, but could possibly have been salicylic acid. From another extraction, dicamba was the main peak and accounted for most of the label (Figure XI). Acid hydrolysis caused a decrease in inten- sity of the peak at Rf .83. This indicated that a con- jugate could have been present. Where the same extracts were enriched with dicamba standard, the main peak increased in intensity and the other peaks remained the same (Figure XII). This indicated that the main peak was dicamba. The peak at Rf .125 corresponded to 5-OH-dicamba. 5-0H-Dicamba spots were detected by iodine color development. Color development was required for the standard since 14C 5-OH- dicamba was not available. When the extracts were chro- matographed in 2:1:1, butanol, ethanol, and water (v/v/v), the dicamba and the 5—0H-dicamba could not be separated. An unknown peak, however, did appear and was observed to diminish when hydrolyzed. Table 10 gives the Rf values for these results. 53 .Hocmusn Huanw cw vmaoam>mp mumaa DAB mo doom .A>\>\>v Munoz paw .mwaoEEm 2.0.10 20¢...— mmEgr—kmo O. ~ ~ to .. to ~ .\ . ~ ... .Hx gunman w c N o .. - - I . o J 3% 4/".\ . z s I x on I s x 00. on. ....... .5 3585-106 .III .amefgoq l'l .36 3:25 CON ' .2m 3635 cam WdO 54 .A>\>\>v nouns 65m .mwaoEEw .Hocmusn Huauw cw comoHo>ov muomhuxo 8528 P: 258% fit. 3.3a one we swam .HHx magma...— z_o_mo 20E mmmhwghzmu w. v. N. o. o @ ¢ N ...... .5 25393-..“ III. .Bm 3:85 + 6.3.123 - I I am 3585 + .3 .255 II .2m 3585 on on. OmN WdO | . I ., F . L 55 Table lO.--Rf values for extracts and standards chromatographed in 2 solvent systems. Solvent System A1 B2 Material Rf Values Rf Values Dicamba .88 .66 5-OH-Dicamba .85 .13 Ethanol Extract Peak 1 - 5-OH-Dicamba - .13 2 - Dicamba .89 .56 3 - Unknown .55 .83 Acid Hydrolysis Peak 1 - 5-OH-Dicamba - .l3 2 - Dicamba .89 .56 3 - Unknown - .88 12:1:1, Butanol, ethanol, and water (v/V/v). 28:1:1, Butanol, ammonia, and water (v/v/v). The discrepancy observed between plants that were exposed for 1 month may have been due to the physiological stage of development of individual plants. Dicamba re- mained as the main compound observed after ethanol extrac- tion and accounted for 62% of the label. After acid hydrolysis, 84% of the recovered label was dicamba. Con- jugates of dicamba plus plant constituents accounted for the increase in dicamba after hydrolysis. This indicated 56 that wild garlic did not have an efficient degradative system for this compound. After an extended period of time, it did metabolize dicamba; however, only a small portion of total compound present was metabolized. These results are supported by the work of Chang (5), Quimby (26,27), and Magalhaes (22) where they found that suscep- tible species metabolized dicamba very slowly while re- sistant species, such as wheat, were capable of rapid dicamba degradation. Quantitative results from the collection of radio- active C02 after combustion in an oxygen atmosphere in- dicated that the percent of material found in the plant as compared to that which remained at the point of treatment increased with time after treatment. Seven days after treatment the maximum amount had left the treatment area. This amount varied from 70-80% with appreciable variability from plant to plant. This was attributed to the physiologi- cal state of the plant at the time of treatment. More material was translocated from the treated area when the plants were leaf treated than when they were treated on the scape. The root systems of the plants did not accumulate appreciable amounts of labeled dicamba. The maximum amount recovered from the root system of the plants was 3% (Table 11). ,1. 57 Table ll.--Percent distribution of recovered 14C labeled dicamba from wild garlic. 