THE RELATIVE TOXICITY OF CERTAIN PHENOLIC DERIVATIVES TO THE ROOTS OF MAJOR CROP AND WEED PLANTS By KEIIH CONVERSE BARRONS A A THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Horticulture 1950 Acknowledgement s I The author Is Indebted to his guidance committee and other members of the staff of Michigan State College for helpful criticism of the manuscript and valuable suggestions during the course of the experiments. Many of the writer's colleagues in various research groups of The Dow Chemical Company have given freely of their assistance and helpful criticism. The cooperative spirit of the company's management in connection with this study and its presentation as a thesis has been greatly appreciated* -Table of Contents I. Introduction and Literature Review II. Experimental Technique A. Sample Size B. Pre-germination of Seed C. Effect of Temperature D. Effect of Light E. Effect of Seed Vitality F. Summary of Technique III. Toxicity Curves for some Representative Species IV. Toxicity Indices for Crop and Weed Species A. Description of Experiments B. Discussion of Data in Table 5 V. Combinations of NaPCP and NH^DNOSBP VI. Toxicity of Parent Phenolic Compounds vs. their Salts i VII. Comparison of NH^DNOSBP with its Para Isomer VIII. Root Toxicity of Alkanol Amine Salts of DNOSBP IX. Root Toxicity of the Sodium Salt of Dinitro-o-cresol. X. Root Toxicity of Tri-sjnd Tetrachlorophenol XI. Notes on the Nature of the Phytotoxicity of Phenolic Derivatives. ' j XII. Suggested Further Research XIII. Summary and Conclusions XIV. Literature Cited Table or Contents I. Introduction and Literature Review II. Experimental Technique A. Sample Size B. Pre-germination of Seed C. Effect of Temperature D. Effect of Light E. Effect of Seed Vitality F. Summary of Technique ii> III. Toxicity Curves for some Representative Species IV. Toxicity Indices for Crop and Weed Species! A. Description of Experiments B. Discussion of Data in Table 5 V. Combi nations of HaPCP and NH^DNOSBP VI. Toxicity of Parent Phenolic Compounds vs. their Salts VII. Comparison of NH^DNOSBP with its Para I sooner VIII. Root Toxicity of Alkanol Amine Salts of DNOSBP IX. Root Toxicity of the Sodium Salt of Dinitaqo-o-cresol X. Root Toxicity of Tri-and Tetrachlorophenol XI. Hotes on the Nature of the Phytotoxicity of Phenolic Derivatives. XII. Suggested Further Research XIII. Sunmary and Conclusions XIV. Literature Cited . -4THE RELATIVE TOXICITY OP CERTAIN PHENOLIC DERIVATIVES TO THE ROOTS OF MAJOR CROP AND WEED PLANTS Introduction and Literature Review The application of residual dosages of herbicides to soil after planting but before crop emergence to prevent the development of many kinds of weed seedlings has been the subject of much field experimentation during recent years. This type of application is called residual pre­ emergence weed control. There have been a number of reports of experiments on the use of 2,4-dichlorophenoxyacetic acid for residual pre-emergence weed control In corn. Also of"current interest among weed-control research workers Is the use of residual dosages of certain phenolic compounds for pre-emergence control of many small-seeded weeds in certain large-seeded crops Including cotton, corn, beans, the cucurbits, and crops grown from roots, tubers, corms.and other vegetative organs. (1)(3)• Pentachloro- phenol (hereafter called POP) and 4,6-dinitro-o-sec.butyl phenol (hereafter called DNOSBP) and their salts are the phenolic compounds that have been used extensively for pre-emergence experimentation because of their known high degree of foliage toxicity. (6 ) (7 )- -5In Hawaii the sodium salt of PCP has been used as a ground spray between rows of establlshe d sugar cane to prevent weed seedlings from emerging. The phenolic compounds » have shown promise as post-emergence re sldual ground sprays between rows of a number of growing crops including cotton, corn and nursery stock. Field observations (l) have lndlcajted that the largeseeded crops which have been successfully treated with residual pre-emergence dosages are protected In part by depth of planting. Roots of weeds germinating near the* surface appear to be markedly affected while the deeper crop plant roots are generally below the level of toxic ! soil. The primary shoot apparently does not absorb toxic i quantities as it pushes up through soli containing the \ toxicant. ! r Weeds that do emerge with the crop have been observed to frequently come from a depth similar to that of the crop. Although depth protection is believed to be an importI ant factor in selectivity, field obseryationsby the writer in 1948 indicated clearly the occurrence of specificity of reaction between the different phenolic compounds and certain plants. Thus lamb1s quarters (Chenopodium album) appeared relatively susceptible to DNOSjJBP and pigweed j (Amaranthus retroflexus) relatively susceptible to PCP. Cotton has been reported to be less sulpject to injury by weed-control dosages of DNOSBP than by equally effective dosages of PCP. (5 ) Exploratory tests with seeds grown in petrl dishes with varying concentrations of the ammonium salt of DNOSBP (hereafter called NH^DNOSBP) and the sodium salt of PCP (hereafter called NaPCP) showed wide species differences in reaction. Thus the field observation that diversity of physiological response to the different chemicals is a probable factor in selectivity was further confirmed. The experiments herein reported were conducted to determine the relative toxicity of several phenolic * compounds to the roots and seeds of crop and weed species under conditions divorced from the variables inherent in field tests. These toxicity data were considered desirable for the purpose of determining with which crops the different compounds would be relatively safe for residual pre-emergence weed control if depth protection under a given, set of soil conditions was not adequate for selective action. It was also considered desirable to determine whether any small-seeded crops that must be planted at a relatively shallow depth possess enough tolerance to the phenolic compounds to permit use for .residual preemergence treatment. Small dosages of these chemicals in general contact formulations have been used as contact pre-emergence sprays to kill tiny weed seedlings that emerge before the crop. Although the necessity for very accurate timing makes this technique of limited use it is very promising for certain slow-emerging crops that will not tolerate larger residual dosages. Even the minimum amounts required for contact kill ing of weed seedlings may sometimes leave enough residue to injure certain crops. Accurate toxicity data *fer_e needed as a basis for determining the relative safety of the different ^phenolic compounds for contact pre-emergence spraying of a given crop. In addition to DNOSBP and PCP and their salts several other phenolic derivatives were included to determine their possible value for residual, soil treatment. To facilitate a better understanding of how these compounds prevent weed seed emergence when used as residual pre-emergence herbi­ cides observations were made on the nature of their toxi­ city to roots and seeds. x Experimental Technique The foliage toxicity of various phenolic compounds has been compared in numerous tests but no experiments on root