increased light penetration was one of the major factors influencing secondary succession. There did not appear to be any relation between herbicidal residue and invading species. For example, several species such as Psychotria berteriana DC. were extremely susceptible to initial application of herbicides, but they were found on all treated plots 18 months after application.

ACKNOWLEDGMENTS

This study was supported by the Advanced Research Projects Agency, Department of Defense. The herbicides were donated by Amchem Products, Inc., Ambler, Pa.; Dow Chemical Co., Midland, Michigan; E. I. DuPont de Nemours & Co., Wilmington, Delaware; Geigy Chemical Corp., Ardsley, New York; and Velsicol Chemical Corporation, Chicago, Ill.

LITERATURE CITED

1. BEARD, J. S. 1948. The natural vegetation in the windward and leeward islands. Oxford Forestry Mem. 21. 192 p. 2. BEVERIDGE, A. E. 1957. Arboricide trials in lowland dipterocarp rain forest of Malaya. The Malayan Forester 20:211225.

3. BOVEY, R. W., F. S. DAVIS, M. G. MERKLE, R. E. MEYER, H. L. MORTON, and L. F. BOUSE. 1965. Defoliation and control of brush. Proc. SWC 18:288-292.

4. DAWKINS, R. C. 1957 Contact arboricides for rapid tree weeding in tropical forests. Trop. Silvicult. FAO Collec tion No. 13(2):109–112.

5. FRISSEL, M. J. and C. H. HOLT. 1952. Interaction between certain ionizable organic compounds (herbicides) and clay minerals. Soil Sci. 94:284-291.

10.

11.

SCHWELZER, E. E. and J. T. HOLSTON, JR. 1966. Persistence of five cotton herbicides in four southern soils. Weeds 14:22-26. SPOSTA. J. W. 1960 Emacion de especies tropicales inferiores por medio de substancias quimicas. Apuntes Forest. No. 4. Trop. Forest Res. Center, Rio Piedras, Ruerto Rico. I p.

12. THIECS B. J. 1962. Microbial decomposition of herbicides. Down to Earth 1827-10.

13. TSCHIRLEY, F. H. 1967 Problems in woody plant control evaluation in the tropics Weeds 15:233–237.

14. TSCHIRLEY, F. H. CLYDE C. DOWLER, and J. A. DUKE. Species diversity in two plant communities of Puerto Rico. A Tropical Rain Forest (In Publication).

15. WATSON, A. J. and B. J. MESLER, JR. 1964. Effect of tordon herbicide as basal frill and tree injection treatments on certain hardwood trees. Down to Earth 19(4):20-23.

16. WATSON, A. J. and M. G. WILTSE 1963. Tordon for brush control on utility rights of way in the eastern United States. Down to Earth 191:11-14.

17. WYATT SMITH, J. 1960. Further arboricide trials in lowland Dipterocarp rain forest of Malaya. The Malayan Forester

23:314-331.

18. WYATT SMITH. J. 1961. Arboricide trials using ammate X. 23-D, 2.45-T, and sodium arsenite. The Malayan Forester 24:81-84.

Abstract. The herbicides 2,4-dichlorophenoxyacetic acid (2,4-D). 2.45-trichlorophenoxyacetic acid (2,4,5-T), and 2-methoxy-3,6-dichlorobenzoic acid (dicamba) each labeled in the carboxyl posi tion were sprayed on a pasture consisting of a mixture of silver beardgrass (Andropogon saccharoides Swartz.), little bluestem (4. scoparius Michx.), and dallisgrass (Paspalum dilatatum Poir) and a sideoats grama (Bouteloua curtipendula [Michx.] Torr.) pasture over a 3-year period. Plant samples were harvested at intervals between 1 hr and 16 weeks after treatment and residues determined by radioassay. No important differences were found in the persistence of herbicides or of different formulations of the same herbicide. Rainfall was the most important factor influencing the persistence of the herbicides. The little bluestem silver beardgrass-dallisgrass samples harvested 1 hr after treatment with the butoxyethyl ester of 2.4.5-T contained both this ester and the acid of 2,4,5-T. One week after treatment, the acid of 2,4,5-T and unknown metabolites were found but no ester.