'I'VVFV *7— 7T ‘Point of kecoveryof 14C Application Days After Dicamba From (%): of 14C Dicamba Treatment Root - Shoot Leaf l 0.2 20.0 Leaf 2 0.2 68.0 Leaf 7 0.6 89.0 Leaf 30 3.0 72.0 Scape l 0.2 43.0 Scape 2 0.2 51.0 Scape 7 0.2 21.0 Scape 30 0.2 39.4 The total percent recovered from the plants decreased with tflme after treatment. This was thought to be due to 2 factors, volatilization of dicamba and decarboxylation. Both processes were slow, but after 1 month could probably account for the loss of the compound from the plant. After 7 days as much as 70% of the label still remained in the plants. After 1 month, 30-50% was recovered. Considering that the major portion of the label recovered was still in the form of dicamba, an effective amount was thought to remain in the plant. SUMMARY Herbicides were evaluated in field studies to find effective control measures for wild garlic, a problem in winter wheat. Translocation and metabolism of dicamba in the plant were studied in the laboratory. Wild garlic was effectively controlled in winter wheat with a treatment of 0.25 pound of 2-methoxy-3,6- dichlorobenzoic acid (dicamba) plus 0.50 pound of 2,4- dichlorophenoxyacetic acid, low volatile ester (2,4-D LVE) per acre. This treatment caused the least injury to wheat over a range of application times. The most effective times of application were at the resumption of spring growth and at the fully tillered stage. In an emergency situation application 10 days prior to harvest would permit removal of the bulbils in the threshing process. This treat- ment reduced wheat yield from physical damage caused by driving through the field during application. Where dicamba was used, the bulbil shrinkage was always more pronounced. Shrinkage was considered necessary if the treatment was 58 59 effective. In non-crop situations, a higher rate of a combination of dicamba plus 2,4-D LVE was effective in control. One-half pound of dicamba plus 1 pound of 2,4-D LVE/A gave 100% control when applied in the spring. The most effective sterilant treatment was 10 lb/A of 5-bromo- 3-sec-bulyl-5-chloro-6-methyluracil (bromacil) which completely controlled wild garlic for 2 years. When wild garlic is setting bulbils or bulbs these structures act as metabolic sinks to which dicamba is translocated. The translocation of dicamba to these areas caused structural changes which allowed the aerial bulbils to be removed in the threshing process. The movement of the labeled compound in the plant depended on the stage of plant develOpment at the time of treatment. Increasing time after treatment allowed for a greater percent of the herbicide to move into the plant. The maximum amount of label was I found in the plant 7 days after treatment. After 1 month, the percent recovered was less than the amount recovered after 7 days. The decrease in recovery was probably due to~ volatilization and decarboxylation of the compound. Wild garlic does not effectively metabolize dicamba. Seven days after treatment, dicamba metabolites were not detected; however, after 1 month metabolites were found. 60 The identity of l of the metabolites is unknown, but could possibly have been a salicylic acid. The other metabolite was 5-0H-dicamba. Both were found in small amounts when compared to the amount of parent compound remaining in the plant. S-OH-Dicamba was found in greater quantities than the other metabolite. A small amount of label was found in conjugation with plant products. This conjugate was hydrolyzed in a 4.0 M HCL solution. After hydrolysis the peak thought to be salicylic acid decreased in intensity. Alteration of wild garlic metabolism with 2-chloro- ethanephosphonic acid (ethrel) prior to treatment with dicamba caused a reversal of dicamba translocation. The flow was now in the direction of the previously dormant basal bulbs. This was probably due to the increased metab- olic activity in these bulbs caused by ethrel treatment which again made them metabolic sinks. 10. 11. LITERATURE CITED Allmond, C. M. 1957. Progress report on wild garlic investigations. Proc. NCWCC. pp. 173-176. Beasley, C. H. 1969. The effects of various chemicals on axillary bud development in Johnsongrass rhizomes. Abstr. WSSA 190. Blackman, G. E., G. Sen, and W. R. Birch. 