and seed toxicity other than pre-emergence tests in soil have come to the writer^ attention. A laboratory -S- technique for determining relative tonicity had to be developed. Exploratory tests In which seeds were planted In dry quartz sand and the containers flooded with the toxic* solutlons did not prove satisfactory. Many seedlings emerged even though the roots were badly stunted and emergence counts alone did not reveal the degree of toxicity. Removal from sand to examine roots proved difficult. The work of Swanson ( 12 ) and Ready and Grant (10 ) suggested In vitro tests using solutions varying In concen­ trations. After preliminary trials petrl dish culture was decided upon as best suited to a rapid screening program. This technique made possible the frequent and critical examination of roots, and space and labor requirements were kept at a minimum. Several.trials with lettuce, tomato, bean and corn as test species Indicated that solutions would have to be re­ plenished to insure uniform contact of the developing roots with the toxicant. Otherwise some roots grew Into the air above the moist filter paper with little evidence of toxi­ city. By flooding the seed with the test solution, remov­ ing the excess after a few hours and then re-flooding and draining at Intervals of 3 0 -to 40 hours thereafter, roots were kept In contact with the toxicant. Variations due to -9» uneven drying were also minimized. After numerous measurements with several species of length and weight of ooth root and top growth, root length was chosen as the best criterion of toxicity. Recording individual root measurements on a calculating machine and enter^^foly the total and mean length for each dish on record sheet was found convenient. Sample S i z e . Root lengths for three replicates of fifty lettuce seedlings each were recorded individually for the purpose of obtaining a statistical estimation of the size of samples and number of replicates needed for a reasonable degree of accuracy. The formula used for this determination and its application has been discussed by Henry, Down and Baten. (8) For the data analyzed 5 dishes containing 14 seeds each or 4 dishes containing 11 seeds each were found to give a degree of accuracy within 10 percent of the general mean the Standard Error of the individual means. x Populations of other species or even other seed samples of the same species could not be assumed to exhibit the same degree of uniformity, and the results -10of this analysis were considered only as a guide pending experience with additional material. Four plates of ten plants each were decided upon for further work until Stan -0 dard Errors of various treatments for a number of species could be calculated in order to determine the range in variability one might expect. Measuring a total of 40 roots per treatment was ndfc considered too time consuming. Pre-germination of Seed. A high degree of varability between plants in the con­ trol of some populations with respect to time of germination suggested selection of samples grown in a germinating dish until radicles emerged as a method of decreasing variabil­ ity. Comparisons were made with lettuce, corn and onion of the method use^L heretofore involving the treatment of dry seed vs. pre-sprouting-and the selection of uniform samples with short radicles for subsequent treatment. Quadruplicate 10-seed samples were treated with distilled water using each of the two methods. dis-colored seeds were eliminated. in Table 1. In all cases small or The data are presented -11Table 1. Mean Root Length and S. E. values for Pour Ten-Seed dishes for Each of Three Crops Treated Dry and After Pre-germination.___________________ Treated Dry Pre-germinated Corn 92.6 + 16.1 96.0 Lettuce 19.4 + 4.6 22.1 + 1.0 Onion 22.5 + 8.5 24.3 + + 6.3 .6 The Standard Error values in Table 1 show that v a r i a ­ bility was decreased with all three species when pregerminated seeds were used. The greater mean root length for the pre- germinated treatments is due largely to the complete lack of seeds showing no germination or slow germination. Such seeds were present in many of the dishes in which dry seeds were treated. All three samples germinated 90 percent or better according to germination test data supplied by the seedsman from whom they were procured. Even a greater d e ­ crease in variability might be expected from the pre-gemina­ tion technique if samples of lower vitality were used. Effect of Temperature. In order to determine whether temperature uniformity was required to obtain comparable results for different tests a comparison was made of Great Laices lettuce at two . temperatures with NaPCP and at four temperatures with NH*DN0SBI -12In addition Red K i d n e y beans were grown at two temperatures following treatment with NaPCP. The temperatures used covered the range through which the species involved will germinate and grow. Root length was expressed as a percentage of the control for each temperature in order to compare relative degrees of Inhibition at different temperatures. Although differences were noted between root length means for different temperatures at a Bingle concentration, consistent trends were suggested only fpr the lowest t e m ­ perature for lettuce and the highest temperature for beans treated with NH*DNOSBP. As these differences were not great and the temperatures involved were extreme, the conclusion was drjawn that the fluctuations in temperature ordinarily encountered under room conditions are unlikely to introduce a n appreciable error in conducting root toxicity tests with phenolic compounds according to the technique outlined above. A constant temperature chamber used for holding petri dishes in some of the later tests proved desirable from the standpoint of leveling the time required for growth of the untreated roots to a given length. Effect of L i g h t . In view of the findings of Weintraub (12) that the inhibitory effect of several substances on the growth of lettuce was lessoned in the presence of light it was consi- -13dered desirable to compare results with cultures grown in light and in dark at a constant temperature. The Great Lakes variety of lettuce was used and the lighted cultures exposed to a 1 0 0 -watt Incandescent lamp at a distance of five feet. The unlighted cultures were placed on the same shelf in a temperature control chamber which varied from 65° to 6 8 °P. and covered with black paper. The light was sufficient to induce marked chlorophyl development in the cotyledons while those of unlighted cultures were yellow. / There were no differences of significance between the lighted and unllghted cultures with either NaPCP or NH^DNOSBP. This test does not preclude the possibility that at greater intensity a light effect might occur. In view of these r e s e t s It seems unlikely that variations in light under room conditions would introduce an appreciable error. Lettuce is among the more light sensitive species with j respect to germination and was therefore oonsidered a good plant for this test. The possibility that light might affect the sensitivity of other species to these compounds was In% vestigated only by cursory observation of bean, corn, and <1 Kentucky blue grass treated with various concentrations of NaPCP. No influence of light on root inhibition