A

INTRODUCTION

VARIETY of herbaceous and woody plants are controlled by 2,4-dichlorophenoxyacetic acid (2,4-D), 2,4,5-trichlorophenoxyacetic acid (2,4,5-T), and 2-methoxy-3,6-dichlorobenzoic acid (dicamba). Although the phenoxy acids have been registered and are used for weed control on lands devoted to forage production, the substituted benzoic acids have restricted usage on these areas. Little direct evidence of the persistence of these compounds in forage grasses has been published. Glastonbury et al. (3) sprayed peas (Pisum sativum L. var. Onward) with the sodium salt of 4-(2-methyl-4-chlorophenoxy)butyric acid (MCPB) and found that the halflife of the retained chemical was 3 days. Gutenmann and Lisk (4) sprayed the diethylamine salt of 4-(2,4dichlorophenoxy)butyric acid (2,4-DB) on a pasture containing birdsfoot trefoil (Lotus corniculatus L.) and timothy (Phleum pratense L.) and found a rapid decrease in herbicide concentration in the forage after rainfall. Concentrations of 2,4-DB in the forage immediately after application of 1.5 and 3.0 lb/A rates were about 70 and 160 ppm, respectively, but were 0.32 and ppm, respectively, after 48 days.

0.80

Klingman et al. (5) sprayed a Kentucky bluegrass (Poa pratensis L.) pasture with either the butyl ester or the 2-ethylhexyl ester of 2,4-D and found that most of the butyl and about 75% of the 2-ethylhexyl ester were hydrolyzed to the 2,4-D acid within 12 hr after spraying. Total concentrations of 2,4-D residues from the butyl and 2-ethylhexyl esters dropped from 58.4 and Received for publication November 7, 1966. Cooperative investigations of the Crops Research Division, Agricultural Research Service and Texas A&M University.

'Research Agronomist, Crops Research Division, ARS, U. S. Department of Agriculture, College Station, Texas; Assistant Range Scientist, Rolling Plains Livestock Research Station, Texas A&M University, Spur. Texas; and Plant Physiologist, Crops Research Division, ARS, U. S. Department of Agriculture, College Station, Texas, respectively.

48.4 ppm 12 hr after treatment to 5.0 and 15.1 ppm, respectively, 7 days after treatment.

The investigation reported herein was conducted to determine the persistence of 2,4-D, 2,4,5-T, and dicamba in range forage grasses, to compare the persistence of amine and acid formulations of 2,4,5-T, and to determine the influence of rate of application on the persistence of 2,4-D and 2,4,5-T.

MATERIALS AND METHODS

Two field sites were fenced for the study. One was at College Station, Texas, in a pasture in which silver beardgrass (Andropogon saccharoides Swartz.), little bluestem (A. scoparius Michx.), and dallisgrass (Paspalum dilatatum Poir.) were the dominant species. The other was at Spur, Texas, in a pasture in which sideoats grama (Bouteloua curtipendula [Michx.] Torr.) was the dominant species. Different areas were treated at each site each year.

Herbicides labeled in the carboxyl position with carbon-14 were mixed with technical grade herbicides in the proportions necessary to give the specified radioac tive levels as well as the specified rate of herbicide per acre. In all experiments, sprays were applied at volumes equivalent to 20 gpa with a compressed air sprayer. Two replications of each treatment were used. In 1962, the plots were 2 by 10 ft and they were 2 by 12 ft in 1963 and 1964.

In 1962, we applied butoxyethyl ester of 2,4,5-T at rates equivalent to 1/2 and 2 lb/A. Sprays were applied June 11 and June 19 at Spur and College Station, respectively, which contained 5 uc of radioactivity per plot. The carrier consisted of 7 parts water and 1 part diesel fuel (v/v).

In 1963, we applied 2,4-D and 2,4,5-T acids to the silver beardgrass-little bluestem-dallisgrass pasture June 14. Each solution contained 50 uc of radioactivity and sufficient herbicide to provide 2 or 2 lb/A rate. The carrier was acetone-water (1:1) containing 0.5% (v/v) surfactants.

In 1964, we applied dimethylamine salt of 2,4-D, dimethylamine salt of dicamba, and triethylamine salt of 2,4,5-T to the silver beardgrass-little bluestem-dallisgrass pasture July 1. We applied both the amine and acid of 2,4,5-T and dicamba to the sideoats grama pasture July 7. The carrier was water containing 0.5% (v/v) surfactants for amine salt formulations and acetone-water (1:1 v/v) for the acid of 2,4,5-T. Each solution contained 60 uc of radioactivity and sufficient herbicide to provide a rate of 1 lb/A.

"Surfactant contained alkylarylpolyoxyethylene glycols, free fatty acids and isopropanol.

litter

Recovers of herbicides from speed plots. The amount of herbicide recovered from grass tissues harvested : hr after treatment calculated as a percentage of the amount applied, varied from 220 green tissue and 87 litter tissue) to 100 42 green tissue and 600 tissue) Table 1. In all but two plots, greater quantities of the herbicides were recovered from the green tissues than from the litter tissues. Although the silver beardgrasslittle bluestem-dalisgrass stands were relatively uniform. the plants and litter did not cover all of the plot areas. The low recovery percentages were due to sparse stands and the higher recoveries to dense plant and litter

cover.