1959. J. of Exp. Bot. 10: 33-54. Broadhurst, N. A., M. L. Montgomery, and V. H. Freed. 1966. Metabolism.of 2-methoxy-3,6-dichlorobenzoic acid (dicamba) by wheat and bluegrass plants. J. Agr. Food Chem. 14: 585-588. Chang, F. Y., and W. H. Vanden Born. 1968. Trans- location of dicamba in Canada thistle. Weed Sci. 16: 176-181. , and . 1969. Dicamba metabolism in tartary buckwheat. Abstr. WSSA 175. Crafts, A. S., and S. Yamaguchi. 1964. The auto- radiography of plant material. California Agr. Exp. Sta. Ext. Ser. Manual 35. p. 143. 1968. Adsorption, translocation, and persistence of herbicides. Abstr. WSSA. p. 9. Davis, F. S., E. J. Peters, D. L. Klingman, H. P. Kerr, and H. O. Fletchall. 1962. Chemical control of wild garlic. Weeds 10: 281-284. , and . 1965. Reproductive cycles of wild garlic and nomenclature of plant and bulb types. Weeds 13: 84-87. , and O. H.Fletcha11. 1965. Effect of 2, 4- D and 2, 3, 6- TBA on bulbs of wild garlic. Weeds 13: 210- 214. 61 N w 62 Freeman, J. F. 1958. Yield and seasonal activity of the reproducing bulb of wild garlic in central Kentucky. Abstr. WSA 34-35. Furrer, A. H., Jr. 1964. Preliminary results on control of wild garlic in grass meadows. Proc. NCWCC. 374-375. Iltis, H. 1949. An immigrant conquers a continent. Story of wild garlic. Sci. Monthly 68: 122-128. Klingman, G. C. 1952. Wild garlic in Arlington oats controlled by competition and 2,4-D. Proc. SWCC 132-134. Lazenby, A. 1961. Studies on Allium vineale L. I. The effects of soils, fertilizers, and competition on establishment and growth of plants from aerial bulbils, J. of Ecology 49: 519-541. 1961. Studies on Allium vineale L. II. Establishment and growth in different intensities of competition. J. of Ecology 49: 543-558. . 1962. Studies on Allium vineale L. III. Effect of depth of planting. J. of Ecology 50: 97-109. . 1962. Studies on Allium vineale L. Effects of cultivation. J. of Ecology 50: I411- 428. . 1962. Studies on Allium vineale L. V. Effect of plant disturbance. J. of Ecology 50: 429-438. . 1963. Studies on Allium vineale L. VI. Effects of cutting. J. of Ecology 51: 55-73. Magalhaes, A. C. F. M. Ashton, and C. L. Foy. 1968. Translocation and fate of dicamba in purple nut- sedge. Weed Sci. 16: 240- 245. Mitchell, J., and L. Sherwood. 1961. Lofe cycle studies of wild garlic (Allium vineale L.). Abstr. WSA. pp. 33-34. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 63 Official methods of analysis of the Association of Official Agricultural Chemists. 1955. Eighth Edition. George Banba Publishing Co. Menasha, Wisconsin. p. 805. Peters, E. J., J. F. Stritzke, and F. S. Davis. 1965. Wild garlic, its characteristics and control. Agr. Handbook No. 298. USDA. p. 23. Quimby, P. 0., Jr. 1967. Studies related to the selectivity of dicamba for wild buckwheat (Polygonum convolvulus L.) and a possible mode of action. Ph.D. Dissertation. North Dakota State University, Fargo. , and J. D. Nalewaja. 1966. Effect of dicamba on wheat and wild buckwheat at various stages of development. Weeds 14: 229-232. Sargent, J. H., and G. E. Blackman. 1962. Studies on foliar penetration. I. Factors controlling the entry of 2,4-dichlorophenoxyacetic acid. J. of Exp. Bot. 13: 348-368. , and . 1965. Studies on foliar penetration. II. The role of light in determin- ing the penetration of 2,4-dichlorophenoxyacetic acid. J. of Exp. Bot. 16: 24-47. Saunders, P. F., C. F. Jenner, and G. E. Blackman. 1962. The uptake of growth substances. IV. Influence of species and chemical structure on the pattern of uptake of substituted phenoxyacetic acid by stem tissues. J. of Exp. Bot. 16: 683-696. Scott, R. H. 1951. Life history of the wild onion and its bearing on control. Ministry of Agr. J. of Great Britain. 51: 162-170. Talbot, M. W. 1929. Wild garlic and its control. U. S. Dept. of Agr. Leaflet 43. p. 8. Wang, C. H., and D. L. Willis. 1965. Radiotracer methodology in biological science. p. 363. Prentice-Hall, Inc., N. J.