was indicated. -llu The Effect of Seed Vitality Considerable variation In time of germination occurred with most seed lots when being pre-germinated. The question arose as to the relative toxicity of the phenolic compounds under Investigation to the early vs. the late germinating seeds within a given sample. Error could, conceivably be Introduced If seeds sprouting quickly and those somewhat delayed In germination were « divergent In their degree of sensitivity. In order to elucidate this point a test was made of onions taken from a germinating dish at Intervals. The first lot was drawn when the seeds sprouting first had radicles 2 to 5 mm. long. A second lot consisting of seeds in a similar stage of development was drawn 30 hours later and a third lot 72 hours after the first. Treatments were made with NaPCP with the concentrations Indicated In Table 2. The petri dishes were placed in a 65°-68°F. temperature con­ trol room and each lot was removed and measured after It had been held 120 hours after the first treatment. Thus each lot had the same opportunity for root elongation. Table 2 presents the results of this test. -15Table 2. Root Length in mm. of Three Lots of Brigham Yellow Globe Onion Varying in Time of Sprouting After Growing for 120 Hours at 60-63 F. Following Treatment with NaPCP. Each Value is the Mean of Three 10-3eed Samples_____________________________ _ 1st. Seeds to Sprout Control 1 ppm. 2 ppm. 4 ppm. 6 ppm. 29 + 1.0 Seeds Sprouting 30 Hours Later 28 + 1.6 .9 14 + 1.0 7 + .6 + + + ± 2.3 .8 18 + 1.6 12 1.1 .2 11 ± 1.6 .4 9 ± . .8 • 16 + 23 15 13 9 0 .8 H 22 Seeds Sprout:ing 72 Hours After First CVJ Concentration (PCP equiv- A variance analysis was run to determine the signi­ ficance of differences between the three lots drawn at successive intervals from the germinating dish. The analysis dataare presented as follows: Source of Variation Total Treatments Sprouting Time Error D.F. 14 4 Sums of Squares . Variance F 694 650 2 27 8 *7 W 13.5 6.4 1 Tabular F value for I% level » 5.28 1 Differences between the first and second lots were obviously of no- significance In view of the highly signi- fleant F value differences between lot three drawn 72 hours after the first radicle emergence must differ significantly from the other lots at the 1$ level. It is of further interest that the Standard Errors of the means of the 72 -16hour samples were consistently higher indicating^greater variability. Based on this test it was decided to draw samples from the germinating dishes no longer than 2h hours after the first radicles emerged. Conceivably seed vitality factors unrelated to speed of sprouting could cause variation in sensitivity to phen­ olic compounds. The question arose as to the likelihood of error resulting from the use of a weak lot of seed hav­ ing limited capacity for root growth. High and low germin­ ating lots of seed were obtained for the same variety of > beans and onions and subjected to a replicated test. N®. differences of significance between means resulted in tests with high and low germinating samples of the same , species. However, Standard Error values for the low- germinating samples tended to be greater than for the highgerminating seed indicating greater variability. Care was therefore taken to procure seed samples having a high germination percentage for subsequent tests. Summary of Technique. Based on experience gained in preliminary tests with several species and on the experiments discussed above the following method was used in subsequent work. -171. A seed sample of good vitality was germinated on moist blotters In covered dishes held at a temperature suitable for germination of the species Involved. When required' for the prevention of mold sodium hypochlorite solution was used for surface sterilization of the seed. 2. Within 24 hour8 after the first seeds showed a radicle, uniform samples werq transferred to petri dishes fitted with a filter paper over which had been poured the solutions to be tested. Radicle length varied from two to ten millimeters depending on the species. Excess liquid was drained from the dishes within one-half hour after transfer. Seeds were placed so all root tips were touch­ ing the moist filter paper. 3. Except as noted ten sprouting seeds were placed In each dish. Three or more replications of each treatment were made according to the circumstances as discussed for each experiment. 4. " Preliminary tests to establish the toxicity range for the species involved were made as required. A geo­ metric progression of concentrations beginning with onehalf part per million was used. Concentrations were calcu­ lated on a parent phenolic compound equivalent basis in all ■ Instances. Each set of^treatments Included a control to which only, distilled water was applied. 5. At intervals of 30 to 40 hours following transfer the solutions were replenished with enough to cover the bottom of the dish. All roots were then wet by tipping the dish from side to side. After 13 to 30 minutes the ex- cess liquid was drained off and the roots brought in contact with the moist paper in so far as possible. The liquid changes were usually made twice between the time of trans­ fer and measuring roots. 6. table. The dishes were held at room temperature on a Measurements were taken after considerable growth of the control had occurred and after an apparent deceleration of growth indicated that food reserves were approach­ ing Repletion. For the most part, roots were measured from five to seven days after transfer. 7. Root measurements, were made by removing one seed­ ling at a time and stretching the root along a millimeter scale. Each root length was entered on an adding machine and the meaiB of the ten plant samples rounded off to the nearest unit were recorded. Accuracy of measurement was estimated to be within ten percent for the species with short roots but undoubtedly increased with root length. -198. Variability was calculated as the Standard Error of the mean and expressed to the first decimal place. When S. E. values indicated excessive variability within a treat ment another replicate for that treatment and the control were grown and measurements made after 'the same interval used in the original test. 9. In order to compare relative toxicities of a compound to different species and to compare tests on the r''" same species that might vary somewhat In the root length of the control rpot length of the various treatment means was expressed as a percentage of the control. The Standard Error of the means were calculated wherever a measure of variability was needed. -20Toxicity Curves for Some Representative Species A suitable index of toxicity was needed to facili­ tate comparisons between chemicals on a given species and i between species with a given chemical. To explore the possi­ bility of estaDlishing such an index, tests were made on eight species with both NaPCP and NH*DNOSBP. Based on field results and preliminary laboratory tests these eight species were believed to vary in their degree of sensitivity to the chemicals. Three members of the Gramineae family having different seed sizes were chosen because of the postulated importance of seed size to selectivity under field conditions. Each of the remaining five species represented a different plant family. Preliminary trials established the toxicity range for each species. A geometric progression of concentrations ranging from a point near the threshold to one near complete inhibition was then employed. was used throughout. The technique described earlier Pour replicates were run and the vari­ ation expressed as the Standard Error of the mean. Root measurement data for each of the eight species are expressed as a percentage of the control in taole 5 a^d are presented graphically in figure 1. V Table 3. Compound and conc. Control ^ Root length data_____________ expressed as a percentage of the control. crop species Kentucky Blue Orass Barley Corn Lettuce Alfalfa Onion IOO+3.3 100+3.3 IOO+5.5 100+5.1 100+5.2 100+2.9" 100+4.5 Cucumber Radish 100+2.; / Ha PCP (PCP equlv.) 