Experiments in 1962. Figure 1A presents a semilogarithmic graph of the concentrations of the ester of 2,45T in green tissues of silver beardgrass, little blucstem and

1

AM

Culteras tarvested at five danes the treatment. The Opera indicating beer of disappearant was Doc affected by THE KAS DUAC Contratons of the ester of 2.45-T TRUCues Oreased rapúr during the second week after treatment when 18 a ranil anered. The ap

3004

2.0 184

HERBICIDE CONC. (PPM)

612

HERBICIDE CONC. (PPM)

RAINFALL (IN.)

012

2.0 LB/A

LOS LBA

B

WKS. AFTER TREAT WKS. AFTER TREAT. Concentrations of ester £43-T residues found in silver serasa tissues harvested at five dates 1960 at 0.5 and 25 DA at College Etter tissues. Solid vertical lines Dave Tata which occurred during the indicated interval

4

parent half-life of the ester of 24.5-T half-life equals average length of time necessary for one-half of herbiodal residue to disappear under the conditions of this experiment averaged 25 weeks. Concentrations of ester of 24.51 in the green tissues 8 weeks after treatment were 23 and 7 ppm, respectively, at the 2.0 and 0.5 lb. A rates

Figure 1B is a semilogarithmic plot of the ester of 245T concentrations in the litter tissues of silver

beardgrass, little bluestem, and dallisgrass. Disappearance was most rapid during the second week after treatment. The rate of disappearance in litter tissues was slower than in the green tissues. The apparent half-life of ester of 2,4,5-T in the litter tissues was about 4 weeks under the conditions of this experiment. Two factors were important in the slower rate of disappearance. First, growth of the green tissues would have diluted the herbicide, but the litter samples were composed of non-living tissues and growth was not a factor in lowering the concentrations in these samples. Second, conditions for microbial decomposition of the herbicide were unfavorable due to the low rainfall.

Figure 2 presents a semilogarithmic plot of the ap parent ester of 2,4,5-T concentrations in the green and litter tissues of sideoats grama. The ester of 2,4,5-T disappeared more rapidly from sideoats grama than from

silver beardgrass, little bluestem, and dallisgrass. More rain fell on the sideoats grama than on the silver beardgrass, little bluestem, and dallisgrass. The apparent halflife of the herbicide averaged 1.6 weeks in the green tissues and 1.7 weeks in the litter tissues. The amount and frequency of the rainfall were conducive to leaching, microbial decomposition of the herbicide, and growth of sideoats grama plants. All of these factors contributed to a rapid reduction in herbicide concentrations. Experiment in 1963. The concentrations of 2,4-D and 2,4,5-T residues found in green and litter tissues of silver beardgrass, little bluestem and dallisgrass harvested at six dates after treatment June 18 are shown in Figure 3. A 0.69-in rain occurred during the first week after treatment, and the concentrations of both herbicides in green and litter tissues decreased rapidly. No rainfall occurred during the second week after treatment and the rate of herbicide disappearance was slower in most of the plots than it was during the first week. During the third and fourth weeks after treatment, 1.58 in of rain occurred and the rate of herbicide disappear

Figure 3. Concentrations of 2,4-D and 2,4,5-T residues found in silver beardgrass-little bluestem-dallisgrass tissues harvested at six dates after treatment June 14, 1963 at 0.5 and 2.0 lb/A at College Station. (A) Green tissues, (B) litter tissues. Solid vertical lines indicate rainfall which occurred during the indicated interval after treatment.

ance was more rapid in most of the plots than during the second week. During the fourth through the eighth weeks after treatment, only 0.08 in of rainfall occurred and relatively small decreases in herbicide concentrations were found. The 2.78 in of rainfall which occurred during the eighth through the fourteenth weeks after treatment probably was the primary factor responsible for the rapid rate of herbicide disappearance during this interval. The average half-life for 2,4-D in green and litter tissues was 2.3 and 2.8 weeks, respectively. The average half-life of 2,4,5-T in green and litter tissues was 2.9 and 3.4 weeks, respectively.

Experiments in 1964. Residues of amine salts of 2,4-D, 2,4,5-T, and dicamba disappeared from silver beardgrass, little bluestem, and dallisgrass tissues at about the same rate (Figure 4). The apparent average half-life for each