1 pp* 2 ppm 78+2.8 61+2.2 4 ppm 86+.4 50+5.0 100+5.2 71+4.2 8 ppm 22+1.7 16 ppm 11+1.7 32 ppm 64 ppm 128 ppm 71+4.6 73+1.8 81+6.2 58+2.5 88+3.4 41+2.7 50+3.3 54+5.4 64+4.0 27+1.8 66+5.2 52+2.4 43+2.0 - 43+3.8 22+1.3 14+2.3 - 38+1.9 14+2.4 - 12+1.4 - - - 23+1.3 14+.8 10+1.5 29+3.7 13+2.5 - 92+3.7 84+.9 40+2.2 2^ 1.2 16+1.1 94+2.0 74+4.0 35+2.7 19+1.3 14+1.4 - - - - - - - - - - - - - - NH^DlfOSBf (DNOSBP equlv.) 1/2 ppm 82+2.3 1 PP« 2 ppm 89+3.9 72+6.1 IO3+3.8 74+5.4 95+3.6 4 ppm 33+2.2 77+2.9 8 ppm 22+3.3 63+4.7 35+2.8 16 ppm 17+2.7 f 32 ppm 64 ppm - 18+1.7 13+1.8 - 52+3.1 23+1.6 16+.6 9+1.3 50+2.7 32+3.6 18+1.8 9+1.3 - 100+5.2 62+5.2 75+2.5 75+7.0 24+2.9 54+2.5 19+2.9 14+2.4 29+2.5 13+. 4 - 95+1.7 81+5.5 59+1.8 40+1.3 21+1.6 74+2.7 45+4.1 21+3.9 13+3.9 6+3.2 • -22The actual threshold was not experimentally determined, but by projection of the curves an approximation may be visualized. Because roots were allowed to begin growuh b e ­ fore treatment the point of complete inhibition was not zero but fell between five and ten percent of the control. Where the highest concentration did not cause complete inhibition extrapolation at the lower end of the curve will provide an estimation of this extreme toxicity level. It will be seen from figure 1 that all regressions tend to be curvilinear with the slope decreasing with an increase in concentration. The equation of the regression appears to follow that for the exponential growth curve: y = axb When the data were plotted on logarithmic paper it was found that the regression tended to be linear, thus further Indicat­ ing that the root toxicity curves for phenolic derivatives probably fit the'exponential equation. * ! -23There is similarity between the curves presented in figure 1 for root toxicity to those presented by Blacktnan, Holly and Roberts (4) for foliage toxicity from dinitro phenolic compounds. Ivens and Blaeitaan. (9) have obtained similar curves by plotting root growth of corn against concentration of pentachloropbenol. Among possible indices of toxicity the concentration in ppm. required for a given reduction in root length was considered most desirable because the sensitivity of differ­ ent species to a given chemical could be readily compared as well as the toxicity of different chemicals to a given species. Root inhibition equivalent to 25 percent of the control f was first considered desirable because this degree of inhibition was generally lethal. However, interpolation at this level was found to be less accurate than at lower S' concentrations because of the decreasing slope of the curve1 . Furthermore, greater variability at higher concentrations, as indicated by higher Standard Error values, increased the liklihood of error in the curves at their lower end. Root inhibition equivalent to 50 percent of the control was finally chosen as a suitable index of toxicity. -24C Thus, the Toxicity Index for a phenolic derivative on a given species equalsf the concentration, In parts per million, required to Inhibit root length to 50 percent of the control as determined by the technique outlined a b o v e . This concentration was determined by graphic interpolation from curves fitted to the data. The toxicity index was estimated to the closest unit for values above 8 ppm. and to the closest five-tenths unit for values below 8 ppm. The Toxicity Indices obtained by interpolation of the graphs in figure 1 for the two compounds are presented in table 4. In order to determine the reproducibility of these tests, solutions equal in concentration to the Toxicity Indices were used on each of the eight species and the actual reduction in root length determined. presented in table 4. These data are -25- Table 4. Root length expressed as a percent of the control for eight species treated with concentrations of NaPCP and NH 4DNOSBP equivalent to the Toxicity Index. Crop Species NaPCRtox. index root length N1UDN0SBP tox. index root leng Kentucky blue grass 3-5 55+3 •0 2.5 58+30 Barley 9 56+2.3 5 49+2.9 Corn 13 58+ 3.8 8 56+2.5 Lettuce 3 42+1.1 1 45+1.8 Alfalfa 9 42+2.3 2.5 69+8.1 Onion 3.3 41+5.3 4.5 53+5•9 Cucumber 7 40+2.7 10 46+2.0 Radish 6 48+3.2 2 0 57+3.7 It will be seen from Table 4 that the root Inhibition obtained varied from the expected value of 50 percent by as much as 19 percent in the case of alfalfa treated with NH 4DNOSBP. Some measure of reliability was obviously needed. Except at ' higher concentrations the resultb of which did not materially effect the interpolated value the Standard Errors In table 3 seldom exceeded ten percent of the mean. The interpolated value of 50 percent was assigned an arbitary S. E. error value of 10 percent on the assumption that this interpolated value should be no more variable than the experimental values. -26Using the conventional test for significance the values in Table 4 differing significantly from +50 + 5 and the pairs differing significantly from each other were determined. It was found that the only value differing significantly from 50 +^ 5 at the 5 percent level was that of alfalfa treated with N H + D N O S B P . The only pair differing-significantly from each other at this level was the alfalfa treatments. It was, therefore,-concluded that the results presented In Table 3 and in figure 1 were repeated within a reasonable degree of accuracy for seven species but not for alfalfa. A further test was made with alfalfa using four concentrations for each chemical. Results did not agree with the first test and new curves were drawn as presented in figure 2 Using concentrations equal to the Toxicity Indices of 5-5 for NaPCP and 3.5 for NH 4DNOSBP treatments were made for the purpose of checking the accuracy of the n e w curves. Root length values expressed as a percentage of the.comparative control were obtained as follows: NaPCP 48 + 2 . 4 , 52 + 1.6. NH 4DNOSBP Thus the second curves were found to be more accurate and the Toxicity Indices of 5.5 for NaPCP and 3.5 for NH 4DNOSBP should be considered reliable rather than those given in Table 3 . No explanation for the obvious error in the first alfalfa test is offered. -27Toxlclty Indices for Crop and Weed Species The Toxicity Indices for a number of crop and weed species were determined by interpolation from curves fitted to data for three or more concentrations of each chemical. Description of Experiments The concentrations used were chosen after a trial run and in each case, bracketed the 50 percent degree of inhibition. Three 10-seed replicates were made for the control and each of the three concentrations except in a few instances as noted. Spinach was grown in 20-seed lots because of extreme variability. Sugar beet which was also quite variable was grown in 20 -seedball lots and the twenty longest seedlings in each dish were measured. Because of germination diffi- culties data for chickweed is based on 5 -seed samples replicated three times and for crabgrass on 5 -seed samples replicated five times. Table 5 presents the Toxicity Indices for the species tested for NH^DNOSBP and NaPCP. The species marked with an asterisk were subsequently tested with solutions equal to the Toxicity Indices In order to check the results of the first test. Root inhibition from these concentrations did not vary significantly from 5 0 + 5 nor did they vary significantly from each other except in the case of tomato and carrot. A n entirely new -26- test of these two was made and curves modified to fit the new data. The Toxicity Index given In Table 5 for these two represents the second test.Toxicity Indices for the eight species discussed earlier are Included. To facilitate comparing the relative sensitivity of different species to the two compounds the quotient of the Toxicity Index for NaPCP divided by the Toxicity Index for NH 4 DNOSBP was calculated for each species. These quotients are presented In the right hand column of Table 5 and will hereafter be designated as the PCP/lDNOSBP quotients. -29Table 5. Toxicity Indices for KaPCP and HH«DNOSBP on several crops and weeds arranged according to family with the PCP/DNOSBP auotlent. Toxicity 'PCP/ Index 'DM0SBP NaPCP NH a DNOSBP*quotient Variety ' 3pqc 1q b ______________ Common name Oramlneae Dlgltarla sanguinalls Crabgrass 3 2-5 . 1.2 Panlcuw ramosua Millet 8 b.5 1.8 Setarla lutsscens Yellow Poxtall 5.5 b l.b •Poa pratensla Kentucky Blue Oraos 3.5 2.5 l.b •Lollum sp. Domestic Rye Orass 7 5.5 1.3 Oat Kent 1* 7 2.0 •Trltleum vulgare Wheat Yorkwln 5.5 3 1.8 Trltleum vulgare Wheat Averarsativa 7 3-5 2 Rye Rosen 7.5 3-5 2.1 eUordeum vulgare Barley Bay 9 5 1.8 ezea mays Corn Ohio M15 13 8 1.6 Zea mays Corn Iona it 8 1.7 Secale cereals Amaranthaceae Amaranthus sp. Tampala 6.5 Amaranthus retroflexus Red Root Pigweed 8 6 •1.3 Common Chlokweed 7 3 2.3 2.^ Alslnaceae Stellarla media Llllaceae Allium cepa Allium oepa Onion Brigham Yellow alobe b.5 3.5 1-3 onion White Portugal 3 3 1.0 b.5 3 1.5 Chenopodlaceae ■Beta vulgaris Sugar beet Splnacla oleracsa Spinach Chenopodlum album Lambaquarters Oiant Thlok Leaved 3 2 • 1.5 10 b 2.5 Streamliner 5 5 1.0 Umbelllferae ■Oaucus carota Carrot Cruclferae Raphanus satlvus Radish Early Scanlet Olobe 6 2.5 2.b Raphanus satlvus Radish Crimson Olant 6 2 Brasslea rapa 3-.0 Turnip Purple Top White alobe 6 2 3.0 Cabbage Qolden Aore 10 3 3.3 Cauliflower Early Snowball 8 2.5 3.2 8 3 2.7 ■Brasslea oleracsa Brasslca oleracsa Brasslca Kaber , Wild Mustard Llnaceae Linum usltatlsslmum Flax Dakota 6 10 .6 Llnum usltatlsslmum Plax Crystal 3 b .7 • -30Table 5 continued. Leguralnosae White cio/er Common 10 16 .6 Tr jfolium pretense Red Clover Mammoth 11 9 1.2 •Trlfollum pretense Red Clover Medium 12 15 .8 8 4 2.0 •Trlfollum repens Trlfollum hybrldlum AlsIke Clover Alfalfa Orlma 5.5 3-5 1.6 Lespedeza Kobe 10 14 .7 Lespedeza Korean 16 20 Viola vlllosa Vetch Hairy 20 19 1.0 PI sum sativum Pea Thomas Laxton 18 8 2.3 Phaseolus vulgaris Bean Hlchellte 21 12 1.7 •Phaseolus vulgaris Bean Strlngless Oreen Pod , 16 11 1.5 Vlgna sinensis Cow pea Clay 9 6 1.5 dlyclne eoja Soy bean Manchu 13 9 1.4 Phaseolus lunatua Lima Bean Henderson Bush 13 11 1.2 Tomato Garden State 7 5 1.4 Pepper California Wonder 8 7 1.1 Eggplant Black Beauty 11 11 1.0 •Medlcago sativa Lespedeza japonlca •Lespedeza stlpulacea a* .8 Solanaceae Lyooperslcuo esoulentum •Capsicum frutsscens Solanum oelongena Polygonaoeae ■Vagopyrum saglttatum Buckwheat 2 3.5 .6 •Polygonum perslscarla Lady's Thumb 2 4 .5 Malvaceae Oossyplum hlrsutum Cotton Delta Pine 15 22 17 !.3 Oossyplum hlrsutum Cotton Stonevllle 2B 25 18 1.4 Hibiscus esoulentus Okra Clemson Spineless 12 12 1.0 a 10 .8 .7 Cucurbltaoeae Cucumls satlvus Cucumber Straight 8 ■Cucumls satlvus Cucumber White Spine 7 10 Cltrullus vulgaris Watermelon Dixie Queen 15 16 .9 Cucurbits maxima Squash Hubbard 12 12 1.0 Muskmelon Honey Rook 8 12 .7 Lactuca sativa Lettuce Orand Rapids 4 1.5 2.7 Lactuca sativa Lettuce Oreat Lakes 3 1 3.0 Clchorlum lntybus Wild Chicory 5 3 1.7 Ragweed 3.5 2 1.7 •Cucumls melo Composltae Ambroslaceae Ambrosia elatlor -31Discusslon of data in table 5 In general there Is a trend toward higher Toxicity Indices for the larger seeded crops,^however there are enough exceptions to preclude generalizations. Buckwheat for example has a fairly large seed yet shows a low toxicity Index par­ ticularly toward NaPCP. Among the small seeded crops white and red clover and the lespedezas show remarkable resistance to both compounds In contrast to other small seeded legumes. The trend toward greater physiologic resistance on the part of large seeded species can hardly account for the marked selectivity noted In field tests. Depth protection of the roots coupled with lack of absorption by top g rowth as It pushes through the toxic surface layer must Btlll be considered as Important factors. Those small seeded crops showing a high Toxicity Index to one or the other of the compounds should be given residual preemergence treatments on an experimental basis. Planting at as great a depth as possible would undoubtedly be desirable. Larger seeded crops not having a high Toxicity / Index may still be adaptable to residual pre-emergence treat­ ment but the risk would be greater under conditions of shallow s. planting, porous soil or heavy rainfall. -32- Probably the most significant information gained from these experiments concerns the marked species differences in response to the different chemicals as indicated by the varying PCP/DNOSBP quotients. Theoretically when this quotient is similar for a given crop and a given weed there would be no advantage of one material over another with respect to physiologic selectivity. If the quotient for the crop is lower than for a given weed NH 4 DNOSBP should show the greatest selectivity. If the crop quotient is greater than that for the weed NaPCP should show the greatest selectivity. Because weeds vary in their relative tolerance to these chemicals and many weed species are often present this line of reasoning has practical value only when the crop quotient is extreme. The cucurbits, onion, flax, okra, white clover, the lespedezas, carrot, egg plant, vetch and buckwheat all have quotients of 1 or less indicating a high relative tolerance to NH 4 DNOSBP. Radish, turnip, cabbage, cauliflower, and lettuce have quotients of 3.0 or more. As several common weeds have lower quotients NaPCP should have greater selectivit for these jsrops with the type of weed complex often encountered Where the PCP/DNOSBP quotient for the crop varies from about 1.5 to 2.5 there Is probably little choice of chemicals for pre-emergence treatments unless the dominant weed species Is more sensitive to nne material than the other. For example if the most prevalent weed is lady's thumb or similar polygonao. -33- weeds, PCF should be used because of the extreme sensitivity ■”*** of this group to this compound. A crop quotient vs. weed quotient favorable to the . use of a given chemical does not mean that a residual treat­ ment has any chance of success. The aetual crop tolerance and possibilities of deep planting must also be considered. It appears unlikely that any small-seeded crops that must be planted within an inch of the surface will prove adaptable to residual pre-emergence methods, however, field fitests with the more resistant should be made. These data do suggest that for contact pre-emergence sprays POP to oil or in emulsions may prove safer on crops with a high PCP/DNOSBP quotient. On the other hand crops with a low quotient should be less likely to show injury from the very small dosages used in contact pre-emergence applications from the application of DNOSBP in oil or in emulsion form. Family trends in the PCP/DNOSBP quotient are rather consistent except for the legumes as will be noted in Table 5 . No marked differences between varieties of a given species were found with respect to the PCP/DNOSBP quotient. The two varieties of flax appeared to differ markedly In their toleranc This difference may In part be due to differences in seed vitality as the Crystal variety was slow in germinating. ( -Jlu Screening of other phenolic compounds at concentra­ tions equal to the Toxicity Index of NH*DNOSBP or NaPCP as a standard for purposes of comparison suggests itself as a quick way to determine' which compounds approach or exceed them in activity. The possibility of variation in tolerance between varieties or seed lots, suggests the desirability of determining the Toxicity Index of the seed at hand for the standard compound before using this seed for routine screening of other compounds or derivatives. Combinations of NaPCP and NIUDNOSBP Because of the wide species differences in response to the two compounds it was evident that a mixture jnight be of value as a residual pre-emergence spray for a weed complex composed of species of divergent sensitivity, in order to determine whether the two compounds are additive or possibly synergistic in their effect tests were conducted with eight species. Triplicated treatments were made with a solution composed of a concentration of each compound equal to onehalf its Toxicity Index as listed in Table 5. of this experiment are presented in Table 6 . The results -35Table 6 . Root length, expressed as a percentage of the comparative control, from treatments w i t h a concentration equal to one-half the Toxicity Index for NaPCP plus onehalf the Toxicity Index for N R 4DNOSBP. Root length expressed as percentage of control Species Kentucky blue grass Barley 55 + 3.5 60 •• 2.9 Corn 58 + 5.0 Lettuce 39 + 3.2 Alfalfa 52 + 1.6 Onion 5^3 + 7.0 Cucumber 46 Radish 57 + 2.7 4.0 None of the values in Table 6 deviated significantly from 50 + 5 at the 5 percent level. It was concluded, there­ fore, that the effects of the two compounds are additive. Toxicity of Parent Phenolic Compounds vs. their Salts ^Favorable field results with formulations of parent PCP and DNOSBP suggested the desirability of comparing the parent compounds with their salts in petri dish tests. 4 -36B e c a u s e of its very weak acid properties PCP is r e l e a s e d from solutions of NaPCP u p o n absorption of C0 2 f r o m the air. W i t h C0 2 being liberated from germinating seeds it appears l ikely that only the parent PCP is present shortly after solutions are poured on the dishes. S o l u b i l i t y of PCP in w a t e r is low but the limits were not exce e d e d except at the h i ghest concentrations used. Comparisons were made of fresh NaPCP solutions vs. a similar solution through w h i c h an excess of C0 2 had b e e n bubbled. Comparisons were also made bet w e e n NH^DN O S B P ^ solutions a n d solutions cf the parent DNO S B P formed by the a d d i t i o n of Just enough dilute HgSO* to the salt solution to cause the yellow color to disappear indicating that the free DNOSBP had been liberated by the displacement of the a m m o n i u m ion. Triplicate petri dish tests were made o n six species at concentrations equivalent to the Toxicity Indices for the respective compounds. in table 7 . These data are presente It will be seen that no differences of s i g n i f i ­ cance were found between either parent compound and its salt. Table 7. Mean root length In millimeters for a three petrl dish test of parent phenols vs. their salts on six species each at a concentration equivalent to the toxicity index. PCP Species DNOSBP *oot length Concen- Cor sodium tration salt Root length for parent PCP Concen­ tration Root length Root length for ammon­ for parent DNOSBP ium salt Buckwheat 2 33+ 1.3 30 + 2.6 4 40 + 3.6 36 + 2.1 Radish 6 37 + 1.9 37 + 2.4 2 42 + 2 . 3 40 + Cucumber 7•5 32 + 1.5 36 + 2 . 7 10 41 + 1.6 38 + 2.0 .9 s 11 14 + .6 13 + 1.0 9' 19 ± .3 18 + 1.0 Lettuce 3 11 + .3 12 + 0 1 15 + .5 17 + Wheat 5 49 + 1.6 43 + 2.8 2.5 56 + 2.0 Red Clover .6 58 + 3-1 - Comparison of NH^DNOSBP with Its para Isomer. Crafts (6) reported markedly greater foliage toxicity from NH*DNOSBP than from its para isomer (ammonium salt of 2,6, dinitro 4 sec. butyl phenol referred to in Figure 5 as DNPSBP). A comparison of the root toxicity of the two isomers was made on four s,pecies using the same technique heretofore employed. These data are expressed graphically in figure 3 * The markedly greater root toxicity of the salt of the ortho isomer (DNOSBP) is evident with a glance at the figure. It is of interest to note that the curve for the salt of the para isomer (DNPSBP) slopes more gradually than that of the ortho isomer. Thus there is a greater percentage difference between the two isomers at higher than at lower concentrations. - -39- Root Toxicity of Alkanol Amine Salts of DNOSBP Pre-emergence field tests with the triethanol amine salt of DNOSBP in 1949 indicated that this derivative is equally effective as the ammonium salt. Because of certain advantages from the formulation standpoint alkanol amine salts are considered the most suitable DNOSBP derivatives for pre-emergence weed control. A commercial formulation placed on the market in 1950 by the Dow Chemical Company under the proprietory name Premerge contains three pounds of DNOSBP per gallon as alkanol amine salts of the ethanol and Isopropanol series. Root toxibity studies were made of several alkanol amine salts in comparison with the ammonium salt using the technique earlier developed. The concentration used for each species was its DNOSBP Toxicity Index. sults are presented in Table 8 . The r e ­ The mixtures indicated in the table were commercially available alkanol amines. It will be seen that no differences of conceivable significance were found between the ammonium salt and any of the alkanol amine 3«lts. I -40- Table 8. Root Length of Four Species Grown in Solutions of Several Alkanol Amine Salts at Concentra­ tions Equivalent to the Toxicity Index of DNOSBP. Root Length is Expressed as a Percentage of the Control.____ ___________________ DNOSBP Amine Salt Radish Wheat Cucumber 2 ppm. 2,5 ppm._____ 10 ppm Lettuce l.b ppm, diethanol pi + 2 . 1 56+3.6 47+3.0 51+2.6 triethanol 54 + 1.0 52+1.3 49+2.1 46+4.1 mixed diethanol and monoisopropanol 51 + 5.6 58 +^ 45 + 6.2 50 + 5.8 mixed mono ethanol and di-isopropanol 47 +.4.1 oO + 1.6 41+3.1 48+2.6 ammonium 52 + 2.6 55 + 4.2 48 + 1.1 55 + 5.2 .8 Root Toxicity of the Sodium Salt of Dlnitro-o-cre 3ol Crafts (6 ) found that dlnitro-o-cresol was the least toxic to foliage of a series of homologous alkyl dinitrophenolic compounds. toxic. In the same tests DNOSBP was the most It was of interest to learn whether a similar di f­ ference between the two compounds exists with respect to root toxicity,. t _ Tests were made with salts of DNOSBP and dinitro-ocresol on four species at a concentration equal to the Toxicity Index for NH 4DNOSBP for each lot of seed used. The results are presented in table 9 . -41Table 9. Compound Root Length of Four Species Grown In Solutions of the Sodium Salt of Dlnltro-o-cresol and NHjjDNOSBP at the Toxicity Index of the Latter. Root Length Expressed as a Percentage of the C o n t r o l . ____________ Radish Wheat Cucumber Lettuce 2 ppm. 3.5 PP®. 10 ppm. 1.5 PP®. Ha Dinltroo-cresol HH^DHOSBP 49 +/l.9 53 + 3.6 86 ± 2.4 52 + 1.5 4 8 + 2.5 115 + 0 .7 59 + 0.5 55 + 2.5 It will be seen from table 9 that the salt of DNOSBP inhibited root growth of wheat, cucumber and lettuce more than did the salt of Dlnitro-o-cresol. With radish there was slightly greater inhibition from the salt of dinltroo-cre3 ol however the difference was not significant. It is of Interest to note an apparent stimulation to roots of lettuce from the concentration used of the salt of dlnltro-o-cresol. This great diversity in relative response of roots of different species to the two homologous compounds is greater than has been noted for foliage reac­ tion; however Barrons and Grigsby (2 ) and Blackman, Holly and Roberts (4- ) have reported specific differences in foliage reaction to the two compounds. -42Root Toxicity of Trl-and Tetrachlorophenols The sodium salts of 2,4,5“t**ichlorophenol and tetrachlorophenol were compared with NaPCP for root toxicity on 4 species. A concentration equal to the NaPCP Toxicity Index for each species was employed for all 3 compounds. The data are presented In Table 10. It will be seen that there tends to be a progressive increase In root toxicity with Increasing chlorination which Is in agreement with observations on foliage toxicity. Table 10. Root Length Expressed as a Percent of the Control for 4 Species Treated with Sodium Salt Solutions of 3 Chlorophenols all at the NaPCP Toxicity Index for the Species Involved._______ 7 Compound Radish ____________ 6 JIM. 2,4,5-tri- 72 + 4.3 chiorophenol sodium, salt tetrachlorophenol, 54 ± 3.6 sodium salt NaPCP 46 + 5.0 Wheat Cucumber 5 PPm._______ 7.5 PP*. 82 + 6.1 Lettuce ___ 3.JPKS*., 78 + 2.9 72 + 3-9 68 + 5.0 60 + 5-3 60 + 3.6 53 ± 4.6 52 + 2.8 46 + 5.0 -43Notes on the Nature of the Phytotoxicity of Phenolic Derivatives Several observations were made during the course of these experiments on the apparent nature of the reaction of seeds and seedlings to the phenolic compounds. Al­ though further experiments would be required to permit firm conclusions on the nature of the toxicity of phenolic JX=> compounds to roots, these observations are recorded here in the event that they will be of interest to those work­ ing with phenolic herbicides. The herblcidal action of these compounds when used as foliar sprays is generally recognized as being of a contact nature. Little movement within the plant beyond the creep­ ing action of the oil used as a solvent is believed to occur. Aqueous solutions of the salts of phenolic compounds used as selective sprays apparently kill tissue only beneath the surface that is actually wet with the spray. A few large droplets result in ’’burning" of holes in the leaf which may otherwise appear healthy. In these petrl dish tests top growth showed few signs of necrosis indicating that at the concentrations used there was little absorption through the cuticle. As discussed earlier in this report reduction in top growth at sublethal concentrations was far below the reduction In root growth. This lack of apparent top effect on plants whose roots were markedly reduced In growth indicates that translocation from <* roots to tops is of little importance. With carrots it was noted that top growth showed signs of a toxic effect after the cotyledons had unfolded and the hypocotyl was two to three centemeters long. Whether this resulted from delayed absorption through the cuticle or translocation through the roots is not known. Lettuce and chicory showed some necro­ tic lesions on cotyledons at higher concentrations indica­ ting cuticular absoration. The lack of top effect except in these instances agrees with field observations on the emergence of healthy seedlings of large seeded crops through a toxified surface layer of soil. NH^DNOSBP consistently caused chlorosis of the cotyledons of the Cruciferae even at dosages no greater than the equivalent of the Toxicity Index. NaPCP did not induce chlorosis on the same species even at concentrations causing an almost complete inhibition of root growth. With no other species was a similar chlorotic effect observed. Concentrations up to those equivalent to the Toxicity Index seldom caused necrosis of roots during the first few days. Rate of root growth was retarded from the time treat- ment was first mad© but roots generally had a healthy appearance, except for the lack of root hairs. Death of root tips was evident at higher concentrations within two days after treatment and at sub-lethal concentrations after a longer period. High concentrations completely inhibited further growth of sprouted seeds once they were treated with the solution. When dry seeds were placed directly in toxic solutions in the first exploratory tests growth was inhibited to about the same extent as when sprouted seeds were treated. As noted earlier, roots had to be wet at intervals and kept in contact with the moist filter paper to insure, uniformity of results. It was observed that sprouting, seeds placed on the paper with the radicle pointing upward produced a root that appeared healthy as indicated by the presence of root hairs and lack of any dis-coloration. Apparently the toxicant was not absorbed through the con­ tents of the seed in contact with the paper but enough water was absorbed for some growth. It was further observed that when only a part of the roots of the branching system of wheat and barley were in contact with the paper those suspended had a healthy appearance and developed root hairs. To study further the localization of the effect of -46phenol 1c compounds on root growth sprouting wheat seeds with roots approximately 1 cm. in length were so placed that a portion of the roots of each seedling grew into water and a portion into an 8 ppm. solution of NH 4DNOSBP. After 4 days the roots in the toxic solution had not elongated and were dls-colored. The roots growing in water had elongated three-to four-fold and showed no signs of injury. Top growth appeared to be in no way affected. See Figure &. The conclusion was drawn from these observations that root toxicity from these compounds is largely of a contact nature. This conclusion was substantiated by the fact that several crops with large seeds produced lateral roots rather freely if not re-wet with the solution even though the primary root had been completely inhibited. This recovery power of the larger-seeded species may be a factor in their ability to survive field treatment. Suggested Further Research Although these tn vitro experiments cannot De used to predict field results with certainty they do indicate probable toxicities in soil. It is noteworthy that lady's thumb was observed to be hypersensitive to NaPCP in 1948 field plots. This observation was verified in petri dish tests. Unless the two compounds are differentially ad­ sorbed on soil colloids there appears to be no reason why the PCP/DHOSBP quotients reported above should not pertain in soil. Leaching studies to determine possible differences in downward movement in soil should prove fruitful. Favorable results with lettuce at low temperatures indicate that practical weed control may be obtained from early spring applications thus confirming 19*8 field ob- servatlons. Further field work on temperature relationships is needed. The diverse PCP/DNOSBP quotients obtained suggest the possible desirability of mixtures of the two compounds for field tests with different weed complexes. For general use a residual pre-emergence herbicide must prove satisfactory for a wide variety of weed species. Field work designed to test the results obtained in these experiments should be conducted with particular reference to crops with extreme PCP/DNOSBP quotients. Small- seeded crops showing outstanding tolerance should be subjected to experiments with residual pre-emergence sprays. In the course of the tests of parent compounds vs. their salts (Table 7 ) it was observed that a slight excess of acid caused a marked Increase in the toxicity of a given solution. Subsequent tests with lettuce and alfalfa c o n ­ firmed these observations. Simon and Blackman (11 ) have pointed out the effects of varying pH on the toxicity of phenolic compounds to fungi. Activity was increased with lower pH and in some experiments relative toxicities were greatly altered by pH changes. The relationship of soil pH to crop tolerance and weed control from residual pre-emergence treatment was beyond the scope of this research but should be investigated further. -4-9Summary and Conclusions 1. A technique was developed for determining the toxicity of phenolic compounds to roots. 2. The relative toxicity to many species of the ammonium salt of 4,6-dinitro-0-sec. butyl phenol (NH^DNOSBP) and of the sodium salt of pentachlorophenol (NaPCP) is presented. 3. A wide variation in species reaction to each of these chemicals was found to exist. 4. Relative toxicity to the two compounds varied from species to species. 5. Combinations of the two compounds gave only an additive effect. 6 . The parent phenolic compounds and their salts were equal in root effect. No differences were found in root inhibition between a number of alkanol amine salts and the ammonium salt of DNOSBP. 7 . NH^DNOSBP was found to be considerably more toxic to roots than its para Isomer which is in agree­ ment with observations on foliage reaction. 8 . The sodium salt of Dinitro-o-cresol was found to be considerably less toxic to roots of three out of the four species tested than NH^DNOSBP. -50- 9. Chlorophenols Increased in root toxicity from tri-through tetra-through penta-substitutions. 10. The nature of toxicity to roots of the phenolic compounds was observed to be essentially of a contact type with little evidence of translocation. r -51literature CITED 1. Barrons, Keith C. Pre-emergence weed control. Down to Earth 4:(No. 3) 2-4 1948. 2. Barrons, Keith C, and Grigsby B.H. The control of weeds in canning peas with chemical sprays. Michigan Agri­ cultural Experiment Station Quarterly Bulletin 28: 145156, 1945. 3. Barrons, Keith C., Mullison, Wendell R., Fitzgerald, C.D., and Coulter, L. L. Certain phenolic compounds for resid­ ual pre-emergence weed control in horticultural crops. Proceedings North Central Weed Control Conference, Spring­ field, 1 1 1 . 5:34-39 1948. 4. Blackman, G.E., Holly, K. and Roberts, H.A., The comparative toxicity of phytocidal substances. . Symposia of the Society for Experimental Biology 3:283-317 1949. 5. Cowart, L.E., Stamper, E.R., and Creasy, L.E. Studies on weed control in cotton. Proceedings of the Southern Weed Conference, Baton Rouge, La. 2:66-67 1949. _ ... 6 . Crafts, A.S., A new herbicide 2;4-dinitro 6 butyl phenol. Science 10-417-418 secondary 1945* 7 . Crafts, A. S. General Contact Weed Killers, .Agricultural Extension Seryice California Circular 137, Revised 19^9. 8 . Henry, G.F., Down, E.E., and Baten, W.D. An adequate sample of corn plots with reference to moisture and shelling percentages. Journal American Society of Agronomy 3 4 :777-781 1942. 9. Ivens, G. W . , and Blackman, G. E., The effects of Phenyl carbamates on the growth of higher plants. Symposia of the Society for Experimental Biology 3: 266-282 1949. .cfM. -52- 10. Ready, Daniel, and Grant, Virginia Q. A rapid sensitive method for determination of low concentrations of 2,4dichlorophenoxyacetic acld_in aqueous solution. Botanical Gazette 109: 39-44 1947. 11. Simon, EeW., and Blackman, G.E. The significance of hydrogen-ion concentration in the study of toxicity. Symposia of the Society for Experimental Biology 3:253265 1949. 12. Swanson, Carl P., A simple bio-assay method for the determination of low concentrations of 2,4-Dichlorophenoxyacetic acid in aqueous solutions. Botanical Gazette 107;507-509 1946. 13. Weintraub, R.L. Influence of light on chemical inhibition of lettuce seed germination. Smithsonian Miscellaneous Collection: 107:1-8 1948. 100%-i RADISH LETTUCE 50 % - v *T 0 2 4 8 0 16 2 4 a 32 16 100% — ) CUCUMBER ONION 50%- 50%- . 0 2 4 B f 16 0 0 2 4 8 16 32 100% - ) KENTUCKY BLUE GRASS 50%- ALFALFA 50%- — > 8 1 Sodium P C P K ey . F ig u r e 16 l 0 2 4 -X 8 * 16 x— = Am m onium 32 DNOSBP T o x i c i t y curves f o r e i g h t re p re s e n ta tiv e sp ecies 1 100 %-| BARLEY 50%- 32 64 32 64 100%-. CORN Sodium Key: Figure 1 P C P - 50 % - — X--- Xx— - A m m o n i u m x x— D N O S B P continued 100%-. ALFALFA (2nd test) 50%- Key F ig u r e Sodium 2 P C P — X = Ammonium D N O S B P R evised curve f o r a l f a l f a baaed on aecond t e a t 100-1 100-1 LETTUCE RADISH 80- 80- 60 - 60- 40 - DNPSBP 20 ONPSBP 40- DNOSBP DNOSBP 20 - 32 lOO-i WHEAT 8060 - DNPSBP 40- DNOSBP 20 - 32 64 100 — 80 CUCUMBER - 60- DNPSBP 40 - DNOSBP 20 - 32 Figure 3. Root toxicity curves for ortho and para dinitrosecondarybutyl phenol. The vertical scale rep­ resents root length expressed as a percentage of the control. The horizontal scale represents the concenrtration in ppm. of DNOSBP and DNPSBP equivalent. 64 / Figure 8 - Wheat seedlings which were grown with the left hand roots In water and the right hand roots In an 8 ppm. solution of NHjjDNOSBP. r While Clover Control 16ppm. NH* DNOSBP /6ppm. No PCP 32 ppm. * Figure 7 Representative white clover seedlings at end of growth period showing marked \ resistance to NH^DNOSBP and slightly greater susceptibility to NaPCP. Radish Control 8 ppm Figure 6 - Representative Radish seedlings at end of growth period showing relative high degree of susceptibility to NH^DNOSBP and slightly lower susceptibility to NaPCP. CohTro! Figure 5- Cucumber seedlings grown in varying concentrations of NlfyDNOSBP. Filter papers were replaced with black paper / to facilitate photography. ✓ Figure 4 - Snapbeans, left sprayed with 9 pounds per acre NlkDNOSBP and right unsprayed control. pigweed. Weeds are primarily lambsquarters and red root